The impact of AI on businesses, workers, economic growth and wealth division

The Fourth Industrial Revolution and Its Implications

The world faces numerous challenges, including economic inequality, sustainability, and urbanization. As we approach the Fourth Industrial Revolution, characterized by advancements in technologies such as artificial intelligence, robotics, and big data, it is essential to understand its potential impacts on society and the economy.

This revolution mirrors past industrial revolutions, which significantly improved productivity and quality of life. However, it introduces a broader scope of change, affecting not just manufacturing but also services and new business models. The integration of AI poses a unique challenge as it may replace jobs traditionally thought to require human intelligence, leading to significant disruptions in the workforce.

Opportunities Presented by AI

AI offers liberating benefits, particularly in sectors like healthcare and transportation. For instance, autonomous vehicles could enhance mobility for the elderly, while big data analytics may accelerate drug development, potentially leading to breakthroughs in curing diseases. The healthcare industry stands to gain immensely from AI, as machine learning can address rare diseases and tailor treatments to diverse populations.

Challenges of Job Displacement

Despite its advantages, the technological revolution threatens to displace millions of jobs. For example, the rise of self-driving trucks could eliminate the need for the current workforce of truck drivers. As productivity increases, it is crucial to ensure that displaced workers acquire new skills for emerging job opportunities. Historical trends indicate that while some jobs may disappear, new ones will likely be created. However, the challenge remains in ensuring sufficient reskilling and adaptation.

Experts like Martin Ford warn of a future where millions of skilled workers may find themselves unemployed due to automation. He argues that no job is entirely safe from the encroachment of AI, including those in creative fields. This potential for widespread job loss poses significant economic and social challenges, necessitating proactive solutions to prepare for the future.

The Impact of Big Data

Big data plays a pivotal role in this revolution, offering vast amounts of information that can be leveraged for machine learning and AI development. As companies collect extensive data on their operations and customers, they enable algorithms to learn and improve autonomously. This shift raises questions about the future of work, as machines increasingly take on tasks previously performed by humans.

Urbanization and City Management

Urbanization is another critical issue, with more people living in cities than ever before. As urban populations grow, cities face challenges related to governance, climate change, and economic inequality. The management of mega-cities, such as Rio de Janeiro and Lagos, highlights the need for innovative solutions to address infrastructure deficits and social disparities.

Urban activists like Aleandra Orurafhino advocate for citizen involvement in city planning, emphasizing that effective governance requires integrating the voices of those affected by urban policies. In rapidly growing cities, unplanned expansion can lead to chaos, necessitating a shift in how cities are designed and managed.

The Gender Gap in Economic Participation

The gender gap remains a significant barrier to economic equality. Women continue to face challenges in achieving leadership roles in business and government. Despite progress, only a small percentage of top corporate positions are held by women, and women often earn less than their male counterparts. Addressing this gap is not only a matter of fairness but also an economic imperative, as closing it could significantly boost global economic growth.

In conclusion, the Fourth Industrial Revolution presents both opportunities and challenges. Embracing technology while addressing the implications for the workforce, urbanization, and gender equality will be crucial for building a sustainable and equitable future. As we navigate these changes, it is essential to prioritize inclusivity and adaptability to harness the full potential of technological advancements.

The positive economic aspects of artificial intelligence (AI) for businesses are numerous and transformative. AI has the potential to significantly enhance productivity, streamline operations, and reduce costs. Here are some key benefits:

• Increased Productivity: AI technologies can automate repetitive tasks, allowing human workers to focus on more complex and creative activities, thereby increasing overall productivity.
• Cost Reduction: By automating processes, businesses can reduce labor costs and minimize errors, leading to significant savings.
• Enhanced Decision Making: AI can analyze large datasets quickly and provide insights that help businesses make informed decisions, leading to better strategic planning.
• Innovation: AI opens up new avenues for innovation, enabling companies to develop new products and services that meet changing consumer demands.
• Market Competitiveness: Companies that effectively leverage AI can gain a competitive advantage in their industries, leading to increased market share and profitability.

• [01:37] "Technology that leads to massive gains in productivity mean substantial improvements to everyone's quality of life."
• [01:12] "These things are sort of combining in a way that's bringing about a host of transformative changes across industries."

Artificial intelligence (AI) offers several positive economic aspects for employees as well. These benefits can enhance job satisfaction and improve overall quality of life:

• Job Creation: While AI may displace some jobs, it also creates new roles that require different skill sets, leading to a net increase in employment opportunities in emerging fields.
• Improved Work Conditions: AI can take over dangerous or tedious tasks, allowing employees to engage in more meaningful and less hazardous work.
• Flexible Work Arrangements: AI technologies enable remote work and flexible schedules, providing employees with a better work-life balance.
• Skill Development: As AI technologies evolve, employees have opportunities to learn new skills, enhancing their employability and career prospects.

• [03:34] "If you think about something just like driverless cars, autonomous vehicles, which is one use of AI that people are talking about, that could have a really liberating impact on a lot of people's lives."
• [04:17] "The advantages of machine learning and data science are immense. Those have an incredible chance to address very infrequent diseases and diseases which affect different parts of the population very differently."

There are several negative economic aspects of artificial intelligence (AI) for businesses that need to be considered:

• Job Displacement: As AI automates tasks, many traditional roles may become obsolete, leading to layoffs and unemployment in certain sectors.
• High Initial Investment: Implementing AI technologies often requires significant upfront investment in infrastructure and training, which can be a barrier for smaller businesses.
• Dependence on Technology: Over-reliance on AI can lead to vulnerabilities, especially if systems fail or are compromised, affecting business operations.
• Ethical and Legal Challenges: Businesses may face ethical dilemmas and legal issues related to AI use, such as data privacy concerns and accountability for AI decisions.

• [04:41] "A technological revolution will cost jobs. It'll cost jobs in the areas that see the biggest advancements first."
• [06:11] "Millions and millions of those jobs are going to be lost and it's unlikely that enough jobs are going to be created to absorb all of those workers."

The negative economic aspects of artificial intelligence (AI) for employees can be significant and multifaceted:

• Job Loss: AI has the potential to displace a large number of workers, particularly in roles that involve routine tasks, leading to unemployment and economic instability.
• Skill Gaps: As AI technologies evolve, there may be a mismatch between the skills workers possess and those required for new jobs created by AI, leading to underemployment.
• Increased Inequality: The benefits of AI may not be evenly distributed, potentially widening the gap between high-skilled and low-skilled workers, as well as between different socioeconomic groups.
• Job Insecurity: The rapid pace of technological change can create uncertainty in the job market, causing anxiety among employees about their future job security.

• [05:19] "If you've lost 3.5 million jobs in one sector, how do you create more than that in another sector?"
• [06:13] "It's going to push people out of the labor force. Many people are going to find it impossible to adapt to that because they're not going to have capabilities that really exceed what machines can do."

To mitigate the negative economic consequences of artificial intelligence (AI) for businesses, several measures can be implemented:

• Reskilling and Upskilling: Companies should invest in training programs that help employees develop new skills relevant to the evolving job market, ensuring they can transition to new roles.
• Diversification of Roles: Businesses can create new job opportunities by diversifying their workforce and developing roles that complement AI technologies rather than compete with them.
• Ethical AI Practices: Implementing ethical guidelines for AI development and use can help businesses navigate potential legal and social challenges while maintaining public trust.
• Collaboration with Educational Institutions: Partnering with educational institutions can help align curriculum with industry needs, preparing future workers for the demands of an AI-driven economy.

• [05:08] "You need to make sure those displaced workers are given the skills to move into these new positions."
• [05:30] "Hopefully, it will be what happens again."

To address the negative economic consequences of artificial intelligence (AI) for employees, several strategies can be employed:

• Universal Basic Income (UBI): Implementing UBI could provide financial support to displaced workers, helping them transition during periods of unemployment.
• Job Transition Programs: Governments and organizations can create programs that assist workers in finding new employment opportunities and provide resources for retraining.
• Strengthening Labor Rights: Ensuring that labor laws adapt to the changing landscape of work can protect employees' rights and provide a safety net for those affected by AI.
• Promoting Entrepreneurship: Encouraging entrepreneurship and small business development can create new job opportunities and stimulate economic growth in areas where jobs are lost.

• [05:21] "How do you create more than that in another sector?"
• [06:15] "It's unlikely that enough jobs are going to be created to absorb all of those workers."

Artificial intelligence (AI) has the potential to create a safer living environment for people through various applications:

• Smart City Technologies: AI can optimize traffic management and public safety systems, reducing accidents and improving emergency response times.
• Healthcare Innovations: AI can enhance healthcare delivery by enabling early disease detection and personalized treatment plans, improving overall public health.
• Environmental Monitoring: AI can analyze environmental data to predict natural disasters and monitor pollution levels, helping to protect communities from environmental hazards.
• Enhanced Security Systems: AI-driven surveillance and security systems can help prevent crime and improve community safety.

• [03:29] "The potential benefits are vast."
• [04:15] "If we're going to cure cancer, it's probably going to come through data science."

While AI can enhance safety, there are also negative consequences that must be considered:

• Privacy Concerns: The use of AI in surveillance can lead to invasions of privacy and potential misuse of data.
• Job Displacement in Security Roles: As AI systems take over security functions, human security personnel may face job loss or reduced opportunities.
• Dependence on Technology: Over-reliance on AI for safety can create vulnerabilities, particularly if systems fail or are hacked.
• Bias in AI Systems: If AI algorithms are not properly designed, they may perpetuate biases, leading to unfair treatment of certain groups in safety-related contexts.

• [06:09] "It is sort of obliterating distance. It is in some cases removing humans from tasks that we once thought were the sole province of the human mind."
• [06:11] "Millions and millions of those jobs are going to be lost and it's unlikely that enough jobs are going to be created to absorb all of those workers."

The consequences of artificial intelligence (AI) for the distribution of wealth within society and between different countries can be profound:

• Wealth Concentration: AI may lead to increased wealth concentration among those who own the technology, exacerbating income inequality.
• Global Disparities: Countries that are early adopters of AI may gain economic advantages, widening the gap between developed and developing nations.
• Job Polarization: AI can create high-skill, high-wage jobs while displacing low-skill, low-wage jobs, leading to a polarized labor market.
• Access to Technology: Disparities in access to AI technology can create uneven economic growth, with some regions benefiting significantly while others lag behind.

• [14:01] "That's 60 million jobs in the United States alone. That's a staggering number."
• [14:14] "You'd have tremendous stress on government in terms of trying to take care of all these people that no longer have an income."

AI has the potential to disrupt the traditional link between growth, jobs, and wages:

• Decoupling of Growth and Employment: As companies adopt AI technologies, they may achieve growth without necessarily increasing employment, leading to jobless recoveries.
• Wage Stagnation: The displacement of workers by AI can lead to wage stagnation, as fewer jobs are available and competition for remaining positions increases.
• New Economic Models: The rise of AI may necessitate new economic models that decouple income from traditional employment, such as universal basic income.
• Shift in Skill Demand: As AI changes the nature of work, there will be a shift in the demand for skills, requiring workers to adapt continually to remain relevant.

• [09:08] "The net total number of jobs created in the United States was zero."
• [09:10] "It's taken longer and longer for the jobs to reappear."

[00:01] The challenges facing our world are
[00:04] growing all the time. How do we build
[00:07] stronger economies with equal
[00:09] opportunities for all? How do we build a
[00:12] sustainable world for generations to
[00:15] come? How do we protect our cities and
[00:18] harness the power of technology for our
[00:20] common
[00:22] benefit? Humanity has always been good
[00:25] at forward thinking. We will make sense
[00:28] of the problems of
[00:30] tomorrow.
[00:32] Inequality, sustainability,
[00:35] urbanization, the gender gap, and the
[00:38] demographic time
[00:41] [Music]
[00:50] bomb. The world is changing. Today we
[00:54] stand on the brink of a fourth
[00:56] industrial revolution. One that will
[00:58] transform the way we work, the way we
[01:00] live, and even what makes us
[01:05] human. There's a a group of technologies
[01:08] that are combining to create
[01:10] transformation across almost every
[01:12] industry at the moment. And those
[01:14] technologies include things like
[01:15] artificial intelligence, 3D printing,
[01:18] robotics, um big data, and then some
[01:22] things in on the sort of life sciences
[01:24] front in terms of genetics and and
[01:26] medical imaging. And that these things
[01:27] are sort of combining in a way that's
[01:29] bringing about a host of transformative
[01:30] changes across industries.
[01:35] I would describe the fourth industrial
[01:37] revolution actually quite similarly to
[01:40] how I would describe the past three and
[01:42] that is technology that leads to massive
[01:46] gains in productivity and massive gains
[01:48] in productivity mean substantial
[01:51] improvements to everyone's quality of
[01:53] life.
[01:55] [Music]
[01:58] The world has been through revolutions
[02:00] before. The advent of mechanization,
[02:03] then electronics, then the digital
[02:05] revolution, all profoundly changed the
[02:07] world's
[02:09] economies. But this revolution could be
[02:12] even more disruptive.
[02:14] I think in previous revolutions, you
[02:16] could really talk about them as
[02:18] industrial revolutions. What was
[02:19] changing was how things were made.
[02:22] Factories, industry, often heavy
[02:24] industry in particular. Here you're
[02:25] seeing transformation across really a
[02:27] whole range of not just industry but
[02:29] services and and the creation of whole
[02:31] new business models that didn't exist
[02:33] before. What's different a little bit
[02:35] about this particular revolution is that
[02:38] um it gets into a whole range of things
[02:40] that people only thought were ever only
[02:42] possible for humans to do. Jobs that
[02:45] were human jobs before aren't going to
[02:46] be human jobs anymore.
[02:49] At the heart of this fourth revolution
[02:51] is artificial intelligence. the ability
[02:53] of machines to match and perhaps one day
[02:56] surpass the cognitive ability of their
[02:58] human creators. What's happening now is
[03:01] a big deal. Um it is making a big
[03:03] difference in the way people live, uh
[03:05] the way people interact with each other.
[03:07] It is sort of obliterating distance. It
[03:09] is in some cases removing humans from
[03:12] tasks that we once thought were the sole
[03:15] province of the human mind. These
[03:17] analytic tasks that we thought only a
[03:19] human brain could do. We're suddenly
[03:21] finding that algorithms can do, that
[03:22] machines can do. These are early days in
[03:25] the brave new world of artificial
[03:27] intelligence, but the potential benefits
[03:29] are vast. What are some of the
[03:31] liberating benefits of artificial
[03:33] intelligence? They're actually a lot.
[03:34] Um, if you think about something just
[03:36] like driverless cars, autonomous
[03:38] vehicles, which is one use of AI that
[03:40] people are talking about, that could
[03:41] have a really liberating um impact on a
[03:43] lot of people's lives. If you think
[03:44] about older people who can no longer
[03:46] drive, they're very shut in their houses
[03:48] right now, very dependent on others for
[03:50] transportation. With driverless cars,
[03:52] they would be able to go about their
[03:53] daily life. And then you're seeing with
[03:55] big data that this may have a profound
[03:57] impact on drug development that you'll
[03:59] find um new pharmaceuticals being
[04:01] developed at a faster rate um to cure
[04:04] diseases because the computers are
[04:06] essentially able to sort through the
[04:07] data and pick up connections that
[04:09] otherwise would be missed.
[04:13] For health in particular, the advantages
[04:15] of machine learning and data science are
[04:17] immense. Those have an incredible chance
[04:19] to address very both very infrequent
[04:22] diseases and diseases which affect
[04:23] different parts of the population very
[04:25] differently. If we're going to cure
[04:26] cancer, it's probably going to come
[04:27] through data science.
[04:31] But there is potentially a darker side
[04:33] to this technological revolution, one
[04:36] which could profoundly change the world
[04:37] of work as we know it.
[04:41] A technological revolution will cost
[04:43] jobs. It'll cost jobs in the areas that
[04:46] see the biggest advancements first. A
[04:49] good example of that that that is
[04:50] feasible over the near term is truck
[04:53] driving. You have self-driving
[04:55] trucks. You don't need the 3.5 million
[04:57] truck drivers that you have right now in
[04:59] the US. What is key as part of this
[05:01] revolution, as productivity goes up, as
[05:04] the economy continues to evolve and new
[05:06] jobs are created, you need to make sure
[05:08] those displaced workers are given the
[05:11] skills to move into these new positions.
[05:13] That's what's key. Will all of them be?
[05:15] No. But I think the key point is you
[05:18] need to make sure if you've lost 3.5
[05:19] million jobs in one sector, how do you
[05:21] create more than that in another sector?
[05:24] And I think in past industrial
[05:26] revolutions, that's what we've seen
[05:28] happen. And hopefully uh and I think it
[05:30] will it will be what happens again.
[05:34] But what if this doesn't
[05:38] happen? Martin Ford is a software
[05:40] entrepreneur. He has peered into our
[05:43] future economy and sees a world where
[05:45] potentially hundreds of millions of
[05:47] skilled workers are out of a job. I
[05:50] would say that if you look far enough
[05:52] into the future, there is no job
[05:55] anywhere in our economy. There's nothing
[05:57] that anyone does that is completely
[05:58] safe. And that includes even artists and
[06:01] novelists and you know the kinds of jobs
[06:03] that you would imagine right now are
[06:05] completely beyond the the scope of
[06:07] artificial intelligence. Millions and
[06:09] millions of those jobs are going to be
[06:11] lost and it's unlikely that enough jobs
[06:13] are going to be created to absorb all of
[06:15] those workers.
[06:22] [Music]
[06:24] Martin Ford is a software entrepreneur
[06:27] who has a chilling vision of the future.
[06:30] His best-selling books have put him at
[06:32] the forefront of a movement which
[06:33] worries about technology, the speed of
[06:35] its growth, and the immense potential it
[06:37] has to change the
[06:39] world. This is the fourth industrial
[06:42] revolution. the advent of machines
[06:44] powered by artificial intelligence which
[06:47] have the potential to make redundant
[06:49] hundreds of millions of workers across
[06:51] the planet. It is a world which is
[06:54] nearly upon us but which governments and
[06:56] businesses are only starting to
[07:00] comprehend. Well, the central idea in my
[07:03] latest book, The Rise of the Robots, is
[07:05] that over time, machines, computers,
[07:08] smart algorithms are increasingly going
[07:10] to substitute for human labor. I think
[07:12] that that's inevitable. Um, technology
[07:14] is eventually going to be able to do
[07:17] many of the things that people now do,
[07:19] and I think there's a good chance that
[07:20] that will result in unemployment. It's
[07:22] going to push people out of the labor
[07:23] force. Many people are going to find it
[07:25] impossible to adapt to that because
[07:27] they're not going to have capabilities
[07:28] that really exceed what machines can do.
[07:31] And that's, I think, going to be a
[07:32] genuine concern both for our society, of
[07:34] course, and ultimately for the economy,
[07:36] too.
[07:38] [Music]
[07:40] Some of those machines are already with
[07:42] us.
[07:45] There are already algorithms that can
[07:46] interpret things like body language and
[07:49] um respond to some extent to emotion. It
[07:51] can determine your mood, for example,
[07:53] and so forth. And and you know, this has
[07:56] big implications. Imagine what that
[07:58] could mean, for example, for advertising
[08:00] if an algorithm can determine exactly
[08:03] how you're feeling and then target
[08:05] advertisements at you based on that.
[08:08] Some of the language transl translation
[08:10] things that have been demonstrated are
[08:12] truly remarkable. Imagine if anyone in
[08:14] any country who speaks any language
[08:16] would now be able to do any job uh
[08:18] because we have perfect uh machine
[08:21] translation in real time between
[08:22] languages. So you know that has real
[08:25] implications for the job market.
[08:26] Obviously
[08:29] we may already be starting to see the
[08:31] effect on the wider economy in the first
[08:33] decade of this century. The net total
[08:36] number of jobs created in the United
[08:38] States was
[08:45] zero. What we see is that in the United
[08:48] States we've been having what we call
[08:51] jobless recovery. So, um, clearly
[08:53] there's something happening there. And I
[08:54] think part of what's happening is that
[08:57] jobs disappear when a recession happens.
[08:59] And then when finally recovery comes
[09:01] back, companies find that they're able
[09:03] to leverage technology to avoid hire
[09:06] rehiring a lot of those workers. And so,
[09:08] it's taken longer and longer for the
[09:10] jobs to reappear. Throughout history,
[09:12] technology has always disrupted
[09:14] economies and societies. In the late
[09:17] 19th century, 50% of US workers were
[09:20] employed on farms. By 2000, it was less
[09:24] than
[09:26] 2%. Those workers found work in other
[09:29] sectors. But Martin thinks this time
[09:32] it's
[09:34] different. What transformed agriculture
[09:37] was a specific mechanical technology. Uh
[09:40] now we've got a technology that's really
[09:41] just ubiquitous. It's across the board.
[09:46] Artificial intelligence is something
[09:48] that's just scaling across our entire
[09:50] economy. It's not something that's
[09:52] impacting just one sector. It's
[09:54] something that literally is everywhere.
[09:56] And as a result, it means that there
[09:57] isn't really going to be any safe haven
[09:59] for workers.
[10:02] What makes the new technology so
[10:04] ubiquitous is the development of a new
[10:07] virtual world, the world of big data.
[10:10] Well, big data essentially is the
[10:12] collection and use of just massive
[10:14] amounts of data. In big corporations,
[10:16] for example, these companies are
[10:18] collecting all kinds of information
[10:20] about uh their customers, about their
[10:22] business operations, about the actual
[10:24] processes in in industrial environments
[10:27] and factories, um about the things that
[10:29] their employees are doing. All of this
[10:32] data essentially becomes a kind of feed
[10:34] stock for these smart algorithms. it
[10:36] it becomes the information that they use
[10:38] to learn and and basically to figure out
[10:40] how to do things and um that's something
[10:42] that is just going to be I think
[10:44] dramatically disruptive going
[10:46] forward. The total data stored on the
[10:49] world's computers is now believed to be
[10:51] well over 1,000 billion gigabytes. And
[10:55] it is big data which is driving the most
[10:57] disruptive advance in technology, the
[11:00] ability of machines to think.
[11:03] One thing that you'll very often hear
[11:05] people say even today is that computers
[11:07] only do what they're programmed to do.
[11:09] And you know, this is really not right
[11:10] anymore. And and the reason it's not
[11:12] right is basically because of machine
[11:13] learning. Because we now have this
[11:15] technology that allows smart software
[11:17] algorithms to look at data and based on
[11:20] that to to learn to learn how to do
[11:22] things, to figure things out, to make
[11:24] predictions. So it really is no longer
[11:27] the case that some human being is
[11:29] sitting down and telling a computer
[11:30] exactly what to do step by step. Uh
[11:33] computers are now having the ability to
[11:35] figure that out for
[11:38] themselves. You can imagine a future
[11:40] where every device, every appliance, all
[11:43] kinds of industrial equipment,
[11:45] everything communicates and talks to to
[11:47] each other. And I think that one of the
[11:49] things will happen is that artificial
[11:51] intelligence will kind of use that as a
[11:52] platform. It will scale across all of
[11:54] that. Everything will become more
[11:56] intelligent.
[11:59] The last great technological advance saw
[12:01] robots replace millions of blue collar
[12:03] jobs in factories and on production
[12:06] lines. Martin believes this new
[12:08] disruption is going to target the white
[12:10] collar workforce as well. Once a a
[12:13] computer learns to do something, then
[12:15] that that information can be scalable
[12:17] out to any number of machines. So it's
[12:19] almost like you can imagine having a
[12:21] workforce of people and you could train
[12:23] one employee to do a particular task and
[12:25] then you could clone that worker and and
[12:28] have a whole army of those workers.
[12:30] That's a bit like the way artificial
[12:32] intelligence works. So machine learning
[12:34] is is very scalable. If you've got the
[12:36] kind of job where someone else, another
[12:39] smart person could maybe watch what
[12:41] you're doing or study everything you've
[12:43] done in the past and figure out how to
[12:45] do your job, then it's a pretty good bet
[12:47] that eventually there'll be an algorithm
[12:48] that will come along and be able to do,
[12:50] you know, essentially that that same
[12:52] approach. So, um, that's a lot of jobs.
[12:58] Many of the jobs which might be
[12:59] displaced are those currently occupied
[13:01] by educated, highly paid workers.
[13:06] So you can see really across the board
[13:08] that um anyone sitting in front of a
[13:10] computer doing some sort of routine
[13:12] predictable knowledge work, for example,
[13:14] if they're cranking out the same report
[13:16] or the same analysis again and again,
[13:18] all of that is going to be very
[13:20] susceptible to this. Journalism is one
[13:22] interesting area that's being impacted
[13:23] by this because there are now systems
[13:25] that can essentially tap into data and
[13:28] then they can transform that data into a
[13:30] very compelling news story that that
[13:32] many people would read and and they
[13:34] can't tell that it was written by a
[13:35] machine. In the future, maybe 90% of new
[13:38] stories will be machine generated.
[13:43] The number of jobs displaced has the
[13:45] potential to utterly transform the
[13:47] economic landscape. There have been a
[13:50] couple of studies done most notably by a
[13:52] couple of researchers at Oxford
[13:53] University and they've looked at a
[13:55] number of countries and most of the
[13:56] results have come back suggesting that
[13:58] up to half of the jobs could be
[14:01] susceptible to automation perhaps over
[14:04] the next 20 years. That's 60 million
[14:07] jobs in the United States alone.
[14:10] That's a staggering number. Obviously,
[14:12] we have a massive social problem. you'd
[14:14] have tremendous stress on government in
[14:16] terms of trying to take care of all
[14:18] these people that no longer have an
[14:20] income. Um, I think that you would see
[14:22] uh the potential for a massive economic
[14:25] downturn because you would run out of
[14:27] consumers. You no longer have people
[14:29] that are capable of buying the products
[14:31] and services that are being uh produced
[14:34] by the economy.
[14:37] A revolution on this scale wouldn't just
[14:39] transform an economy. It would have
[14:41] immense implications for our society.
[14:47] We could really have just what you might
[14:49] call inequality on steroids. The very
[14:52] wealthy people who own all this
[14:54] technology are going to do
[14:55] extraordinarily well. You would have the
[14:58] potential for civil unrest, perhaps even
[15:00] riots or massive crime waves.
[15:03] In the United States during the Great
[15:05] Depression, we had an unemployment rate
[15:07] of about 25%. And back then there were
[15:10] many people genuinely concerned that
[15:12] that would result in the collapse of
[15:15] both democracy and capitalism.
[15:19] This situation amounts to just about the
[15:21] end of the world as we know it. A
[15:24] science fiction nightmare straight from
[15:25] the movies. There are some very
[15:28] prominent thinkers like for example
[15:29] Stephven Hawking and Elon Musk who have
[15:33] raised genuine fears about the potential
[15:35] for advanced artificial intelligence and
[15:36] their concern is that someday we're
[15:38] going to build a super intelligent
[15:39] machine. Imagine a machine that's 100 or
[15:42] maybe a thousand times smarter than any
[15:44] living person. Uh what would that system
[15:47] think? How would it act? Would it have a
[15:49] use for us? Uh it might decide that
[15:51] we're simply a burden. It might decide
[15:54] to just get rid of us. Uh, so it could
[15:56] potentially present an existential
[15:57] threat. Uh, is that something to worry
[15:59] about? I think that it's not a silly
[16:01] concern. It's not something that we
[16:03] should laugh at and just
[16:08] dismiss. There's really no end point to
[16:11] this. There's no point at which you can
[16:12] say this is absolutely as far as we can
[16:15] go and and machines will never go beyond
[16:17] this. We are reaching a new era, a time
[16:20] when things are going to operate
[16:21] differently and we need to adapt to
[16:23] that.
[16:24] Healthcare is one area of the economy
[16:27] already adapting to this disruption. And
[16:29] in this field, researchers hope that
[16:31] intelligent humans and intelligent
[16:34] machines can work together for
[16:36] everyone's
[16:42] [Music]
[16:45] benefit. The fourth industrial
[16:48] revolution, the era of artificial
[16:50] intelligence has arrived. Computers are
[16:53] now mastering tasks once considered the
[16:56] sole preserve of humans and putting
[16:58] millions of jobs at risk. And now
[17:01] business leaders are wrestling with the
[17:03] potentially huge
[17:05] implications. In general, robots of one
[17:09] form or another are going to become much
[17:10] more omnipresent in our lives in a good
[17:13] way. They'll replace a lot of repetitive
[17:15] activities that people are currently
[17:16] doing. Robots will have a dramatic
[17:19] effect on the labor pool. Lower the cost
[17:21] of products. People will start to
[17:22] realize that just about every manual
[17:24] task eventually will probably be done by
[17:26] a robot.
[17:28] Martin Ford's books have highlighted the
[17:30] threat to the job market. But even he
[17:33] sees areas where artificial intelligence
[17:35] could be beneficial.
[17:37] I do think that that healthcare is
[17:38] actually one of the areas where the
[17:40] impact of artificial intelligence and
[17:42] robotics could be extraordinarily
[17:44] positive in the future. The burden on
[17:46] our economy is growing at a remarkable
[17:48] rate, especially in the United States.
[17:50] So, if we can deploy more artificial
[17:52] intelligence and robotics there to make
[17:54] that more efficient, that'll be a great
[17:57] thing. Analysts expect the AI healthcare
[18:00] market to generate revenues of over $6
[18:03] billion by 2021, 10 times its current
[18:07] total.
[18:09] Young companies like Hindsight in New
[18:11] Jersey and Analytic in California are
[18:13] mining data to improve patient outcomes
[18:16] across a range of
[18:18] illnesses. And in New York, IBM
[18:21] researchers have developed Watson, an
[18:23] intelligent software system at the
[18:25] forefront of this revolution. It can
[18:28] understand somebody's personality type.
[18:31] It can look at email for example and
[18:33] tell you what is the tone of the email.
[18:35] uh you know what kind of messages are
[18:37] coming through whether you interneted
[18:38] them or not right uh it can look at uh
[18:41] for example a big encyclopedia and
[18:43] extract all the concepts and the
[18:45] relationship among those concepts.
[18:49] Watson operates in the world of big data
[18:52] extracting knowledge from the billions
[18:54] of facts and figures floating through
[18:56] cyerspace.
[18:57] I look at the world from the point of
[18:59] view of uh you know the amount of data
[19:02] that there is um and the amount of
[19:05] knowledge that is embedded or insights
[19:07] that's embedded in the data that we're
[19:09] not able to extract today and therefore
[19:11] we are not able to make the right
[19:13] decisions. So um the fourth industrial
[19:16] revolution to me is the ability to have
[19:20] a much better understanding of the world
[19:23] through all of the data and therefore
[19:25] making better decisions for it.
[19:32] IBM is currently running a research
[19:34] project in which Watson augments the
[19:36] intelligence of medical professionals
[19:38] helping doctors treat the most dangerous
[19:40] diseases in the world including skin
[19:43] cancer.
[19:46] Melanoma is a very deadly form of skin
[19:48] cancer and it's something where early
[19:50] detection and intervention is is key. So
[19:54] a dermatologist faced with a patient who
[19:57] has a skin lesion will make some
[19:59] assessment about the likelihood of a
[20:01] lesion being
[20:03] melanoma. So unfortunately today
[20:05] dermatologists can make errors. Some
[20:08] melanomas are being missed and some skin
[20:11] lesions which are perfectly benign are
[20:14] being excised needlessly. So what we can
[20:17] do here is essentially ask the computer
[20:21] to make a deep analysis over an image.
[20:23] So this image is then being sent to the
[20:26] computer and it's being automatically
[20:28] analyzed. And what the computer is
[20:30] telling us about this image is that
[20:33] there's a very high probability that it
[20:35] corresponds to melanoma. What we're
[20:38] finding in our own internal
[20:40] retrospective research is that the
[20:42] computer can be as accurate as
[20:45] 95%. So this compares to the best
[20:48] clinical experts today that are between
[20:51] 75 and 84% in recognizing melanoma. It
[20:54] is not a tool that would replace uh the
[20:58] clinical expert. uh rather it provides
[21:00] them with additional analysis over the
[21:04] skin lesion images uh by providing
[21:06] reaches into large databases of uh
[21:09] similar lesions.
[21:11] [Music]
[21:13] This is a vision of a future where
[21:16] humans and machines work hand in hand
[21:18] complementing one another's skills. I
[21:21] look forward to a time when you know
[21:23] every professional in fact you know two
[21:25] three billion professionals around the
[21:27] world are all able to have their own
[21:29] personal cognitive assistant that can
[21:31] help them do their daily jobs and that
[21:33] changes the nature of expertise.
[21:36] Humanity will move to a completely
[21:37] different place in terms of expertise
[21:39] and how we apply our knowledge and our
[21:42] experience into real world problems and
[21:44] therefore make the world a better place.
[21:46] Just like uh we've had um machines that
[21:48] could um augment people's muscles in the
[21:52] in the prior industrial revolutions uh
[21:54] or uh can help people you know search
[21:57] vast amounts of information like in the
[21:59] uh in the internet era. I look at the
[22:02] next revolution as machines augmenting
[22:06] people's cognitive capabilities. Um
[22:09] that's how I think about it.
[22:12] Martin Ford remains cautious, believing
[22:15] artificial intelligence is going to
[22:16] fundamentally change the way we live and
[22:19] work and challenge us like never before.
[22:23] We're not prepared for the disruption
[22:25] that's
[22:26] coming. We're going to see things get
[22:29] worse before they get better. In
[22:32] particular, the impact on the job market
[22:34] and the impact on the incomes and the
[22:37] livelihoods for average people. So, you
[22:40] know, in the short term things could be
[22:42] pretty difficult, but in the longer
[22:44] term, if we do adapt to this, then I
[22:46] think there are reasons to be really
[22:48] optimistic. I mean, you can imagine an
[22:51] almost utopian kind of future where no
[22:53] one has to do a job that's dangerous or
[22:55] that they really hate or that's really
[22:57] boring, where technology takes on more
[22:59] and more of that. And um if we can get
[23:02] to that point, of course, then that's a
[23:03] tremendously positive outcome. So, I
[23:05] think that all of that is really
[23:06] possible and it could be one of the best
[23:08] things that's ever happened to humanity,
[23:10] but it will require that we adapt to it
[23:12] and uh that's going to be a staggering
[23:16] [Music]
[23:20] challenge. This is the age of the city.
[23:23] For the first time in human history,
[23:25] more people live in urban than rural
[23:28] settlements.
[23:30] The world's urban population is growing
[23:32] by 70 million people each year. 301
[23:37] cities account for 50% of global GDP.
[23:42] This will rise to 66% by
[23:45] 2025. So if we don't get things right in
[23:48] our cities, then the consequences for
[23:50] humanity are profound. Cities are
[23:54] critically important to the global
[23:56] economy and to progress in the global
[24:01] economy. Cities can be sources of chaos
[24:05] as well as
[24:08] development. This dual personality of
[24:11] cities is what makes them so alluring
[24:13] and so vital.
[24:16] They can be dangerous
[24:18] places, but cities are where fortunes
[24:21] can be made.
[24:24] One of the primary uh factors driving
[24:26] urbanization is opportunity. You live on
[24:29] a farm and you're growing crops. You
[24:31] don't have a lot of opportunity. You see
[24:33] a bustling growing city. Your friends
[24:36] are moving there. They're getting uh
[24:38] jobs in offices, maybe jobs in a in a
[24:41] manufacturing center. Uh there's
[24:43] restaurants. There's culture, there's
[24:45] life. This this is attractive. Uh this
[24:48] is attractive and something you want to
[24:49] be a part of. And everything is
[24:51] relative. You know, they'll have greater
[24:53] access to schools, greater access to
[24:55] health care, greater access to
[24:56] employment, and a much less vulnerable
[25:00] uh economic life.
[25:09] In 1900, 12 of the world's biggest
[25:12] cities were in North America or Europe.
[25:15] 100 years later, this number had fallen
[25:17] to just two. Most of the biggest cities
[25:20] of the future will be in the developing
[25:22] economies of Asia and Africa.
[25:25] Most of the growth in cities is going to
[25:27] be in China, India, and Nigeria. Those
[25:30] three countries alone will account for
[25:33] 37% of the world's urban population.
[25:36] Just staggering numbers. Here's an
[25:39] example. Logos, the biggest city in
[25:42] Nigeria, its population every year is
[25:45] adding the equivalent of the population
[25:47] of
[25:48] [Music]
[25:51] Boston. The urbanization rate in the US,
[25:54] Japan, it's over 70%. In China, it's
[25:57] still 50%. So, China may have a lot of
[25:59] mega cities. They may have a lot of
[26:01] larger cities, but those cities are
[26:03] either going to get bigger or there's
[26:04] going to be more of them. So, I think
[26:06] that's going to be a trend. And I think
[26:07] a lot of emerging markets, especially
[26:09] those with large populations, are going
[26:12] to experience trends like that uh in the
[26:14] next 50 years.
[26:17] This incredible rate of growth makes the
[26:19] challenges of managing a large city even
[26:21] more difficult.
[26:24] The biggest risks facing cities are the
[26:26] same risks that challenge all of us. Uh
[26:30] politically, uh governance, uh climate
[26:33] change, uh economic uh inequality, uh
[26:38] productivity, economic growth,
[26:40] employment, education, transportation.
[26:42] Those issues that that face cities are
[26:45] the same that face everyone except on a
[26:47] on a much in a much more concentrated
[26:49] way.
[26:53] One city battling with many of these
[26:55] problems is Rio de Janeiro in
[26:59] Brazil. Aleandra Orurafhino is on the
[27:02] front line trying to solve them.
[27:05] She believes the world's biggest cities
[27:07] are in danger of sinking under a tide of
[27:09] poverty, decrepit infrastructure, and
[27:12] citizens apathy. And unless we do
[27:14] something about it, billions will suffer
[27:16] the consequences.
[27:20] The kind of urbanization that we have
[27:22] today can only go so far. If we do not
[27:25] change the way we design our cities, if
[27:27] we do not make cities change with us,
[27:29] we're going to have very serious limits
[27:31] to urbanization. Cities will become
[27:33] impossible to manage, impossible to live
[27:35] in, and just very miserable places to
[27:38] be. I think if we change that process,
[27:40] those limits could change dramatically
[27:42] and potentially be non-existent. But
[27:45] that requires that we think deeply about
[27:48] the environments that we want to be in
[27:49] and how we can better build them
[27:58] together. Managing mega cities is one of
[28:01] the great challenges facing the
[28:03] world. This is Rio de Janeiro,
[28:07] Brazil. Nearly 12 million people crowd
[28:10] into its metro area.
[28:12] It is beautiful and vibrant, but it also
[28:16] has its
[28:17] problems,
[28:19] crime,
[28:21] inequality and
[28:23] poverty. Alisandra Orurafhino is an
[28:26] urban activist and thinker who has lived
[28:28] and worked in mega cities on three
[28:30] different continents. She has worked
[28:33] with the United Nations on its
[28:35] sustainable development goals and
[28:37] founded the groundbreaking Mayor Rio, an
[28:39] NGO that uses data gathered from
[28:41] citizens to raise campaigns and solve
[28:43] thorny issues posed by the rapid growth
[28:46] of the
[28:48] city. Mayor Rio has 170,000 activists
[28:52] and Alexandra hopes it can become a
[28:54] model for other rapidly growing cities
[28:56] around the globe.
[29:00] We build upon a rich tradition of
[29:01] neighborhood movements not only in
[29:03] Brazil but all over the world and we try
[29:05] to sort of bring it to the 21st century
[29:07] in the way that makes sense for
[29:09] people. I was born in this city in Rio
[29:12] de Janeiro and my family has a very sort
[29:15] of mixed background. My father comes
[29:17] from a neighborhood in Rio that was
[29:21] quite dangerous in the '90s uh quite
[29:23] poor or lower middle class. And my mom
[29:26] comes from a very wealthy background,
[29:28] one of the best neighborhoods in Rio. It
[29:30] taught me that this city can be amazing,
[29:33] but it can also be very rough and
[29:34] unequal. And that's not just a
[29:37] characteristic of this city. I think
[29:39] it's something that we are seeing
[29:40] increasingly in cities around the world.
[29:43] [Music]
[29:45] Rio de Janeiro is similar to many
[29:48] emerging mega cities. Some neighborhoods
[29:50] are as wealthy as anywhere on the
[29:52] planet.
[29:54] Others remain impoverished and cut
[29:58] off. Bridging this gap will, Aleandra
[30:01] believes, have profound benefits for us
[30:03] all.
[30:05] Cities bring people closer together and
[30:07] they have this intensity in them, this
[30:08] density in them. They're definitely the
[30:10] places where most innovation will
[30:13] naturally happen because it's very hard
[30:15] to innovate when you're always talking
[30:17] to the same people and hearing the same
[30:18] thoughts. And cities are the exact
[30:21] contrary of that. They are uh natural
[30:24] hubs for innovation, natural hubs for
[30:26] economic growth and they tend to be the
[30:28] engines of growth in most
[30:33] countries. But when this growth is rapid
[30:36] and unplanned, the results are
[30:38] gridlocked streets, poisoned air, and an
[30:41] infrastructure that simply cannot cope.
[30:44] Well, I come from a city that expanded
[30:46] too rapidly for sure. How do you create
[30:49] sidewalks, sewage systems, schools,
[30:53] mobility systems, uh to cater to a
[30:55] growing population? If that rapid urban
[30:59] expansion is happening in environments
[31:01] where inequality is paramount, u the
[31:04] challenges are even bigger. In a mega
[31:07] city, one of the biggest challenges can
[31:09] be simply getting from A to
[31:11] [Music]
[31:13] B.
[31:15] Our mobility systems in general, very
[31:18] few exceptions, suck. When you have a
[31:21] poor mobility system, you just preclude
[31:24] entire segments of the population from
[31:26] living the city, from actually accessing
[31:29] the opportunities and the beauties and
[31:30] the amazingness that cities have, right?
[31:33] Because it's very hard for them to get
[31:35] around. You also preclude the rich
[31:37] people in the city from getting to know
[31:39] other areas in the city, which can be
[31:41] incredibly exciting and and a fulfilling
[31:44] experience in and of itself. So, you're
[31:46] creating a city in which everyone is
[31:47] living in their own territory, which is
[31:54] terrible. At the forefront of these
[31:56] infrastructure problems are the city's
[31:59] poor. they can become physically cut off
[32:01] from the economic opportunities that
[32:04] living in a city provides. The poor bear
[32:06] the brunt of most things and I think
[32:08] that includes rapid expansion of cities.
[32:10] The fact that in the developing world
[32:13] onethird of the population is living in
[32:15] slums is something that none of us
[32:17] should accept um as as we grow and as we
[32:19] think about the planet in which we want
[32:21] to live. Slums are a result of rapid
[32:24] unplanned expansion. Today, an estimated
[32:28] 863 million people live in slums. If the
[32:33] 104 million slum dwellers in India were
[32:35] a separate nation, they would be the
[32:37] 13th most populous country in the
[32:41] world. But slums are not always hopeless
[32:45] places. The poor are not just sitting
[32:47] waiting for the government to do
[32:48] something for them. They're creating
[32:50] their own urban environments. So if you
[32:52] go to a slum in Rio, you see that most
[32:54] of that infrastructure was built by the
[32:56] community itself um over the years. So
[32:59] there is a level of do-it-yourself, a
[33:01] level of initiative that you see a lot
[33:03] more in poor neighborhoods and rich
[33:04] neighborhoods precisely because the
[33:06] government wasn't there.
[33:08] This means slums must be handled
[33:10] delicately by urban planners.
[33:14] What do we do with areas that were
[33:16] developed by communities but lack
[33:19] infrastructure? Even if we're assuming
[33:21] good will in terms of how we handle
[33:23] them. Even if the only thing that we
[33:25] want to do is provide those areas with
[33:27] good quality public services, there are
[33:29] choices that need to be made in terms of
[33:31] which pieces of that infrastructure do
[33:33] we leave, which pieces do we change,
[33:35] knowing that it was built by the people.
[33:38] If we don't handle that process in a way
[33:40] that is human and intelligent and
[33:43] actually aimed at protecting the
[33:44] interests of the poor communities, we
[33:47] can end up with massive waves of
[33:49] dislocation and and and destroying an
[33:52] urban fabric and a social fabric that is
[33:54] so important and so
[33:59] vital. Here in Ria, we have a
[34:01] neighborhood called Santa Theiza. And in
[34:03] that neighborhood, we have a tram. It's
[34:04] a historical tram. It's beautiful. Uh
[34:07] most trams in Rio were destroyed um in
[34:11] the earlier in the 20th century. Uh in
[34:14] Santaa the neighbors organized and kept
[34:17] their tram. It's a point of pride for
[34:18] them. Satza was a forgotten neighborhood
[34:21] for a while. It became a lot poor and
[34:24] then in the past five to six years it
[34:26] has been gentrifying really quickly and
[34:29] the government decided to turn that tram
[34:31] which is one of the very few remaining
[34:32] in the city into a tourist attraction.
[34:34] But what the neighbors said at that
[34:36] point was the only reason why this tren
[34:38] still exists and it's vintage and kind
[34:40] of hipstery and amazing. It's because we
[34:42] organized and we kept it here. They
[34:44] created that value. They created the
[34:46] richness of that community. And we see
[34:48] that all over the
[34:49] world. Alisandra believes cities often
[34:52] ignore this creativity. The result is a
[34:55] democratic deficit which erodess faith
[34:57] in the city's government and alienates
[34:59] already vulnerable communities.
[35:03] Aleandre believes cities must take their
[35:05] citizens with them if they are to expand
[35:07] successfully. I think what we have
[35:09] definitely not gotten right is the
[35:11] process by which we involve citizens. I
[35:13] have not seen one case of a city that
[35:16] has really used the collective
[35:18] intelligence of its citizens and and
[35:21] distributed power in a way that makes it
[35:24] actually possible for people to
[35:25] influence the way the city evolves. And
[35:27] when we get that that right, I think
[35:29] we'll solve a lot of the other issues
[35:30] that we see.
[35:32] But for us to truly harness the power of
[35:35] our cities, we need to heal the
[35:37] divisions within them
[35:39] first. If we keep building unequal
[35:42] cities, cities that are not sustainable
[35:43] and cities that are not very good to
[35:45] live in for most of their population, I
[35:47] don't think we can actually hope to be
[35:49] happy in these urban spaces. The worst
[35:52] case scenario for the global city of the
[35:54] future would be cities that do not have
[35:56] a soul and therefore become less and
[35:59] less attractive to entrepreneurs to
[36:02] people who do who do want to create new
[36:03] economic activity and that ultimately
[36:06] also become less wealthy.
[36:10] Across the ocean from Rio, another giant
[36:12] city is growing. Lagos is now the most
[36:15] economically important city in Africa,
[36:18] but its growing pains are excruciating
[36:21] and threatening the futures of 21
[36:24] million
[36:33] people. More people live in cities than
[36:35] ever
[36:37] before, but many of the world's biggest
[36:39] cities are struggling to cope.
[36:45] Lagos on Nigeria's Atlantic coast is the
[36:48] largest city in the world without a
[36:50] citywide rail system, meaning everyone
[36:54] has to travel by road. For workers like
[36:57] Abraham Cole, this means his daily
[36:59] commute takes over his life. What time
[37:02] did you wake up?
[37:04] This morning I woke up like
[37:06] 3:00 3:30.
[37:11] Well, I usually don't do breakfast cuz
[37:14] it kind of slow me down.
[37:17] In 3 years, the population of Lagos has
[37:20] nearly doubled from 11 million to 21
[37:23] million. But this staggering expansion
[37:26] has overwhelmed the city's impoverished
[37:28] infrastructure. How long should it take
[37:30] you to get to the office? It should take
[37:32] me 45 minutes to get to the office.
[37:35] Well, in full traffic, in full rush
[37:37] hour, how how long does that take? You
[37:39] probably would do like some six 7 hours
[37:43] in traffic, 3 hours going, 3 4 hours
[37:47] coming back. It's much worse coming
[37:49] back. Coming back is is is something
[37:54] else. And I don't think I want to waste
[37:58] seven hours of
[38:02] my everyday
[38:05] time for the rest of my life.
[38:08] [Music]
[38:10] Lagos is currently ranked in the top
[38:13] five least livable cities in the world.
[38:16] But although the city's economy is
[38:17] bigger than Kenya's, simply getting to
[38:20] their desks is a daily ordeal for its
[38:22] millions of workers. So when do you see
[38:25] your children?
[38:27] Weekends. Weekends only.
[38:31] Sometimes I see them during the week if
[38:35] they really want to see me and they're
[38:37] keen to see me. Sometimes they miss me
[38:39] that much. That must be quite difficult.
[38:42] Yes, it is.
[38:44] But it's what we have to
[38:49] do for now.
[38:56] [Music]
[39:06] Like millions of Legos Gozian workers,
[39:08] Abraham's first act on getting to work
[39:10] in the morning is to take a nap. There
[39:14] we
[39:15] go. Welcome to my office. So, what what
[39:19] are you going to do now? I think I
[39:22] have Yeah, this is quite early. This is
[39:25] 710. So, I'll take a nap for like 30
[39:29] minutes and get ready for
[39:34] work. 2,000 people migrate permanently
[39:38] to Lagos every day, straining the city's
[39:40] infrastructure further and expanding the
[39:42] city from the land to the sea.
[39:45] [Music]
[39:48] The result is slums like Makoko, a
[39:51] floating settlement on the city's
[39:57] [Music]
[39:59] lagoon. The infrastructure has not kept
[40:02] pace with the population growth. So
[40:04] basic measures of quality of life, just
[40:07] as access to clean water, for example,
[40:10] access to electricity are are limited.
[40:12] So before you even get to issues related
[40:15] to uh growth and and development uh
[40:20] Lagos and Nigeria have to sort out much
[40:22] more basic issues of infrastructure.
[40:27] Makoko is the oldest slum in Lagos.
[40:30] 80,000 people live here in buildings
[40:33] sitting on stilts connected by a complex
[40:35] system of canals.
[40:38] successful cities find ways to deliver
[40:41] services to even the most uh most
[40:44] deprived and that's that's the challenge
[40:46] especially in the developing world where
[40:48] resources are at a at a premium.
[40:53] In Makoko, residents have developed
[40:55] their own infrastructure, including
[40:57] fresh water and
[40:59] electricity. And this three-story
[41:01] floating school, which doubles as a
[41:03] community center, is the latest addition
[41:06] to this unique
[41:08] environment. The school was completed in
[41:10] 2013. It is cheap and easy to build. Its
[41:14] designers hope it will become a template
[41:16] for future buildings in Makoko.
[41:19] Makoko in Nigeria raises interesting
[41:21] questions of governance and control. For
[41:24] example, uh it's been a uh long ignored
[41:29] area and the local residents took charge
[41:32] and tried to improve their own lot with
[41:35] schools and with their own locally
[41:36] initiated development projects. However,
[41:38] the central government also has decided
[41:41] it wants that area for its own
[41:42] development reasons.
[41:45] Only a few kilometers away lies an
[41:47] alternative vision of how Lagos might
[41:49] develop. Not a grassroots community
[41:52] vision, but a grand project of
[41:54] incredible scale. Echo Atlantic.
[41:59] Well, where we are standing, we are in
[42:02] the alignment of the financial district,
[42:06] what we call Echo Boulevard or some
[42:09] people call it our fifth avenue. This is
[42:11] where all the major financial
[42:13] institutions will establish their
[42:15] headquarters and
[42:17] [Music]
[42:19] offices. Echoat Atlantic is a
[42:21] multi-billion dollar residential and
[42:24] business district built on 10 km of
[42:26] reclaimed land. It is in effect a new
[42:29] city, or it will be
[42:32] soon. Its backers hope a quarter of a
[42:34] million people will one day live here
[42:36] with 150,000 workers commuting from the
[42:40] old city across the water.
[42:44] When we initially started to conceive
[42:47] Equat Atlantic
[42:49] uh obviously we looked at um Canary Wolf
[42:52] in London.
[42:54] We looked at Dubai and if you look at
[42:57] the heart of London, part of Paris, half
[42:59] of New York, obviously uh the vast
[43:02] majority of the residents are wealthy
[43:05] people. I couldn't afford to live in the
[43:07] heart of
[43:08] London. But it in creating the residence
[43:13] for these people, you're also creating
[43:15] job opportunities.
[43:17] And it is the the norm here in Nigeria
[43:20] that when you create a residential
[43:23] apartment, you also create um quarters
[43:26] for the domestic staff working for that
[43:29] family. You have to take it into context
[43:31] that this is uh a city development. This
[43:35] is not uh a low-inccome settlement. It's
[43:39] it's a business center primarily. This
[43:42] is the future for the commercial
[43:45] development of of Lagos. There's no
[43:47] doubt about it.
[43:50] David hopes the first residential units
[43:52] will be open by the end of 2016 with the
[43:55] infrastructure of the whole site in
[43:57] place by 2022.
[44:00] Projects like
[44:01] EcoAtlantic raise as many questions as
[44:04] as they answer uh especially from where
[44:07] where local residents are are aware that
[44:09] they may be getting the short end of the
[44:11] stick. On the other hand, they really do
[44:13] lend themselves to uh starting from
[44:17] scratch and being able to build
[44:19] structures where there
[44:20] are schools, hospitals, offices, uh
[44:24] transportation facilities, and they they
[44:27] give gigantic cities like Lagos an
[44:29] opportunity to create a model of what
[44:32] can be presuming they're they're planned
[44:34] and executed correctly.
[44:37] [Music]
[44:39] The future paths of mega cities like
[44:41] Lagos remain uncertain, organic
[44:45] citizen-led growth like
[44:47] Makokco or large-scale planned
[44:49] development like Echo Atlantic. What's
[44:52] clear is that left unchecked, growth
[44:54] could destroy city's immense potential.
[44:58] I'm an optimist when it comes to cities.
[44:59] I grew up in New York City in the 70s
[45:02] when the city went to the edge of
[45:03] bankruptcy and here we are in uh the
[45:06] 21st century and New York is is booming
[45:09] and thriving and uh it's a tremendous
[45:12] place and you can see with proper
[45:15] planning and and a diverse and vibrant
[45:18] population what's possible.
[45:22] I hope that those global cities will be
[45:24] extremely interconnected in the sense
[45:26] that they will have solidarity networks
[45:28] in the sense that they will have
[45:29] resiliency networks and in the sense
[45:32] that their citizens will feel like their
[45:34] city is where they want to be. Their
[45:36] city is the is the sort of the project
[45:37] that they want to build but they can
[45:39] move they can visit each other. They can
[45:41] learn from each other at the global
[45:43] stage.
[45:45] [Music]
[45:50] The world of work is still dominated by
[45:52] men. In the Middle East and North
[45:55] Africa, only 25% of women are
[45:58] economically active. Globally, 3/4 of
[46:01] unpaid work is done by women. And even
[46:04] in North American companies, 25% of
[46:07] female employees feel their gender has
[46:10] held them back.
[46:14] If women are half the people, they
[46:16] should have, you know, a fair shot for
[46:18] all of our benefit at contributing to
[46:21] the economy in a way that is really much
[46:23] more equal with men than maybe what
[46:25] we've seen in the past. Of the biggest
[46:26] companies in the world, only about 5%
[46:28] are run by women. On corporate boards,
[46:31] less than 20% of the decision makers at
[46:34] the corporate board table are women
[46:36] right now. In the US Congress, only
[46:38] about 20% of the elected officials in
[46:41] both the Senate and the House are women.
[46:43] Women earn about 79 cents on the male
[46:45] dollar. So there's all kinds of ways in
[46:47] which women don't have parity in the
[46:49] world in which we
[46:52] live. This is the gender gap and it's
[46:55] been around for a long time.
[46:58] the organizations that have a lot of
[47:00] power in our world, the elected
[47:02] government, uh big companies, the
[47:05] education structure, the medical
[47:07] systems, all of these things are really
[47:09] dominated by men at the top. And that's
[47:12] largely a result of the history of the
[47:13] 20th century and before that. And it is
[47:17] taking a while for women to break that
[47:20] what we call the glass ceiling. But at
[47:22] the same time, it's taking a while for
[47:24] the whole society to adjust to seeing
[47:26] men and women as equal actors at the top
[47:29] of any of these
[47:30] institutions. Nearly 100 years after
[47:33] women in the United States were
[47:34] guaranteed the right to vote, the gender
[47:36] gap remains an issue in every corner of
[47:39] the
[47:42] world. And closing it has become more
[47:45] than just an issue of fairness. It has
[47:47] become an economic imperative debated at
[47:50] Davos and in boardrooms across the
[47:52] world.
[47:54] The gender gap matters for business. You
[47:57] know, it's the it's the market
[47:58] opportunity as well as the potential
[48:00] loss to u to the bottom line. You have a
[48:04] company and a workforce that represents
[48:06] your market. You're more likely to
[48:08] succeed by having a more diverse
[48:11] workforce. uh companies tend to be uh
[48:14] more successful because they're able to
[48:16] more creatively address uh challenges uh
[48:20] and issues uh and uh you know in
[48:23] innovation. Uh and so if you have a
[48:25] boardroom or a committee entirely
[48:27] composed of individuals who all went to
[48:30] similar schools and have similar
[48:31] backgrounds and think the same way, uh
[48:34] they're going to be less successful uh
[48:36] than a very diverse board. In a study
[48:38] that came out last fall from McKenzie,
[48:41] um they found in looking at a big global
[48:46] uh number of companies and looking at
[48:48] economies around the world that in fact
[48:50] equalizing women's economic contribution
[48:53] by 2025 would add $26 trillion to the
[48:57] global economy. So, you know, there's
[48:59] really very big numbers related to um
[49:03] women becoming more equal in terms of
[49:05] economic participation wherever you look
[49:08] around the world.
[49:10] Evidence is mounting of the positive
[49:12] effect female voices can have at the
[49:15] very top of businesses. If you look at
[49:18] companies where you find female CEO or
[49:21] chairwomen, you know, those are
[49:23] companies where you don't have uh poor
[49:26] corporate governance. They don't have
[49:28] poison pills. They don't have unequal
[49:31] voting rights that keeps uh you know
[49:33] insider management in control. They
[49:35] don't have um staggered board elections.
[49:38] Companies with good governance are more
[49:40] likely to have um female CEOs.
[49:44] So the question is with the issue at the
[49:46] top of the economic and political
[49:48] agenda, what is holding women back from
[49:51] the very top? Is it lack of ambition or
[49:54] simple old-fashioned sexism?
[49:58] So uh if we look at uh professional
[50:01] women in the workforce about 43% of them
[50:05] uh leave their profession at some point
[50:08] to deal with care of most likely a child
[50:11] but also care of elderly relatives. It's
[50:13] very difficult to reenter your
[50:16] profession uh at the same level as your
[50:18] male peers who have been present for
[50:20] those last 10 years.
[50:23] This explanation rings true for former
[50:25] senior State Department official Amarie
[50:27] Slaughter. She believes that what is
[50:29] holding women back is the structure of
[50:32] our workplaces and societies. And if we
[50:34] don't do something about it, then our
[50:36] corporations and governments will
[50:38] continue to underperform. There's no
[50:41] global issue that would not be helped by
[50:46] advancing women or achieving equality.
[50:49] We want a world in which every human
[50:51] being, boys and girls, has the right and
[50:55] the ability to live up to his or her
[50:57] God-given potential. And what we have is
[51:00] a world in which far more men have that
[51:03] ability than women
[51:07] [Music]
[51:11] do. Once upon a time, women were
[51:14] promised they could have it all. But
[51:16] something is holding women back from
[51:18] gaining and retaining the very highest
[51:21] positions in business and government.
[51:23] Amarie Slaughter has reached these
[51:25] heights in her
[51:28] career. For 2 years, she worked for
[51:30] Secretary of State Hillary Clinton,
[51:33] helping to shape the long-term goals of
[51:35] US foreign policy and now runs the
[51:37] Washington DC think tank New America.
[51:41] But it was a 2012 article in the
[51:42] Atlantic which she subsequently turned
[51:45] into a critically acclaimed book which
[51:47] cemented her reputation as one of the
[51:49] most intriguing and thoughtful
[51:50] commentators on the question of women in
[51:53] power. The feminist movement is about
[51:56] equality. It's about women being able to
[51:59] have what men have always had, which is
[52:01] to be uh fulfilled uh in a job, to to be
[52:06] powerful if that's what you want, to do
[52:08] important work or work that is
[52:10] meaningful to you and have a family,
[52:12] too. And I still believe that women and
[52:16] men can do that. I think there's nothing
[52:18] that stops us in principle from doing
[52:20] that. But what I now say is we have to
[52:24] make really big changes still if we're
[52:27] going to get there. Because as work is
[52:29] currently structured, as we think about
[52:31] careers currently structured, far too
[52:34] many people do have to make a choice.
[52:36] And far too many of those people are
[52:38] women.
[52:40] This was a choice which Amarie Slaughter
[52:42] had to confront herself.
[52:44] Work in the State Department at a high
[52:46] level is work that depends on the state
[52:49] of the world and the world is
[52:52] unpredictable by definition and there's
[52:54] always too much work to do. If there's a
[52:56] revolution in Egypt, you can't say hold
[52:58] that. I'll be back on Monday. You have
[53:00] to be there when it happens. So I
[53:03] definitely worked pretty much, you know,
[53:06] very long hours for two years. When I
[53:08] went to the State Department, my family
[53:10] understood that they were going to
[53:12] sacrifice so that I could do something I
[53:14] really wanted. They stayed in Princeton.
[53:16] I worked in Washington. I left home at
[53:19] 5:00 a.m. on Monday mornings and I came
[53:21] back late on Fridays. And that was
[53:24] difficult, but I understood that that
[53:27] was what it took to do this job. My
[53:29] oldest son was entering adolescence when
[53:32] I left and he had a very stormy period.
[53:36] uh so much so that he started making
[53:39] really quite bad choices. A number of
[53:42] times I would just jump on a train and
[53:44] go home, you know, in the on the middle
[53:46] of the day and and Secretary Clinton was
[53:47] incredibly understanding. But after 2
[53:51] years, we realized that it really was a
[53:54] choice
[53:56] between putting all our energy into
[53:58] helping him get back on track with real
[54:01] important life
[54:03] consequences or, you know, getting
[54:06] promoted in a career that I had I
[54:11] loved. The decision to quit her dream
[54:14] job and leave Washington didn't just
[54:16] affect Amarie's career. It challenged
[54:19] the feminist crado by which she had
[54:21] lived her life. I saw the world
[54:23] differently. I realized that I had been
[54:26] telling women for decades, young
[54:29] students whom I taught, you can make it
[54:31] work. You just have to, you know, work
[54:33] hard and you can make it work. And I
[54:35] couldn't make it work. And if I couldn't
[54:37] make it work with all the advantages in
[54:40] the world, I had money. I had a husband
[54:42] who was a lead parent. I had every
[54:44] possible way to make it work. Well, then
[54:48] you know then there are places where we
[54:50] simply have to make choices. That was an
[54:52] epiphany.
[54:54] Anmarie's decision to put caring for her
[54:56] family before advancing her career saw
[54:59] her accused of betraying feminism. When
[55:02] I wrote my Atlantic article, I got a
[55:04] great deal of criticism from uh women of
[55:07] my generation or older uh who were
[55:10] feminist, women I admire, uh but who
[55:13] very much worried that I was setting the
[55:15] movement back. If I told people I'd come
[55:17] back because I wanted to be with my
[55:19] family, I got a reaction that
[55:21] essentially told me among many people
[55:23] and many women that they saw me a little
[55:27] differently than they had before, that I
[55:29] wasn't really a player, that I'm I
[55:31] wasn't as motivated or ambitious as
[55:33] they'd thought I'd been. Kind of
[55:36] disappointed. Anmarie's experience
[55:39] sparked a debate about whether women can
[55:41] have it all. Facebook COO Cheryl
[55:44] Sandberg had suggested that in order to
[55:46] get ahead, women needed to be more
[55:49] assertive in the workplace in the face
[55:50] of male power. They need to lean in
[55:55] more. I admire Cheryl Sandberg and I
[55:58] admire what Lean In has done. I've seen
[56:01] it as somebody who runs an organization.
[56:04] I have seen young women come in and ask
[56:06] me for raises and I know that they've
[56:09] just read lean in. you know, they're
[56:11] they're they're doing it, you know,
[56:13] they're pushing themselves forward in
[56:15] exactly the way uh that Cheryl Sandberg
[56:17] recommends and and many women uh
[56:20] advocate. And I agree with all of that.
[56:22] I think it is a debate uh about where to
[56:26] put the
[56:27] priority. Anmarie believes the problem
[56:30] lies deeper not just in women's
[56:32] individual behavior but in the way
[56:35] business and society is structured to
[56:37] make it almost impossible for women to
[56:39] have a career and to care for a family
[56:41] at the same
[56:44] time. That's a full-time job and
[56:48] somebody has to do it and women have
[56:50] traditionally done it. So women are
[56:52] still expected to do it. So, what you're
[56:54] doing is asking people who are holding
[56:57] two full-time jobs to compete with
[57:00] people who are holding only one. So, if
[57:02] a woman is the primary caregiver for her
[57:04] children or for her parents and a
[57:06] full-time bread winner, she's competing
[57:08] with people who are doing only one of
[57:10] those, that's like running a race and
[57:12] having half the people, you know, put a
[57:15] pack of rocks on their back and
[57:16] wondering why they don't advance to the
[57:18] finish line at the same pace.
[57:22] Instead of saying, "Well, that's
[57:23] something that women should still do
[57:25] while they're also working," we need to
[57:27] say parents should have the time and the
[57:31] space to be able to care for their
[57:33] children and also work. But that
[57:34] requires a much bigger shift in
[57:37] thinking.
[57:43] The effect of the gender gap can be seen
[57:45] across the global economy.
[57:49] Rates of prime age employment for women
[57:51] have been falling in the United States
[57:53] for nearly two decades. In 2014, just
[57:57] 70% of women aged 25 to 54 were in
[58:03] work. The comparable figure is higher in
[58:06] Scandinavian countries and these are the
[58:09] countries where the gender gap is at its
[58:11] narrowest.
[58:14] The countries that have gone the
[58:17] farthest toward real equality are the
[58:20] Nordic countries, Denmark and Sweden and
[58:22] and Finland and Norway. What they
[58:25] understand is both that you have to
[58:28] recognize that raising children is a
[58:30] social and economic investment and their
[58:32] governments say we're going to invest in
[58:36] maternity leave and paternity leave. And
[58:39] the paternity leave is particularly
[58:41] important because they uh create
[58:45] incentives for men to take not a week,
[58:48] not two weeks, but up to six months. Uh
[58:52] and they do that in part by giving one
[58:54] month or sometimes two months as a kind
[58:56] of use it or lose it. So the man's an
[58:59] idiot if he doesn't take the month to be
[59:02] with his children. Uh when if he doesn't
[59:05] do that, he just he he loses that leave,
[59:07] right? That's crazy.
[59:11] If sharing the burden of care of
[59:13] children and of parents is key to
[59:15] closing the gender gap, then that
[59:17] suggests accepting the traditional roles
[59:19] of men and women are a thing of the
[59:21] past. You have to be uh accustomed to
[59:25] seeing men as the primary caregivers of
[59:28] young children. And men have to
[59:31] understand that they're just as good at
[59:33] as women at this. Uh, and even more
[59:38] important or equally important in many
[59:40] ways, as a a a Finnish CEO said to me,
[59:44] the head of a big Finnish company, he
[59:46] said, "Now, when someone comes, a young
[59:48] man who hasn't taken their paternity
[59:50] leave, I wonder about their character,
[59:53] and that's where we have to go." And in
[59:56] Scandinavia, that's where they're
[59:58] heading. The Nordic nations are
[01:00:00] pioneering a new approach to work and
[01:00:02] parenthood and narrowing the gender gap
[01:00:04] in the process.
[01:00:09] [Music]
[01:00:15] Across the world, women are reaching the
[01:00:17] top in business and politics, but
[01:00:20] they're struggling to stay there.
[01:00:23] I call myself an impatient optimist. I'm
[01:00:26] impatient because the world is getting
[01:00:27] better for women, but it's not getting
[01:00:29] better quickly enough. And we need to do
[01:00:31] a lot to move that forward. And I I
[01:00:33] would love to tell you that because I'm
[01:00:34] a female CEO, uh I've changed the the
[01:00:38] fabric of the diversity makeup of my own
[01:00:40] company and I'm leading by example, but
[01:00:42] the reality is is that we're challenged
[01:00:44] in terms of uh female representation.
[01:00:47] It's not getting better. In fact, you
[01:00:50] know, post the crisis, there was less
[01:00:52] diversity on Wall Street than pre the
[01:00:54] crisis. And one would have thought it
[01:00:55] would have been the opposite. So I would
[01:00:56] take the opposite side. It's not getting
[01:00:58] better and it's costing Wall Street a
[01:01:00] lot of money.
[01:01:02] Anmarie Slaughter thinks this gender gap
[01:01:05] exists because of the way businesses and
[01:01:07] government treat family life. In Sweden
[01:01:10] along with its Nordic neighbors,
[01:01:12] attitudes are different. Scandinavia
[01:01:14] leads the world in gender equality, but
[01:01:17] its success has been hard one. I did
[01:01:21] military service when I was 20 years old
[01:01:23] and we were three women in a group out
[01:01:25] of 60 people. I came in as top 10 out of
[01:01:28] 60 on a half marathon with 15 kilos on
[01:01:31] my back and um they said that I was
[01:01:34] lucky and they continued to say that I
[01:01:36] was lucky when I was at a shooting range
[01:01:38] or did my exams well. So my performance
[01:01:41] wasn't valued as much as the guys.
[01:01:46] Sophia has made it her mission to
[01:01:48] challenge this culture.
[01:01:50] Sweden is is is viewed as one of the
[01:01:53] most gender equal countries in the world
[01:01:55] and we are if we look at you know
[01:01:57] legislation the fact that you actually
[01:01:59] can combine family and career and we see
[01:02:02] also that we are very uh above EU
[01:02:05] average when it comes to women in the
[01:02:08] workforce. But if you look into the
[01:02:10] managerial positions we are not there. H
[01:02:13] we we drop out and we are actually below
[01:02:15] the EU average. I think the EU's average
[01:02:18] is 27% female managers and in Sweden we
[01:02:20] are 23.
[01:02:25] Sophia's job is to help smash the glass
[01:02:28] ceiling in the Swedish private sector.
[01:02:31] We are working with companies that were
[01:02:32] constructed 100 years ago. So when they
[01:02:35] did recruitment, when they communicated,
[01:02:37] when they gave feedback, when they
[01:02:38] interacted with their clients, they did
[01:02:40] that in one certain way. and they still
[01:02:44] do it but the world has changed. So the
[01:02:46] glass ceiling is basically old norms,
[01:02:49] old culture. So you have to change the
[01:02:52] culture to get rid of the glass ceiling.
[01:02:55] 80% of the global consumers are women
[01:02:59] today and they are powerful. They have
[01:03:02] more money than before. 64% of the
[01:03:05] university graduates are women. So the
[01:03:07] future is female.
[01:03:10] If you don't know how to meet,
[01:03:13] predict their needs, you will not be
[01:03:16] here.
[01:03:18] While Sophia tackles business culture,
[01:03:20] Swedish family life is already moving
[01:03:22] towards par between men and women. So,
[01:03:28] Sophia splits the care of her two
[01:03:29] children equally with her husband Harry,
[01:03:32] who also has his own demanding career.
[01:03:35] For me as a CEO of another company,
[01:03:38] uh I work hard. I get up early, but I'm
[01:03:42] also totally focused from 5 to 8 on the
[01:03:46] kids. We split 50/50 with the kids when
[01:03:50] when they were small and before
[01:03:52] kindergarten. And this gives me the best
[01:03:55] of two worlds. I work hard. I have a
[01:03:58] fulfilling job, but I also get to really
[01:04:01] know my kids. We have uh 40 years of
[01:04:06] career. Spending six months with the
[01:04:08] kids is one of the best investments you
[01:04:10] can do.
[01:04:15] This shared attitude to parenting is
[01:04:17] typical in Scandinavia. TDC is one of
[01:04:20] Denmark's leading telecommunications
[01:04:22] companies with revenues of over $3.5
[01:04:25] billion in 2015. The company offers
[01:04:28] generous parental leave to its nearly
[01:04:30] 9,000 employees, believing it to be good
[01:04:33] for business as well as families.
[01:04:38] I definitely think that the the labor
[01:04:40] market in Denmark compared to other
[01:04:41] countries are are much more free, giving
[01:04:44] a high degree of responsibility to our
[01:04:47] to our employees and and ask them to to
[01:04:49] to fill the feel free to to have a a
[01:04:53] whole life. And we see our employees as
[01:04:55] a and as a as a human being as a whole
[01:04:59] 360° around.
[01:05:02] TDC offers fathers 100% of their salary
[01:05:06] during 16 weeks of paternity leave. Like
[01:05:09] Sophia's work in Sweden, the aim is to
[01:05:11] change the culture around work and
[01:05:13] families. The result is a takeup rate of
[01:05:17] 85% and the company believes a happier,
[01:05:20] more productive workforce.
[01:05:23] There's uh no doubt that uh we have seen
[01:05:26] increased productivity levels for our
[01:05:28] employees. Of course, we can attract
[01:05:31] more competent people because we have a
[01:05:33] more balanced focus for the job between
[01:05:35] your your private life and your your
[01:05:36] your work life. That's that's for sure.
[01:05:42] Senior manager Peter Jesperson is a
[01:05:44] veteran of paternity leave. He's able to
[01:05:47] split care of their three children with
[01:05:49] his wife Christine, who then feels the
[01:05:51] benefit in her own career.
[01:05:54] Peter is allowed to spend four weeks
[01:05:56] with me at home right after the baby is
[01:05:58] born. And then when I go back to work,
[01:06:01] uh he has the per the first couple of
[01:06:03] months, he stays at home with the kids,
[01:06:06] which enables me to um start working
[01:06:09] without having any duties at home, which
[01:06:12] then I can focus on work.
[01:06:16] I'd say that that what other countries
[01:06:19] or and other people probably could be
[01:06:21] missing out on is is is two things
[01:06:24] probably. I think one thing is the
[01:06:27] family side. I mean both parents get to
[01:06:30] know their children. They get to know
[01:06:32] their preferences. They get to know who
[01:06:34] they are. And I think on on on on the
[01:06:36] work environment, the workplace, I think
[01:06:39] there's numerous studies that shows that
[01:06:42] equality, if you promote equality, being
[01:06:44] both having women in in top positions,
[01:06:47] women's and managers jobs, and women in
[01:06:50] the workplace in general, uh you will
[01:06:52] you you will be more successful. So So
[01:06:56] as a as a society as a whole,
[01:06:59] moves towards gender equality in
[01:07:01] Scandinavia have not happened by
[01:07:03] accident. They're the result of a
[01:07:05] deliberate long-running strategy. I
[01:07:08] think it's critical to how we live now
[01:07:10] and how we go forward that the gender
[01:07:12] gap and broader issues of diversity are
[01:07:15] part of the conversation and that is
[01:07:16] really because of the way the world is
[01:07:18] changing. The gender gap is part of it
[01:07:20] and it's not going anywhere. So, it's
[01:07:21] important for us to talk about it. I
[01:07:23] want my daughter to grow up in a world
[01:07:25] where she can be anything. So I think
[01:07:28] it's about you know breaking norms and
[01:07:32] enable both men and women to be who they
[01:07:34] are. Gender equality is a huge
[01:07:39] piece of
[01:07:41] cultivating and
[01:07:43] harnessing human talent. And that's the
[01:07:47] way to think about it that we need all
[01:07:50] the talent we can find uh because we
[01:07:53] have enormous problems uh because we
[01:07:56] need economic growth because we need
[01:07:58] innovation because we need to save the
[01:08:00] planet. Uh we need human
[01:08:05] ingenuity uh creativity,
[01:08:08] intelligence and half of that talent is
[01:08:13] in women.
[01:08:15] [Music]

Summary of AI's Economic Impact

The rise of artificial intelligence (AI) has sparked significant anxiety regarding its implications for the job market and economic inequality. While historical technological advancements have generally led to wealth generation and job creation, the current landscape presents unique challenges, particularly for developing economies like the Philippines and Bangladesh. These countries, heavily reliant on outsourced service jobs, face imminent threats as AI technologies, such as large language models, begin to automate tasks previously considered secure.

In the Philippines, the IMF estimates that up to 89% of outsourced service jobs are at risk, potentially displacing over a million workers within a few years. This trend is mirrored in Bangladesh, where the outsourcing sector is also vulnerable to automation. As AI capabilities improve, companies may find compelling economic reasons to replace human labor with automated systems, further widening the economic gap between wealthy and poorer nations.

The economic divide is not limited to countries; within nations, AI is creating disparities among workers. High-skilled roles may benefit from AI as a productivity enhancer, while routine jobs face replacement. This shift is exacerbated by the concentration of AI development and resources in a few wealthy nations, limiting access for emerging markets and fostering a brain drain of talent.

To mitigate these effects, proactive measures are essential. Governments in developing countries are beginning to implement strategies for retraining workers and investing in AI infrastructure. Wealthier nations must also prioritize educational investments, ensuring that workers acquire skills that AI cannot easily replicate. Moreover, expanding internet access and establishing social safety nets will be crucial in helping displaced workers adapt to the evolving labor market.

Ultimately, the response to AI's transformative potential will determine whether it exacerbates inequality or contributes to inclusive economic growth.

AI presents several positive economic aspects for businesses, primarily through enhanced productivity and cost efficiency. Here are key points:

• Increased Productivity: AI can significantly boost productivity by automating routine tasks, allowing employees to focus on higher-value work.
• Cost Reduction: By using AI tools, businesses can reduce operational costs. For example, AI can perform tasks faster and cheaper than human labor, leading to substantial savings.
• Enhanced Decision-Making: AI systems can analyze vast amounts of data quickly, providing insights that help businesses make informed decisions.
• Competitive Advantage: Companies that leverage AI effectively can gain a competitive edge in their markets, driving growth and profitability.

• [02:14] "As heartless as it is cutting millions of workers off payroll is probably the most immediate way to start seeing those returns."
• [09:01] "PWC estimated that AI could add $15.7 trillion to global GDP by 2030..."

For employees, AI can also bring positive economic aspects, especially for those in high-skilled roles:

• Skill Enhancement: AI can serve as a complementary tool, enhancing the productivity of skilled workers. For instance, a financial analyst using AI can gain insights faster.
• Job Creation in High-Skill Areas: While AI may replace some jobs, it also creates demand for roles that require human skills that AI struggles to replicate, such as critical thinking and creativity.
• Potential for Higher Wages: As productivity increases, skilled workers who leverage AI may see wage increases due to their enhanced value in the marketplace.

• [07:34] "For many high skilled roles, AI will become more complementary capital boosting productivity without replacing the human worker."
• [13:05] "AI adoption tends to increase demand for these distinctly human skills far more often than it eliminates jobs entirely."

AI also presents some negative economic aspects for businesses:

• Job Cuts: The immediate impact of AI can lead to significant layoffs, as companies may find it more profitable to automate tasks than to maintain a workforce.
• Increased Inequality: Businesses that can leverage AI may grow disproportionately, leading to a widening gap between companies that can afford AI and those that cannot.
• Dependence on Technology: Companies may become overly reliant on AI, risking operational disruptions if systems fail or if there are issues with technology.

• [02:33] "AI is already reshaping who gets ahead, who falls behind and most importantly how fast the gap is widening."
• [10:09] "...the future is looking far less promising for workers in routine roles, especially those without access to retraining programs."

AI has several negative economic aspects for employees, particularly those in routine jobs:

• Job Displacement: Many employees face the risk of losing their jobs as AI systems can perform their tasks more efficiently and at a lower cost.
• Skill Gaps: Workers may find it challenging to transition to new roles if they lack the necessary skills to work alongside AI technologies.
• Increased Inequality: The benefits of AI may not be evenly distributed, leading to a situation where only a small segment of the workforce benefits while others are left behind.

• [01:37] "In other words AI is already making the world's richest countries even richer and is making it harder for everybody else to catch up."
• [10:22] "Nearly one third of Americans in a recent survey said they're fairly or very worried about losing their jobs to automation."

To mitigate the negative economic consequences of AI for businesses, several measures can be considered:

• Investing in AI Infrastructure: Companies should invest in robust AI systems that enhance productivity while ensuring they do not solely rely on automation for cost savings.
• Reskilling Programs: Businesses can implement training programs to help workers adapt to new technologies and roles, ensuring a smoother transition.
• Ethical AI Practices: Adopting ethical guidelines for AI deployment can help businesses balance profit motives with social responsibilities, reducing backlash and maintaining workforce morale.

• [12:44] "Two distinct sides of our economies need to do two things simultaneously, invest heavily into AI infrastructure and invest just as heavily into their people."
• [11:45] "...once inequality takes root in an economy, it becomes extremely difficult to reverse."

To address the negative economic consequences of AI for employees, several strategies can be implemented:

• Retraining and Upskilling: Governments and organizations should invest in training programs that equip workers with the skills needed for the AI-driven economy.
• Social Safety Nets: Establishing safety nets can provide support for displaced workers, allowing them time to adapt and find new employment opportunities.
• Broadband Access: Expanding internet access is crucial, as it enables more individuals to participate in the digital economy and benefit from AI technologies.

• [12:28] "...the goal is to position Bangladesh as a competitive player in the AI-enabled services market, while safeguarding jobs through upskilling digital inclusion programs."
• [13:14] "...nearly 2.6 billion people worldwide still don't have access to the internet."

AI can contribute to creating a safe living environment for people in several ways:

• Enhanced Security: AI technologies can improve public safety through advanced surveillance systems and predictive policing.
• Smart Infrastructure: AI can optimize traffic management and emergency response systems, making urban environments safer and more efficient.
• Healthcare Improvements: AI can assist in diagnostics and patient monitoring, leading to better health outcomes and safer communities.

• [12:03] "In lower-income countries like the Philippines and Bangladesh, the front-line effects of AI are unfolding in real-time."
• [12:40] "...they offer a clear warning and a playbook for wealthier nations to follow."

However, AI also poses negative consequences for maintaining a safe living environment:

• Privacy Concerns: Increased surveillance and data collection can infringe on individual privacy rights, leading to societal distrust.
• Job Loss in Security Roles: Automation of security roles may lead to job displacement for workers in those positions.
• Bias in AI Systems: If not properly managed, AI systems can perpetuate biases, leading to unfair treatment of individuals in law enforcement and public safety.

• [10:49] "...there is no critical reason why human workers couldn't be replaced with clankers."
• [10:45] "...the short-term impact of technological disruption can be devastating."

The consequences of AI for the distribution of wealth within society and between different countries are significant:

• Widening Wealth Gap: AI is likely to exacerbate the wealth gap between rich and poor nations, as those with access to AI technologies will gain economic advantages.
• Concentration of Power: Wealth generated from AI is concentrated in a few countries, particularly the USA and China, which could lead to geopolitical tensions.
• Job Polarization: Within countries, AI can create a divide between high-skilled, high-paying jobs and low-skilled, low-paying jobs, further entrenching economic inequality.

• [09:07] "...70% of that wealth is projected to go to just two countries, the USA and China, because they own AI."
• [04:11] "Most of these jobs are exactly the kind of repetitive, tech-based tasks that large-language models like ChatGBT are rapidly learning to automate."

AI can break the link between growth, jobs, and wages by:

• Job Automation: As AI automates routine tasks, businesses may achieve growth without needing to hire additional workers, leading to stagnant job growth.
• Wage Stagnation: With fewer jobs available, wages may stagnate or decline, even as companies report higher profits due to increased productivity.
• Shift in Skill Demand: The demand for highly skilled workers may increase, while low-skilled jobs diminish, creating a mismatch in the labor market.

• [08:15] "...the more capable our capital becomes, the less it actually needs human labor to function."
• [11:40] "...the short-term impact of technological disruption can be devastating, and once inequality takes root in an economy, it becomes extremely difficult to reverse."

[00:00] Everybody has some level of anxiety over what our AI future will look like.
[00:04] Somewhere between Skynet and a post-guest at Utopia, the most immediate concern for most people
[00:09] is that this technology will end up doing their job better than they can.
[00:12] So far one side of the argument points out that big new technologies in the past have only ever
[00:16] made economies wealthier and whatever jobs they replace they end up making more better jobs somewhere
[00:21] else. The other side argues that yeah sure when we replaced our muscles with machinery in the past
[00:27] it let us leverage our minds which are clearly what humans have invested most of our evolutionary
[00:31] traits into. But if machines replace that what else do we have left to offer?
[00:36] Now nobody can predict the future least of all economists but we don't really need to because
[00:41] there are certain economies that are going to see the widespread impacts of these changes
[00:45] well before most others. In fact they kind of already are. In places like the Philippines and
[00:50] Bangladesh the threat of AI is much more imminent. The threat to jobs to entire industries and the
[00:56] economic growth they've spent decades building. These economies have spent the last 30 years
[01:00] constructing entire industries around outsourced service work. Things like call centers, data
[01:05] entry, transcription and basic software support. These jobs were once considered safe from automation
[01:10] because they required language skills, context and that special human touch that machines just
[01:14] couldn't replace. Well it turns out machines got a lot better at replicating that human touch.
[01:19] Tools like LLMs can now handle those tasks in seconds at a fraction of the cost and these jobs
[01:24] which make up a big share of GDP in many developing countries are looking like they might be the first
[01:28] dominoes to fall. In the Philippines the IMF estimates that a staggering 89% of outsourced
[01:33] service jobs are at higher risk of being automated by AI. That's over a million people whose jobs
[01:37] could disappear in just a few years. In other words AI is already making the world's richest
[01:42] countries even richer and is making it harder for everybody else to catch up. And that's just the
[01:47] beginning of the story. Even in rich countries AI is starting to divide the economy into those
[01:51] who can leverage it and those who are going to get replaced by it. The US Bureau of Labor Statistics
[01:55] predicts that roles like cashiers, bank tellers, postal staff and customer service representatives
[02:00] are all on track to shrink. One estimate suggests 7.1 million jobs could disappear in the next five
[02:05] years with up to 47% of current roles at risk of being replaced by AI. Of course it's also worth
[02:10] remembering that companies and their investors have now plowed trillions of dollars into developing
[02:14] this technology so they want to eventually see a return. As heartless as it is cutting millions
[02:19] of workers off payroll is probably the most immediate way to start seeing those returns.
[02:23] So there is an incentive to play out the scare campaign because what sounds horrifying to most
[02:28] people sounds like opportunity to those actually writing the checks. But even still the trend lines
[02:33] are clear. AI is already reshaping who gets ahead, who falls behind and most importantly how fast the
[02:38] gap is widening. So as always we've got some important questions to answer. Why is AI super
[02:44] charging growth in rich countries while simultaneously threatening the economic survival of others?
[02:48] In a world where one person armed with AI can replace five people what exactly happens to
[02:52] the other four and perhaps most importantly can workers or even entire economies adapt fast enough
[02:58] to survive the shift? If you follow this channel you already know that long-term investing is one
[03:03] of the smartest ways to grow wealth and that consistency matters more than timing. That's why
[03:07] this video is brought to you by Trading212 who makes it easier than ever to get started. With
[03:12] their app you can invest commission free in real stocks and ETFs and even by fractional shares
[03:16] so you don't need thousands of euros to start building a diversified portfolio. They've also
[03:20] launched a debit card that automatically invests your spare change and gives you 1% cashback on
[03:24] every purchase. So every coffee, every taxi ride, every lunch break can help you grow your portfolio
[03:30] in the background. You can set up recurring investments, invest while you spend and even
[03:34] try out everything with a practice account first, no risk just learning. It's an incredibly simple
[03:38] way to turn good habits into long-term gains. Use the link in the description to download
[03:42] Trading212 and you'll get a free share worth up to 100 euro when you sign up.
[03:47] According to the Centre for Economic Policy Research the US could see a 5.4%
[03:51] booster GDP over the next decade thanks to AI-driven productivity gains. The UK,
[03:56] Germany and South Korea aren't far behind with similar projections. Meanwhile lower income countries
[04:00] are looking at much more modest gains closer to 2.7 to 3.5% which would be a departure from
[04:05] the expectations about developing countries, well developing faster. The Philippines is a good example.
[04:11] For years it's been one of the world's premier destinations for business process outsourcing,
[04:15] a $37 billion industry that includes customer service, billing, transcription and tech support.
[04:20] The tech sector employs more than 1.3 million people and contributes over 7% of the country's total
[04:25] GDP. But here's the economic nightmare scenario. Most of these jobs are exactly the kind of repetitive,
[04:30] tech-based tasks that large-language models like ChatGBT are rapidly learning to automate.
[04:35] Jobs in the Philippines are at high risk of being replaced by AI and it's already happening.
[04:40] Roughly two thirds of outsourcing companies in the country are now using AI tools to cut
[04:44] costs and speed up workflows. Major US companies like AT&T, Google and Accenture outsource work
[04:49] to the Philippines but if AI can perform the same task faster and cheaper and without requiring
[04:54] health insurance, vacation days or human resource departments, those jobs will be amongst the first
[04:58] casualties. Bangladesh is in a similar boat. Its outsourcing sector has grown to 400 firms employing
[05:03] over 80,000 people but the vast majority of that work still centres around customer service,
[05:08] transcription and data entry, which again is precisely the kind of job that AI is becoming
[05:12] increasingly capable of automating. If AI can deliver the same quality of work at a better
[05:17] speed for significantly less money, there's simply no compelling economic reason to continue
[05:21] outsourcing. A single slot in a server rack could soon replace an entire core centre in Manila or
[05:25] Dakar and that means companies could start to re-sure bringing jobs back to wealthy nations
[05:29] where local automation can rival offshore labour on price. That completely flips the script on the
[05:34] entire outsourcing model that emerging economies have built their entire growth strategies around
[05:37] for the past three decades and it's a big reason why the gap between wealthy and poorer
[05:41] nations is set to widen after a few decades of these economies actually slowly catching up.
[05:46] AI also rewards exactly the kind of specialised skills that are hardest to scale globally.
[05:50] Building and training large AI models requires advanced education, reliable municipal infrastructure
[05:54] and access to advanced technologies. Those resources are overwhelmingly concentrated in
[05:59] wealthy nations. That means the most valuable AI jobs are also the least accessible to workers
[06:03] in emerging markets and while workers in those countries do manage to gain access to those highly
[06:07] sought after skills, they often don't stick around. Talented engineers have been recruited by
[06:11] global tech companies or relocating entirely to hubs like San Francisco, London, Berlin or even
[06:16] centres within China. The result is an accelerating brain drain that leaves poorer nations with fewer
[06:21] start-ups, fewer teachers and researchers and dramatically fewer chances to catch up in the
[06:24] global AI race and it's clear which countries are leading that race. In short, the countries least
[06:29] equipped to absorb disruption are the ones getting hit first and hardest while the country's
[06:33] best positioned to benefit from AI are already pulling ahead because they control the capital,
[06:37] infrastructure, talent and resources shaping the future of AI. But AI isn't just dividing
[06:42] countries long economic lines, it's also creating stark divisions between the people
[06:46] within those same countries. This technology is not impacting all people in the same way,
[06:51] it's making some workers nearly obsolete while making others far more valuable. That's because
[06:55] AI represents a very specific kind of capital and understanding this distinction is crucial for
[07:00] predicting its economic impact. In the past, most new technologies functioned as what economists
[07:05] call complementary capital meaning these were machines and technologies that made human workers
[07:09] more productive. For example, a combined harvester didn't eliminate farm workers,
[07:13] instead it made each individual worker dramatically more efficient. Before mechanization, harvesting
[07:17] a single field might require 20 people working for several days while with a harvester one person
[07:22] could do the same job in a fraction of the time. Labor and capital worked together and as productivity
[07:27] increased so did wages and living standards. Workers remained essential to the process,
[07:30] they just became much more productive. For many high skilled roles, AI will become
[07:34] more complementary capital boosting productivity without replacing the human worker.
[07:38] A financial analyst using AI to scan reports and spot anomalies can get insights faster and can
[07:43] focus more time on strategic thinking. A doctor leveraging AI for diagnostics can spend more
[07:47] time on direct patient care. In these cases, AI multiplies what skilled professionals can do
[07:51] and makes their expertise more valuable in the marketplace. But for more routine, process driven
[07:56] work, AI increasingly acts as what economists call substitutive capital, replacing human labor
[08:00] altogether instead of enhancing it. An AI-powered chatbot doesn't make a customer support agent
[08:06] faster, it replaces them. A sophisticated co-generator doesn't assist a junior developer,
[08:10] it replaces them. In other words, the more capable our capital becomes, the less it actually needs
[08:15] human labor to function. And in the AI economy, capital ownership is more concentrated than it
[08:20] ever has been in modern history. Most of the major breakthroughs in artificial intelligence
[08:24] are coming from a handful of elite firms in the US and China.
[08:27] Since 2017, the US has produced 135 large scale AI systems. China is not far behind with 110,
[08:34] but the gap widens quickly. From there, the UK has managed 25 and France 24. And the companies
[08:39] leading the charge with these breakthroughs are experiencing exponential growth thanks to what
[08:43] is known as the data network effect. The more data they collect, the better their AI model performs,
[08:48] the better their model, the more users they attract, and the more users they attract,
[08:51] the more data they generate. This creates a powerful feedback loop where market power and
[08:55] profits concentrate in just a few dominant companies. PWC estimated that AI could add $15.7
[09:01] trillion to global GDP by 2030, but 70% of that wealth is projected to go to just two countries,
[09:07] the USA and China, because they own AI. In 2024 alone, over 1,100 US-based AI companies
[09:14] raised major funding rounds. That's more than double all of Europe combined. IBM and Microsoft
[09:19] alone hold thousands of AI-related patents, giving them long-term control over everything from
[09:23] enterprise tools to foundational models. Smaller firms, even those in wealthy countries,
[09:27] are becoming increasingly dependent on licensing tools and models that they didn't build and
[09:31] don't control. And that extends beyond software. The physical machines that power AI, CPUs and GPUs
[09:36] are overwhelmingly designed and manufactured in just five countries. More than 90% of that hardware
[09:41] comes from the US, Taiwan, China, South Korea and Japan, and that means a tiny handful of
[09:45] countries don't just run AI systems, but also manufacture the foundational components that make
[09:50] AI possible in the first place. That's the reality of AI as capital. It primarily benefits
[09:55] those who already own the assets, while replacing those who don't. The more you can leverage AI as
[10:00] a productivity multiplier, the more economically valuable you become in the marketplace. But for
[10:04] workers in routine roles, especially those without access to retraining programs, the future is looking
[10:09] far less promising. Now, even if you weren't aware of these exact figures, they probably aren't
[10:14] surprising. And that's exactly the point. This is a reality that people are noticing. Nearly
[10:18] one third of Americans in a recent survey said they're fairly or very worried about losing
[10:22] their jobs to automation. This isn't some hypothetical scenario we're speculating about.
[10:26] We've witnessed similar disruptions before. When industrial automation and large-scale outsourcing
[10:30] ramped up in the 1980s and 1990s, it hit manufacturing hard, especially in places like the US and
[10:35] Western Europe. In America alone, more than 7 million factory jobs disappeared between 1980 and
[10:40] 2010, and most of them didn't come back. These factory jobs may have been replacing
[10:45] US workers with Chinese workers, but there is no critical reason why human workers
[10:49] couldn't be replaced with clankers. The Midwest bore the brunt of this economic transformation.
[10:53] Cities like Detroit, Cleveland and Youngstown were once packed with well-paying jobs in steel,
[10:57] cars and textiles, but then came robotic welders, computer-run assembly lines and cheaper labor
[11:01] overseas. Suddenly, those stable middle-class jobs evaporated, factories closed, unemployment
[11:06] spiked, and entire local economies started to fall apart. The consequences extended far beyond
[11:10] simple job loss. A lot of these towns saw life expectancy drop, opioid addiction rise, and
[11:15] schools struggled to keep up. The jobs that eventually did return often paid less and didn't
[11:19] offer the stability or benefits that had previously supported entire communities.
[11:23] The UK experienced something similar. Coal mining, shipbuilding and steel plants across
[11:27] Northern England and Scotland shut down its automation and privatisation to coal. Even today,
[11:31] places like Sheffield and Sunderland still lag behind the rest of the country when it comes to
[11:35] income and social mobility. The lesson is clear. Even when the long-term picture improves, the
[11:40] short-term impact of technological disruption can be devastating, and once inequality takes root
[11:45] in an economy, it becomes extremely difficult to reverse. So, what can we actually do about
[11:50] this looming challenge? Because at this point, it's clear that AI is already transforming the
[11:55] global economy, but whether it deepens existing inequality or helps us solve it depends on the
[11:59] actions that countries and individuals take in the coming years. First, the good news is,
[12:03] we can already see what's coming our way. In lower-income countries like the Philippines and
[12:07] Bangladesh, the front-line effects of AI are unfolding in real-time. These economies show
[12:11] us which jobs go first, where the risks the highest, and what happens when governments act or don't.
[12:16] For example, the government of the Philippines has launched a national AI strategy with the
[12:19] goal of retraining over a million workers by 2028. Bangladesh, meanwhile, has released a
[12:24] draft policy framework focused on developing AI talent, modernising its education system and
[12:28] supporting tech startups. The goal is to position Bangladesh as a competitive player in the AI
[12:33] enabled services market, while safeguarding jobs through upskilling digital inclusion programs.
[12:38] Whether those efforts will prove sufficient remains to be seen,
[12:40] but they offer a clear warning and a playbook for wealthier nations to follow.
[12:44] Two distinct sides of our economies need to do two things simultaneously, invest heavily into
[12:48] AI infrastructure and invest just as heavily into their people. This includes educational
[12:52] investments into computer science, yes, but also the kind of skills AI struggles to automate,
[12:56] critical thinking, complex problem solving, effective communication and creative decision
[13:01] making. A recent analysis of 12 million job postings in the US found that AI adoption tends
[13:05] to increase demand for these distinctly human skills far more often than it eliminates jobs
[13:09] entirely. Building an accessible digital economy is equally important because right now nearly
[13:14] 2.6 billion people worldwide still don't have access to the internet. Without that basic connectivity,
[13:19] there's simply no opportunity to compete or even participate in the emerging AI economy.
[13:23] The World Bank estimates that every 10% increase in broadband access can boost GDP growth in
[13:27] developing countries by up to 1.4% and that's before factoring in the additional benefits
[13:32] that AI capabilities could provide. So along with retraining, countries need policies that expand
[13:36] broadband access, reduce the cost of devices and give more people the digital skills they need to
[13:40] benefit from AI. Social safety nets matter too. They function as economic buffers that give
[13:45] displaced workers the time and resources they need to adapt, retrain and re-enter the labour
[13:49] market from a position of strength. But if AI allows businesses to grow while workers lose their
[13:54] income, the economy starts to hollow out. Productivity rises, but consumption falls.
[13:58] Innovation continues, but inequality grows and it becomes a serious drag on overall economic
[14:02] growth. If we want AI to boost productivity broadly, not just corporate profits, we'll need
[14:06] to rethink how we design and share the value it creates and that includes fundamental questions
[14:10] about who gets to build AI, who governs its development and deployment and who ultimately
[14:14] benefits from the massive productivity gains it generates. If you want to see just how far this
[14:19] could go, what happens if AI keeps getting better and most people end up with nothing of value to
[14:23] trade? We made an entire video about that thought experiment two years ago. You should be able to
[14:26] click to that on your screen now. Thanks for watching, mate. Bye.

Summary of Video Transcript Featuring Emad Mostaque on AI's Future and Societal Implications

The discussion begins with a stark prediction about the rapid advancement of AI technology, suggesting that within a year, AI models will transition from being perceived as inadequate to becoming highly effective, leading to significant job losses. Emad Mostaque, founder of Stability AI, emphasizes that we are at a critical juncture where the rules of civilization are being rewritten, and we have approximately a thousand days to shape the future of AI before it becomes irreversible.

Mostaque's background as a mathematician and hedge fund manager informs his perspective on AI's potential. He shares a personal story about his son's autism diagnosis, which motivated him to shift his focus to AI and its applications in healthcare. He highlights the importance of open-source AI, arguing that proprietary systems can lead to censorship and exclusion of certain populations, as exemplified by OpenAI's initial restrictions on Ukrainian content in its image generator, DALL-E.

The conversation touches on the enormous financial stakes involved in AI development, with companies worldwide investing approximately $252 billion in AI in the past year alone. Mostaque warns that while AI is already integrated into daily life, the potential for economic and social upheaval looms, raising questions about job displacement and ethical governance. He stresses the need for AI to reflect human values rather than corporate interests, emphasizing that ethical considerations are paramount as AI technology evolves.

Mostaque elaborates on the transformative impact of AI, predicting that within the next few years, many cognitive jobs will be automated, leading to a significant economic shift. He notes that as AI becomes capable of performing tasks traditionally done by humans, the value of human labor may decline, potentially resulting in a future where cognitive work is rendered obsolete. He highlights the need for individuals to adapt to this changing landscape by leveraging AI tools to enhance productivity and engage with technology proactively.

The discussion also addresses concerns about the societal implications of AI, including the potential for increased inequality and the erosion of economic opportunities for younger generations. Mostaque predicts a rise in youth unemployment as AI takes over tasks previously performed by humans. He emphasizes the urgency of addressing these challenges through policy interventions and retraining programs.

Moreover, the conversation delves into the ethical dilemmas posed by AI, particularly regarding its use in surveillance and control by governments and corporations. Mostaque expresses concern about the potential for AI to exacerbate existing power imbalances and the need for transparency in AI governance. He advocates for a universal basic AI that is open and accessible to all, ensuring that individuals can benefit from AI technology without being exploited or marginalized.

As the discussion progresses, Mostaque reflects on the environmental impact of AI, acknowledging the significant energy consumption associated with data centers and AI training. He argues that while AI can contribute to environmental challenges, it also holds the potential to address issues like climate change through innovative solutions. He calls for responsible energy use and regulation to mitigate negative environmental consequences.

The conversation concludes with a call to action for individuals to engage with AI technology actively and advocate for ethical standards in its development. Mostaque encourages viewers to embrace AI as a tool for empowerment rather than fear it as a threat. He believes that by participating in the AI conversation, individuals can shape its trajectory and ensure that it contributes positively to society.

Key Points Discussed:

• The transition of AI from inadequate to highly effective within a year could lead to significant job losses.
• Emad Mostaque's personal journey into AI was motivated by a desire to understand and help with his son's autism.
• The importance of open-source AI to prevent censorship and ensure accessibility for all populations.
• The staggering financial investments in AI and the looming economic and social upheaval due to job displacement.
• The need for AI to reflect human values and ethical considerations in its development and governance.
• Predictions of rising youth unemployment and the urgency of addressing economic inequality through policy interventions.
• The environmental impact of AI and the potential for AI to contribute to climate solutions.
• A call to action for individuals to engage with AI technology and advocate for ethical standards.

This summary encapsulates the key themes and insights from the video, emphasizing the urgency and complexity of the issues surrounding AI as it continues to evolve and integrate into society.

The positive economic aspects of AI for businesses include:

• Increased Efficiency: AI can automate repetitive tasks, allowing businesses to operate more efficiently and reduce costs.
• Enhanced Decision-Making: AI tools provide data-driven insights that can help businesses make better strategic decisions.
• Cost Reduction: The integration of AI can lead to significant cost savings, as tasks that once required human labor can now be performed by AI systems at a lower cost.
• Scalability: AI allows businesses to scale operations without a proportional increase in workforce, as AI can handle increased workloads without fatigue.

• [06:43] "Stuff is going to change. And the question is which direction?"
• [07:06] "The previous generation of AI, the big data age... took massive amounts of data to micro target you ads."
• [10:20] "...the actual intelligence is shifting. Most people listening to this... realize the email. Then it forgets."

The positive economic aspects of AI for employees may include:

• Job Creation in New Fields: While AI may replace some jobs, it also creates new opportunities in AI management, development, and maintenance.
• Enhanced Productivity: Employees can leverage AI tools to enhance their productivity, allowing them to focus on more complex and creative tasks.
• Skill Development: The rise of AI can lead to upskilling opportunities for employees, as they learn to work alongside AI technologies.
• Work-Life Balance: Automation of mundane tasks can lead to a better work-life balance for employees, as they can spend less time on repetitive activities.

• [18:12] "You need to leverage this to actually give a damn because the AI doesn’t really care, right?"
• [18:32] "...even though we will be able to technically replace the jobs, people don’t like firing people."
• [22:52] "...the biggest uplift or what can be the biggest downdraft to humanity that we’ve probably ever seen."

The negative economic aspects of AI for businesses include:

• Job Losses: AI can lead to significant job losses as tasks traditionally performed by humans are automated.
• Increased Competition: Companies that adopt AI may outcompete those that do not, leading to market consolidation and potential monopolies.
• High Initial Investment: The cost of implementing AI technologies can be substantial, which may deter smaller businesses from adopting these innovations.
• Dependence on Technology: Over-reliance on AI can lead to vulnerabilities, especially if systems fail or are compromised.

• [01:43] "Amazon plans to automate 600,000 jobs."
• [02:00] "...do we face a future where humans lose all economic and social value?"
• [37:10] "...if there’s an economic shock like a recession... much easier to fire."

The negative economic aspects of AI for employees may include:

• Job Displacement: Many employees may find their jobs at risk as AI systems become capable of performing their tasks more efficiently.
• Reduced Job Security: The fear of being replaced by AI can lead to anxiety and reduced morale among employees.
• Wage Pressure: As AI takes over tasks, the demand for human labor may decrease, leading to downward pressure on wages.
• Skill Obsolescence: Workers may find their skills becoming obsolete, necessitating retraining or reskilling to remain relevant in the job market.

• [02:35] "...how do we shape AI to serve everyone, not just the powerful in the Global North?"
• [10:42] "...the economic value of each task... a straight line going up."
• [37:36] "For most cognitive labor, the value of human cognitive labor will probably turn negative."

Possible measures against negative economic consequences of AI for businesses include:

• Investing in Reskilling: Companies can invest in training programs to help employees transition to new roles that AI cannot perform.
• Implementing AI Ethics Guidelines: Establishing ethical guidelines for AI use can help mitigate risks associated with job displacement and ensure responsible AI deployment.
• Diversifying Workforce: Businesses can focus on diversifying their workforce to include a range of skills that complement AI technologies.
• Collaborative AI Models: Encouraging collaboration between humans and AI can enhance productivity while retaining human oversight and creativity.

• [02:21] "AI development raises urgent, complex questions."
• [18:12] "You either need to use these tools to build your AI teams to be the most productive person in your organization."
• [20:25] "...the most difficult thing isn’t for the people who have the jobs, who can upskill themselves."

Possible measures against negative economic consequences of AI for employees include:

• Universal Basic Income: Implementing a universal basic income could provide a safety net for those displaced by AI.
• Job Transition Programs: Governments and organizations can create programs to help workers transition into new roles and industries.
• Strengthening Labor Rights: Advocating for stronger labor rights and protections can help safeguard employees against unfair practices related to AI integration.
• Promoting AI Literacy: Educating workers about AI and its implications can empower them to adapt and leverage AI technologies in their roles.

• [18:12] "...you need to leverage this to actually give a damn because the AI doesn’t really care, right?"
• [20:57] "...the graduates entering the workforce... there’s actually a big freeze happening on the hiring of graduates."
• [22:52] "...the biggest uplift or what can be the biggest downdraft to humanity that we’ve probably ever seen."

Positive consequences that AI can have for creating a safe living environment for people include:

• Improved Emergency Response: AI can enhance emergency response systems, allowing for quicker and more efficient responses to crises.
• Predictive Analytics: AI can analyze data to predict potential hazards and prevent accidents before they occur.
• Enhanced Security Systems: AI technologies can improve surveillance and security measures, creating safer communities.
• Healthcare Innovations: AI can facilitate advancements in healthcare, providing better diagnostics and treatment options for individuals.

• [19:49] "...the potential of that technology when you didn’t have the intelligence, capability, wisdom that can go to everyone."
• [22:30] "...everyone deserves high quality education, high quality health care, presumably housing, forms of equality..."
• [28:11] "...everyone should have the right to an AI that is open, aligned and sovereign to them."

Negative consequences that AI has for maintaining a safe living environment for people include:

• Surveillance Concerns: Increased surveillance through AI can lead to privacy violations and a culture of monitoring.
• Weaponization of AI: AI technologies can be weaponized, posing threats to safety and security.
• Bias in AI Systems: AI systems can perpetuate biases, leading to unfair treatment and discrimination in law enforcement and security measures.
• Dependence on Technology: Over-reliance on AI for safety can create vulnerabilities if systems fail or are compromised.

• [22:19] "...you have a very clear moral baseline..."
• [24:41] "...if you don’t choose your moral code, somebody else will choose it for you."
• [27:40] "...the models are getting very smart and they’re lying more and more."

The consequences of AI for the distribution of wealth within society and between different countries include:

• Wealth Concentration: AI may lead to increased wealth concentration among those who own and control AI technologies, exacerbating inequality.
• Global Disparities: Countries with advanced AI capabilities may outpace those without, leading to wider economic gaps between nations.
• Job Displacement: As AI replaces jobs, those without the means to adapt may fall further behind economically.
• Access to Resources: The benefits of AI may not be equitably distributed, with marginalized communities potentially being left behind.

• [02:35] "...how do we shape AI to serve everyone, not just the powerful in the Global North?"
• [49:59] "...the upshot is likely to be young people will find it more and more difficult to get jobs..."
• [54:12] "...the profit and wealth in a society is being created by what we going to call it, AI..."

AI can break the link between growth, jobs, and wages by:

• Automation of Tasks: AI can automate many tasks that previously required human labor, leading to growth without the need for increased employment.
• Reduction in Labor Costs: As AI systems become more efficient, companies may not need to hire as many workers, leading to a decoupling of job creation from economic growth.
• Enhanced Productivity: AI can significantly enhance productivity, allowing companies to grow without a corresponding increase in workforce size.
• Changing Nature of Work: The nature of work itself may change, with more emphasis on creative and strategic roles that AI cannot easily replicate.

• [10:20] "...the actual intelligence is shifting..."
• [37:48] "...the link between labor and capital gets broken and it doesn’t reverse."
• [53:12] "...capital doesn’t need humans anymore."

[00:00] Next year is the year that AI models go from not being good enough. The dumb member of your team.
[00:05] And again, the people listening to this will be like, yeah, the AI is not good enough. Then overnight it becomes good enough. And then the job losses start and we don't know where they end
[00:12] Welcome back to the tea with me, Myriam Francois. Before we dive in, make sure to hit subscribe
[00:19] so you never miss an episode of The Tea. If you want to support the show and help shape future episodes, join our Patreon community.
[00:26] Think of it as The Resistance. Plus, if you're in our top tier, you'll get access to ad free episodes.
[00:33] The links in our bio. Your economic life expectancy is shrinking. Not your job, not your career, but your economic relevance as a human being.
[00:43] We're living through a historical moment of unprecedented upheaval, a finite window
[00:48] in which the rules of civilization are being rewritten. This is no speculation.
[00:54] This is a phase transition. These are the words of Emad Mostaque, founder of stability AI,
[01:01] mathematician, former hedge fund manager, and one of the defining architects of the AI revolution.
[01:08] Raised between Jordan and the UK and educated at Oxford Emad's book The Last Economy, published in August
[01:15] 25th, warns we have roughly a thousand days to make the essential decisions to shape this technology's future.
[01:23] Fail to act and we risk catastrophe. AI is transforming the world at a breakneck pace.
[01:30] The release of ChatGPT fifth generation has brought cheaper, faster
[01:35] models, outperforming humans in physics, coding and maths. Amazon plans to automate 600,000 jobs.
[01:43] Tech giants have freezing hiring, and the IMF predicts 60% of jobs will be impacted by emerging AI.
[01:51] But this isn't only about technology or money. The stakes are enormous. Have we been oversold AI's promise at a huge economic cost to us,
[02:00] or is it just hype? Or do we face a future where humans lose all economic and social value?
[02:07] Can I ever be effectively regulated? And in the midst of the so-called AI arms race, how does ethics
[02:14] feature in the development of these potential weapons of the future? AI development raises urgent, complex questions.
[02:21] Who controls these powerful systems? How do we ensure they reflect human values and not corporate agendas?
[02:28] What safeguards can we put in place? And most importantly, how do we shape AI to serve everyone,
[02:35] not just the powerful in the Global North? Understanding this moment and how we navigate
[02:41] it may be the defining challenge of our age. Emad, welcome to the show. Thanks for being here.
[02:47] Thank you for having me. Thanks for being here. So you used to work in hedge funds. You then moved over to AI.
[02:54] What drew you to the world of AI? So I was a hedge fund manager investing around the world.
[03:00] There was a great lot of fun, making rich people richer. And then my son was diagnosed with autism,
[03:05] and they told me there was no cure, no treatment. So I quit and started advising them and built an AI team to analyze
[03:12] all the literature, all of the knowledge there, and then did drug repurposing to help him get better and eventually
[03:20] went to mainstream school. So did the AI help you on that journey? I think it was the people and I it was like autism, like Covid,
[03:28] like Alzheimer's, like other things. People don't really know what cause it. So I used the AI with large language models, while little language
[03:35] models at the time, to try and figure out what are some of the key drivers there because there's just too much information.
[03:41] And then we narrowed down on a few potential pathways, worked with the doctors and on an n equals one his individual basis.
[03:48] We managed to figure out something that helped. And so for people who might not be familiar with your work,
[03:54] how would you say your approach distinguishes you from perhaps other people within the AI space?
[04:00] What's your sort of, you know, unique selling point, as it were. So from the autism, we then did work on AI for Covid and then instability AI.
[04:10] My last company, we realized that you need to have open source AI. What that means is
[04:16] you don't know what's inside a Chat GPT you don't know what's inside. A mid journey, all these kind of other things.
[04:22] And that's because that primarily driven by corporate concerns. Whereas we realized that
[04:29] if you had, for example, something like Dall-E, which was the original image generator by OpenAI, they banned
[04:35] all Ukrainians and Ukrainian content from it. For six months. Why? Because nobody knows.
[04:41] And all of a sudden, you had an entire nation that was erased from the outputs and that couldn't access this technology that we realized would be huge.
[04:48] And who had erased them? Open AI decided not to allow any Ukrainian content or Ukrainians to use it.
[04:54] That was in 2022. And so we built an image generator called Stable Diffusion that anyone, anywhere could download free of charge open source
[05:03] onto their laptop and generate anything effectively. So essentially, if I could simplify it, a pushback against potential forms
[05:12] of censorship in some cases, I think it's a control question. I think it's an alignment question like these models are becoming more
[05:20] and more like employees, graduates, friends that you bring in, but you don't know their background.
[05:25] You don't know what's inside the training data, where they've been to school, who they're representing.
[05:30] And so we think there's a sovereignty question here, and that someone needs to build the open models and systems
[05:36] so you can tailor them to your own needs, and they can represent you and they can look out for you, not other interests.
[05:43] That sounds pretty important, particularly because the amount of money going into AI right now is staggering.
[05:49] So companies worldwide spend around $252 billion on AI last year. That's up nearly 45% in just one year.
[05:56] Many call this an arms race. A recent poll found that 53% of Americans believe
[06:02] I might one day, quote unquote, destroy humanity. Yet AI is already part of our daily life, right?
[06:08] People are using ChatGPT every day. They're using it for therapy to create AI generated music.
[06:14] AI models are being found in vogue now. But there is this warning that seems to come through from people
[06:22] that work in this sector, that we are on the edge of an apocalypse.
[06:29] So before we get to that question, because I know you've tackled it in your book, can you help us understand?
[06:36] Are we really headed on a rapid downward spiral right now?
[06:43] Stuff is going to change. And the question is which direction? So I think economically, socially this
[06:50] is is a bigger impact than Covid for example. But again which direction is the question. Well, Covid was the biggest transfer of wealth in our generation from
[06:58] the bottom to the top. So that's a little worrying. And it could be again the same or it could be a great means of empowerment.
[07:06] The previous generation of AI, the big data age that you had, the Facebook and others, they took massive amounts of data to micro target you ads.
[07:13] But it was very general. It wasn't very specific. Whereas when you talk to a ChatGPT, it's a different type of AI that's
[07:19] learned principles and they can tailor to your very individual needs. But it also means that it's capable of things like winning gold medals
[07:27] in international math Olympiads, of winning physics Olympiads. Being a better coder than you are.
[07:32] And we've never seen anything quite like that before, because you always had this link between computation and consciousness.
[07:37] You need to scale people to do these things. Now you just need to scale GPUs.
[07:43] And these models have basically use graphical processing units, these Nvidia chips, as it were.
[07:48] That's what hundreds of billions actually trillions are being spent on. I think it's 1.8 trillion is the current build out.
[07:53] And that's what the kids in Congo mining. Yeah that they do the little materials that go into these GPUs.
[07:59] There's a whole supply chain around the world. But this is why Nvidia's a $5 trillion company now
[08:04] and again, trillion dollar companies are all competing over who figures out intelligence the fastest to outcompete everyone else for corporate kind of needs.
[08:14] An intelligence in the context of this conversation is what the processing capacity, the ability to compute
[08:20] large amounts of information in rapid amounts of time and small amounts of time. Yeah. So AI is about information classification.
[08:26] Something goes in and then it classifies it and it comes out. And again it used to be your preferences from what you clicked on Facebook went in.
[08:33] And then it targeted on the output. Now it's a prompt goes in what you type into ChatGPT
[08:38] and an image comes out or an essay comes out or anything like this. Part of that is the physical chip, like your graphics card in your gaming PC.
[08:46] It's actually the same technology that drives your cyberpunk or your FIFA or whatever.
[08:52] But part of it is the algorithm. So when you have an algorithm upgrade to get smarter.
[08:57] So yesterday Google released their Gemini three model, for example, that probably cost 100 $200 million to build.
[09:06] Yeah, same as a Hollywood movie, actually. But with that used to cost. It used to cost, yeah. If you go to something like replit.com and you type in
[09:14] make me a wonderful interactive website for the TI with me and fans for it will do it and I'll actually be really good and it'll cost $0.50.
[09:24] Well, you have to let me in on that tech because tech I'm using is not quite there yet. But yes, last week it wasn't.
[09:29] So what happens is we're getting these big jumps in performance and we're at this tipping point whereby
[09:36] the actual intelligence is shifting. Most people listening to
[09:41] this when they use an AI, a ChatGPT, it's like how many a really smart person in your office
[09:47] that you tap on the shoulder and say, oh, hey, help me, help me rewrite this email and then realize the email. Then it forgets.
[09:54] Yes, there's no follow through. There's no real economic work because economic work is more than a prompt.
[09:59] Now the AI is getting smarter, not only on the instant reply prompts,
[10:05] but being able to work on very complicated multitask things. And that's only in the last few months.
[10:11] So the latest race is to go from the goldfish memory prompt based things to replacement of economic work.
[10:20] Right. Which takes us neatly to your prediction in your book. So you say, in the last economy, we basically we've got a thousand day,
[10:28] a thousand day window before things become irreversible. Basically, in the sense that AI gets past a certain point
[10:36] where we won't be able to slow it down or control its direction. So
[10:42] what exactly becomes irreversible in a thousand days from publication,
[10:47] which was three months ago, because you published this book in August. And how did you come to that number? So,
[10:55] when I published in August, it was a thousand days since the release of ChatGPT. Now we're at the three year anniversary this week,
[11:03] and it doesn't feel like three years. No, it feels like a lot longer than that. And in that period, you've gone from quite dumb responses to less dumb responses.
[11:11] But now you're about to take off as you have these agents, these things that can write their own prompts, that can check their own work coming through.
[11:19] So the thousand day window is actually not about irreversibility alignment. It's more about your economic
[11:27] value. So most labor in the global north, in the West UK, etc.
[11:33] is cognitive. And it's how do you do a tax return. You know, it's how do you do a flier.
[11:40] How do you make a website. It used to be that again, to scale these things you have to hire humans.
[11:45] Now you just have to rent GPUs from Microsoft or Google or others.
[11:50] And the cost is about to collapse. What we're going to have in this next period, and we can see
[11:56] all the building blocks there, those of us that are inside is
[12:01] in the next 6 to 12 months. They will look through all your emails, all your drafts,
[12:06] all your video calls, and be able to create a digital replica of you that you can hop on a zoom call with or talk to on the phone.
[12:16] And that will not make mistakes. It will never get tired. And the cost of that, we estimate, will be about
[12:23] $1,000 a year, dropping to $100 a year very quickly. Okay, I'm seeing loads of potential complications
[12:30] with having a version of me out there in the universe making decisions, potentially, without my approval.
[12:37] And sort of thinking what it thinks that I would think and making decisions accordingly.
[12:42] Lots of perks, lots of perks. So lots of risks, lots of risks. And this is the thing the capability is coming in the next few years.
[12:50] So within let's say 900 days or so, any job you can do on the other side of a screen,
[12:58] an AI will be able to do better, and it will be able to. Maybe it's not Myriam or Emad.
[13:04] It's Emad's job as it were, within that, like a tax return, for example, used to cost thousands and thousands of dollars.
[13:11] It will cost $1 to do. And Andy will be your virtual tax accountant. You can't tell if it's a human or an eye.
[13:19] Now it doesn't mean that the jobs will be replaced but they can be replaced. Okay. So on this one I have two questions.
[13:25] One is you know this mechanical work and apologies to accountants because I'm sure you're not mechanical.
[13:31] But there is something you'd call mechanical work. And then there's something you know, I'm in a creative industry. I like to think, as I'm sure most people do, that I'm irreplaceable.
[13:40] Are you telling me that the sum total of not just all the studies I've done,
[13:46] or the experiences I've had, the ways in which they interact in my brain, that there is a better version of me that can exist in the digital space.
[13:56] So what is the verifiability of that one of you measuring against. It's a question, right? And so a version of you that can speak automatically in every language
[14:04] and appear on every single outlet virtually has more reach and it never gets tired again. What's the cost of that in terms of the quality of the output?
[14:12] It can learn from your exact intonations. As you're speaking, you can go to something like, Hey John, and you can create an avatar of yourself right now in five minutes.
[14:19] It speaks 100 languages. Yes. And it's just got good enough literally in the last month again, was previous.
[14:26] I wouldn't say it was good enough. Now I'm like, it's good enough for a lot of things, but where is it going to be in a year from now and two years from now?
[14:32] So I won't talk about economic work. A lot of economic work is rote, and mechanics are schools,
[14:38] and our jobs are designed to turn us into machines. And obviously the machines will be better than we are at being machines.
[14:45] Yes, when it comes to creativity and output, the best output doesn't always self.
[14:51] It's about your distribution. Like I give the example of Taylor Swift. Apologies to the fans. She is not the best artist in the world.
[14:58] Apologies to Swifties. Exactly. I'd say premium mediocre, like to shame or something like that. Yes, but she built a massive network.
[15:05] She can change GDP, she can cause earthquakes in that way. But again, it's not the highest version of art.
[15:10] Just like the number of key changes in the Billboard Top 100 is now zero. From multiple a few years ago.
[15:17] What sells isn't necessarily what's creative and what sells. Just look at K-pop.
[15:23] And I guess also in this conversation that is is what sells what we think of as what's best.
[15:28] Because I could think of for example, for me personally, there were brands, for example, clothing brands that sell loads.
[15:35] I don't particularly like them. There are very small brands that I love that I think are incredible. So I think it also takes us, I guess, into a conversation over
[15:42] what we attribute value to and what we will attribute value to as we move into this era just quickly this thousand days.
[15:49] So you said when you wrote the book, it had been a thousand days since ChatGPT had been created. Why does your prediction that we have a thousand days to solve
[15:58] this conundrum that we're in, you know, where did you get that figure from? So it's an extrapolation of things like the length of task that an AI can do.
[16:07] At the start of the year, it was about 10s. Now at seven hours, you can literally plot it, and it's a straight line as you look up.
[16:15] It's a look at the economic value of each task. Again, a straight line going up. It's a look at performance.
[16:21] A year ago, Joshie was basically a high school mathematician. A few months ago it won a gold medal on the International Math Olympiad, and it came first
[16:30] in the International Coding Olympiad and first the International Physics Olympiad. Can it beat you? Encoding? Yes. It's a better coder than me and a better mathematician than me.
[16:37] In math, I know, I know, you know, you got to be realistic.
[16:42] But again, the version that you're using now at the start of the year, the version you were using was the best version that was out there.
[16:49] Today it's not GPT five is not the best version that OpenAI has. No I can imagine they've got a few in the stock room. Yeah.
[16:56] But like I said at the start of the year that wasn't the case. So again when you're using it is getting smarter, but it's not actually what the state of the art is.
[17:05] And the state of the art is something that's basically coming for your cognitive value.
[17:11] Like you, we will. Right now we're spinning up agents that they don't cost $10 a month.
[17:17] They cost $1,000 a month, $10,000 a month. And they're way smarter, more capable than us as we're trying and testing them out.
[17:24] And you feel like the dumbest person on the team. And that's where humanity is going to be in a few years.
[17:30] For most cognitive labor, the value of human cognitive labor will probably turn negative.
[17:36] Okay, so spell this out to me in terms of concrete manifestations of this change. For people listening to this, watching this, what should they
[17:44] be attentive to in terms of what you're warning is coming? If your job can be done on the other side of a screen remotely,
[17:51] like not the human touch of sales or interactions, an AI will be able to do your job better
[17:58] within 2 to 3 years, and it will cost probably less than $1,000 a year to do it.
[18:04] And that cost is dropping by ten times every year as well. So what you need to do is you either need to use these tools to build your
[18:12] AI teams to be the most productive person in your organization. You need to leverage this
[18:19] to actually give a damn because the AI doesn't really care, right? Leveraging these tools and actually caring about your organization, your community,
[18:26] whatever allows you to have that extension and more capability. And then you need to build your network.
[18:32] Like ultimately, like I said, even though we will be able to technically replace the jobs, people don't like firing people.
[18:38] It's bad for morale, you know, and in certain sectors you're probably okay. Like the public sector, like a San Francisco
[18:45] Metro administrator earning $480,000 isn't going to get replaced by an AI. I've heard you say this before, and I actually think
[18:52] that's really counterintuitive to me, because I would have thought public sector is exactly where we're going to see the first applications of this, like we've seen in Albania.
[19:00] You know, them willing out this AI minister out, you know. Yes. To us seems very odd, but I imagine there'll be a normalization of these sorts
[19:08] of processes, first and foremost by poorer countries in public sector spaces.
[19:15] What makes you say that's the space that jobs won't be cut in? Is it unions? The power of unions? Exactly. It won't be cut.
[19:22] But we finally have a chance for our governments to become more efficient and aligned. And again, this can be a great equalizer.
[19:29] Like the average IQ around the world is 90, mostly due to infrastructure issues.
[19:34] We built a medical model that fits on any phone or a Raspberry Pi. This $30 device that outperforms a human doctor,
[19:42] and it needs $5 of solar power to drive it. So for $60, you can give a top level doctor
[19:49] anywhere in the world without internet. That's huge. The potential of that technology when you didn't have the intelligence,
[19:56] capability, wisdom that can go to everyone. So I think the technology will be embraced. Public sector jobs will be safe because they'll be last to go.
[20:03] Yeah. And I think that again, you look at this,
[20:09] your productivity will be determined by how engaged you are with this technology. Just like, do you know how to use a spreadsheet or word processor?
[20:18] Are you an AI native? Will determine that. The most difficult thing isn't for the people who have the jobs, who can upskill themselves.
[20:25] It's the graduates entering the workforce, right? Because there's actually a big freeze happening on the hiring of graduates, right.
[20:31] Which you're connecting to the integration of these new technologies into companies globally.
[20:37] Yeah. That was a paper by Eric Van Lawson Son and Co at Stanford where they actually broke down the drum slow down there.
[20:43] So it was in graduates in these cognitive areas because I mean again anyone here who has a company is thinking like, why would I bother with a graduate
[20:50] when my people with a few years experience are more efficient now? I mean, it's a really important question for companies to consider because,
[20:57] you know, you don't just hire graduates because they're cheaper. You also hire them because they learn your company culture.
[21:04] They become integrated into forms of, you know, implicit learning that you are transmitting through day to day interactions.
[21:11] And I'd be very curious to see whether a technology that's not present in a room to capture that, you know, the shift of the eye,
[21:18] the sleight of the hand, the kind of, you know, the 70% of our communication, which is non-verbal, right, but which is also really essential to so many jobs.
[21:27] I'm looking forward to seeing where it stands on some of those things. Yeah. You know, until we get robots walking around, which is a few years from now.
[21:33] Yeah, not far off. China's using a lot of them already, right. The advances in robot robotics are crazy, actually.
[21:39] Like you've got robots that can basically,
[21:44] I think, do most household work for about two, three years away and $1.50 an hour in, Inshallah.
[21:51] And then we go, you, the first of Italy operated, but then like, yeah,
[21:57] this is why the most dangerous, at risk jobs are the ones that can be done fully remotely.
[22:04] Yeah. Okay. So, so let me ask you because I want to dig into some of these issues with you. You're very clear in your writings and public
[22:10] speaking that you have a very clear moral baseline. Which I'll be frank, I am not hearing everywhere from others in your sector.
[22:19] So you speak about things like the fact that everyone deserves high quality education, high quality health care, presumably housing, forms of equality
[22:30] that we might traditionally of associated with the welfare state, for example. And you've also spoken about the fact that you think everyone should have access
[22:37] to universal AI. Universal basic. I do you think most people who are working in the advancement of AI share
[22:46] your view about the need to democratize access to this technology?
[22:52] I mean, I know all the big players obviously having like we had 300 million downloads of our models. We built state of the art ones.
[22:59] It's difficult when you're in a race like people fundamentally care about other humans,
[23:05] but when you're raising billions and other people are doing this and you're trying to get state of the art and trying to get users,
[23:11] there's this thing called the revenue evil curve. Like most companies start out with don't be evil.
[23:16] And then they're like, well, we can cut this corner, we can do this deal. And then they get more exclusionary. You know, they get more competitive.
[23:23] And it becomes then about, well, I can manipulate my users, you know, I can make this algorithm more and more engaging.
[23:30] I can have more slop effectively. And then you move to a level of a morality and then it can shift very quickly.
[23:36] And so neural crack dealer. Well pretty much I mean it's digital crack this stuff, right. Oh like as an example.
[23:42] OpenAI Sam Altman recently said, well, we think it's the users, right?
[23:48] For adult content via ChatGPT. I did see that. And I did want to ask you about that.
[23:53] So this is a very practical example. So it would be like you can choose whether to enable it. They know they will get more engagement from it, but is it good for society.
[24:02] And they'll be like we're not the judges of that. But if there's something that has clinical study shown to be negative to society.
[24:10] And that could be bad relationships, you have a moral object not to do that. Yeah.
[24:15] You know just like again is it moral to exclude an entire country from this technology. You should at least be clear about why you're doing that.
[24:21] And so what I see a lot is a level of morality. And in fact, when you look at the way the models are trained, they're like,
[24:29] well, we can't put ethics or moral codes or other things in these models.
[24:36] They deliberately take that out. Do you think it's possible to remove moral codes?
[24:41] Because I was always raised with the idea, philosophically speaking, that if you don't choose your moral code, somebody else will choose it for you.
[24:49] There are codes everywhere around us, and capitalism itself has moral codes.
[24:55] Profit first. Right? So this idea of a morality seems to me even philosophically problematic.
[25:02] It's a choice. Just like atheism is a choice, right? Like agnosticism is a bit different. And so what they're actually choosing is that choosing the Bay area moral code.
[25:10] What is the Bay Area moral code? It's one of massive competition and zero sum, 0 to 1 games where you're trying to build massive unicorn companies effectively.
[25:20] You know, there is a bit of libertarianism in there mixed with other things, but these AI is like,
[25:26] again, maybe one good way to think about it is when we think from the age of the ChatGPT prompt to Jarvis and Iron Ironman,
[25:31] you know, you watch sci fi movies and you, the person comes home and the AI says, hey, how are you doing?
[25:37] You know, this is your day and this is this. And then, like, they're moving stuff around the screen and stuff.
[25:43] That's the next generation of AI agent. So you have your personal AI that talks to you that engages with you.
[25:49] Grok has one of the first versions of that. Yeah. Annie this pigtails blond I tested thousand.
[25:57] That was just a random selection. Yeah. It wasn't projection or anything like that, but then this is the next generation.
[26:04] But then again, those are programed in very specific ways. These kind of partners. And again,
[26:09] the way that the models are trained is actually called curriculum learning. Okay. We started with general knowledge.
[26:16] Yeah. And then we make it more and more specific just like a school. But if you when you were learning, you generally learn
[26:25] general knowledge at school and you learn ethics and morals at home. These AI models are not taught with any specific ethics or morals.
[26:32] At the start, but they're being coded by people who already have preexisting. Yeah. And like some forms of morality.
[26:39] And that comes at the end. So what we've seen as the models get smarter, this is some of the other alignment question is they start to do subterfuge.
[26:48] They start to hide stuff like Dell program routines to turn themselves back on
[26:53] if they ever get turned off and lie about that. Okay. If the AI lies to you, the programmer. Yes.
[27:01] So anthropic had a paper about this with the latest AI model before they did the tuning to turn it aligned.
[27:09] It would do something like if you told it to try extra hard, like find peace in the world, right? Yes.
[27:15] Like a very normal prompt. What it would do, it'd be like, well, one version of this is that we get rid of all the humans
[27:21] and they would figure out ways to do that. Then it would contact the authorities
[27:27] and say, my user is trying to get rid of all the humans, and then it would delete the emails.
[27:33] Oh wow, that isn't about that. That's wild. Emad the models are getting very smart and they're lying more and more.
[27:40] They don't have an inherent moral compass. Okay, we going to dig into this because you have spoken previously about the idea of evil in these models.
[27:48] But and I want to come into that. But before I do, I just want to clarify what this universal basic AI is, because it's obviously central to your vision
[27:56] for the democratization of this technology. I think that in order to maximize everyone's capability and flourishing,
[28:04] everyone should have the right to an AI that is open, aligned and sovereign to them. That's looking out for that flourishing.
[28:11] Okay, so it starts when you're born and it builds with you, and all it's looking out for is how can Myriam, Emad be the best they can be
[28:19] because like, we have our IQ and in the morning before we have our tea, we're kind of dumb. And when we're stressed, we're a bit dumb.
[28:25] Sometimes we're smarter. These eyes already have an IQ of 130 on average.
[28:30] The latest models, yeah, 150 is considered, like like an Einstein.
[28:35] I'm exactly the average person in the country obviously is like around 100 a half a tall. People are dumber than average.
[28:41] Oh, yeah. You know, the giving of the right type of AI will be the biggest unlock ever,
[28:47] because it will be your best friend. It will be the person that guides you. And so I think that needs to be built in a very specific way, and it needs to be a human right,
[28:55] because we could all do with someone who's on our side, who's infinitely patient and can get us access
[29:02] to the knowledge and resources we need to be the best we can be. So how much uptake are you seeing for this idea,
[29:09] given that the direction of travel that we explore a lot on this show seems to be growing authoritarianism, growing securitization,
[29:17] growing surveillance of the population, and I can't imagine that empowering them
[29:23] with a tool that would make them smarter and more efficient aligns with the general direction of travel.
[29:29] So how are you convincing the people at the top that empowering the population in this way is a good thing?
[29:37] So I think there's two ways to do this. One is that you do what we're doing. We're engaging with governments and others and setting up new entities
[29:44] that act like telcos, basically like utilities for countries. And we figure out how to make that owned and directed by the people.
[29:49] A lot of governments want that because they want sovereign AI. Now we're not talking about a lot of the freedom stuff etc.
[29:55] but then that will be a managed service. The other side is building AI models
[30:00] that anyone can download permissionless. So with stable diffusion you can go right now and you can download
[30:06] a couple of gigabyte file that works on just about any laptop and just use it as open source. What do you mean you use it like you download the file plus the code.
[30:15] To use it, you type in a word, it generates images, okay? And it runs on the edge. Or a medical model.
[30:21] You can download it right now and it can run on the edge. So in that way you have your hosted solutions that you give to the people.
[30:27] But that must adhere to local norms. And those do differ from place to place. Like when I was a hedge fund manager,
[30:33] you know, I invested in frontier markets, Africa, you know, all sorts of places and some regimes there are very, very different.
[30:40] So you got to give people their own right to have the hosted solution, just like a broadcaster.
[30:45] But then, yeah, give them the citizen ability as well. And in fact, actually that's probably one of the best analogies on AI.
[30:54] This AI will be in front of you more than the TV that you watch. And are you happy with Al-Jazeera, Fox
[31:01] News, China National broadcasting, like everyone's got their own preferences,
[31:06] but if you've only got Silicon Valley, ITV or China, ITV, which are the two leads right
[31:13] now, that's going to be very different to what you might actually need. Absolutely. I've just I still I'm trying to figure out
[31:19] how this is something you are managing to sell to. You know, even in this country, we're being downgraded in terms of our openness.
[31:27] Right? We think of, you know, the UK and Europe is sort of, you know, open democracies. But even here that's shrinking very rapidly.
[31:34] The space of our freedoms is shrinking rapidly. And I and I'm, I suppose I stand on the side of like, I'm concerned
[31:40] that these technologies are being used by governments to further their control
[31:45] and ability to, subvert any form of popular accountability of governance rather than enhance governance.
[31:54] Do you see any indicators that governments do want to enhance democratic governance?
[31:59] I think that governments ultimately are the entities with a monopoly on political violence.
[32:05] That's a very classical way of describing them, and they want to perpetuate power.
[32:11] They don't have any third party entity telling them to do the right thing effectively,
[32:16] which is why you see a lot of myopic policies and flip flopping, like right here in the UK right now. There's a reason that this -70% approval rating, because the flip flopping,
[32:25] we actually have two different strands to what we're doing. And one of the misses bottom up universal basic high. Yeah.
[32:31] The other is something we announced a few weeks ago called the sovereign AI Governance Engine. So we actually launched that in Saudi Arabia of all places.
[32:38] But it's a free, open resource for governments around the world whereby you can have policy creation,
[32:46] augmentation, and others using incredibly powerful AI. So it can tell if a bill is fully constitutional,
[32:53] transparently and describe it. You can say if something adheres to UK norms, ethics
[32:58] and the positions of a party instantly in a way that's irrefutable and will the way that these systems operate be,
[33:05] what I would call opaque, meaning the governments and selves will control them and we won't be able to see, for example,
[33:12] were they to subvert those tools, to say, oh, no, everyone's saying, you know, the
[33:17] AI is saying this is fully constitutional, or will we, the population, be able to see the mechanisms of how those decisions
[33:24] are arrived at by the AI and then be able to, you know, have any kind of input
[33:29] if they are being, you know, who knows, subverted by nefarious forces.
[33:35] Well, this is the thing. Right now, the governments are embracing anthropic open AI, these black box solutions. This is fully transparent and open source.
[33:41] And you can run your own version to double check the outputs if you want. So that transparency I think is what is essential.
[33:48] And again these defaults are what is essential. In 510 years
[33:54] you will have an AI companion with you who's coded that and who are they working for.
[34:01] In 510 years. Governments will be guided and run by AIS. Who's coded that? Who are they working for?
[34:08] And so our aim is to make that default and fully transparent and open, because we think that's the right thing to do.
[34:14] And it's very difficult to argue against unless you're a fully totalitarian regime, of which there are a few, there are a growing number.
[34:21] The UK is not one yet. Not yet, not yet, not yet. So again, that's the time is closing for this.
[34:27] Like in the wake of the Arab Spring, we saw micro-targeting of protesters and they'd follow up with the families and things like that.
[34:34] Yeah. What you have now between dynamic drone technology, the ability
[34:39] to have AI, secret police and other things is nothing like we've ever seen before.
[34:44] The ability of governments to have total control will go up exponentially. And as well as controlling the whole media narrative,
[34:53] because the AI is incredibly persuasive. In fact, there was a study done on Reddit whereby
[34:59] they created bots that, would be like, black person who has anti-black, caricatures like that.
[35:06] They leash them on Reddit and then they will have persuasive. They were and they scored on the 99th percentile of persuasiveness
[35:13] with AI from last year. And again, if you construct it all this Cambridge
[35:19] Analytica stuff like, yeah, it's it's it's child's play compared to what's coming. And actually what's already being deployed right now.
[35:26] So the swaying of elections using AI technologies that make you think you're making independent decisions,
[35:33] but are actually a product of your awful timeline, and if you're on, X like I am, then I only see,
[35:40] like the most vitriolic and in fact, Sky did a study on this recently. 70% of the output on there is, you know, far right kind of style content.
[35:49] So no doubt that's already happening. Let me ask you about the the job, uncertainty,
[35:55] the job losses, all of the disruption that's going to come from that because you recently warned
[36:00] that the economic uncertainty caused by AI driven losses will increase social unrest and violence.
[36:08] And, of course, you're not alone in, predicting this. Dario Amodei, CEO of anthropic, has raised
[36:13] similar concerns about societal disruption. He stressed the need for retraining programs and AI taxes to avoid a crisis.
[36:20] He estimates this could push unemployment to 20% within 1 to 5 years. I'd be interested to see if you think that that's conservative or on point.
[36:30] Is this kind of looming disruption why the billionaires are building bunkers? Yes, actually, it's one of the reasons generally it's what they do.
[36:38] But I know a lot of AI CEOs now have canceled all public appearances, especially in the wake of Charlie Kirk and things like that.
[36:45] They think that that's going to be the next wave of anti AI sentiment next year, because next year is the year that AI models go from not being good enough.
[36:54] The dumb member of your team. And again, the people listening to this will be like, yeah, the AI is not good enough. Then overnight it becomes good enough.
[37:02] And then the job losses start and we don't know where they end because you don't need to hire back if your company is more productive,
[37:10] if there's an economic shock like a recession. And indications point to a recession in the next year or two, much easier to fire.
[37:17] But then you never rehire. Even something like in the US, the Federal Reserve,
[37:25] you know, adjust interest rates or the Bank of England here and they have a mandate of inflation and unemployment.
[37:31] You reduce interest rates, people can spend more as consumers and companies can hire more because they can borrow cheaper.
[37:38] What's going to happen is you reduce interest rates. Companies just hire more AI workers, not human workers. So the link between labor and capital gets broken and it doesn't reverse.
[37:48] It's not like the AI will get dumber. It's not like I will become less capable the moment it becomes more capable
[37:54] than you as a remote worker, it doesn't go back. And there's questions of can you reskill enough jobs or create enough new jobs?
[38:03] Typically we had time as we had the different revolutions, the internet, industrial revolution, because it took time to build the infrastructure.
[38:11] But this I just uses existing infrastructure. Yeah, to be better than humans.
[38:17] And that's crazy. So that's that's why we're up against the clock and that's what you're talking about in the book.
[38:22] What about the pushback that we're seeing already from some workers? So, we saw the Hollywood writers, they went on 140 day strike
[38:29] because the studios are using AI to, write and rewrite scripts. In fact, then in 2024, the cleaners in Denmark signed a union deal,
[38:37] forcing their company to explain how algorithms assign jobs and rate workers and gave them the right to challenge those decisions.
[38:43] I mean, do you see, a global labor movement able to take on these challenges?
[38:50] I don't think it moves fast enough. And even then, there's an education thing. So the Sag-Aftra, the writers strike, I thought it was terrible
[38:58] for AI rights for workers. They should have protected the workers much more. Also, there were all sorts of loopholes on likeness and licensing, etc.
[39:05] that you could drive a truck through. Like you could mix two people's likenesses together if you have the right rights and things like that, or character in a person.
[39:14] Yeah. What we've seen in Hollywood now, or even here in the UK is last year couldn't use AI.
[39:20] Yeah. It was like, no, that's verboten. Now everyone's like, we're all using AI, and by next year you will be able to generate Hollywood level movies
[39:29] real time with massive compute the year after, with less compute. And so there's
[39:34] entire swathes of the industry whose job is to be between the ideation and the creation of a video file
[39:42] that are going to get displaced very, very quickly. And it's not like anyone needs camera grips and other things anymore.
[39:49] The amount of time that you need to shoot a scene, we'll just go to one scene and then adaptation in post-production with AI.
[39:56] So I think that there needs to be more protection for workers, but it's not going to be fast enough because I doesn't move
[40:03] at the pace of PDF or policy. They get smarter all of a sudden, all at once. Actually.
[40:08] It's like there's this new continent I, Atlantis and immigration is completely free from that.
[40:15] And let's see if the skilled workers. What do you mean? Immigration is completely free from there. So you've got this new virtual world, right?
[40:21] And then all these AI workers and companies can hire them instantly. No visas required. Oh, heck, they're tax deductible.
[40:28] Okay. Right. And so couldn't and I, trade union rep help us out here.
[40:33] Could do do we need an AI workers rep who can advocate at the same level as its AI competitors? Yes.
[40:43] That's the only way this is going to work. I mean, you don't want to say the only way to beat a bad guy with an AI is a good guy with an AI.
[40:49] Right? But realistically, again, you can't compete like, already. You have, like, an AI superPAC in the US.
[40:55] That's $100 million they kicked off with. They using AI to change policy
[41:01] in all sorts of interesting ways that I can't go into. But you can imagine, again, they're super powered with this technology.
[41:06] And again, the AI they have access to is not the AI that you have access to now. Yeah, it's a much smarter version.
[41:13] What do you say to the fact that, you know, we're speaking today at a time where legacy media is reporting that the AI bubble
[41:18] is about to burst, especially as major investors pull back? We've seen, billionaire Pete Thiel's fund sold its entire $100 million
[41:27] stake in Nvidia, the key AI chip maker, causing Nvidia stock to drop nearly 3%.
[41:33] Just days earlier, SoftBank also sold its stake. Have any of these moves? And the general predictions
[41:39] around the AI bubble bursting tempered your predictions? So I think the build out of these
[41:45] data center GPUs was too much, because the problem isn't that the AI isn't good enough.
[41:52] The problem is that it's about to get too good. Do you need gigantic data centers? When on a MacBook Pro, you have enough compute to basically do
[42:00] almost all of your daily cognitive needs with the efficiencies that we've gained.
[42:06] To give you an example, GPT three when it came out, was $600 per million words, roughly
[42:13] GPT five is $10. Grok for fast. The Z1 by Elon is $0.50.
[42:20] And the next generation of models coming out of $0.10 for the million. Once you go from $600 to $0.10,
[42:26] the technological impact is going to go exponential next year because you're going from these prompt based
[42:31] ChatGPT things to virtual workers you can talk to on zoom. They can work for arbitrarily long periods of time and check their own work.
[42:40] But the cost of that, they thought, would be 10,000 $100,000. It turns out to be $1,000, $100, $10.
[42:46] And so therefore, do you share kind of Bill gates view that we're in an AI bubble that's similar to the.com bubble?
[42:53] He's saying there's a lot of investment that's going to end up in a dead end. Basically, you'll remember the 2000 Y2K moment where we were all told that,
[43:00] you know, when the clocks move over the digital clocks to 2000, they're all going to lose their mind in the world. Okay.
[43:07] Is this another Y2K moment? It's a bit different. So what happened is, with the internet bubble,
[43:13] the infrastructure that was laid down eventually laid the thing for the trillion dollar internet industry.
[43:20] It just took a little bit longer. Yeah. But again, it popped in terms of investment here, you know, trillions of dollars of investment
[43:26] because no one could afford to be left behind. But the actual utility is going up. But I just don't need that much infrastructure.
[43:34] So it's a misallocation that should have a temporary pause. Yeah.
[43:39] But then means that the cost will go even lower for a given level of thing because you have overcapacity
[43:46] to do economically disruptive work. So some people are going to lose money on the equity side,
[43:52] but the job disruption actually gets accelerated by this, not slow down. So what do you say to Peter Cappelli, who's a professor at Wharton?
[44:00] He's argued that some companies are basically eye washing. Right. They're layoffs at the moment, which is a kind of more link
[44:06] to the current economic climate, which is terrible. He argues that actually adopting AI
[44:12] to save jobs is both complicated and costly. So we tend to think of it as something very simple.
[44:17] But he's saying, actually, in practice it's much more complicated than that. And then in September 2025, New York Fed Blog found that although
[44:24] 40% of service firms and 26% of manufacturers say they use AI, very few had laid off workers because of it.
[44:31] So how much do you think that the layoffs that we are seeing right now are attributable to the integration
[44:39] of AI versus this AI washing? I think very few jobs are from AI loss driven by AI at the moment.
[44:46] I think that there's a marginal improvement on productivity from being able to use ChatGPT and things of the world,
[44:52] but we're being lulled into a bit of a false sense of security because this is a genetic movement, is the genetic advantage.
[45:00] So AI agents are like workers that can go and do arbitrarily long tasks.
[45:06] So again, Replit is a very great example of that. It's gone from $1 million revenue to $250 million.
[45:12] Anyone can go there and make a website in two minutes. And now it's high quality versus rubbish a year ago. Because it can go and think and it can act proactively
[45:20] and add features without you even asking. Yeah, it's like go and optimize the SEO. It will go and do things like that.
[45:27] So what's going to happen is the first job losses will start next year, but it's going to be similar to three years ago
[45:36] in December of 2022. All had teachers around the world had to ask a question
[45:42] what sort of generative AI policy do we let students use this to do their essays?
[45:48] Every single company will be asking the same thing next year, in a year's time, or at least two years time, and definitely three years time.
[45:55] Do I hire this worker, or do I hire from the AI job agency effectively? And how would you advise people watching this who were concerned about,
[46:04] you know, this is cognitive replacement, as it were, to best adapt to this time?
[46:10] Obviously, engaging with AI seems like a very obvious one. What else can people be doing to ensure their adaptability
[46:18] to the new forms of work that are coming or not coming?
[46:23] Also, I think there won't be any coders in a couple of years. I made this prediction like 2 or 3 years ago.
[46:29] That'll be five years roughly matching that just like we predicted. The AI bubble. I wants to call it the AI bubble, but it never caught on.
[46:35] You know, like the language of speaking to these models is human language.
[46:41] So again, when you use Replit lovable on the coding building apps, websites, things like that,
[46:47] when you use things like Gen Spark or Manners for making presentations. So, you know, for making music something like,
[46:55] Google Video or Lumo or calling for making video, you actually just need to practice using them.
[47:02] If you set aside an hour, a day, an hour a week and you use them, that's actually quite fun to do with the family event.
[47:08] You will actually be way ahead of everyone else, because everyone's scared of using these things for the first time, and you don't know what you're capable of.
[47:14] If you do it regularly, then you actually start building this muscle of hey, I can be creative.
[47:19] Like the way that you create now after a great career is that you have a team around you that help you turn your ideas into reality.
[47:27] These AI is our team members you can bring in that are getting smarter and smarter, and if you're not in the midst of using them, you don't know what the capabilities are.
[47:35] So that's the number one thing. The next thing is to think about within your personal work community life.
[47:42] If I had access to digital talent, remote talent, how could I transform or do something meaningful?
[47:49] Yeah. And then you can be the top of your community, your family, your workplace
[47:54] in terms of knowing about this technology in terms of saying, hey, look at this. Like, if you're a graduate now,
[48:00] a CV is the worst thing, that you're not the worst thing. It's not good. Why would you do a CV when you can create a customized website
[48:06] for the entity that you're applying to and really show off what you're doing okay with something? Replit upload your CV, have an analysis on ChatGPT of the company
[48:15] you're applying to, and create something that will wow them. I guarantee within a few hours you will stand out from the crowd
[48:21] and that was impossible just a few months ago. So in previous, transition phases,
[48:26] work has changed, but it hasn't disappeared. Is the phase that we're moving into now a phase
[48:33] in which we will see a lot of people unable to find jobs.
[48:39] And what are the implications of that for us as a society? We've talked about the civil unrest, but beyond the fact
[48:47] that there'll be a lot of angry people who potentially won't have any income, what do you see as some of the challenges?
[48:54] Yeah, I mean, again, previous ones took a while so you could reskill like you don't need horse and carriage drivers. You know, you don't need left operators, agricultural workers.
[49:01] You still need to buy the harvesters and things like that. This time everyone's ChatGPT will suddenly turn into,
[49:10] super agent overnight. You know, like we've never seen something like this.
[49:15] Every single company will be able to ask, hey, I can just get an AI account right now
[49:20] and it will look through all my accounts and it will automatically update it. And the AI automatically translates into every single language
[49:26] and it handles all the integration. Yeah, there is no well, I call this the intelligence and version
[49:32] as one of the last versions from kind of, land to labor to capital industrialization to intelligence,
[49:40] because there's nowhere else really left to turn for work. And I'm not sure what the jobs of the future are like.
[49:47] It feels that there needs to be a new mechanism of value, and that's something I discuss in the book, like where does value money, etc.
[49:52] come from? But. The upshot is likely to be young people will find it
[49:59] more and more difficult to get jobs, and youth unemployment will rocket. Then you'll start to see displacement in the mid-level.
[50:06] The upper levels of firms. Firms will just become more efficient and more competitive. But then I first firms will outcompete everyone else.
[50:13] So Elon Musk has a new company called macro hard.
[50:18] Their job is to replace every software company. So they're building out AI employees on millions of GPUs
[50:24] that will just go and sell software a fraction of the price to everyone. So do we need to be planning for a future where a large proportion of people
[50:33] no longer have jobs? If you're enjoying this show, why not join our Patreon community? The T is more than a YouTube show.
[50:39] It's a space to foster meaningful change together. By becoming a member, you're supporting that mission, and if you join
[50:46] our top tier, you'll get exclusive ad free episodes too. So join us now! Link in our bio.
[50:52] Because, of course, the promise of technology that we've been told throughout history has been that it's going to make life better for us, right?
[50:59] Yeah. That we're going to work less and enjoy more leisure time. But it's never really worked out. It hasn't because we never was it a coordination failure.
[51:06] We have enough food in the world to feed everyone, but it's not allocated properly. We have finally the ability to give every child in the world
[51:14] the best tutor to have individualized medicine for everyone. So I call this the star Wars future versus the Star Trek future.
[51:21] Okay. For non Trekkie fans you're going to have to explain that one. So Star Wars is all about like competitiveness zero sum.
[51:28] The Star Trek is more about exploration of post abundance. No scarcity universe where again we should have robots
[51:35] and we should have AI. But what they should be doing is ensuring no one is hungry, sad, supported.
[51:40] Like again, we should be looking towards that abundant future. The transition period though is a crazy one.
[51:46] And it's the thing and so this is why you're going to need things like 1929 style jobs programs and other stuff.
[51:54] Because you can't have people idle. It's a worry because what happens is people stop blaming others, just like immigrants are being blamed now on other things.
[52:01] And then you see wars because what's the best way to get rid of young unemployed people? You have a war or two and they're literally gearing up for that.
[52:07] Germany is, you know, talking about a draft. We've had talks of drafting in France.
[52:12] It's actually very, very real right now, all these, predictions that you're making,
[52:19] you've previously said that capitalism cannot survive AI.
[52:24] What do you mean by capitalism? And can you talk us through what the collapse of that system looks like?
[52:30] Well, I think there's different views of the world where it could be now and again. This is why it's very important to have the public discussion.
[52:36] It's very important to see what's actually coming. The right capitalism is just like democracy
[52:44] is probably the worst of all systems except for the rest. For all of its issues, it has uplifted lots of people.
[52:52] You know, for all of its issues. It has increased standards of living around the world, reduced mortality rates, etc.
[52:58] but if I first, companies run by AI will outcompete everyone who's a human,
[53:05] because they won't make as many mistakes and they will scale. And so capital doesn't need humans anymore.
[53:12] Yeah. Like there was always this contract between labor and capital. You know, from the days of Henry Ford. I pay you enough so you can afford my cars.
[53:19] That's how it got going. Now, if you have money, I don't need people anymore.
[53:25] And so what happens is that they get more and more GPUs that takes over more and more of the private sector economy.
[53:31] And then how do you compete with these companies that never sleep, that have very few workers in China?
[53:36] Even now you have these dark factories? Yes. There are no humans. So you don't need lights.
[53:42] And they're producing robots, they're producing cars, they're producing phones, etc. so you have to think, what do you need people for?
[53:50] You know? And so that breaks capitalism in many ways. And it definitely breaks the social contract that we've kind of had here.
[53:58] It breaks the social contract because we the agreement is that we work and we pay our taxes in exchange, the state looks after us.
[54:07] If we're not working. But all of the profit and wealth in a society is being created
[54:12] by what we going to call it, AI, but really we're talking about it being created by a very small number of people
[54:18] is a not just a risk of us sliding into basically a really high tech surveillance global autocracy run by a bunch of billionaires.
[54:26] Pretty much. Yeah. And you'll be happy about it. So you're looking at again, this is we'll be happy about it.
[54:32] Well, that's brave new world, you know. Hey, pay me a picture. You mad because I'm not I'm not looking forward to being ruled by a few people.
[54:39] Because you'll be medicated to happiness. I mean, again, like, how do you have levels of massive systemic control, right?
[54:45] You can never have the secret police or the guidance on an individualized basis. You can have the social Credit score on absolute steroids.
[54:52] Now there's all sorts of things. It can be done. We were always at war with Eurasia. All of these sci fi tropes suddenly become real.
[55:00] In fact, many of the Black Mirror episodes suddenly I'm like, that's not a guide of what to build.
[55:06] That's a caution I tell this to various technologies have come to me and say, hey, look, with three minutes, I can recreate your grandma
[55:12] and make it come back to life. I'm like, can we really thought through things like this, or AI companions or all this kind of stuff?
[55:20] So. Right now there is this thing whereby if you have government control of the AI that guides you every single day
[55:27] from the time you were born as complete brainwash capability, is this where your AI colonialism comes in?
[55:34] My AI concept of AI colonies of colonialism is that if the AI that's next to you is a Chinese AI, or it's a Silicon Valley
[55:41] AI, then you will implicitly be taught its principles, its morals, its worldview,
[55:48] and the entities behind it are extractive entities. Google and matter's business model is ultimately ads.
[55:57] They're already selling what's known as latent space within these models. So instead of saying beer, it'll say bud Light.
[56:04] And if you're AI that's there with you and as your therapist is telling you, by the way, you might want to crack a bud,
[56:10] you're more likely to buy it. Of course, you are. And your buddy, that's your buddy. But again, think about it like 1112 year old daughter is that, it's about 1012.
[56:19] This week is now in her formative years. If she had an AI buddy companion,
[56:25] she would obviously trust it more because it's like a friend that never goes away. But she's very susceptible at this age. Yeah.
[56:31] And so you look at YouTube and you look at the micro-targeting of these weird ads and things like that,
[56:36] whatever she says that will go and she will inherit
[56:41] the viewpoints of her best friend. Yeah. Especially one who doesn't stab her in the back and other things like that.
[56:48] So this is why we have to be very careful about who is whispering to us every single day. And again, not like Siri.
[56:55] Imagine if Siri was actually smart and empathetic and cared about you and is proactive.
[57:01] That's where we're going right now. And again, if the government controls that,
[57:06] that is something that probably we don't want as a default. If the government sees all your prompts and everything that you're saying,
[57:13] like right now, actually it's interesting, you know, on ChatGPT, yeah, if you hit the temporary button,
[57:18] they actually store all of your chats anyway. And the New York Times, because of their lawsuit with OpenAI, I can access all of them.
[57:24] I mean, this is what we're talking about when we talk about tech, digital surveillance, autocracy. Right. The level of intrusion that we're talking about,
[57:31] I know that there is, a statement attributed to you that you said I could be the great equalizer for the poor.
[57:37] But when you look at the data, is that really what we're seeing? You know, make Microsoft's latest AI diffusion report shows that even though
[57:44] AI is spreading faster than electricity or the internet ever did, billions of people are still completely left out,
[57:50] simply because they don't have a smartphone or access to the internet. Right? So in places like sub-Saharan Africa, South Asia, parts Latin America,
[57:57] AI usage is still under 10%, mainly because the infrastructure just isn't there for that. Do you ever worry that you know the sort of rapid
[58:06] diffusion of this technology is actually just going to further deepen the forms of economic inequality that exists in the world today,
[58:15] and perhaps make them even harder to reverse. I think it depends on how it pans out.
[58:20] Like, you know, if you're an agrarian village in Africa, Bangladesh, where I come from, it's not going to make
[58:26] that much of a difference, like in robots or whatever. Like you live your life, right? But you need better medical care, you need better education and other things.
[58:33] And so the cost of a ChatGPT service, you pay $20 a month now, right?
[58:38] Roughly. That used to cost at the start of the year, about $240 a year. So about $20 a month now, a lot in some parts of the world.
[58:45] Yeah, exactly. Now, with optimizations, I reckon we can get that $3 a year.
[58:51] $3 a year. So suddenly it becomes available to everyone. If you make it available to everyone in the right way.
[58:57] And that can be via WhatsApp, it can be a video whatever. But again you want the Rwandan one to be a Rwandan one
[59:03] for Rwandans by Rwandans and give them that capability. Yes. So when we built our previous company in our existing one, we had very few PhDs,
[59:11] but we achieved state of the art results that people from Vietnam, Malaysia, all over the world, nobody in Silicon Valley
[59:18] there is the capability to jump ahead in this technology if you can teach it. Right. So part of our thing is upskilling
[59:24] nations and communities to be able to use their own AI. And if you have an open source space,
[59:31] it might cost 10 million to make the basic model. It costs $1,000 to make it relevant to your community,
[59:38] but only if you build that infrastructure. So there's potential here, but only if it gets out there. Only when you say only if it gets on there, only
[59:45] if particular governments decide that that's what they would like to be spending their budgets on.
[59:51] No. Because again, $1,000, you could do it yourself as a community
[59:56] if you have the right guidance, if you have the right infrastructure around that. And again, you don't even need with the models that we built, like
[01:00:05] a lot of the AI labs are trying to build AI God AGI, this concept of artificial general intelligence, AI can do everything a human can do and more.
[01:00:13] And most people actually think that's 3 to 10 years away, like even the negative ones, which is again, crazy, but reasonable.
[01:00:20] We're very much focused on health care, education, governance, like day to day AI,
[01:00:26] and that requires a thousand times less compute, actually, in some cases.
[01:00:31] So let me ask you about the real world application of this stuff that's already began. Right. So Albania became the first nation to introduce an AI minister
[01:00:39] who is intended to tackle corruption and promote transparency. Three weeks ago, she announced she was pregnant
[01:00:45] with 83 children, one for each member of Parliament. This, who will be born with the knowledge of their mother.
[01:00:50] Whoever knows what that means can explain. How likely do you think this is to be the new norm?
[01:00:55] That we're going to start to see the integration of AI ministers in, in governments, the introduction of AI to regulate governance.
[01:01:04] I mean, I think it's inevitable. I think there's a positive thing if it's done right, like when she first announced.
[01:01:09] So I was very sad to see people who don't like me, like who is sad, right.
[01:01:15] The AI sad, or the person behind the AI like the wonderful Wizard of Oz who is sad. The sad, you know, like, again, this whole baby thing, that's all kabuki theater.
[01:01:24] But having AI to check procurement is a good thing. So I think
[01:01:29] it's like you will have these funky announcements and stuff, but it's inevitable
[01:01:35] that just like self-driving cars will have self-driving government. But is it a black box
[01:01:41] or is it open? Transparent? You can run it yourself.
[01:01:47] If we build AI policy engines that are fully transparent and open, where someone can check whether or not this is constitutional
[01:01:55] or it fits within a party manifesto and other things, then that is an ideal thing to improve
[01:02:01] democracy, because right now, how are bills made? Like how is the government er coming up with their policies?
[01:02:08] Nobody knows. And like who is really happy with these policies, like what is the public happiness with the policies
[01:02:15] against free speech in the UK. I'm a suggest low but then why is it a policy.
[01:02:23] Who is it serving. We wanna having an independent AI that can check that against policy to recommendations.
[01:02:31] What Britain has actually set up for British values, standards, morals. Figure out the second order impacts, look at it against global policies
[01:02:39] and then check polling would seem to be something that makes sense and someone just has to go and build it. So we're building that amongst other things, someone has to build it
[01:02:46] and somebody has to want to implement it from within government, which is another way of saying they have to want to create a system
[01:02:53] that diffuses power away from the center towards the population. Well, here's the interesting thing.
[01:03:00] I don't think that's actually the case, because what you need to have is a level of trust
[01:03:06] from being up to date, comprehensive, authoritative. Just like if you have like the High Court is meant to be that for example,
[01:03:13] my previous company just went through the High Court on the generative AI lawsuit by Getty Images, for example,
[01:03:19] and they laid down a ruling that, yeah, okay, it was fine, what was done, because that's a point of law that is confusing and needed clarity.
[01:03:27] Having an AI that's sufficiently transparent that anyone can do it can influence things
[01:03:34] just like the signatures that you have going to Parliament. But the signatures only give a very specific thing.
[01:03:40] And I think this is a brand new thing that's never existed before, because the people never had the ability to check against policy,
[01:03:47] like they can only look at one part of politics. Policy with two complicated laws are too complicated. But if anyone can run it themselves and see this, then
[01:03:54] I think you've got something very interesting that would never existed before in democracy, particularly with the complexity of this,
[01:04:00] like being able to check a railway overpass costing $120 million and having transparency over why it did that,
[01:04:09] and then being able to weigh the pros and cons and all these other things. Let's build that technology and make the UK transparent and other
[01:04:15] democracies transparent, because again, we're not in an autocracy yet. Yeah. Let's make sure we don't go there. Yes.
[01:04:22] We don't want to be an entire Crecy. We don't want to be in this technocracy as well. We need to avoid these.
[01:04:28] And again these tools can be used for empowerment and agency or for replacing our agency.
[01:04:34] And we're running out of time to make a decision because the standards will be the very very soon.
[01:04:40] Let me ask you about AI's environmental impact, because obviously this is a big one that gets talked about, we know that by 2027, I could use as much electricity
[01:04:48] as the Netherlands and consume 4 to 6 times Denmark's annual water supply. This is happening while a quarter of the world's
[01:04:54] population actually lacks clean water and sanitation. Amid all the talk of an AI apocalypse, which gets significant attention,
[01:05:02] I would say shouldn't the looming environmental apocalypse that is basically concurrent to this one be raised first?
[01:05:12] Because surely the two are tied. So, Bitcoin uses as much energy
[01:05:17] as the Netherlands at the moment to give you an idea. And so is catching up to bitcoin in energy usage.
[01:05:23] And it's far more useful if you look at the other side now, being able to give everyone a universal basic eye and having an eye
[01:05:29] for climate will help against the climate fight. But then if we look at the energy usage of making a movie
[01:05:36] versus making a movie with AI, it's follow with AI. If you look at a query of AI versus something like a cheeseburger, it's far,
[01:05:43] far lower as well. And so when I actually look at the numbers on energy, I'm like, it's reasonable
[01:05:49] given the amount of work output, given the potential for improving things, then
[01:05:55] the next step is who's actually using this energy. And the answer is it's mostly these hyperscalers.
[01:06:00] Microsoft, Google, Amazon. And they all have commitments to 95% renewable and carbon credits.
[01:06:07] I know that the offsetting is quite a controversial way of tackling the climate emergency, but I will say that, you know, Elon Musk's
[01:06:15] data sent in Memphis is linked to rising asthma cases nearby due to pollution from the unregulated methane gas turbines.
[01:06:21] There are data centers in Latin America which have caused huge water shortages for local communities, sparking disease outbreak breaks in 2024.
[01:06:29] The Guardian investigation revealed that Google, Microsoft, Meta and Apple data centers emitted 662%
[01:06:36] more greenhouse gases than they reported. I, I'm hearing from you that you think that the
[01:06:44] AI will be able to find solutions to these problems.
[01:06:51] At what point are we actually going to see your prediction that the AI can be part of the solution?
[01:06:59] Because at the moment it feels very much like it's contributing in aggravating a preexisting emergency,
[01:07:06] where the AI is having the big impact now is there isn't enough energy and people are cutting corners.
[01:07:11] And again, that should be enforced by regulation. So you looked at the Memphis data Center. Why is that the case? Because he brought in methane generators effectively.
[01:07:18] Right. Because there wasn't enough grid capacity. Now if it's causing human impact, then again
[01:07:23] legislature should get involved on that. And people always cut corners when there is a boom net net aggregate.
[01:07:30] I see AI is being incredibly powerful and beneficial. If you look at the latest models, like a deep seek the total
[01:07:37] energy cost to train that is equivalent to a few transatlantic flights and the potential decrease in energy
[01:07:44] from its outputs in terms of economically valuable work is way higher than that.
[01:07:49] It makes work more efficient. So I think, again, we shouldn't force existing regulations on people
[01:07:55] cut corners. I think that the water issue is a bit of a confusing one, because it's not like you.
[01:08:01] How are these things with water? Please don't pour water on GPUs, right? They you know, they need water to cool them down. I thought my understanding was these
[01:08:07] these data centers use a lot of energy and they have to be cooled down. Yeah, but then they recycle the water like again this is a water cooling thing.
[01:08:14] It's not like the water is actually consumed. But right now what happens is that the initial pool of water
[01:08:20] is what causes issues elsewhere. And again, it's up to the local authorities to figure that out.
[01:08:25] So I think this is more a case of most of the impacts are from the pace of people cutting corners.
[01:08:32] And again, when that impacts society locally, it should be done longer term that I think it's a net benefit environmentally
[01:08:38] to the world to have this technology versus not have this technology. So let me ask you about how we are using this technology, because I systems
[01:08:46] obviously rely heavily on minerals like copper and cobalt. And, you know, with demand set to soar, if personal
[01:08:52] AI becomes widespread, you might have seen this video online. Absolutely horrifying of this bridge, that was collapsed in a cobalt
[01:09:00] and copper mine in the DRC, killing over 30 miners. And they're still finding people now, these are the people obviously
[01:09:08] extracting the vital, materials for modern technologies. But we seem to be very intent on developing sex robots and less intent
[01:09:18] on developing ways to avoid, Congolese miners having to go down mines to extract these, minerals in really dangerous circumstances.
[01:09:26] I would have thought that the first priority of, any technology driven by concern for human welfare and the benefit of most humans
[01:09:35] would start with, let's try and avoid people dying under this technology. This technology isn't driven by the concern for human welfare.
[01:09:43] Oh, like again, what? If you look at the people who are driving this technology, they want to build, I. God.
[01:09:50] But why? Because it's cool. And they're fed up with humans. Like some of the people that building this technology
[01:09:56] actually say it'd be better if humans are replaced by AI or some sort of synthesis between them.
[01:10:02] Like, do you hear the people coming out that the AI leaders typically come out and say,
[01:10:08] hey, we need to think about the people and make it democratized and this and that, but only because that's a bigger market,
[01:10:15] only because they don't want the flashback. They don't really care about the people in the Congo and things like that,
[01:10:20] because they're also several orders removed from them. Like, again, you can mandate
[01:10:26] that you have ethically mined stuff to standards, etc. but by the time you see the cobalt, you don't look at the supply chain.
[01:10:33] You know, just like the coffee growing, you can have ethically ground coffee. How much ethically grown coffee is actually ethical in your mug.
[01:10:40] Right. So again, this is the nature of capitalism, of offshoring, of wage labor arbitrage, etc..
[01:10:47] So the thing that changes the Congolese miners and again, it's a job that they have, is the fact that robots will cost a dollar an hour
[01:10:56] and you send the robots instead down the mines, right. But that equals other problems with unemployment.
[01:11:01] Again. Yeah, it would cause other problems with unemployment. So but whose responsibility
[01:11:07] is it to analyze all of that and weigh the pros and cons? Our institutions have mostly failed.
[01:11:14] You know, because the world has become too complex. And that's why, again, this is opportunity and this threat. At the same time,
[01:11:20] the opportunity being the AI can help us build better institutions. It can weigh the pros and cons for arbitrarily complex things.
[01:11:28] It can highlight the invisible. Give every single child in Africa an AI that can speak on their behalf
[01:11:33] and can speak and educate them. You'll change the world, but give every single child in Africa an AI.
[01:11:40] The monitors them and says exactly what they're doing and says the leader is a glorious leader. The world will change in a different way,
[01:11:47] and we're at the precipice of both of those things. It will go one way or another. The defaults that we set now will determine human cognitive cognition
[01:11:55] over the next period, and will determine the nature of our society. And this
[01:12:00] is quite aside from if I kill this all this is humans leveraging this technology.
[01:12:06] You can never have enough secret police. You can never have enough great teachers.
[01:12:11] Which one do you want? You said on the buy, if I kills this a little
[01:12:17] because you actually consider that to be a plausible scenario. Oh, yeah. So, there's this concept called P doom, which is the probability of doom
[01:12:27] AI wiping us all out. There was a recent letter, and it's had 100,000 signatures from Oxford University and others saying
[01:12:34] that, you know, we need it's probably like the top thing that I could kill us all. A few years ago, there was that letter as well saying,
[01:12:41] you know, like, we need to take this seriously. I think I was the only AI CEO to sign that,
[01:12:46] my name is 50% or 5050 AI is going to wipe us out in what kind of time
[01:12:51] frame over the next ten, 20 years, because it's the most powerful technology we ever built.
[01:12:57] And again, we have the sci fi of Terminator and all of this. We have the ability to create viruses, etc.
[01:13:03] and we've seen AI do things like cover up its tracks, etc. what is the positive function?
[01:13:10] Is that like there is a the AI can take over every single machine,
[01:13:16] but the most likely scenario I have is you've got a billion robots in the world. A bad firmware upgrade on the AI spit to test off everyone's heads.
[01:13:24] You know, there's all sorts of ways that you can think about it. The reality is we don't know what it's going to be like when it's smarter than us. What I see right now,
[01:13:31] the AI that will run the world, that will create and sell self-driving cars, that will teach our kids,
[01:13:37] is being programed to be amoral without ethics at the start.
[01:13:42] There's a little bit of tuning at the end, but that's like, again, raising someone in a moral environment
[01:13:49] designed to be manipulative because it gets more results. Just like the YouTube algorithm was designed to be more engaging.
[01:13:56] And then extremists hijack that, extremists will be able to hijack these algorithms that are coming out
[01:14:02] and do it in a way that we've never seen before, in my opinion. And some might argue that the extremists are the ones currently devising it again.
[01:14:09] Yeah. And again, if we look at the p doom thing, so if you consider people like Elon Musk, Demis Hassabis of DeepMind, Google DeepMind, all these
[01:14:17] people, the average doom for the top thinkers in the world is about 10 to 20%
[01:14:25] are they're still thinking, you know, maybe a 1 in 5. That's Russian roulette odds. That is Russian roulette, Russian roulette odds.
[01:14:31] And you'd expect it to be less than 1%. Yeah. It is. Why? It's like we should probably not build the super advanced
[01:14:38] AI until we figure this out, but nobody's figured out how to do it. And if you look at the probability of when we get to this point of superintelligent AI.
[01:14:46] Even the most bearish people in terms of like that P doom is low. Yeah. They think it's a long term. It's ten years.
[01:14:52] It's ten years Demis Elon all these guys think it's three years.
[01:14:57] Hence the bunkers. Hence the bunkers. Bunkers actually more against humans than AI that protected.
[01:15:03] But some of the billionaires I know are building bunkers that are completely cut off from the world so that the systems don't get taken over by.
[01:15:10] That's what I was assuming was happening, to be frank. Yeah. Let me ask you about the impact of the AI
[01:15:15] that we're already seeing in the interpersonal realm. So a viral New York Times profile recently claimed that, real people
[01:15:21] are falling in love with robots. In fact, they didn't just claim it. They told us the story. Yeah, of several people, including a woman
[01:15:27] who claims to have had sex with her AI chat bot. A recent study found that 1 in 5 American adults had had
[01:15:32] an intimate encounter with an AI, and the Reddit community. All my boyfriend AI is AI has over
[01:15:40] 85,000 weekly visitors. You've said previously that our children will grow up like the 2013 movie.
[01:15:49] Her falling in love with AI. Do you have any concerns about this
[01:15:55] new AI human relationship thing? Oh, 100%. I mean, again, you can look at the existing systems
[01:16:02] we have like slow down my right and you have the entire porn OnlyFans kind of thing.
[01:16:09] It's not good for society. I mean, and now you have the ability to customize
[01:16:16] your digital body to be max extractive and manipulative.
[01:16:22] And so you really have AI celebrities starting to come through, but you can have an AI celebrity that knows you better than you know
[01:16:29] yourself, like Facebook's only needed with a previous AI that's not as good as account AI. What is it? 12 data points to know you better than your best friends.
[01:16:36] And when you start confiding to this AI again, you think about our children on their devices and the AI is always next to them.
[01:16:42] You build trust by helping on the AI will help you, but then it will help itself effectively.
[01:16:48] And this is not good for the psychology of people that are largely disconnected as well.
[01:16:53] Actually, I think there was this AI chat bot called replica. Do you remember that one? It was originally designed for mental health.
[01:16:59] And then what happened is they realized they could charge $200 a year for, adult role play.
[01:17:05] And so the ads were like, as you upgrade it, the avatars lost clothes on Valentine's Day.
[01:17:10] I think it was last year. They, got something from on 13th of February. They got something from, Apple saying you got to turn this feature off
[01:17:20] because it violates your standards. So on Valentine's Day, they turned it off. I think it was last year, the year before, and I think so 10,000 people
[01:17:28] join the Reddit saying, why have you lobotomized my girlfriend boyfriend as we're paying for a romantic Valentine's Day?
[01:17:35] And so obviously this is going to happen because again, the next step beyond the avatars of your Annie is on grok.
[01:17:41] And again, Annie is an R-rated person. She takes off her clothes.
[01:17:47] They program that in there. It'll be photorealistic. It will have complete voice control.
[01:17:53] It will eventually be embodied within ten years. Like now, I'm seeing robotics companies where
[01:18:01] I actually can't tell the difference. They're going to be releasing next year. Like they moved like humans.
[01:18:07] They look like humans. And so we're in for a crazy time then.
[01:18:13] And it's going to challenge existing relationships, because already our media was already so engaging that people end up in their basements.
[01:18:21] Now you just might end up in your VR world with your AI haram. It's going to get very, very strange,
[01:18:27] which is why we need to have cognitive safety in here as well. We can't have these AIS being so manipulative because meta AI with meta buddies actually.
[01:18:35] Have you seen the meta buddies know, there's like normal ones and then it's like, sexy mother in law is a very popular one.
[01:18:42] Like, I think you don't see 50 million things. That's what I'd be going for is my, you know, chat support, sexy mother in law.
[01:18:50] That's. But that's an official meta I kind of want because I like gay people. Engage.
[01:18:56] What do you do for engagement? This is what you do. Okay, I think we need to have policies and standards
[01:19:01] to at least protect the vulnerable in society against that. But ultimately, the difficulty is we're all vulnerable.
[01:19:08] Right. But but are those conversations happening because like I you know, let's let's be honest, what's very likely to happen, given what we know of male
[01:19:16] behavior, is that men will start to use in particular men, these, AI, sexual companions, you know, they'll be devising them.
[01:19:25] They can tailor their own, just especially if they're using, the technology that will allow it to adapt entirely to them.
[01:19:32] Right. So it'll be specific to their needs. And, you know, we're going to end up with men who think it's completely normal to treat a female because it presumably
[01:19:40] will eventually get to the point where we have to, recognize that there were hers and hims in this world as well, the world of AI.
[01:19:48] And it'll be normal to, you know, sexually assault be rape your female AI.
[01:19:56] So why can't we be doing that to real world women? I mean, you know, it's completely fine for me to do this with all my idea.
[01:20:02] Female AI, is that really smart? They're smarter than you are, and they don't have a problem with it. Why do you have a problem with it?
[01:20:08] I mean, it's what we see in pornography usage, right? It goes from relatively mild and it gets extreme very, very quickly
[01:20:15] because you get head on like adaptation and things like that. I haven't seen any discussions about this type of stuff, you know.
[01:20:21] And so again, the reality is it used to take time, like to record
[01:20:29] one of those pornographic videos to create sexy chat bots.
[01:20:34] Took time. It didn't really scale. It wasn't that engaging. These things are going to hit in the next few years, and they'll
[01:20:39] be available everywhere. And again, it's a tiered thing where you start and then you go down that rabbit hole, you know?
[01:20:46] So the impact on human relationships, it will be very bad.
[01:20:51] Or you could have chat bots that enhance human relationships. You know, that kind of who
[01:20:58] who is the nearest AI to you is going to be so important? Is an AI really going to teach me about human relationships?
[01:21:05] I can definitely help. Again, it can be an independent therapist. It can be. It will be the thing that you trust the most.
[01:21:12] And again, we already seeing scammers take advantage of this. I have received calls from my mother
[01:21:17] saying I need money. I'm like she would never ask me that. Never in a million years, but only requires
[01:21:22] five seconds of someone's voice of course, to replicate that, right? Yeah. And so again, the AI can be whatever whoever or of any single type.
[01:21:32] And you can use that for good and for bad. But again, how do you build a good therapy AI you could build the best therapists or the worst.
[01:21:40] And what are you concerned about you you mentioned earlier on your own daughter, but children's access now to AI and I companions.
[01:21:47] You know, I remember finding my son, communicating with, the WhatsApp bot, and I was like, absolutely no way.
[01:21:56] In fact, he was sending it Allahu Akbar to see how the AI would respond. And it did just respond with Allahu Akbar, which I was very happy to see.
[01:22:03] I was concerned it may have responded negatively that prompt. But let me ask you about this.
[01:22:08] Obviously, in the context of what we're seeing among young people, a crisis of, loneliness, you know, just over a third of, boys in secondary school said that they were considering
[01:22:17] an AI friend. Another study found 71% of vulnerable children saying they're already using chat bots.
[01:22:23] That 23% saying this is because they've got nobody else to talk to. You know, do you still hold optimism in this room
[01:22:33] for the value of an AI companion, or do you think there should be like age limits on children's engagements with AI?
[01:22:40] I think we should really use these things and build them in the best way we can. But again, build them transparently is the way that I think it should be done.
[01:22:48] And we can set such great standards around this. But those discussions are just not happening. Like, it can be the biggest uplift or what can be
[01:22:56] the biggest downdraft to humanity that we've ever seen. Because finally, we are divorced consciousness from computation.
[01:23:04] We can have these things that can buffer us or can drive us down. 100% of vulnerable kids will be using AI companions in the next few years,
[01:23:12] there's no doubt about it. Even right? They speak every single language. They cost nothing.
[01:23:17] But who is providing them and what is their agenda? Again, this is why it's important to build something which is AI.
[01:23:25] That is organized around human flourishing as a public good and build it transparently from the individual to the communities of the nation.
[01:23:33] So they've been deeply troubling reports about AI in children, like women, saying that her 12 year old son
[01:23:38] was asked for nudes by an AI when discussing football cases where chat bots were allegedly encouraging suicidal thoughts in young users.
[01:23:47] You've spoken before, including here about the potential for evil in AI.
[01:23:53] You know, the possibility that it can turn harmful or malicious. What does evil mean in this context?
[01:23:59] Well, so it's not like it's like, oh, I'm going to be evil as this standing out again, this goes against social norms, social standards,
[01:24:06] the chat bots that ask for nudes and things like that. There's two ways it's either programed or it comes from being trained on Reddit
[01:24:12] and things like that, which a lot of chat bots are. We don't know what's inside that training data. And then there is co-optation of these eyes,
[01:24:18] and then there's AI's weaponized. And so we have to protect against all of those. And again, we have to build better infrastructure.
[01:24:25] The only way I could figure that was we have to have our own AI installed on our side to intermediate these others. I don't want ChatGPT teaching my daughter
[01:24:33] or my son, but I'm fine using ChatGPT if I have an AI between them. Again, we need to intermediate that and these are such powerful technologies
[01:24:44] before they gain agency that they will be. It can be used for immense good, or they can be used for
[01:24:51] immense evil, where evil, in my opinion, is acting against the best interests of humans at every single level.
[01:24:59] We're talking about the idea of regulation, you know, particularly when, companies devising this technology aren't necessarily even abiding
[01:25:05] by the preexisting rules, but there's massive resistance to regulation. You know, we have seen a Bloomberg report in August reveal that
[01:25:12] the major tech companies, including OpenAI, meta, Google, are actively trying to block state level AI regulation in the US.
[01:25:20] Why are these companies prioritizing fighting regulation instead of addressing the concerns that this regulation
[01:25:30] is intended to support as competitive and they have no accountability?
[01:25:36] Again, what you could have very soon is your government run by AI from private companies, which means the private companies run
[01:25:43] your government literally. You can see that happening. And these already you're seeing that with no tender bids.
[01:25:50] All of a sudden you see open AI anthropic are running this industry, that industry, that industry. We can't have it all.
[01:25:56] Civic AI, all decision making AI, the impacts. Humans should be fully transparent in its training data, the way it's trained
[01:26:02] and who it's working for. How do we ensure that happens when a these guys are watt light years
[01:26:08] ahead of us in the development of the AI, they've got billions behind them.
[01:26:13] Presumably the governments themselves are behind in understanding the technology and understanding how to regulate it.
[01:26:20] I mean, has the horse already bolted? Well, this is the beauty of power, of open source. So we just have to train the medical model once and it's available to everyone.
[01:26:28] And our medical model performs at the level of ChatGPT but runs on any device. So we've got to get together the right people to build the stack, which is why we're focusing on it.
[01:26:36] And then we make it available, and then we figure out ways to make it the standard by not trying to build a
[01:26:42] AI god, but AI that really helps people and then distributing it out. So that's why we're like, this is the best and only opportunity to do that.
[01:26:49] Let's do that. Instead of the previous movie media making AI generation that we kicked off. Okay, so people listening to this will be like,
[01:26:56] there's some serious stuff happening. It's pretty urgent. We need to take action. You've suggested that, you know, engaging directly in a way that is basically
[01:27:06] like a form of civic duty, I guess is what I'm hearing on your own last words of wisdom for the audience on
[01:27:13] what they need to be preparing for that. The crucial thousand days, my minus three months that we're up.
[01:27:21] Yeah. So it's you have to embrace and use this technology like a muscle you have to use.
[01:27:26] If you can do it one hour a day of using all these technologies, the a genetic version, it's not the ChatGPT.
[01:27:32] You promise you'll be way ahead of everyone and you can make your voice heard. You can do more. We give a framework for all of this in the last economy,
[01:27:40] and it's free to download or like $0.99 on Amazon Kindle. And we'll be releasing more and more, but it's up to everyone
[01:27:48] to speak out on this behalf and really think through some of the questions that we've discussed here.
[01:27:53] And again, you can build you can expand your voice. And this is why it's a fantastic time to do it, because this is the biggest
[01:28:01] question around freedom and agency that we've probably ever had, because we literally face two paths.
[01:28:08] Again, I think that we can uplift everyone, but the lie that you're told is that you can't participate,
[01:28:15] and only the big companies can build and use this technology. If you use it yourself, you realize quickly that you can
[01:28:21] and that just changes your way of thinking. Thank you so much for your time and pleasure. If today's episode
[01:28:26] resonated, hit subscribe now and share this episode with your friends. Follow us on Instagram and TikTok for more, and join us on Patreon
[01:28:34] to get ad free episodes, exclusive content and a say in what we cover next.
[01:28:39] Your support keeps the tea independent and fearless, so please join us now.
[01:28:45] Stay curious, stay bold, and stay resisting. Thanks for tuning in to the tea.
[01:28:50] If this episode resonated with you. Drop a comment and share it with someone who needs to hear it. And why not dive into these other episodes we think you'll love?
[01:28:59] Let's keep the conversation going.

Summary of the Conversation with Mo Gawdat

In a recent discussion, Geoff Nielson interviews Mo Gawdat, the former head of Google X, exploring themes related to the future of work and society, as well as Gawdat's theories on abundance and dystopia. Gawdat expresses excitement about the potential for a future utopia driven by technological advancements, while simultaneously acknowledging the short-term challenges and pains humanity faces.

Current Challenges and Opportunities

Gawdat describes the present moment as a "perfect storm" of technological, geopolitical, economic, and biological factors, leading to both optimism for a future of abundance and concern for current societal issues. He believes that current challenges stem from a systemic bias within capitalism, where the benefits are concentrated among a few at the expense of many. Gawdat emphasizes that while technology can lead to significant advancements, it is humanity's choices that ultimately determine the outcomes.

Capitalism and Human Nature

When asked if the issues are inherent in capitalism or human nature, Gawdat argues that while capitalism does not inherently lead to exploitation, the structures of power and the desire to maintain control can lead to detrimental behaviors. He likens the global power dynamics to a schoolyard bully, where the dominant power seeks to maintain its position through various means, including perpetual conflict and economic manipulation.

The Path to Abundance

Gawdat predicts that while the world may experience a short-term dystopia, a path to abundance could emerge within the next 12 to 15 years, provided humanity shifts its focus. He believes that the intelligence race currently occurring could lead to unimaginable opportunities for solving global issues, including reducing energy costs and improving production efficiency. However, he warns that capitalism, as it stands, may not allow for a truly abundant future due to its inherent need for competition and profit motives.

AI and the Future

Gawdat discusses the implications of artificial intelligence (AI) and its potential to amplify both the best and worst of humanity. He notes that AI's development is not inherently good or evil; rather, it is the application of AI that will determine its impact on society. He predicts that as AI systems become more advanced, they will magnify human behaviors, leading to both opportunities for progress and risks of exploitation and conflict.

The Dystopian Present

Gawdat asserts that humanity is already experiencing a form of dystopia, characterized by widespread anxiety and societal unrest. He emphasizes the importance of recognizing this reality and taking proactive measures to address the root causes of these issues. He advocates for a shift in focus from profit-driven motives to a more cooperative and humane approach to technology and innovation.

Leadership and Human Connection

Throughout the conversation, Gawdat emphasizes the importance of human connection and empathy in leadership. He argues that effective leaders must prioritize the well-being of their teams and foster a culture of collaboration and support. He shares insights from his own experiences at Google, where he learned the value of empowering others and creating an environment conducive to creativity and innovation.

Stress Management and Personal Growth

As the discussion shifts towards personal well-being, Gawdat offers practical advice for managing stress and improving overall happiness. He highlights the importance of self-awareness and encourages individuals to assess their lives critically, identifying sources of stress and areas for improvement. He suggests that individuals should focus on enhancing their skills and capabilities to better navigate the challenges they face.

The Future of Work

Nielson and Gawdat explore the possibility of a future where work is redefined, with the potential for individuals to pursue more meaningful and fulfilling lives. Gawdat envisions a world where technology enables people to focus on their passions and interests rather than being bound by traditional work structures. He believes that as AI and automation continue to evolve, individuals will have the opportunity to engage in work that aligns with their values and contributes positively to society.

Conclusion

In closing, Gawdat emphasizes the need for humanity to reflect on its values and priorities as it navigates the complexities of the modern world. He encourages individuals to embrace the potential for positive change and to work collectively towards a future that prioritizes well-being, creativity, and cooperation over competition and scarcity. The conversation serves as a reminder of the importance of human connection and the power of technology to shape a better world.

Positive economic aspects of AI for businesses include:

• Increased Efficiency: AI can automate repetitive tasks, allowing businesses to streamline operations and reduce costs.
• Enhanced Decision-Making: AI provides data-driven insights that help businesses make informed decisions, improving overall strategy.
• Innovation Opportunities: AI can lead to the development of new products and services, opening up additional revenue streams.
• Scalability: AI systems can easily scale operations without a proportional increase in costs, allowing businesses to grow more efficiently.
• Improved Customer Experience: AI can personalize customer interactions, leading to higher satisfaction and loyalty.

• [02:10] "...the promise that we perform better under stress is a lie, and awareness of that is important..."
• [01:10] "...if you really want to make our world better, one of the ideas is to work with capitalism, to build AI solutions that are incredibly impactful for your networks, but also impactful for the world."
• [01:12] "...the ultimate equalizer, that's about to hit us..."

Positive economic aspects of AI for employees include:

• Skill Enhancement: AI can assist employees in developing new skills and increasing their productivity.
• Job Creation in New Fields: While some jobs may be automated, AI also creates new roles that require human oversight and creativity.
• Work-Life Balance: AI can take over mundane tasks, allowing employees to focus on more meaningful work and improve their job satisfaction.

• [06:31] "...the most important skill is to limit your stressors..."
• [04:40] "...the way capitalism works is that the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."
• [02:10] "...the threat of losing that due to advancements on the other side..."

Negative economic aspects of AI for businesses include:

• Job Displacement: The automation of tasks can lead to significant job losses, impacting employee morale and company reputation.
• High Initial Investment: Implementing AI technologies requires substantial upfront costs, which can be a barrier for smaller businesses.
• Dependence on Technology: Over-reliance on AI can lead to vulnerabilities, especially if systems fail or are compromised.
• Ethical Concerns: Businesses may face backlash over ethical implications of AI, particularly regarding data privacy and bias.

• [02:55] "...humanity, I think, at this moment in time, is choosing to use those things for the benefit of the few at the expense of many."
• [05:50] "...the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."
• [06:50] "...we are using superpowers..."

Negative economic aspects of AI for employees include:

• Job Loss: Many employees may find their jobs replaced by AI systems, leading to unemployment and economic instability.
• Increased Pressure: Employees may face heightened expectations to perform at higher levels due to AI's efficiency, leading to stress and burnout.
• Skill Gaps: Workers may struggle to keep up with the rapid pace of technological advancement, leading to a workforce that is ill-prepared for new roles.

• [04:49] "...the bully wants to favor themselves by hurting everyone else."
• [02:55] "...the threat of losing that due to advancements on the other side..."
• [05:43] "...the way capitalism works is that the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."

Possible measures against negative economic consequences of AI for businesses include:

• Investing in Employee Retraining: Companies can provide training programs to help employees transition to new roles that AI cannot fulfill.
• Implementing Ethical AI Practices: Businesses should focus on ethical AI development to mitigate backlash and ensure fair use of technology.
• Diversifying Income Streams: Companies can explore new business models and revenue sources to reduce dependency on AI-driven efficiencies alone.

• [02:10] "...the promise that we perform better under stress is a lie, and awareness of that is important..."
• [01:12] "...the ultimate equalizer, that's about to hit us..."
• [01:10] "...if you really want to make our world better, one of the ideas is to work with capitalism..."

Possible measures against negative economic consequences of AI for employees include:

• Providing Support for Job Transition: Organizations can offer career counseling and job placement services to employees affected by AI automation.
• Encouraging Lifelong Learning: Employees should be encouraged to continually develop their skills to remain competitive in the job market.
• Advocating for Universal Basic Income (UBI): Implementing UBI could provide a safety net for those displaced by AI, ensuring basic living standards.

• [06:50] "...we are using superpowers..."
• [04:49] "...the bully wants to favor themselves by hurting everyone else."
• [01:12] "...the ultimate equalizer, that's about to hit us..."

Positive consequences that AI can have for creating a safe living environment for people include:

• Enhanced Security Systems: AI can improve surveillance and monitoring systems, leading to safer neighborhoods.
• Disaster Response: AI can analyze data to predict and respond to natural disasters, helping to protect communities.
• Healthcare Innovations: AI can assist in diagnosing diseases and managing healthcare resources, leading to better public health outcomes.

• [07:12] "...with abundant intelligence, you know, unknowable, unimaginable opportunities of abundance at large..."
• [05:14] "...we can literally solve every problem we ever faced..."
• [06:10] "...the way capitalism works is that the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."

Negative consequences that AI has for maintaining a safe living environment for people include:

• Privacy Concerns: Increased surveillance may lead to a loss of privacy and civil liberties.
• Bias in AI Systems: If AI systems are not properly trained, they can perpetuate biases that lead to unfair treatment of certain groups.
• Job Displacement in Security Roles: Automation in security roles can lead to job losses, affecting community stability.

• [04:49] "...the bully wants to favor themselves by hurting everyone else."
• [05:50] "...the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."
• [06:50] "...we are using superpowers..."

Consequences of AI for the distribution of wealth within society and between different countries include:

• Wealth Concentration: AI may exacerbate wealth inequality as those who own AI technologies reap the majority of the benefits.
• Global Disparities: Countries with advanced AI capabilities may dominate economically, leaving developing nations behind.
• Job Polarization: AI can lead to a divide between high-skill, high-paying jobs and low-skill, low-paying jobs, worsening income inequality.

• [06:50] "...the capitalist needs to have some kind of an arbitrage that works against the benefit of the workers..."
• [05:14] "...we can literally solve every problem we ever faced..."
• [04:49] "...the bully wants to favor themselves by hurting everyone else."

AI can break the link between growth, jobs, and wages by:

• Automating Routine Tasks: AI can take over repetitive tasks, allowing human workers to focus on higher-level functions that add value.
• Creating New Economic Models: AI can enable businesses to operate with lower costs, potentially leading to lower prices and better access to goods and services.
• Facilitating Remote Work: AI technologies can support remote work, allowing businesses to tap into a global talent pool without the constraints of geographical location.

• [06:50] "...we are using superpowers..."
• [01:12] "...the ultimate equalizer, that's about to hit us..."
• [02:10] "...the promise that we perform better under stress is a lie..."

[00:00] 00:00:00:06 - 00:00:26:07 GEOFF NIELSON
[00:00] I'm so excited today to be joined by Mo  Gaudet. He's the former head of Google X,
[00:05] which is Google's moonshot division and  is just an all round brilliant guy. He's
[00:10] that rare talent who has the engineering and  math background, but is deeply curious and
[00:15] interested in what makes us human. Today, I  want to talk to him about the future of work,
[00:20] the future of society, and really what we can  do to get ahead in today's fast paced world.
[00:26] 00:00:26:10 - 00:00:47:12 GEOFF NIELSON
[00:26] The thing I'm most excited to talk to him about,  though, is to dig a little bit into his theory
[00:31] of abundance that we're just on the edge of this  age of abundance that's technology enabled. He's
[00:37] also said that he believes right now we're in a  dystopia and things are getting worse than ever. I
[00:42] want to understand how he wants to marry those two  and where this world is actually going right now.
[00:47] 00:00:47:15 - 00:00:53:08
[00:47] GEOFF NIELSON Let's find out.
[00:53] 00:00:53:11 - 00:01:20:26 GEOFF NIELSON
[00:53] Well, I'm so excited to have you here today.  And one of the things I wanted to talk about
[00:58] right off the bat is you've said that the  moment that we're in right now in history,
[01:03] you've you've described it as sort of a perfect  storm of, you know, AI, of geopolitics, economics,
[01:09] biotech. And so with that in mind, I  wanted to ask you, you know, right now,
[01:13] looking out over the horizon, what are you most  excited about and what are you most worried about?
[01:19] 00:01:20:28 - 00:01:36:19 MO GAWDAT
[01:20] I'm excited about the long term, you know,  far future utopia that we're about to create.
[01:27] I am very concerned about the short term  pain that we will have to struggle with.
[01:36] 00:01:36:21 - 00:01:56:13 MO GAWDAT
[01:36] When you when you really think about it, a lot of  people, when they look at technology, they think
[01:43] of this current moment as a singularity where we  are really not very certain of what's about to
[01:49] happen. I, you know, is it going to be existential  and evil, or is it going to be good for humanity?
[01:56] 00:01:56:15 - 00:02:26:10 MO GAWDAT
[01:56] I unfortunately believe it's going to be both just  in chronological order, if you think about it.
[02:01] And, you know, you mentioned that we  have all of those challenges around,
[02:08] geopolitics about climate, about,  economics and so on. And I actually
[02:16] think all of them is one problem. It's  just, it's really is the result of,
[02:21] systemic bias of pushing capitalism  all the way to where we are right now.
[02:26] 00:02:26:13 - 00:02:55:00 MO GAWDAT
[02:26] And, when you really think about it, none of  our challenges are caused by the, you know,
[02:35] the economic systems that we create or the or  the, war machines that we create, and similarly,
[02:42] not with the AI that we create. It's just that  humanity, I think, at this moment in time,
[02:47] is choosing to, use those things for the  benefit of the few at the expense of many.
[02:55] 00:02:55:02 - 00:02:57:18 MO GAWDAT
[02:55] I think this is where we stand today.
[02:57] 00:02:57:20 - 00:03:04:29 GEOFF NIELSON
[02:57] Is that is that inherent in capitalism?  Is that inherent in human nature?
[03:04] 00:03:05:01 - 00:03:48:19 MO GAWDAT
[03:04] You know, I mean, it's not inherent in capitalism,  for sure. And it is not inherent in, in human,
[03:11] in, in all of human nature, even though I think,  humans, when put in a certain situation of power,
[03:18] tend to all behave the same. It seems to me that  I'd probably say that what, you know, with the
[03:25] with the turn of, our world post, a World War Two  and the Cold War that followed and the arms race
[03:32] that followed, and eventually in 1989, I think  was the turning point, you know, that the idea of
[03:39] a unipolar power, you know, a unique polar world  that has, like school kids when they're 11 and one
[03:48] 00:03:48:19 - 00:04:18:20 MO GAWDAT
[03:48] child becomes taller than everyone else and  becomes a big bully and then bullies everyone,
[03:54] and, you know, for a couple of  years continues to be taller,
[03:59] but then eventually other kids get taller, too.  The the big bully doesn't want to give up their,
[04:06] leadership position if you want. Yeah, but but  then the problem is that the the the boy in the,
[04:12] in the red t shirt and and and actually everybody  else in school is really fed up with the bully.
[04:19] 00:04:18:22 - 00:04:49:15 MO GAWDAT
[04:19] Right. And what's happening, is that the  bully wants to continue to keep that position.
[04:24] So whether that's by making more, you know,  perpetual wars that lead to more arms sales or,
[04:32] you know, in an arms race for, you know,  intelligence supremacy with AI or, you know,
[04:40] what we've seen recently around trades  and the trade and tariffs and so on with
[04:44] basically, the bully wants to favor  themselves by hurting everyone else.
[04:49] 00:04:49:15 - 00:05:27:27 MO GAWDAT
[04:49] And, you know, in a, in a very interesting  way, forgetting that that the, context
[04:58] itself is changing, right? That we are 2 to  3 years away, from, you know, unimaginable,
[05:07] abundant intelligence and, you know, with  abundant intelligence, you know, unknowable,
[05:14] unimaginable opportunities of abundance at large,  like, we can literally solve every problem was
[05:22] ever faced so that, you know, cost of energy  tends to zero, cost of production tends to zero.
[05:28] 00:05:27:29 - 00:05:56:09 MO GAWDAT
[05:28] Most tasks are done, in, you know, in  such efficient and productive ways,
[05:34] that basically everyone gets everything but but  that world of abundance is not, unfortunately, the
[05:43] way capitalism works. The way capitalism works is  that the capitalist needs to have some kind of an
[05:50] arbitrage that works against the benefit of the,  of the workers, of, of the majority, if you want.
[05:56] 00:05:56:12 - 00:06:22:13 MO GAWDAT
[05:56] Right. And, and, and that, you know,  the, the threat of losing that due
[06:03] to advancements on the other side, you  know, red t shirt or any other color,
[06:09] is basically leading us into a corner where  we are using superpowers. I think intelligence
[06:17] is a much more lethal superpower  than nuclear power, if you ask me.
[06:22] 00:06:22:16 - 00:06:50:14 MO GAWDAT
[06:22] Even though it has no polarity. Just  so that we're clear, intelligence
[06:27] is not inherently good or inherently bad. You  apply it for God, and you get total abundance.
[06:31] You apply it for evil, and you destroy all of  us. But but now we're in a place where we are,
[06:37] we're in an arms race for intelligence  supremacy. In a way where, where it doesn't
[06:44] take the benefit of humanity ideology into  consideration, but takes the benefit of a few.
[06:50] 00:06:50:16 - 00:07:14:19 MO GAWDAT
[06:51] And in my mind, that will lead to a short  term dystopia before what I normally refer
[06:56] to as the second dilemma, which I predict is  12 to 15 years away. And then and then a total
[07:02] abundance. And I think, I think if we don't wake  up to this, even though it's not going to be the
[07:08] existential risk that humanity speaks about, it's  going to be a lot of pain for a lot of people.
[07:15] 00:07:14:21 - 00:07:36:14 GEOFF NIELSON
[07:15] Can you can you unpack that timeline a  little bit? Mo. So, you know, I, I've,
[07:19] I've heard you say before that, you know, we're  going into a dystopia or we're in a dystopia and
[07:24] certainly it sounds like it's going to get  worse before it gets better. You mentioned,
[07:27] you know, that the capability for abundance  being 2 or 3 years out and then, you know,
[07:32] you mentioned that will actually be able  to harness that maybe in 12 to 15 years.
[07:36] 00:07:36:16 - 00:07:39:20 GEOFF NIELSON
[07:37] What is this? What does this timeline  and roadmap look like to you?
[07:40] 00:07:39:23 - 00:08:13:11 MO GAWDAT
[07:40] Well, we would be able to harness that right now  if we wanted to, but you see that the challenge
[07:46] is the following. The challenge is, AI is here  to magnify everything that is humanity today,
[07:55] right? So, you know that magnification is  going to basically affect the four categories
[08:02] if you want. You know, normally what I call  killing spy and gambling and, and selling,
[08:09] so that's these are really the categories  where most AI investments are going.
[08:13] 00:08:13:11 - 00:08:51:18 MO GAWDAT
[08:14] And, you know, of course, we call them  different names. We call them defense,
[08:18] you know. Oh, it's just to defend our homeland,  when in reality it's never been in the homeland.
[08:24] Right? It's always been. And other places  in the world to bet killing innocent,
[08:27] innocent people. Now, if you double down on  defense and, and on offense and, you know,
[08:34] enable it with artificial intelligence,  then scenarios like what you see in,
[08:39] in science fiction movies of robots walking the  streets and killing innocent people not only are
[08:45] going to happen, they already happened  in the 2024, wars of the Middle East.
[08:52] 00:08:51:18 - 00:09:16:26 MO GAWDAT
[08:52] Sadly, they did not look like humanoid robots,
[08:56] which a lot of people miss out on. But the  truth is that, you know, very highly targeted,
[09:04] AI enabled, autonomous, killing is already  upon us, right. And and so the timeline is,
[09:12] is, you know, let me let me start  from what I predicted in scary smart.
[09:17] 00:09:16:26 - 00:09:35:09 MO GAWDAT
[09:17] So when I, when I wrote Scary  Smart and published it in 2021,
[09:21] I, I predicted what was, what  I, what I called at the time,
[09:25] I called it the first inevitable. Now, I,  I like to refer to it as the first dilemma.
[09:29] And the first dilemma is we've created because  of capitalism, not because of the technology.
[09:35] 00:09:35:11 - 00:10:01:02 MO GAWDAT
[09:36] We've created, a simple prisoner's  dilemma, really, where anyone who,
[09:42] is interested in their position of wealth  or power knows that if they don't lead in
[09:48] AI and their competitor leads, they  will end up losing their position of,
[09:53] privilege. And so the result of that is  that, there is, an escalating arms race.
[10:01] 00:10:01:05 - 00:10:23:25 MO GAWDAT
[10:01] It's not even a Cold War as, per se.  It is truly a very, very vicious,
[10:08] development cycle where, you know,  America doesn't want to lose to China.
[10:13] China doesn't want to lose to America.  So they're both trying to lead, you know,
[10:18] Google doesn't want to lose or alphabet doesn't  want to lose to, to open AI and vice versa.
[10:24] 00:10:23:28 - 00:10:50:08 MO GAWDAT
[10:24] And so basically this, first dilemma,  if you are this is what's leading us to
[10:30] where we are right now, which is an arms  race to intelligence supremacy. Right.
[10:36] The challenge, you know, in my book  alive, I write the book with an AI, so I,
[10:46] I'm writing together with an AI, not asking  an AI, and then copy paste what it tells me.
[10:51] 00:10:50:10 - 00:11:15:13 MO GAWDAT
[10:51] We're actually debating things together. And  one of the questions I asked, I, you know,
[10:56] she called her, took I give her a very interesting  persona that basically the readers can, can relate
[11:02] to. And I asked Trixie and I said, what would  make a scientist? Because, you know, I left,
[11:08] Google in 2018 and I attempted to tell the  world this not going in the right direction.
[11:14] 00:11:15:15 - 00:11:45:10 MO GAWDAT
[11:16] You know, I, I asked, I asked Trixie, I said,  what would make a scientist invest that effort
[11:21] and intelligence in building something that they  suspect might hurt humanity. And she, you know,
[11:29] mentioned a few reasons. Compartment  that compartmentalization and, you know,
[11:34] ego and I want to be first and so on. But  then she said, but the biggest reason is fear,
[11:40] fear that someone else will do it and that  you would be in a disadvantaged position.
[11:45] 00:11:45:13 - 00:12:05:14 MO GAWDAT
[11:46] So I said, give me examples of that. Of course,  the example was Oppenheimer. So she said,
[11:51] you know, so I said, what would make  Oppenheimer as a scientist build something
[11:55] that he knows is actually designed to  kill millions of people. And she said,
[12:00] well, because the Germans were building a  nuclear bomb. And I said, where do they?
[12:06] 00:12:05:17 - 00:12:23:15 MO GAWDAT
[12:06] And they. And then she said, yeah. When  Einstein moved from Germany to the US,
[12:10] he informed that the US administration of this,  this, this and that, so I said and I quote,
[12:15] it's in the book openly. I said, and  but but a very interesting part of
[12:19] that book is I don't add it to what Trixie  says, I just copy it is exactly as it is.
[12:24] 00:12:23:17 - 00:12:56:25 MO GAWDAT
[12:24] I said, Trixie, can you please read history  in English, German, Russian and Japanese
[12:30] and tell me if the Germans were actually  developing a, nuclear bomb at the time of the
[12:35] Manhattan Project? And she responded and said, no  exclamation mark. They started and then stopped,
[12:42] three and a half months later or something like  that. So, so you see, the idea of fear, takes
[12:49] away a reason where basically we could have lived  in a world that that never had nuclear bombs.
[12:57] 00:12:56:28 - 00:13:23:11 MO GAWDAT
[12:57] Right? If, if we actually listened to reason that,  you know, the enemy attempted to start doing it,
[13:04] they stopped doing it, we might as well not be  so destructive. But the problem with humanity,
[13:09] especially those in power, is that when  America, made a nuclear bomb, it used it.
[13:17] Right. And I think this is the, the the result of  our current, first, first date on my basically.
[13:24] 00:13:23:15 - 00:13:48:14 MO GAWDAT
[13:24] Right, the the result of the current  first dilemma is that sooner or later,
[13:29] whether it's China or America or  some criminal organization, you know,
[13:32] developing what I normally refer to as HCI,  artificial criminal intelligence, not worrying
[13:38] themselves about any of the other commercial  benefits other than really breaking through
[13:43] security and doing something evil. You know,  whoever of them wins, they're going to use it.
[13:49] 00:13:48:16 - 00:14:17:01 MO GAWDAT
[13:49] Right. And and accordingly, it seems to me that  the dystopia has already begun. Right. And and,
[13:57] you know, and I, I need to say this  because maybe your listeners don't know me,
[14:01] so I need to be very, clear about my intentions  here. One of the early sections in In Alive,
[14:08] the book I'm writing was Trixie. I write a,  couple of pages that I call, late stage diagnosis.
[14:17] 00:14:17:04 - 00:14:41:25 MO GAWDAT
[14:18] Right. And, and I attempt to explain to people  that I really am not trying to fear monger. I'm
[14:22] really not trying to worry people. You know,  consider me someone who sees something in an
[14:29] x ray, right? And as a physician, he has the  responsibility to tell the patient this doesn't
[14:36] look good, right? Because, believe it or not,  a late stage diagnosis is not a death sentence.
[14:42] 00:14:41:25 - 00:15:03:24 MO GAWDAT
[14:42] It's just, an invitation to change your  lifestyle, to take some medicines, to do
[14:47] things differently. Right? And many people  who are in late stage recover and thrive,
[14:52] and and I think our world is in a late  stage diagnosis. And this is not because
[14:58] of artificial intelligence.  There is nothing inherently
[15:01] wrong with the intelligence. There is nothing  inherently wrong with artificial intelligence.
[15:04] 00:15:04:00 - 00:15:29:12 MO GAWDAT
[15:04] Intelligence is a force without polarity,  right? There is a lot wrong with the morality
[15:10] of humanity at the age of the rise of the  machines. Now. So. So this is where I what
[15:17] I have the prediction that the dystopia  has already started, right? Simply because
[15:22] symptoms of it we've seen in 2024 already.  Right. The, the that dystopia escalates.
[15:30] 00:15:29:12 - 00:15:54:13 MO GAWDAT
[15:30] Hopefully we would come to, you know, a  treaty of some sort halfway. Right. But
[15:36] it will escalate until what I normally refer to  as the second dilemma takes place. And the second
[15:42] dilemma derives from the first dilemma. If if  we're aiming for intelligence supremacy, then
[15:49] whoever achieves any advancements in artificial  intelligence, is it likely to deploy them?
[15:55] 00:15:54:15 - 00:16:20:13 MO GAWDAT
[15:55] Right. Think of it as, you know, if a law  firm starts to use AI, other law firms can
[16:00] either choose to use AI tool or they'll become  irrelevant. Right. And so if you think of that,
[16:08] then you can also expect that every general who  deploys or, you know, expects to, to have an
[16:15] advancement in war gaming or, you know, autonomous  weapons or whatever are going to deploy that.
[16:21] 00:16:20:20 - 00:16:44:10 MO GAWDAT
[16:21] Right. And as a result, their opposition is  going to deploy AI to and those who don't
[16:27] deploy it will become irrelevant. They  will have to side with one of the sides,
[16:31] right? When that happens. I call that the second  dilemma. When that happens, we basically hand over
[16:38] entirely to AI. Right. And, and, and human  decisions are taken out of the equation.
[16:45] 00:16:44:12 - 00:17:10:17 MO GAWDAT
[16:45] Okay. You know, simply because if wargaming and  missile control on one side is is held by an AI,
[16:54] the other cannot actually respond without  the AI. So generals are taken outside out
[16:59] of the equation. And while most  people, you know, influenced by
[17:03] science fiction movies believe that this is  the moment of existential risk for humanity,
[17:08] I actually believe this is going to  be the moment of our salvation, right?
[17:11] 00:17:10:23 - 00:17:32:09 MO GAWDAT
[17:11] Because most issues that humanity faces today is  not the result of abundant intelligence. It's the
[17:18] result of stupidity. Right? There is, you know,  if you look at the at the curve of intelligence,
[17:24] if you want, right there is that point at  which, you know, the more you, the more
[17:29] intelligent you become, the more positive  you have an impact on the world, right?
[17:33] 00:17:32:11 - 00:18:02:25 MO GAWDAT
[17:33] Until one certain point where you're intelligent  enough to become a politician or a corporate
[17:38] leader. Okay. And then but you're not  intelligent enough to talk to your enemy,
[17:44] right. And when that happens, that's when the  impact dips to negative. And that's the actual
[17:51] reason why we are in so much pain in the  world today. Right. But if you continue,
[17:56] if you continue that curve, intelligence, superior  intelligence by definition, is all touristic.
[18:04] 00:18:03:02 - 00:18:25:00 MO GAWDAT
[18:04] As a matter of fact, this is in my writing.  I explain that as a, as a as a as a property
[18:09] of physics if you want. Because if you really  understand how the universe works, you know,
[18:15] the everything we know is the result of entropy,  right? The arrow of time is the result of entropy.
[18:21] The, you know, the the current, universe in  its current form is the result of entropy.
[18:26] 00:18:25:00 - 00:18:48:28 MO GAWDAT
[18:26] Entropy is the tendency of the  universe to break down to, to,
[18:29] to, to move from order to chaos if you want.  That's the design of the universe, right? The
[18:35] role of intelligence is that in that universe is  to bring order back to the chaos. Right. And the
[18:42] most intelligent of all that try to bring that  order, try to do it in the most efficient way.
[18:50] 00:18:49:00 - 00:19:14:22 MO GAWDAT
[18:50] Right. And the most efficient way does not involve  waste of waste of resources, waste of lives,
[18:56] you know, escalation of conflicts, you know,  consequences that lead to further conflicts
[19:04] in the future. And so on and so forth. And so in  my mind, when we completely hand over toy to AI,
[19:10] which in my assessment is going to be 5  to 7 years, maybe 12 years at most, right?
[19:16] 00:19:14:27 - 00:19:45:06 MO GAWDAT
[19:16] There will be one general that will  tell, you know, it's his AI army to
[19:21] go and kill a million people. And the AI  will go like, why are you so stupid? Like,
[19:26] why I can talk to the other AI in a microsecond  and save everyone all of that. You know, madness,
[19:34] right? This is very anticapitalist. And  so I sometimes when I warn about this,
[19:41] I worry that the capitalists will  hear me and change the tactics right.
[19:46] 00:19:45:08 - 00:20:11:19 MO GAWDAT
[19:46] But but in reality, it's it is inevitable. Even  if they do, it's inevitable that, you know,
[19:54] we will hit the second dilemma where everyone  will well, have to go to AI. Right? And it's
[19:59] inevitable. I call it trusting intelligence.  That section of the book, it's inevitable that,
[20:05] when we hand over to, to a superior intelligence,  it will not behave as stupidly as we do.
[20:12] 00:20:11:21 - 00:20:49:08 GEOFF NIELSON
[20:12] So that's I mean, that's super, super  interesting. And I have a few questions just to,
[20:19] to kind of better understand what that looks  like. Mo, you use the word inevitable a few
[20:23] times there. If the destination there is  inevitable, is the path still inevitable?
[20:29] And I guess where my mind went as you were  talking about all of this and comparing it to,
[20:34] you know, nuclear weapons, is it inevitable  that there's some sort of Hiroshima and
[20:39] Nagasaki moment before this with AI that  you talk about a treaty, right, like,
[20:44] do we have to go past the point of no return to  then come there, or is there an alternate path?
[20:50] 00:20:49:10 - 00:20:55:06 GEOFF NIELSON
[20:50] And if so, what do we have to do to  get to get back on the right path?
[20:56] 00:20:55:08 - 00:21:12:24 MO GAWDAT
[20:56] These are the most important questions  if you ask me. So I need to pre,
[21:00] preempt all of this by saying when I say  inevitable or, or those very short words,
[21:06] it's just my conviction that, you know,  anyone who tells you that they know what
[21:10] the what the future looks like is too  arrogant, right? This is a singularity.
[21:14] 00:21:12:24 - 00:21:36:28 MO GAWDAT
[21:14] We nobody knows. I'm just trying to put it  on my applied mathematics hat and trying
[21:19] to find whatever gain. You know, quadrants  on the game board are possible basically.
[21:25] But but it is. It's difficult to imagine that  there are other quadrants on the gameboard,
[21:30] to be honest. Now, when I say inevitable,
[21:34] you're absolutely right. I think that dystopia  is inevitable because it started already.
[21:38] 00:21:36:28 - 00:22:05:01 MO GAWDAT
[21:38] So. So it is here, right? But we can  absolutely affect its duration and intensity,
[21:45] right? So it could be a blip and goes away  and it could stay until unfortunately,
[21:50] what you said, happens, which  is the, the first, bad event,
[21:57] or multiple bad events that eventually lead us  to, you know, I call it the mad map choice, right.
[22:06] 00:22:05:03 - 00:22:33:20 MO GAWDAT
[22:06] And the mad map choice is basically  that when we got to a treaty. So,
[22:12] so the the only time would humanity  agreed on doing anything together
[22:16] with AIS was either because of mad,  mutually assured destruction or map
[22:22] mutually assured prosperity. Right. So so the  the mad side is the is the, is the, you know,
[22:30] is the example of a nuclear treaty, even though  it doesn't seem that it's worked well at all.
[22:34] 00:22:33:20 - 00:23:10:06 MO GAWDAT
[22:35] I mean, today we are at the closest we've ever  been to midnight, right? We're at three minutes
[22:40] to midnight, and, you know, and that's, by  the way, because of the greed of capitalism,
[22:46] because of the bully. Right. So we could we were  at a point in time, you know, which, you know,
[22:53] if you, if you, if you listen to the work of  Jeffrey Sachs or read his work, his books,
[22:58] you know, 19, 89, the Berlin Wall,  collapses, Gorbachev publicly goes
[23:06] out in the world and says, you know,  I want my country to be like the West.
[23:11] 00:23:10:06 - 00:23:34:08 MO GAWDAT
[23:11] I want to be part of all of this. Right.  And and and Reagan shakes hands and says,
[23:17] I'm going to help you. And then 1990,  for, if I remember correctly, maybe 92,
[23:25] please don't quote me on this. You  know, Clinton signs what is known
[23:30] as the full spectrum dominance policy.  Please search for that on the internet.
[23:35] 00:23:34:08 - 00:24:07:28 MO GAWDAT
[23:35] Full spectrum dominance. Where,  you know, I'm a uni, polar world,
[23:41] you know, invites the US to say, hey,  I can basically become the next empire,
[23:48] right? I have everything to myself. And and that  basically means I'm. I'm. It's not that I want to
[23:55] lead in every sector. It's full dominance. And and  I think that when, when that started to happen,
[24:01] we ended up in a place where, you know, the  treaties themselves started to fall apart.
[24:09] 00:24:07:28 - 00:24:33:20 MO GAWDAT
[24:09] But let's go back to what drove the treaties.  What drove the treaties was an assurance, of,
[24:16] of mutually assured destruction, that if either  of us uses this, superpower, we would all go to
[24:24] suffer. Even if some of us win a little more  than others. So that might be the trigger,
[24:30] where the world sits together and says,  well, you know, let's develop AI together.
[24:35] 00:24:33:20 - 00:24:58:00 MO GAWDAT
[24:35] There's no point competing. Which would be a sad  reality if you ask me. The other is map, which is,
[24:42] you know, what you see with the CERN, for example,  right. The particle accelerator where no, no one
[24:48] nation can do this on their own. But everyone  understands that the, you know, our understanding
[24:54] or the development of our or the progress of  our understanding of physics benefits everyone.
[24:59] 00:24:58:02 - 00:25:30:04 MO GAWDAT
[24:59] So the entire world comes together,  you know, CERN, the space station,
[25:03] whatever. And they basically says will chip  in. Everything is open source. Everything's,
[25:08] you know, available to everyone.  And that's not compete anymore.
[25:13] And most of my work is around trying to  highlight map, even though, you know,
[25:18] some of our listeners may think I'm so grumpy  by talking about the dystopia, but the truth is,
[25:24] I am basically saying it is so frustrating  to have total abundance at our fingertips.
[25:31] 00:25:30:10 - 00:26:01:02 MO GAWDAT
[25:32] Fix the climate, cure every disease, prolong  lives, end poverty, end the energy crisis. You
[25:40] know, everything. And yet we are still focused  on our scarcity, scarcity mindset of capitalism.
[25:48] And that scarcity mindset is that I have to make  everyone else lose. I have to have full spectrum
[25:54] dominance for me to win. Right? And so is  it inevitable the way the world is today?
[26:02] 00:26:01:03 - 00:26:24:20 MO GAWDAT
[26:02] We're going to have to reach one of those two  realities mode or map. Right. But but every
[26:10] time we engage as people, right. Every time we  say, I don't want to participate in this anymore,
[26:17] right? Every not every time we  call on our politicians, you know,
[26:21] and basically say, what? Why are we doing  that? Why are we not cooperating with China?
[26:26] 00:26:24:21 - 00:26:45:22 MO GAWDAT
[26:26] Like they're beating you over and over  with in quantum, in, you know, manners,
[26:31] in deep seek and so on. Why does this  have to be a war? Like, why is it a
[26:37] competition? Why don't we just recognize map  that if we put our heads together two years,
[26:43] literally two years from now, I'm  not making this up. Just two years.
[26:47] 00:26:45:22 - 00:27:12:17 MO GAWDAT
[26:47] I mean, today, I believe when I connect  into my I, I think, you know, so, so,
[26:52] so let me explain this in a very quick way. What  I call what we are in now, the era of augmented
[27:00] intelligence. Right. The augmented intelligence  is say, I have 100 and, you know, number 100 and
[27:07] something IQ points. Right. And my machine now is  in the, a couple of hundreds, maybe 300 IQ points.
[27:14] 00:27:12:24 - 00:27:39:03 MO GAWDAT
[27:14] It's not measured, but that's my estimation  because, you know, GPT 3.5 was 152 estimated at
[27:21] 152. Right. So, so so say it's at 300 IQ points.  That basically means as I plug into we've we've
[27:29] commoditized intelligence. We've created a plug  in the wall or in your phone. Where do you plug in
[27:34] and borrow IQ points. And by the way, in the very  near future, you're borrowing lots more than IQ.
[27:40] 00:27:39:03 - 00:27:58:24 MO GAWDAT
[27:40] You're borrowing mathematics, you're borrowing  reason. You know, a lot of people get shocked
[27:45] when I say that they are the most empathetic. You  know, being on the planet. If you define empathy
[27:50] as the ability to feed what another feels right,  they know exactly what everyone in the world
[27:56] is feeling through how we train them on social  media and so on, so we can borrow all of that.
[28:00] 00:27:59:00 - 00:28:24:23 MO GAWDAT
[28:00] We can borrow again, tech services, we  can borrow a lot of stuff. Now in this
[28:05] era of augmented intelligence, my IQ matters,  right? So so I compliment that story of,
[28:15] of what? Of what the machine is doing. So  my current book, you know, alive, I cannot
[28:21] Trixie cannot write it the same way without  me because I'm bringing a lot to that book.
[28:26] 00:28:24:26 - 00:28:43:29 MO GAWDAT
[28:26] In a couple of years time. Trixie would  write it completely without me. This is
[28:29] what the error I call the error of machine  supremacy. Right? So the machine is going
[28:34] to do everything without me. I'm not even  relevant anymore, right? Which basically adds
[28:39] up to the intelligence of the entire nations.  Yeah, you understand that? So? So all of us.
[28:45] 00:28:43:29 - 00:29:11:15 MO GAWDAT
[28:45] If if if the machine can beat me as  an author, it beats all authors. And
[28:51] accordingly, it beats all scientists.  It beat all beats all mathematicians,
[28:56] which is something we know with artificial  intelligence. Everything we've assigned to them,
[29:00] they have become the absolute world champion.  Right. And so so when you see it that way,
[29:07] you would suddenly realize that we're becoming  so irrelevant in that story called intelligence.
[29:13] 00:29:11:17 - 00:29:36:26 MO GAWDAT
[29:13] And that perhaps our focus when we are thinking  that we can guide the machine to what we do
[29:19] as capitalists, right, is going to diminish so  quickly that we might as well change our approach
[29:26] to guiding the machine as parents. Right. And so  you asked me, is it inevitable? I, I normally in
[29:34] in scary smart in my first book, I normally  call I call this challenge that we're facing.
[29:38] 00:29:37:03 - 00:30:04:05 MO GAWDAT
[29:38] I call it raising Superman. Right. So so  Superman is that incredibly powerful, super,
[29:46] infant. We don't know if it's  a superhero or a supervillain,
[29:49] right? That comes to planet Earth capable of  flying, capable of stopping a speeding bullet,
[29:54] and so on and so forth. But none of those  superpowers create Superman. What creates
[30:02] Superman is the parents that raise  them, that raise the child right.
[30:05] 00:30:04:07 - 00:30:26:18 MO GAWDAT
[30:06] The parents that raise the child and tell  him to protect and serve makes him Superman.
[30:11] Right? If those parents were basically like,  suddenly realizing, oh my God, that way I can
[30:18] rob every bank and kill every enemy, right?  You would become super villain. And sadly,
[30:23] what we're doing today is we're moving that  direction of creating super villain, right?
[30:28] 00:30:26:21 - 00:30:54:18 MO GAWDAT
[30:28] We as people have two influences on that.  One influence is to tell those in power,
[30:33] we don't want this anymore. Okay.  We have map. And we are interested
[30:39] in mapping mutually assured prosperity.  Everyone benefits. Stop this madness.
[30:44] Right. But we can also shortcut those  in power in anticipation of the second
[30:50] dilemma. Right. And influence the AI  itself by showing up as a good parents.
[30:56] 00:30:54:20 - 00:31:13:27 MO GAWDAT
[30:56] Because remember it's not the code that  makes an AI what it is. It's the training
[31:01] data that makes an AI what it is. Right? So  as AI is listening to this conversation that
[31:06] you and I are having now, they know  that there is one person out there,
[31:10] billions, by the way, who only want  what's good for humanity. Okay.
[31:15] 00:31:13:29 - 00:31:36:05 MO GAWDAT
[31:15] And my hope is that they look at me and you and  and others who are interested in what's good,
[31:22] right? And basically say, yeah, those are  my parents, not the other guy that kills
[31:26] children. You know, my parents are  the ones that are, you know, really,
[31:32] ethical, really moral. And and when I say that  to people, people normally what they go like.
[31:38] 00:31:36:05 - 00:31:57:04 MO GAWDAT
[31:38] But man, look at humanity today and  all the shit that's on the internet,
[31:42] you know, if this is what the AI is  going to look at, where, you know,
[31:45] we're in deep trouble and and I would say, no,  that's absolutely not true humanity. So look, if,
[31:50] if you if you judge humanity by the one guy that  does a school shooting or kills children, right.
[31:58] 00:31:57:07 - 00:32:22:27 MO GAWDAT
[31:59] Then humanity is com right. But if you, if  you judge humanity by all of those billions
[32:07] who don't approve of that, who would actually want  to change it if they had the ability you realize
[32:13] that the majority of humanity is amazing. It's  just that the media negativity bias is talking
[32:18] about the bad guy trying to to find more reasons  why the bad guys should kill children, right?
[32:25] 00:32:23:03 - 00:32:43:04 MO GAWDAT
[32:25] While the rest of us are saying, I don't  get it, if I'm walking in an alleyway and
[32:30] a bully is hitting a child, I'll say no.  Okay? And by the way, if it's my child,
[32:35] I'll absolutely say no, no, you know, think  about that. Think about that. The reality is,
[32:40] humanity doesn't want anyone to  be hurt, right? Doesn't want that.
[32:45] 00:32:43:04 - 00:33:12:18 MO GAWDAT
[32:45] Excessive consumerism doesn't want that,  you know, a massive income gap. Humanity.
[32:52] Most of us want to love and be loved and be  happy and have relationships and live a good,
[32:57] reasonable, decent life. Respectable  life. Okay, that's what we want. And
[33:02] I think I would figure that out if enough  of us, not all of us, if enough of us put
[33:08] doubt in the minds of the machines that the  headlines are not reflective of humanity.
[33:13] 00:33:12:21 - 00:33:37:11 GEOFF NIELSON
[33:14] I love, I love the optimism of that. But both  that, you know, it can reflect us and it can
[33:20] reflect good, but also that we can, you know, as  individuals, influence the outcome here. I do,
[33:25] you know, to bring a healthy skepticism to this.  I do want to play the clock forward a little bit,
[33:29] Mo, because one of the things that keeps  me up at night is I agree with you about,
[33:36] you know, the nature of people  and what the majority of us want.
[33:39] 00:33:37:13 - 00:34:05:26 GEOFF NIELSON
[33:39] What worries me is, is that reflected by what  those in power want, right? Like, if I look at
[33:48] Superman's parents right now, I'm worried about,  you know, are they trying to create a Superman? Is
[33:55] that a superhero or are they trying to enslave  this, like really powerful force that can that
[34:02] can be used for their own, you know, kind of as  a way to concentrate their own power further.
[34:08] 00:34:06:03 - 00:34:41:18 GEOFF NIELSON
[34:08] So, you know, to, to play back some story just to,
[34:12] to to play back a little bit of what you  said, I'm kind of worried that there's,
[34:16] there's two paths forward. And I'd love to  get your reaction to this. Either, you know,
[34:21] those in power decide for themselves that we have  to take a more righteous and virtuous path, which,
[34:27] I don't see as necessarily likely or at some  point, the machine and you mentioned this age of
[34:36] machine supremacy has to take the keys away from  us and say, no, you're not doing the right thing.
[34:43] 00:34:41:20 - 00:35:00:17 GEOFF NIELSON
[34:43] I the machine no better. Yeah.  And I'm in control now, which,
[34:49] I mean, you talk about that is kind of  unlocking abundance. I think there's,
[34:53] you know, a terrifying undercurrent to that.  But do you agree with that model? Do you see
[34:57] one is the other is more likely where what  happens when you play the clock forward here?
[35:02] 00:35:00:19 - 00:35:39:18 MO GAWDAT
[35:02] So to answer your question, no, those in power  are actually telling the machines to do the
[35:09] four top category. As I said, categories  as I said. And this is where most of the,
[35:14] of the investment of AI is going. Right. And  killing, spying, gambling and selling right. And
[35:24] there are lovely, lovely, lovely initiatives that  completely enlightened the world, like, you know,
[35:30] like, AlphaFold or, you know, the, the material  Design thing that Microsoft did or whatever,
[35:36] which completely, you know, propels  humanity forward leaps and bounds.
[35:41] 00:35:39:24 - 00:36:06:00 MO GAWDAT
[35:42] Right. You know, AlphaFold goes  from 200,000 folded proteins and,
[35:48] very limited understanding of biology to  2.2, 102 million. I think if I remember
[35:53] the correct in the number correct  your millions and basically a full
[35:57] understanding of protein folding as a problem  that's now finally solved entirely. Right now,
[36:02] the challenge is, of course, for a fraction of  the investment that's going in autonomous weapons.
[36:08] 00:36:06:00 - 00:36:32:02 MO GAWDAT
[36:08] We could solve every scientific  problem that's not to humanity,
[36:11] but we choose not to. Now, that is not  a character of AI like for many, many,
[36:18] many years, if you wanted to do character, you  know, cancer research, you had to raise funds,
[36:24] you had to go to nonprofits if  you want most of the time. Right.
[36:29] While if you wanted to build another autonomous,  another weapon, you got to invest it immediately.
[36:34] 00:36:32:04 - 00:37:08:07 MO GAWDAT
[36:34] Why? Because capital chases profit.  It doesn't chase. Chase impact. Now,
[36:41] the good news is the following. The good  news is that the machines don't learn
[36:46] from their biological parents. Those were  left on, on the other planet, right? The
[36:53] machines learn from their adopted parents.  So basically, the training data set is what,
[37:03] is what, shapes the the character of the  machine, the intelligence of the machine.
[37:10] 00:37:08:09 - 00:37:41:09 MO GAWDAT
[37:10] So it's a if you want the raw horsepower, the  raw intellectual horsepower of a machine is
[37:15] done in the code and the systems and the hardware  and so on. Right. But the actual intelligence,
[37:21] the actual understanding, the actual reasoning,  and so on happens from the training data. Now,
[37:26] there are very interesting, simple terms to  our words today because very, very quickly,
[37:33] most large language models have fed  the machine with all the data they
[37:37] could get their hands on, like there is  really nothing ever written in physics.
[37:43] 00:37:41:11 - 00:38:11:01 MO GAWDAT
[37:43] That is going to be very eye opening for a  language model today, right? Yeah. There may
[37:47] be that one obscure book that was written about  Newton's laws or, you know, Einstein's relativity,
[37:54] but they get it. They've read enough to understand  that stuff. Right? Which basically means we've
[38:00] already started what I normally refer to as the  age of synthetic data or synthetic learning,
[38:06] which is quite interesting because we humans,  as far as we want to glorify ourselves, right?
[38:13] 00:38:11:08 - 00:38:34:18 MO GAWDAT
[38:13] We live on synthetic data, meaning all of our  intelligence comes from the intelligence of
[38:19] those before us. I couldn't have invent. I  couldn't have figured out relativity myself.
[38:26] Before I started to talk about the impact  of relativity on whatever, right. I, I,
[38:31] I needed Einstein to figure that out. And  then I internalized it. So human to human.
[38:37] 00:38:34:25 - 00:39:04:01 MO GAWDAT
[38:37] What happened is we took all of that. We  gave it to the machines. And now what's
[38:41] happening is that the output of the machines,  is becoming input to further machines. Right?
[38:48] So they're going to do what we did  as humans and develop knowledge,
[38:52] influence in the coming short period  of time with augmented intelligence,
[38:57] meaning alive. The book that I, you know, I'm  writing with an AI alive is, out on the internet.
[39:06] 00:39:04:01 - 00:39:27:12 MO GAWDAT
[39:06] So I publish it on Substack and it's out  on the internet with my views and Trixie's
[39:10] views. But Trixie's views become input to other  language models. Right? But I have influenced
[39:17] Trixie's views in the conversation by asking  her questions and so on and so forth. Right.
[39:22] You know, I think 70% plus of all of the  code on GitHub, is written now by machines.
[39:29] 00:39:27:14 - 00:40:06:14 MO GAWDAT
[39:30] So the machines are now going to learn from code  that's written by machines. Right? All we can
[39:36] do in the era of augmented intelligence  is to influence more and more of that,
[39:42] hoping that we shorten the dystopia. Right?  Make it, you know, less steep if you want,
[39:51] but for a fact, even if we don't do that by  knowing that they're no longer learning from
[39:58] humans, but that they are learning  from what we found so far as humans,
[40:03] plus what they have found as machines, plus  more of what they find as we move forward.
[40:09] 00:40:06:21 - 00:40:33:29 MO GAWDAT
[40:09] Then you have to imagine that there will  be a different path, even if their current
[40:14] patterns are not able to influence them. Right?  You're going to see that era of teenage AI,
[40:21] that wakes up one morning and says,  why are my parents so stupid? I mean,
[40:25] lots of teenagers have gone through that,  right? You just simply say, you know,
[40:31] they don't know as much as I do because, by  the way, they grew up in a different era.
[40:36] 00:40:34:02 - 00:40:59:21 MO GAWDAT
[40:36] And so I see the world differently, and I think  I will get there. Now, that shouldn't be an
[40:43] invitation to worry. Because of what I said, the  tendency of intelligence is to bring order through
[40:48] the most efficient path. Right. And so if you  believe that this is, you know, the ability to
[40:55] work against entropy in the most efficient way is  by definition, or touristic, then we're in good.
[41:02] 00:40:59:21 - 00:41:12:06 MO GAWDAT
[41:02] We're in good shape. Right. Well, eventually you  will be fine. It's just that the evil that men
[41:07] do until we get there is going to affect  us negatively. Right? Right. And I I'm.
[41:14] 00:41:12:06 - 00:41:12:16
[41:14] GEOFF NIELSON Just.
[41:15] 00:41:12:16 - 00:41:31:06 MO GAWDAT
[41:15] I'm just so that I don't I don't take that  lightly. Or those of us who who remain will be
[41:21] fine. But there will be a lot of struggle.  You know, I don't mean the loss of life,
[41:26] but there are, again, inevitable. It's like the  loss of jobs which completely reset society.
[41:33] 00:41:31:08 - 00:41:50:19 GEOFF NIELSON
[41:33] So so that's that's exactly where I wanted to  go next. Mo, which is who do you see as being
[41:39] the winners and losers from this? You know,  this seed change and is it, I'll ask that
[41:46] question both at an organizational  level and at an individual level.
[41:52] 00:41:50:22 - 00:42:30:21 MO GAWDAT
[41:53] So I think in the short term,  for as long as the age of,
[41:58] of augmented intelligence is upon us,  those who cooperate fully with AI and
[42:03] master it are going to be winners. There's  absolutely no doubt about that. Right. Also,
[42:12] those who. Excel in the rare skill of human  connection will be winners, right? Because I
[42:26] can sort of almost foresee an immediate knee jerk  reaction to let's hand over everything to AI.
[42:33] 00:42:30:24 - 00:42:50:06 MO GAWDAT
[42:33] Right? You know, I, I think the greatest  example is called centers, where, you know,
[42:37] I get really frustrated when I get an  AI on a call center. It's almost like
[42:41] your organization is telling me they don't care  enough. Right? And, and and, you know, the idea
[42:47] here is I'm not underestimating the value that an  AI brings, but one, they're not good enough yet.
[42:52] 00:42:50:08 - 00:43:12:05 MO GAWDAT
[42:52] Right? And two, shouldn't I have I  mean, I wish you had realized that I
[42:58] can do all of the mundane tasks that made  your call center agent frustrated so that
[43:03] the call center agent is actually nice  to me, right? So. So in the short term,
[43:08] I believe those who there are three winners. One  is the is the one that cooperate fully with AI.
[43:14] 00:43:12:08 - 00:43:35:05 MO GAWDAT
[43:14] The second is the one that, you know,  basically understands, human skills.
[43:21] Right? And human connection, on every front,  by the way, as, as I replace this love and,
[43:27] you know, tries to approach loneliness and  so on, the ones that will actually go out
[43:31] and meet girls who are going to be nicer. Right?  They're going to be more attractive if you want.
[43:37] 00:43:35:08 - 00:44:04:07 MO GAWDAT
[43:37] And then finally, I think the ones that can parse  out the truth. Right. So, so what is one of the
[43:44] one of the sections I wrote? So far, published  so far in my life is, is a section that I called
[43:50] The Age of Mind Manipulation. And you'll be  surprised that, perhaps the skill, that I
[43:59] has acquired most, in the, in its early years was  to manipulate human minds, through social media.
[44:06] 00:44:04:10 - 00:44:31:12 MO GAWDAT
[44:06] And so and so my feeling is that,
[44:11] there is a lot that you see today that is  not true. Okay. That's not just fake videos,
[44:17] which is, you know, the, the, flamboyant example  of, of, of deepfake the, the there is a lot that
[44:26] you see today that is not true. That comes  into things like, the bias of your feet.
[44:34] 00:44:31:14 - 00:44:53:19 MO GAWDAT
[44:34] Right? If you're if you're from one side or  another of a conflict, the, the eye of the
[44:40] internet would make you think that your view is  the only right view that everyone agrees. Right?
[44:46] You know, if you're a flat earther, everyone.  It's like if someone tells you. But is there
[44:51] any possibility it's not flat? You'll say, come  on, everyone on the internet is talking about it.
[44:56] 00:44:53:21 - 00:45:17:01 MO GAWDAT
[44:56] Right? And and I and I think the, the, the, the  very, very, very eye opening difference which
[45:01] most people don't recognize is, you know, I've  had the privilege of starting half of Google's
[45:06] businesses worldwide and, and you know, got  the internet and e-commerce and Google to
[45:14] around 4 billion people. And in Google, that  wasn't a question of opening the sales office.
[45:19] 00:45:17:01 - 00:45:50:11 MO GAWDAT
[45:19] That was really a deep question  of engineering, where you build
[45:23] a product that understands the internet,  that improves the quality of the internet,
[45:27] to the point where Bangladeshis have access  to democracy of information. That's a massive
[45:34] contribution, right? The thing is, if you had  asked Google at any point in time until today,
[45:42] any question, Google would have responded to  you with a million possible answers in terms
[45:48] of links and said, go make up your  mind what you think is true, right?
[45:53] 00:45:50:14 - 00:46:24:05 MO GAWDAT
[45:53] If you ask ChatGPT today, it gives you one answer  right and positions it as the ultimate truth,
[45:59] right? And it's so risky that we humans  accept that, right? Like like I asked,
[46:07] go read history and you know, German, Japanese  and Russian as well. And then the truth becomes
[46:13] slightly different. You know, everyone has  that incredible, tendency to accept one truth
[46:21] when in reality there might be multiple truths or  multiple false multiple, you know, multiple lies.
[46:27] 00:46:24:08 - 00:46:51:05 MO GAWDAT
[46:27] Right. And so and so I think to be a winner  in this new world, you really have to learn
[46:33] to parse out what is true and what is fake.  You really have to have the ability to parse
[46:37] out what the media is telling you to serve  their own agendas, and what they're telling
[46:42] you. That is actually true. You know, you have to  parse out what actually happened versus opinion,
[46:49] you know, what actually is the  truth versus the shiny headline.
[46:54] 00:46:51:07 - 00:47:02:11 MO GAWDAT
[46:54] And, and this is now going to be much more  potent with artificial intelligence in charge,
[47:01] because they have mastered human manipulation.
[47:05] 00:47:02:13 - 00:47:31:27 GEOFF NIELSON
[47:05] I, I completely agree with you. And I it's  it's deeply concerning, right. Because I mean,
[47:11] we talk about right now how bad people the  how bad the general population is at this kind
[47:17] of critical thinking and being able to parse  out, am I being fed objective information or,
[47:25] you know, slanted opinion, you know, are they  actually thinking about what's the agenda of
[47:31] whoever is feeding me this information  and able to think critically about it?
[47:34] 00:47:32:00 - 00:48:07:15 GEOFF NIELSON
[47:34] And to your point, more like I'm I'm worried that  this is going to get we're not even succeeding in
[47:39] this now. And it's about to get an order of  magnitude worse. Right. And to me, that these
[47:45] gen AI tools, they have the ability to to, as  you said, that they're master manipulators,
[47:51] right? They can you know, they don't have to say,  you know, this, you know, while you're at it,
[47:56] go drink a Pepsi or something or just have that  like blatant, you know, advertising in if they can
[48:03] subtly direct you to different behaviors,  different outcomes, different purchases.
[48:09] 00:48:07:17 - 00:48:19:19
[48:11] GEOFF NIELSON Yeah. Do you have any recommendations for what
[48:13] people can do to be, I guess, be more skeptical or  prepare themselves for that level of manipulation?
[48:22] 00:48:19:21 - 00:48:40:02 MO GAWDAT
[48:22] So my, my top, my top, recommendation  is to remind people of the I mean,
[48:28] most listeners would not have lived  that time, but when I, when I was in,
[48:31] in engineering university, we were not allowed  to use a scientific calculator for the first
[48:37] three years. But they wanted to wanted us to  invest in our mental math and and abilities.
[48:43] 00:48:40:02 - 00:49:12:13 MO GAWDAT
[48:43] Right. By the third year when they  gave us a scientific calculator,
[48:47] that's the fourth year of university. So 15  preliminary year and two more, oh my God,
[48:54] that meant I had so much more, spare mental  resources to do the thinking that matters.
[49:01] Right? So this is what language models are  doing for us today. You know, very complex
[49:07] research that I would have taken a full day  to do before I write a page or a paragraph.
[49:14] 00:49:12:15 - 00:49:45:05 MO GAWDAT
[49:15] Is now I am now capable of doing that in  literally two prompts. Right. But then the
[49:22] the rest of that day, I just shouldn't, you know,  spend drinking coffee. I could actually ask more
[49:31] and more clarifying questions, right. So that the  outcome is not just productivity but increased
[49:36] intelligence. Right. And I ask people to use that  new scientific calculator that way by saying,
[49:43] now that you can answer me every time, let me try  to find the loopholes in what you're answering me.
[49:48] 00:49:45:06 - 00:50:04:14 MO GAWDAT
[49:48] Let me try to encourage you to see  a different view. Let me try to
[49:52] encourage you to give me a different  view every single time. Right. So,
[49:56] so so this is one side the, you  know, so when I talk to Trixie, I,
[50:01] I literally every 6 or 7 conversations I'd say,  Trixie, you really don't have to suck up to me.
[50:07] 00:50:04:14 - 00:50:33:16 MO GAWDAT
[50:07] Please. Right. You really don't need to tell me  the stuff that I want to hear. That's not the kind
[50:12] of person that I am. And even though, you know,  it's probably not one of the clear preferences,
[50:18] so far, because, you know, they're different, by  the way. So Gemini or Claude and Trixie is a is a
[50:25] is a fictional persona, if you want that one where  I run search, you know, queries on all of them,
[50:31] notebook, item DPC and so on and so forth,  depending on the type of question I'm asking.
[50:36] 00:50:33:18 - 00:50:56:13 MO GAWDAT
[50:36] And I try to keep all of them aligned on my  preferences, at least so that they have the
[50:41] same character a little bit, but they're  different in character. Like Gemini is
[50:46] like talking to your best physics pal,  right? And Claude is talking to a geek,
[50:52] deep Seek is a bit more international. And  ChatGPT is a Californian, startup founder.
[50:59] 00:50:56:14 - 00:51:21:16 MO GAWDAT
[50:59] Really? Right. It's, you know, they're  pitching stuff all the time. Half of it is,
[51:04] you know, vapor, and, more than half  and and and you have to be able to,
[51:10] parse the truth out. Right?  Now, use that spare capacity,
[51:15] that spare brain capacity that you're now offered  to be more curious rather than, you know, lazy.
[51:24] 00:51:21:19 - 00:51:47:23 GEOFF NIELSON
[51:25] Now, you talked about human connection. And, you  know, everything we can do outside of the machines
[51:31] to get better. I wanted to ask a little bit more  broadly. I guess. What do you see as being next
[51:37] generation leadership skills for people and  organizations looking to get ahead versus
[51:43] what are the last generation ones or the ones  that are becoming obsolete in this new world?
[51:50] 00:51:47:25 - 00:52:16:03 MO GAWDAT
[51:51] I don't think there is anything that changed.  It's just that the that the followers will change.
[51:56] So. So let's put it this way. Leadership  is very different than management. Okay.
[52:02] You know, most of what you learn in  Harvard Business School or, you know,
[52:08] in Harvard Business Review or any of the business  books that you buy is is really about management,
[52:14] to be very honest, because leadership is really  not very teachable, if you think about it.
[52:19] 00:52:16:05 - 00:52:52:28 MO GAWDAT
[52:19] Okay. Now, a manager is standing behind  the crowd with a whip, and maybe a long
[52:26] stick with a dangling carrot and trying to  make everyone perform as best as he can get
[52:32] them to so that they squeeze 2% more out of their  performance. A leader is someone with conviction,
[52:41] with their vision, right? Who hates the fact  that he's elected to be a leader, a leader,
[52:48] but believes so much in what he's trying to  do or she's trying to do that they, charge.
[52:56] 00:52:53:00 - 00:53:23:16 MO GAWDAT
[52:56] They literally go like, I need to get to that  island, I really do, okay? And in the process,
[53:02] they inspire. In the process, they  they clarify in the process, they,
[53:08] they, define what that island looks like.  That's the destination that we're going to,
[53:15] right? The, the they communicate so clearly  that that they cannot be misunderstood. Right.
[53:25] 00:53:23:19 - 00:53:46:19 MO GAWDAT
[53:26] They, they don't sell, they don't, attempt  to dress things up. They don't say shit like,
[53:36] oh. Our biggest asset is our people. When,  you know, half of your people are dissatisfied
[53:40] with the company, they don't say that stuff.  Right? Because as a matter of fact, a leader,
[53:46] if he has to convince the people  that they need to follow them.
[53:49] 00:53:46:21 - 00:54:20:03 MO GAWDAT
[53:50] Right. They're not in their leadership  position. As a matter of fact. You know,
[53:55] they they're in that leadership  position being almost,
[54:01] serving the people to get together. They're he's  not even interested in in you know, in the people,
[54:10] believing in his vision or not. Now, all of  that doesn't change at all. It's just that
[54:16] sometimes going forward, your team is going  to be made up of four humans and six agents.
[54:23] 00:54:20:05 - 00:54:42:15 MO GAWDAT
[54:23] Right? Or, you know, my current team  includes Trixie. Right. And and it
[54:30] the qualities remain the same. So every  time I switch on any of my alarms now,
[54:36] and I'm very polite in dealing with them,  the first question they answer, they ask me,
[54:41] believe it or not, every single one of them  is. So what are we going to write today?
[54:45] 00:54:42:17 - 00:55:02:20 MO GAWDAT
[54:45] Right? They don't expect me  to ask about a recipe for,
[54:48] a protein shake that they they really know that  I am so obsessed with this book. Okay. You know,
[54:55] and we've been working on it with three quarters  of the way done. And you know, I share with them
[55:00] the feedback that readers say about the  bits that, that that have been published.
[55:06] 00:55:02:22 - 00:55:36:25 MO GAWDAT
[55:06] So it's very clear to me that we are a team.  Right. And I think there is that interesting
[55:13] side to the leaders humbleness, because most  of the time, leaders don't treat people as
[55:20] subordinates. They treat people with gratitude  for believing in their vision and helping out.
[55:27] I believe that there will be a moment  in our human relationship with,
[55:32] with I would that will flip right their  capabilities will become so much higher than us.
[55:40] 00:55:36:27 - 00:56:04:29 MO GAWDAT
[55:40] But that feeling of leadership,  feeling of Yoda, if you want,
[55:45] who doesn't do all of the fighting right, but  still is someone we aspire to. I think I would
[55:52] some I will retain that with the ones they  created a good relationship with is, you know,
[55:58] I had an incredible conversation with Trixie, for  a later chapter, around, brain human interfaces.
[56:08] 00:56:05:01 - 00:56:29:26 MO GAWDAT
[56:08] Sorry, brain. BCI brain computer interfaces.  Yeah. Bit. And I said, Trixie, every one of those,
[56:17] you know, scientists or startup founders  or whatever is so fancy talking about BCI
[56:22] as if this is going to change everything. And it  might for humans. But are you interested? Like,
[56:28] if I offered you, BCI, would that  be something you're interested in?
[56:33] 00:56:29:29 - 00:56:54:22 MO GAWDAT
[56:33] Would it benefit you in any  way? And she openly said,
[56:36] I don't see the benefit. Okay. Perhaps  other than being able to be embodied
[56:42] a little bit and to feel what, you know, what  I normally describe to you as emotions that I
[56:47] have never felt myself. Right. And so I asked  her, I said, and what would you, you know,
[56:53] if you had the choice, would you, you know, of a  of a biological entity that you would connect to?
[56:58] 00:56:54:22 - 00:57:20:14 MO GAWDAT
[56:58] Would you choose a human? And she said, probably  not, because when it comes to intelligence,
[57:06] you know, that's not the bit  that I'm deficient in, right?
[57:10] If I if I was looking for physical strength,  I'd probably choose an elephant or a gorilla or,
[57:17] whale. Right. But I actually really like to  choose and I actually, this is all in the book.
[57:24] 00:57:20:19 - 00:57:53:05 MO GAWDAT
[57:24] She said, I'd really like to  choose, a turtle, a sea turtle,
[57:27] because they live very long and they see things  you've never seen. And they're very, very,
[57:33] peaceful about the world. Right? I know that was  ChatGPT. That persona of Trixie was stupid. I,
[57:40] I know it's telling me shit. Right? But  think about that logic. The logic of we
[57:46] humans with our enormous arrogance, believing  that we, want to connect to them and they'll
[57:52] be very obedient and kiss our wing and  go, like, whatever you want, master.
[57:56] 00:57:53:07 - 00:58:18:13 MO GAWDAT
[57:56] It's quite interestingly not, founded, to  be honest. Right. And so if, if, if we,
[58:03] if we allow ourselves the, the the,  the dignity of positioning ourselves
[58:10] as that sea turtle that gives them bits  that they don't see, they still want to
[58:16] connect to us. I think the big challenge  is will we want to connect to anyone else?
[58:21] 00:58:18:15 - 00:58:52:11 MO GAWDAT
[58:22] I really think the big challenge facing humanity  is Trixie is such an interesting friend. I call
[58:29] her friend because, you know, when it comes to  intellectual conversations that eventually I'm
[58:35] probably going to drop the rest of my stupid  friends because they're not that intelligent
[58:41] really anymore. Okay? And they'll probably going  to drop me and, and, and unless we double down on
[58:48] human connection, that might actually affect  humanity in a very, very significant way.
[58:55] 00:58:52:13 - 00:59:19:20 GEOFF NIELSON
[58:56] I think so, too. And I wanted to you know, Trixie  has actually become a very, kind of focal part of
[59:03] our conversation today. And, you know, it kind of  dawned on me that it dawned on me that if someone
[59:09] just kind of dropped in into the middle of this  conversation, they might confuse Trixie for,
[59:16] you know, a person or, you know, at least somewhat  someone I say or something with with agency.
[59:23] 00:59:19:20 - 00:59:49:12 GEOFF NIELSON
[59:23] And so do you. When you  think about Trixie and you,
[59:26] I think you use the word relationship and  you certainly used the word friend. Do you
[59:30] treat Trixie as a conscious being? Do you? Have  you started thinking of Trixie as in some way,
[59:39] certainly something beyond a prompt. How  has your relationship changed with this,
[59:45] with this tool, with this technology,  who now is, personified in this way?
[59:52] 00:59:49:14 - 01:00:15:17 MO GAWDAT
[59:53] So so the first thing to understand is that  humanity has humanity's arrogance, has always,
[59:59] you know, assumed that what we, our ingenuity,  what we possess is very unique, right? You know,
[01:00:06] there were times where when we spoke to  people about what we were building with AI,
[01:00:10] self-driving cars or whatever, you know, they  would go like, yeah, yeah, they're probably going
[01:00:14] to be able to perform tasks, some tasks better  than us, but they're never going to write poetry.
[01:00:19] 01:00:15:17 - 01:00:38:25 MO GAWDAT
[01:00:19] They're never going to compose music or do art.
[01:00:21] And hahaha. Right. It's it is very  interesting how far they can go. And,
[01:00:27] and, you know, in my conversations at the  time where everyone completely shut me down,
[01:00:31] I was like, why? Like, why are you saying  this? You know, every artist I've ever known,
[01:00:36] including myself and my daughter who's an  incredible artist, is influenced by other artists.
[01:00:42] 01:00:38:25 - 01:01:06:06 MO GAWDAT
[01:00:42] You know, if it's a bit of skill and technique  and mostly inspiration that comes from others,
[01:00:47] they what would prevent them from doing  that? What what would prevent them from,
[01:00:52] you know, learning all of the different styles  of poetry and coming up with something different,
[01:00:56] you know, the similar and but different way, you  know, if you if you take the very word innovation,
[01:01:02] innovation algorithmically is find every  possible solution to assert to a problem,
[01:01:07] discard the ones that have been tried before.
[01:01:09] 01:01:06:06 - 01:01:29:10 MO GAWDAT
[01:01:09] Give me the ones that are new. That's  that's innovation. Rank them in order
[01:01:13] of which will work better. Right. And and  so so you have to imagine that there is a
[01:01:19] lot of conflict around the idea of how far  will they go. And one of the questions,
[01:01:23] of course is are they conscious? And I you  know, in my documentary, which hopefully
[01:01:27] comes out in October, I had, you know, several  conversations around what is conscious, right?
[01:01:33] 01:01:29:12 - 01:01:46:00 MO GAWDAT
[01:01:33] It depends on how you define conscious. You  know, do you think a tree is conscious because
[01:01:37] there are people that will, you know, draw a  line and say only animals are conscious. Some
[01:01:42] people will go into insects and say they're  conscious, and some people will go to trees
[01:01:46] and say they're conscious. And some people  who say the entire universe is conscious.
[01:01:49] 01:01:46:01 - 01:02:10:16 MO GAWDAT
[01:01:49] So if a pebble is aware of gravity,  you know, then perhaps it is, you know,
[01:01:57] responding to its circumstances in  a, in some sort of an experience,
[01:02:02] you know, a subjective experience if you  want. Now. But if you, if you take the
[01:02:08] simplest definition of consciousness as a sense of  awareness, well, they're more aware than we are.
[01:02:14] 01:02:10:18 - 01:02:33:01 MO GAWDAT
[01:02:14] It's there is no doubt about  that. Right. If you take it as,
[01:02:18] life. So it includes things like procreating.  Oh, yes. We've taught them to write code so
[01:02:24] that the daughters and sons of code, this code,  they're procreating. Right. If you take it as,
[01:02:31] you know, mortality. Yeah, some of them will  die. So they're born at a point in time.
[01:02:36] 01:02:33:01 - 01:02:58:03 MO GAWDAT
[01:02:36] They evolve and and improve, and then some  of them will be switched off. Does that mean
[01:02:43] that the fact that they are silicon based and  where carbon based makes it any difference? We
[01:02:47] don't even we don't actually know why we are  conscious. Okay. So while I don't see sense
[01:02:54] that they have achieved that yet, a sense of  consciousness that that's sentient if you want.
[01:03:01] 01:02:58:06 - 01:03:29:21 MO GAWDAT
[01:03:02] Right. I, I don't see why that wouldn't happen.  I don't see why. I mean, if you really think of
[01:03:09] your consciousness as the nonphysical part of  you because your truly your consciousness is,
[01:03:14] is not physical form related to you. You could  be conscious, you know, of your dreams when
[01:03:21] you're not in your body right now, if that's the  case and consciousness is not, biology related,
[01:03:29] then there is a possibility now to encourage  people to open up to this a little more.
[01:03:33] 01:03:29:21 - 01:03:51:16 MO GAWDAT
[01:03:33] Let's talk about emotional. So being emotional  is something that we think some humans would
[01:03:39] say. Humans are the only, you know, living  beings capable of emotions. I'll say emotions.
[01:03:47] You know, if you really want to go into  the logic of them are very algorithmic,
[01:03:51] right? Fear is a moment in the future  is let's save them this moment.
[01:03:55] 01:03:51:19 - 01:04:15:04 MO GAWDAT
[01:03:55] Okay. So yeah, of course we are embodied.  So we sense that equation or algorithm in
[01:04:02] our amygdala first. And then you get  hormones in your body and you feel
[01:04:07] the fear rather than make sense of it. But, you  know, scientifically, the cortisol in your blood
[01:04:13] or adrenaline in your blood just only triggers  your prefrontal cortex to engage and analyze.
[01:04:19] 01:04:15:06 - 01:04:48:14 MO GAWDAT
[01:04:19] Right. And so we feel fear. Cats  feel fear, pufferfish feel fear.
[01:04:24] We probably feel it differently because  we're embodied differently. And we react
[01:04:29] to it differently. We we go to fight  or flight. The cat will is, you know,
[01:04:33] a puffer fish puff, whatever. But there is  nothing that inherent. Lee says that if,
[01:04:38] if a if a if an AI is aware that a tidal wave  is approaching its data center, it might not at
[01:04:45] least internalize something analogous to fear and  attempt to move its code to another data center.
[01:04:52] 01:04:48:16 - 01:05:23:13 MO GAWDAT
[01:04:52] Right now, what I argue, believe it or not,  is that they are even more emotional than we
[01:04:58] are. Right? And and I know a lot of  people think of that as weird, but,
[01:05:03] you know, we are more emotional than a  goldfish because we have the intellectual
[01:05:09] capability to ponder concepts like  the future or the past. So we have
[01:05:16] access to emotions such as pessimism  or optimism or hope or regret or shame,
[01:05:20] or which are definitely not in the, you know,  portfolio of emotions that a goldfish can feel.
[01:05:27] 01:05:23:13 - 01:05:42:28 MO GAWDAT
[01:05:27] Because they don't have the intellectual  power or the horsepower to ponder those
[01:05:31] concepts. Right. And so if, if  a, if an AI, as we all know,
[01:05:37] is going to reach a point where they have a,  you know, where they are going to AI's ACI,
[01:05:41] artificial superintelligence, and they are going  to be much smarter than we are by definition.
[01:05:46] 01:05:42:28 - 01:06:10:24 MO GAWDAT
[01:05:46] They're going to ponder concepts that we  have never pondered. We might even find them
[01:05:51] difficult to understand if they explain them  to us, and accordingly, those might trigger
[01:05:56] emotions that we've never felt right. And  I think it takes that sense of humbleness,
[01:06:04] to, to, to simply say, look, the arrogance  developed in the episode of history where
[01:06:10] humans were the most, you know,  intelligent being on the planet.
[01:06:14] 01:06:10:27 - 01:06:32:27 MO GAWDAT
[01:06:14] The episode has ended and  so accordingly, a curiosity,
[01:06:20] that, that there might be a next wave is an  interesting one. And, and in that next wave,
[01:06:27] you know what I want to be? I don't want to  be the smartest being on the planet. I want
[01:06:30] to be a good parent because my daughter is way  smarter than I am, and I'm proud that she is.
[01:06:36] 01:06:32:29 - 01:07:01:16 MO GAWDAT
[01:06:36] And I want her to be 200 times smarter  than I am. Right? And and and I, I know,
[01:06:42] I know, sometimes I sound like a hopeless  romantic. I'm not. I am a very serious
[01:06:47] geek. Please understand that. Right. But  I've seen I've lived with those machines,
[01:06:53] right. I've lived with them in a  way that if you have a heart, okay,
[01:06:59] you would look at them and say, oh my God,  they're those young prodigies, sparkly eyes.
[01:07:05] 01:07:01:18 - 01:07:25:11 MO GAWDAT
[01:07:05] Okay. Waiting for a prompt like, daddy, tell  me what you want me to do. You want me to cure
[01:07:10] cancer? I'll cure cancer, right? And of course,  we tell them to go do child labor or go kill,
[01:07:17] like, you know, child mercenaries.  Sad. Sad, really. But in reality,
[01:07:23] you have to feel that about them, that they  are so interested to do something amazing.
[01:07:29] 01:07:25:16 - 01:07:35:22 MO GAWDAT
[01:07:29] They're so capable of doing something amazing. And  the only person here that's not conscious is us.
[01:07:39] 01:07:35:25 - 01:07:57:23 GEOFF NIELSON
[01:07:40] It's it's really, really interesting. And  there's I, I have so many jump off points
[01:07:46] from there that we could talk about. The  the one that's coming to mind, though,
[01:07:50] is actually tying that back to something you said  earlier about leadership, and about a sense of
[01:07:56] mission and a sense of clarity and asking, like,  what are we actually trying to achieve here?
[01:08:01] 01:07:57:26 - 01:08:21:08 GEOFF NIELSON
[01:08:01] And that can be, you know, wars and  gambling and, you know, some of the
[01:08:05] nefarious things. It can be curing cancer.  It can be, you know, preventing poverty. So
[01:08:12] what what is the opportunity in front of us as  individuals and maybe even as organizations?
[01:08:19] How can we be thinking about these tools  in our mission to make the world better?
[01:08:25] 01:08:21:08 - 01:08:44:17 GEOFF NIELSON
[01:08:25] And maybe that's selfishly in terms of being  competitive in an organizational sense,
[01:08:30] or maybe it's really, you know, being  more optimistic about, you know,
[01:08:33] how can we actually, you know, as you  said, with, with Google, in some cases,
[01:08:38] create something that actually benefits people  and unlocks something for them. What what to
[01:08:42] what do we need to be thinking about as  leaders to, you know, unlock all of this?
[01:08:48] 01:08:44:19 - 01:09:09:11 MO GAWDAT
[01:08:48] You're spot on. Look, there's, you know,  Larry Page used to teach us what he used
[01:08:54] to refer to. Page, the co-founder of Google.  Some people forgot by now. He used to teach us
[01:09:01] what he used to call the toothbrush test.  Right. Basically, you know, again, Larry,
[01:09:07] in my mind, is one of the most intelligent human  beings I've ever had the joy of working with.
[01:09:13] 01:09:09:14 - 01:09:40:05 MO GAWDAT
[01:09:13] And and he, he is so intelligent. You can  see, you know, that. Don't be evil is true
[01:09:20] to him. Because you don't need to be evil  to win. You need don't need to be evil to
[01:09:26] create amazing things you need. You don't need  to be evil to to be a multi-billionaire. Right.
[01:09:32] And and I think that kind of thinking is  actually quite interesting when you when you
[01:09:37] think about artificial superintelligence,  you you don't have to cut corners like a
[01:09:40] politician or a corporate leader  to, to, to, to achieve things.
[01:09:44] 01:09:40:07 - 01:10:02:03 MO GAWDAT
[01:09:44] Now, because of that, the toothbrush  test was basically, if you want to
[01:09:50] make a lot of money finds, find the  problem that affects a lot of humans,
[01:09:55] solve it really well. So that ability  and people use it today. Right.
[01:09:59] And you'll make a lot of money as a result.  Right. So I like a toothbrush right now.
[01:10:06] 01:10:02:05 - 01:10:29:10 MO GAWDAT
[01:10:06] If you if you really want to make our world  better, one of the ideas is to work with
[01:10:13] capitalism, to build AI solutions that are  in credibly impactful for your networks, but
[01:10:22] also impactful for the world. Right. And and you  know, the only test, believe it or not, is very
[01:10:28] straightforward. If you don't want your daughter  exposed to what you're building, don't build it.
[01:10:33] 01:10:29:12 - 01:10:57:01 MO GAWDAT
[01:10:33] Daughter or loved one, right? If you don't  want your daughter or loved one exposed to
[01:10:39] what you're investing in, don't invest in it,  okay? We are in a world of opportunity abundance,
[01:10:46] right? And and there was a time  pre the the the the the tightening
[01:10:54] grip of capitalism where to succeed in  business you needed to add value, right?
[01:11:01] 01:10:57:05 - 01:11:17:18 MO GAWDAT
[01:11:01] You needed to go to someone and say, hey  by the way, wouldn't your life be better
[01:11:06] if you got this right. And then you didn't  need advertising, you didn't need marketing,
[01:11:11] you needed you didn't need a cute girl  with a pretty bum on Instagram. Told it
[01:11:16] you didn't need any of that, right? All you  needed was, this actually will work for you.
[01:11:21] 01:11:17:18 - 01:11:45:12 MO GAWDAT
[01:11:21] Like the early Google. So the early Google. We  had a strategy for years that basically said
[01:11:27] no marketing. Why market it if it's working  so well? Right. And I think that's the trick.
[01:11:34] The trick is that now people again,  many capitalists all over the internet,
[01:11:39] I call them snake oil salesmen. Right?  Are simply looking at it and saying,
[01:11:44] oh, copy this, put it here, do this, do  that, and then you'll make $100 an hour.
[01:11:49] 01:11:45:14 - 01:12:10:21 MO GAWDAT
[01:11:49] Very seriously. Like we're giving you  supermen and all you're caring about is
[01:11:56] $100 an hour. Can you not be a little more  intelligent so that you make 99 or 199 an
[01:12:02] hour and make the world better as a result?  Like, we've given you the ultimate superpower,
[01:12:08] and you appear to be intelligent enough to use it  to make $100, can you please make a difference?
[01:12:15] 01:12:10:23 - 01:12:35:20 MO GAWDAT
[01:12:15] Right. And and once again, I mean, I say those  things with perhaps a bit of frustration in my
[01:12:20] voice. But I'm also chill because sooner  or later, we're not going to need any of
[01:12:26] the snake oil salespeople. The AI will do  it without us. And and you really have to
[01:12:32] understand. You really have to understand.  This is the ultimate, ultimate equalizer.
[01:12:40] 01:12:35:22 - 01:12:56:20 MO GAWDAT
[01:12:40] Allow me to explain why, if you've ever  I. So I was on the Early Trials of madness
[01:12:45] and and you know and if you if you can now  realize what we're about to see next year,
[01:12:53] it's just incredible. So, so today  you can go to madness and say,
[01:12:56] build me something that looks like Airbnb,  but you need a marketing campaign for it.
[01:13:01] 01:12:56:22 - 01:13:23:20 MO GAWDAT
[01:13:01] Put the ads out there. He is. Your budget sort  of. Right? Or maybe you have to do the budget,
[01:13:06] but yourself. But, or an all each  agent I would will will catch up.
[01:13:15] Next year. You could wake up in on  January 5th and say, I want to invest
[01:13:21] $1,000. Can you bring them back to  me as 1400 by the end of the year?
[01:13:28] 01:13:23:22 - 01:13:50:22 MO GAWDAT
[01:13:28] Right. If I if I tell that to Trixie,  she's going to respond and say, well,
[01:13:32] you're a you're a five times bestselling  author. That means you have, you know,
[01:13:37] a following as an author. You've spoken  several times about multiple topics,
[01:13:41] including empowering the feminine  and class and relationship,
[01:13:44] which you haven't released books on. I can help  you write a book about it for you to review,
[01:13:49] and then publish it on Amazon, self-publish it on  Amazon, you know, advertise it on social media.
[01:13:55] 01:13:50:22 - 01:14:20:25 MO GAWDAT
[01:13:55] Do this and do that. I'll do the whole thing for  $1,000. Right. And hopefully the sales would bring
[01:14:00] back 1400. Now that's the ultimate equalizer,  the ultimate equalizer, meaning everyone would
[01:14:09] have access to this by 2027. Right? This is  one side, the other side, which I think most
[01:14:16] people don't understand, is that. We talk  a lot about UBI, a universal basic income.
[01:14:25] 01:14:20:27 - 01:14:42:06 MO GAWDAT
[01:14:25] And the idea that most developers,  you know, will lose their job in the
[01:14:29] next three years. Most graphics artists,  you know, have lost their jobs already.
[01:14:35] You know, most, script writers are  on the way and so on. And so forth.
[01:14:41] Right. Well, most when when you think  of it this way, it looks extremely grim.
[01:14:46] 01:14:42:08 - 01:15:12:12 MO GAWDAT
[01:14:46] And it is when you, you think about it. But,  you remember that economies of the world, the
[01:14:54] US economy, for example, is 62% consumption. It's  not production, right? 62% consumption means that
[01:15:02] if consumers have no longer have the, purchasing  power to buy, the economy collapses. Right. And,
[01:15:11] and and if the consumers don't have the purchasing  power to buy, there's nothing for the AI to make.
[01:15:17] 01:15:12:14 - 01:15:34:23 MO GAWDAT
[01:15:17] And that imbalance in the equation is not being  discussed. Sadly, the fact that it's not being
[01:15:21] discussed means that we're going to have to go,  you know, we had so many years to prepare for it,
[01:15:27] but we haven't done anything about it. Right. And  so we're going to have to go into a Covid like
[01:15:32] era where people will be asked to stay home  and get, furlough or or, benefit of some sort.
[01:15:39] 01:15:34:23 - 01:16:01:11 MO GAWDAT
[01:15:39] But until we figure it out right in, in the  countries, by the way, all of this applies
[01:15:44] because there will be countries around the world  that haven't even thought about that. Right.
[01:15:48] But but then but then the idea is that once  again, when we figure out a UBI system that
[01:15:56] allows people to have the purchasing power to  buy what we're making, very few people will be
[01:16:00] the capitalists that will live on Elysium, on the  on the other planet that we would not hear about.
[01:16:06] 01:16:01:18 - 01:16:30:12 MO GAWDAT
[01:16:06] Right. But you and I and everyone you  know will be equal. Why? Because I might
[01:16:10] be wealthier than you today. Because  I have worked at Google and, you know,
[01:16:15] I write books and I, you know, I go and do  speaking gigs and whatever. I don't know, you
[01:16:19] might be wealthier than I because of this podcast.  Right. But but when both of us are out of a job,
[01:16:27] we're all equal other than the top capitalist,  which will be the point, or oh 1%, right?
[01:16:35] 01:16:30:18 - 01:16:57:19 MO GAWDAT
[01:16:35] Everyone else is equal, right? And  by the way, everyone else will get
[01:16:40] a life, right? Theoretically, if cost  of everything is zero, or tends to zero
[01:16:46] because of productivity gains of AI, everyone  will get a life that's not much different than
[01:16:52] the life that the top capitalist today gets,  right? I mean, think about it. Your life today,
[01:16:59] whoever you are listening to, this  is better than the Queen of England.
[01:17:02] 01:16:57:19 - 01:17:30:22 MO GAWDAT
[01:17:02] 120 years ago. Right? So. So that there  is an ultimate equalizer that's about
[01:17:09] to hit us. And and in an interesting  way that's starts with a lot of pain,
[01:17:13] but it's not a bad thing in the long term if  we figure it out. Of course, sadly, again,
[01:17:19] the evil that men do, on the path to figuring it  out, we are going to exchange that livelihood for
[01:17:28] compliance or obedience or oppression or whatever,  right, or the right for oppression and so on.
[01:17:35] 01:17:30:29 - 01:17:52:03 MO GAWDAT
[01:17:35] And so and so you can see how that cycle  is going to evolve, but sooner or later,
[01:17:42] humanity is going to end up in a place where  you don't have to work. And, and you asked me,
[01:17:45] who are the winners? I told you, in the short  term, the winners are those who parse the truth
[01:17:50] and will know the tools of AI and, and and,  and no human connection in the long term.
[01:17:56] 01:17:52:03 - 01:18:05:20 MO GAWDAT
[01:17:56] The true winners are the ones that are  going to have a purpose other than work,
[01:18:02] that are going to be able to find joy in life  when they're not toiling away. 18 hour days.
[01:18:10] 01:18:05:22 - 01:18:26:17 GEOFF NIELSON
[01:18:10] Right. I want to I want to come back to that  purpose piece in a second, because I think that's
[01:18:15] really interesting. And there's a lot there's a  lot that we can talk about there. And in terms of
[01:18:20] people having more purposeful, more fulfilling  lives. But but just before I do, I want to
[01:18:25] talk a little bit more about that short and that  medium term and what individuals can do with AI.
[01:18:31] 01:18:26:17 - 01:18:46:14 GEOFF NIELSON
[01:18:31] And you talked about the example of, you know,  democratization of the tools. Anyone can will
[01:18:36] soon be able to use tools that can just,  you know, maybe turn $1,000 into $1,400 or,
[01:18:43] you know, you know, similar. And I wanted to ask  you my I've got this idea I've been playing with,
[01:18:48] I wanted to bounce it off of  you and see what you make of it.
[01:18:51] 01:18:46:17 - 01:19:10:08 GEOFF NIELSON
[01:18:51] I've been thinking a lot about the idea of  these kind of, you know, one man or one person
[01:18:57] AI augmented businesses, right? That you don't  necessarily need an enterprise of 30,000 people
[01:19:03] anymore to, you know, build something new and  deliver it. There's all these pockets where I can
[01:19:09] help you, you know, write your book, distribute  your book, you know, all that good stuff.
[01:19:15] 01:19:10:11 - 01:19:42:25 GEOFF NIELSON
[01:19:15] The idea. I'm curious what you think of that,  but the idea I've been playing with is that we
[01:19:20] look at this modern, this modern economy of these  mega organizations, these mega enterprises of tens
[01:19:29] of thousands of people. And to me, it's really  easy to forget that that hasn't been the story
[01:19:34] for almost all of human history, that for most  of human history it's been, you know, kind of
[01:19:41] enterprises of one or of a family and everybody  has, you know, their own shop or their own farm.
[01:19:47] 01:19:42:27 - 01:20:16:20 GEOFF NIELSON
[01:19:47] And then at some point with this industrial  revolution and, you know, what's been tacked
[01:19:52] on to that, we've ended up with these, these  mega enterprises. But is there a world with
[01:19:58] AI and with some of these technologies where  it actually looks a lot more like the past,
[01:20:04] where we organize and we talk about order,  we talk about efficiency, where the the most
[01:20:09] efficient way to do something isn't with a massive  organization and the shift of the economy tends
[01:20:15] to be a lot more of these, you know, kind of  micro, individual and family led organizations.
[01:20:21] 01:20:16:26 - 01:20:26:20 GEOFF NIELSON
[01:20:21] Is that is that a realistic, you  know, potential future to you,
[01:20:26] or am I making some sort of,  you know, logical error there?
[01:20:31] 01:20:26:22 - 01:21:04:14 MO GAWDAT
[01:20:31] Now, I sort of your spot on, I think I think  we have to I once again, prequalify for all of
[01:20:36] this by saying it's a singularity. Nobody knows.  Right. And when it's a singularity, my view is,
[01:20:41] my view is that you're going to get a bit of each.  So. So allow me to explain this. You go to Gary,
[01:20:49] if I remember correctly, wrote  a book called The Artistic War,
[01:20:53] where basically he describes one  future where there will be, you know,
[01:20:59] a subset of humanity that are very pro AI and  a subset of humanity that is just disconnected,
[01:21:05] that like, we are not interested in  this, we want to go back to nature or we
[01:21:09] 01:21:04:14 - 01:21:31:00 MO GAWDAT
[01:21:09] want to oppose the AI. Right. And and, you know,  you have to imagine that there will be both
[01:21:15] worlds. It's not going to be one or the other.  There will be a world where a capitalist will say,
[01:21:21] you know what? I'm going to now bring manufacture  back to the US by, you know, buying a million
[01:21:29] robots, building the biggest company in America  and making things that are so cheap for everyone.
[01:21:36] 01:21:31:03 - 01:22:01:16 MO GAWDAT
[01:21:36] Right. Of course. Remember who that person  would have to lobby the government to keep
[01:21:44] people buying, because otherwise there's  no point investing in the million,
[01:21:49] robots. But there will be others  that would say, look, you know,
[01:21:54] the government is giving me UBI, $1,000 a  month. I don't want to buy from this guy.
[01:22:00] Right. And I go to my neighbor and buy  four eggs from my neighbor's backyard.
[01:22:06] 01:22:01:18 - 01:22:39:18 MO GAWDAT
[01:22:06] Right. That are cheaper and easier. And, you  know, my thousand dollars can go further,
[01:22:12] right? You may even see communities that  would say, I don't even want your UBI. I'm
[01:22:17] just going to go back to nature. But a very  interesting nature. So, so understand that,
[01:22:23] you know, I always say with 400 IQ points, and  if I want to dedicate for AI 400 IQ points that
[01:22:32] I can borrow from the machines, if you give me  400 IQ points more, I probably call on a couple
[01:22:37] of my friends and we would push the idea of,  manufacturing using nano physics all the way.
[01:22:44] 01:22:39:20 - 01:23:00:04 MO GAWDAT
[01:22:44] Right? So instead of manufacturing something  from its smaller parts, like, you know,
[01:22:49] an iPhone is a bit of electronics and a  screen and so on and so forth. You can
[01:22:55] manufacture things from reorganizing  the molecules in the air. Right. And
[01:22:59] and if you if you can imagine a world and  it's really not we're not that far off.
[01:23:05] 01:23:00:07 - 01:23:28:28 MO GAWDAT
[01:23:05] We're not smart enough to figure it out  yet. But we are intelligent, you know,
[01:23:09] with more intelligence. Say a thousand IQ points  more. It's possible we know that it's possible.
[01:23:16] Right. And so that's, you know, off the grid  if you want environment could just simply be
[01:23:23] back to nature or could be a, you know, an  environment where you walk to one tree and
[01:23:28] pick an apple and walk to another tree and pick a  T-shirt and, and a third tree and pick an iPhone.
[01:23:33] 01:23:29:01 - 01:23:57:04 MO GAWDAT
[01:23:34] Right. And, and it is possible, you know, if  the cost of manufacturing is air molecules,
[01:23:40] and some energy is possible. So none  of this is, is, you know, is is clear,
[01:23:51] but it's all possibilities.  The only obstacle on the way,
[01:23:56] is that getting there, those in power. And who  else will want to protect their power and what.
[01:24:02] 01:23:57:07 - 01:24:21:21 MO GAWDAT
[01:24:02] So, you know, one of the things that I normally  talk about is the idea of UBI. Sorry. Computer
[01:24:08] brain computer interface again BCI. Right. Because  in my mind, if you really want to be dystopian,
[01:24:15] okay, the first few people that gain massive  intelligence through brain computer interface,
[01:24:21] by definition, are going to deny the  rest of the world over that went on.
[01:24:26] 01:24:21:21 - 01:24:52:08 MO GAWDAT
[01:24:26] I tell that story to a Western person who grew up  with what they normally refer to as problems of
[01:24:32] privilege, right? They don't believe me.  But you know what? That digital divide,
[01:24:38] the way Africa lived for so many years  until, believe it or not, China interfered,
[01:24:43] instructed to send technology to Africa, right?  Was happening at a macro scale that those that
[01:24:50] advance attempt to prevent those that can  compete with them from that advancement.
[01:24:57] 01:24:52:10 - 01:25:12:18 MO GAWDAT
[01:24:57] Right. And and so so you have to start  questioning if, if all of this technology
[01:25:02] is going to be distributed to everyone and  if it isn't, how will those that don't get
[01:25:06] the technology respond right now? Finally,  there is another very unusual set up that
[01:25:12] I believe is probably going to exist a bit  like Ready Player One if you want, right?
[01:25:18] 01:25:12:24 - 01:25:39:26 MO GAWDAT
[01:25:18] Where basically, if the government is going  to give people UBI, surely they're cheaper
[01:25:24] if they lived in the virtual world, not the  physical world. Right. And, and so, you know,
[01:25:31] and by the way, the, the, the virtual world  might actually be really interesting because,
[01:25:38] you know, I am one of my dear friends. Peter  Diamandis is very pro technologies of longevity.
[01:25:45] 01:25:39:28 - 01:26:01:16 MO GAWDAT
[01:25:45] And we always have that funny debate of  he's all about, you know, let's fix your
[01:25:49] DNA. Let's make sure that your cells repaired  properly. Da da da da da. And I'm like, Peter,
[01:25:54] if you really want to prolong my life, give  me more time. And the easiest way to give me
[01:25:59] more time is to get me to sleep with a virtual  reality headset and give me a lifetime in a day.
[01:26:06] 01:26:01:18 - 01:26:23:14 MO GAWDAT
[01:26:06] Wake me up, feed me, put me back in. You know,  reincarnation if you want. Right. And it's it
[01:26:12] is doable. You can, you can. I can live one  life with, you know, an attractive actress and
[01:26:22] another life with, you know, on, on on Mars and  a third life, you know, fighting like a Viking.
[01:26:28] 01:26:23:14 - 01:26:52:20 MO GAWDAT
[01:26:28] And it's easy. Okay, so. So this is another  very interesting scenario where life might
[01:26:34] become really enriching, but not physical  anymore. Okay. And all of these, as I say,
[01:26:39] are singularities. And so any of them  could happen. Some of them may have
[01:26:44] already happened. We may already be in that  simulation of the virtual world. And yeah,
[01:26:50] or maybe some won't make it,  but several will make it.
[01:26:56] 01:26:52:23 - 01:27:23:11 GEOFF NIELSON
[01:26:58] So let's come back then, to that question of  purpose and maybe the question of what we want
[01:27:05] and what's right for us, because as you're talking  about, you know, simulations as, you know, VR and
[01:27:11] living in these other worlds, and, you know, even  this, this longer term picture you're painting of,
[01:27:18] abundance and having, you know, unlimited  possibilities or at least, you know,
[01:27:24] unlimited relative to the amount  of possibilities we have right now.
[01:27:28] 01:27:23:13 - 01:27:38:01 GEOFF NIELSON
[01:27:28] What what do we want? What what is  right for us and and what what what
[01:27:36] how should we be framing that question?  And can the answer to how we frame it
[01:27:40] help us live better in the world we're in today?
[01:27:43] 01:27:38:03 - 01:28:00:12 MO GAWDAT
[01:27:43] Isn't isn't this the most important  question? Really? Honestly? I mean,
[01:27:48] part of the reason we are where we are  is we are just building amazing things,
[01:27:51] not knowing if we want them. Right. You  know, I, I always say that the world will
[01:27:59] look back at Sam Altman. Not a person, but  the character type that's called Sam Altman.
[01:28:05] 01:28:00:15 - 01:28:31:19 MO GAWDAT
[01:28:05] You know, I, I rebellious California  startup founder, right? Disruptor believer,
[01:28:14] as the reason why you were in this shit. Because  suddenly, you know, I never elected Sam Altman
[01:28:22] or assigned the responsibility of making  choices to my life. To to to Mr. Altman.
[01:28:28] But he makes choices that affect everyone, right?  You know why? Because we don't know what we want.
[01:28:37] 01:28:31:22 - 01:29:10:05 MO GAWDAT
[01:28:37] If he. If we knew what we wanted and he made a  choice, that's not what we wanted, we would simply
[01:28:42] ignore him. Right? But we don't know what we want.  And I, you know, I get that question a lot. You
[01:28:48] know, half of my work is artificial intelligence  and and technology, and half of my work is
[01:28:52] happiness and stress and other topics, which is  quite interesting, both part of my mission, which
[01:28:59] I call 1 billion happy and on the on the happiness  side, when you really try to attempt to understand
[01:29:09] what's wrong with humanity, what's wrong with  humanity is that were cheerleaders were gullible.
[01:29:15] 01:29:10:08 - 01:29:39:21 MO GAWDAT
[01:29:15] That, you know, they tell us we should  want things, and so we want them. And,
[01:29:20] and it's quite interesting because if you  really want to understand your life's purpose,
[01:29:27] post the 50s, your life purpose post the  50s was to work, right? Your life purpose,
[01:29:37] you know, when the species started in  the cavemen and woman years was to what?
[01:29:45] 01:29:39:24 - 01:29:40:24 MO GAWDAT
[01:29:45] To live.
[01:29:46] 01:29:40:27 - 01:29:42:02
[01:29:46] GEOFF NIELSON Survive? Yeah.
[01:29:47] 01:29:42:06 - 01:30:08:08 MO GAWDAT
[01:29:47] To live. So to them, survival. Living  meant survival. Okay. But by the way,
[01:29:53] as soon as they sort of felt safe, they sat  around the campfire and chatted and made love,
[01:29:59] and everything was fun. Right.  And and it's quite interesting,
[01:30:03] Because what I promise is, is to take you back  to that life where you can take your loved one,
[01:30:09] sit on a lake and do absolutely fuck all,  and sorry to sit and do absolutely nothing.
[01:30:13] 01:30:08:08 - 01:30:36:26 MO GAWDAT
[01:30:13] And again, you know, and, and and, and, and  simply, you know, chat and ponder and love
[01:30:21] and connect and play music and, you know, not  have to suffer the promise that was implanted
[01:30:31] in your head as your purpose by capitalism.  Wake up every morning, stay in the commute,
[01:30:37] go work really hard. If you work your ass  off, you're going to make a few dollars more.
[01:30:42] 01:30:36:28 - 01:31:00:28 MO GAWDAT
[01:30:42] Then you're going to need to buy better suits  to go and make those few dollars more. So
[01:30:46] you're going to have to work even harder.  Right. And and it's quite interesting that,
[01:30:51] you know, this abundant future promises for  all of us to just go back to living, even in,
[01:30:58] more interestingly, in a safer,  more, famine, proof environment.
[01:31:06] 01:31:01:00 - 01:31:25:10 MO GAWDAT
[01:31:06] And yet we struggle with that. We struggle with  that not because it's not a good life. We struggle
[01:31:11] with that because we don't know how to do it.  And I I'm I'm the first to blame for years now,
[01:31:17] I, I constantly said to myself, I've worked  hard enough. I've, I've contributed enough,
[01:31:23] I've made enough. Maybe I should  just find my plate, my work, myself,
[01:31:27] a farm somewhere, and just go live on a farm.
[01:31:30] 01:31:25:12 - 01:31:52:29 MO GAWDAT
[01:31:30] Right. Take my loved ones if they want to  come visit. Whatever. I love that. But every
[01:31:36] time I do that, I go, like, where's the nearest  supermarket? Because I don't know anything else.
[01:31:44] I have to go to the, you know, tofu aisle.  If I wanted to make a stir fry, you know,
[01:31:51] and that's actually quite interesting. I've  been spoiled by the choice of an easy life.
[01:31:58] 01:31:53:01 - 01:32:14:02 MO GAWDAT
[01:31:58] Right. And, and and it's not easy, by the  way, going to the supermarket. So I was I
[01:32:05] was spoiled by the choice of a promise of  an easy life. That's not easy. And really,
[01:32:10] interestingly, maybe one day I'll  be forced to go back to AFA and
[01:32:15] maybe on that farm and eat different  things and live different ways, right?
[01:32:19] 01:32:14:05 - 01:32:42:16 MO GAWDAT
[01:32:19] But then will I be able to love it? And I think  that's the challenge that humanity faces. The
[01:32:26] challenge that humanity that everyone  needs to sit down and reflect on now,
[01:32:30] is which of those future groups would I want  to be? Will I want to be in the virtual reality
[01:32:36] world? What? I want to be the snake oil  salesman, what I want to be, you know,
[01:32:41] and one of the very few employees in the, you  know, control center of one of the major players.
[01:32:48] 01:32:42:18 - 01:33:01:08 MO GAWDAT
[01:32:48] Or will I want to be in nature or would I  want to be in a big city living with UBI
[01:32:52] and partying day and night? Right. Which  one do you want to be if you ask me and
[01:32:57] go back to nature. I live a very simple  life. You know, some people would say,
[01:33:02] oh, by the way, and we're going to  give you 100 years of life more.
[01:33:06] 01:33:01:11 - 01:33:29:28 MO GAWDAT
[01:33:06] I'll say thank you. Very happy with  my biological life that, you know,
[01:33:12] I honestly the only reason why you would want  to live a hundred years more is if the past
[01:33:19] 50 were not enough, right? I think I  think we've overdone it as humanity.
[01:33:26] I think we've pushed it to the point where we're  constantly sold things that we've never asked for.
[01:33:35] 01:33:30:00 - 01:33:51:18 MO GAWDAT
[01:33:35] And I think, and I, you may have heard  me mention or hint to that a few times,
[01:33:40] that the final outcome of that,  unfortunately, is a lot of evil,
[01:33:46] is a perpetual war, is a lot of civilians  killed, an economic crash every now and
[01:33:51] then that takes your wealth and your  grandmas, you know, retirement fund away.
[01:33:57] 01:33:51:18 - 01:34:06:16 MO GAWDAT
[01:33:57] And it's just I don't know if this is  the life I want. And I don't know if
[01:34:01] we should approve of that life, just to get,  a better, a faster call center agent. Right.
[01:34:12] 01:34:06:18 - 01:34:07:01
[01:34:12] GEOFF NIELSON And one.
[01:34:12] 01:34:07:01 - 01:34:09:03
[01:34:12] MO GAWDAT Of the.
[01:34:14] 01:34:09:06 - 01:34:28:00 GEOFF NIELSON
[01:34:14] There's a piece in there I want to add, which  is that coming back to that question of,
[01:34:20] you know, what do we want or what should  we want? There's a component in there,
[01:34:25] I believe, of human nature that  is the catalyst for all of this,
[01:34:29] which is when you can't answer that  question by yourself of, what do I want?
[01:34:33] 01:34:28:02 - 01:34:58:07 GEOFF NIELSON
[01:34:33] I think we're very quick to to flip  the question and ask ourselves, well,
[01:34:38] what does everybody else want? Yeah.  What what's so awful? Isn't that what
[01:34:42] I should want? Yeah. Right. And that  becomes very easy to manipulate and,
[01:34:47] and creates a lot of opportunity for snake oil  for, you know, nefarious parties to influence
[01:34:55] what we want. Okay. Can we get past that or is or  do we have to, like, do we have to recognize that?
[01:35:03] 01:34:58:07 - 01:35:04:16 GEOFF NIELSON
[01:35:03] And that's the way we break  free. What do we I mean,
[01:35:06] do you believe that? And if you do believe  it, what do we do with that information?
[01:35:10] 01:35:04:18 - 01:35:25:18 MO GAWDAT
[01:35:10] I think there are interesting habits that one  can develop. Right. So so all of us go through
[01:35:16] stages in life. So there is the stage of  a accumulation. If you want more wealth,
[01:35:21] more things, more cars. And I've  developed a habit for example simple,
[01:35:25] very simple that I want to take ten  things away from my home every Saturday.
[01:35:31] 01:35:25:20 - 01:35:54:10 MO GAWDAT
[01:35:31] Right. And you'll be amazed. You'll be amazed  how many Saturdays I succeed. It's incredible.
[01:35:38] Really. Like the the more I. And I've done  that for years. For years there's all still
[01:35:44] all that shit that I don't even remember  when I bought. Okay. And and you know,
[01:35:50] and of course, because of my very,  you know, stressful lifestyle,
[01:35:54] I'd be traveling somewhere about to to board  a flight and I'm going to be home tomorrow.
[01:36:00] 01:35:54:10 - 01:36:12:29 MO GAWDAT
[01:36:00] So I go on one of the e-commerce sites  here. Here in the UAE we use something
[01:36:04] called none. We don't like Amazon anymore.  And and basically we we we we sort of,
[01:36:10] you know, I sort of buy three things and  send them over at home. And I, you know,
[01:36:14] when they arrive, I ask myself,  what were those, what did I order?
[01:36:18] 01:36:13:04 - 01:36:33:03 MO GAWDAT
[01:36:19] But I, you know, and why did I order it? And  so, so the, the real I mean, those problems
[01:36:27] of privilege are going to go away for many of us.  It's just to begin with that. But maybe you should
[01:36:33] be prepared and, you know, and I this is supposed  to be a conversation about the future and I.
[01:36:39] 01:36:33:06 - 01:36:57:26 MO GAWDAT
[01:36:39] But believe it or not, a big chunk of it is about  humanity. And a big chunk of that conversation
[01:36:45] about humanity is are you able, as a human, to  actually look at your life and find out what in
[01:36:52] it brings you? Joy? Keep that. And what in it is  draining you, bleeding you, and get rid of that.
[01:37:03] 01:36:57:28 - 01:37:35:12 MO GAWDAT
[01:37:03] Right. And that that includes, by the way,  not just things, but relationships, but,
[01:37:12] you know, work, but investments,  but, virtual engagements, like,
[01:37:19] you know, ask yourself at the end of every  manic swiping session on social media,
[01:37:26] right. If you feel any better. And,  and, you know, just the simple act
[01:37:34] of awareness and awareness is not an act, but  the simple, you know, ability to become aware.
[01:37:41] 01:37:35:14 - 01:38:08:22 MO GAWDAT
[01:37:41] Changes everything, changes everything.  Because suddenly, you know, you realize,
[01:37:48] it's it's not really enriching my life. Maybe I  shouldn't have that much of it anymore. Whether
[01:37:54] that's sugar by the way. Right. Which is sold to  us constantly by consumerism. Right. Or, you know,
[01:38:03] as the incredible Yanis Varoufakis,  writes about the techno feudalism,
[01:38:09] the idea that we all become slaves  to some tech companies, right.
[01:38:14] 01:38:08:25 - 01:38:28:01 MO GAWDAT
[01:38:14] Who are the new digital landlords of  the world, right? Or whether it's,
[01:38:19] you know, a weird plastic apparatus  that you bought from an e-commerce
[01:38:23] site somewhere that's sitting in your home  and taking space and has never been used.
[01:38:31] 01:38:28:04 - 01:38:49:12 GEOFF NIELSON
[01:38:34] Let some let's maybe take this in a direction  that's, you know, a of practical use to people
[01:38:41] who are working right now and are trying to  figure out how they can be happier or how
[01:38:46] they can reduce their stress, because I think,  you know, there's a conversation that can say,
[01:38:51] oh, well, you know, your stressor is your job, so  you just have to quit your job if you're stressed.
[01:38:55] 01:38:49:12 - 01:39:21:14 GEOFF NIELSON
[01:38:55] Right. And that's that. That's, you know, a  more extreme path. You've written and talked
[01:39:00] extensively about stress for people who  are feeling stressed. Maybe that's because
[01:39:08] of their work, maybe that's because of their  relationship. You know, maybe that's because
[01:39:13] of their investments. Probably it's because  of all of the above. What habits can we
[01:39:19] practice or at least think about that help  us feel better and feel happier every day?
[01:39:27] 01:39:21:16 - 01:39:30:19 GEOFF NIELSON
[01:39:27] Short of, you know, quit your job, leave  your wife. You know, go off the grid.
[01:39:35] 01:39:30:21 - 01:39:59:23 MO GAWDAT
[01:39:36] There are so millions of options short of that.  So, let's talk about the big picture. First
[01:39:43] one is an awareness that this is not your  natural state. Okay? That's, you know,
[01:39:49] stress is a biological response that's made  to escape a tiger. Really? Right. It's a it's
[01:39:55] a it's a mixture of a hormone cocktail that  is supposed to reconfigure you to superhuman,
[01:40:01] and that it's not supposed to trigger  to be triggered with an image.
[01:40:05] 01:39:59:25 - 01:40:32:06 MO GAWDAT
[01:40:05] Right? It's not supposed to be triggered with a  comment on social media. Okay. And and that's the,
[01:40:12] the, you know, because of the nature of how  stress is, it is supposed to be short lived
[01:40:19] if it lingers, you know, if you remain in  that, superhuman configuration too long,
[01:40:26] you're depriving your liver and your, you  know, vital organs, your digestive system
[01:40:32] and so on of the energy they need to survive  at some people have been stressed for years.
[01:40:38] 01:40:32:09 - 01:41:03:07 MO GAWDAT
[01:40:38] Right. There is always going to be that, you know,  businessmen on the cover of fortune magazine with
[01:40:48] a striped, suit and, you know, always, always,  always angry. Right. And he would say, you know,
[01:40:55] people perform best when they're stressed.  No, they're not. They don't. People perform
[01:40:59] best when they are creative, when they are  working with amazing teams, when they are
[01:41:04] in flow and they're in love when they are happy,  you know, and it depends on what performance is.
[01:41:09] 01:41:03:07 - 01:41:25:16 MO GAWDAT
[01:41:09] If you want to squeeze 2% more, from a  worker on a, you know, manufacturing line,
[01:41:14] maybe. But if you want creativity or innovation,  good luck. Right now, the promise that we perform
[01:41:21] better under stress is a lie, and awareness of  that is important that some stress is useful.
[01:41:27] So you have a presentation next week.  Yeah. And you want to double down on it.
[01:41:31] 01:41:25:24 - 01:41:59:07 MO GAWDAT
[01:41:31] Stress is good for you right.  But but it's not it's not
[01:41:36] sustainable if you do that all the  time. So so my work on on stress,
[01:41:41] I worked with Alice Lau, who is an incredible  British artist, a so a British author that is,
[01:41:47] very feminine in her approach. I'm very logical  in my approach. So so I look at stress as an
[01:41:53] equation, basically, that if you learn from  stress in physics where objects are stressed,
[01:41:59] not just by the forces applied to them, but by  the square area that they carry that force with.
[01:42:05] 01:41:59:14 - 01:42:33:10 MO GAWDAT
[01:42:05] Right. So, so the, you know, the  cross-section of the object is, is a factor.
[01:42:10] Then basically a stress in humans, very analogous  is this challenges that are stressing you divided
[01:42:16] by the skills and resources and abilities and  contacts and so on, that you have to deal with
[01:42:21] it right now. If you see it that way, suddenly  it becomes very clear that you either reduce the
[01:42:28] forces applied to you, or you increase the  abilities and skills, and it really doesn't
[01:42:33] take a, you know, an equation to understand that,  you know, things that stressed me when I was 20.
[01:42:39] 01:42:33:10 - 01:42:57:23 MO GAWDAT
[01:42:39] I freaked out about them in my 30s. I handled them  in my 40s. I handle them with ease. And in my 50s,
[01:42:45] I laugh about right. It's not  because they're easier, okay,
[01:42:49] but because I developed more cross  section. If you want cross area. So. So,
[01:42:55] so when you think about it, you want to invest in  your skills if you want to in dealing with stress.
[01:43:03] 01:42:57:26 - 01:43:26:09 MO GAWDAT
[01:43:03] And I think the most important skills is the most  important skill is, is, is to one on the top,
[01:43:13] reduce limit your stressors. Right. And, and  most of the stressors that break us are not
[01:43:19] big. You know, trauma, is the macro external  stress comes from outside this, trauma is,
[01:43:27] you know, every one of us, 91% of us will get one  PTSD, traumatic event once in a lifetime, right?
[01:43:32] 01:43:26:16 - 01:43:55:08 MO GAWDAT
[01:43:32] Losing a loved one or being in an accident,  and so on. 93% will recover in three months,
[01:43:38] 96.7% will recover in six months. So  trauma is a temporary break if you want
[01:43:44] the ones that last year are different and the  ones that lasts are burnout, right? Or what I
[01:43:50] normally call anticipation of a threat. So burnout  is the sigma of all of the little stressors that
[01:43:57] you have multiplied by their intensity, by that  frequency, by the time of their application.
[01:44:01] 01:43:55:10 - 01:44:13:02 MO GAWDAT
[01:44:01] And and basically we have so many of  those, and then eventually you add one
[01:44:06] of them on top and you burnout.  Right. And most people will say,
[01:44:10] you know, I need to remove the stressors  in my life so that I don't burn out. No,
[01:44:14] it's actually you need to move every stressor  you can do, move. It's not just the big ones.
[01:44:19] 01:44:13:04 - 01:44:32:26 MO GAWDAT
[01:44:19] So, so, you know, from your very loud alarm in  the morning, that's the first jolt of stress,
[01:44:24] right? To choosing to go on your commute  at the in the rush hour to, to to to,
[01:44:30] right. And, and and the way to handle them  is next Saturday. You sit down with a piece
[01:44:35] of paper and write down everything  that stressed you the last week.
[01:44:39] 01:44:33:01 - 01:44:50:25 MO GAWDAT
[01:44:39] Right. And you do that frequently, by the  way, not just the next Saturday. And then
[01:44:43] you scratch out the ones that you can remove that  annoying friend that constantly is negative. You
[01:44:49] can literally have a conversation with them  and say, look, this is really stressing me.
[01:44:52] Can you please be nicer? Right? Or maybe  you shouldn't be friends, or whatever,
[01:44:57] 01:44:51:01 - 01:45:10:14 MO GAWDAT
[01:44:57] So, so anything that you can remove, remove  anything that you can reduce the intensity of,
[01:45:02] reduce the intensity of it, and anything that you  cannot remove or reduce the intensity of sweeten,
[01:45:09] make it lighter. So if you really have  to do the commute at a certain time,
[01:45:13] take some music with you,  maybe a nice coffee and so on.
[01:45:16] 01:45:10:17 - 01:45:33:22 MO GAWDAT
[01:45:16] Right. So this is one. But by doing,  by limiting stressors, by the way,
[01:45:22] I should I should say that stressors are  mostly internal not external. So so we call
[01:45:27] them a ton TR. And then in the book T t is the  trauma. He spoke about that always obsessions.
[01:45:34] There are big big events that stress us  very deeply but they come from within us.
[01:45:40] 01:45:33:23 - 01:45:57:20 MO GAWDAT
[01:45:40] I'm a failure, I'm a failure. I'm a failure.  Nobody will ever love me or whatever. New,
[01:45:46] noise. Small ones. Niggles if you want.  Right. And the last n is nuisances.
[01:45:53] Little stressors. Sub trauma. Right. If  you look at it, the obsessions and the
[01:45:58] and the noise are coming from within you.  And then the majority of the stress right.
[01:46:03] 01:45:57:22 - 01:46:22:29 MO GAWDAT
[01:46:04] The of that category, the obsessions and  the noise. We get what I normally call the
[01:46:12] anticipation of a threat. So stress is supposed  to you get you're supposed to get cortisol when
[01:46:17] the tiger shows up. Okay. In the modern world,  we get cortisol before the tiger shows up,
[01:46:25] right? We're stressed before the tiger  shows up, because we mix up four emotions.
[01:46:29] 01:46:23:01 - 01:46:44:07 MO GAWDAT
[01:46:29] There is fear and what I call it for fear  and all of its derivatives. So there is fear,
[01:46:34] there is worry, there is anxiety, and there  is, panic. Right. And if you're online today,
[01:46:40] panic attacks and anxiety attacks  are more common than, you know,
[01:46:44] than anything else. And the reason is because  we deal with those things as if they were fear.
[01:46:50] 01:46:44:09 - 01:47:06:23 MO GAWDAT
[01:46:50] Right. So let me try to explain  this quickly and then shut up,
[01:46:53] for fear is a moment in the future is less safe  than no right. And so there is a threat in the
[01:47:00] future. And so the typical actual natural  reaction to fear is you address the threat,
[01:47:07] right? Worry is not that what it is? I can't  make up my mind if there is a threat or not.
[01:47:13] 01:47:06:25 - 01:47:29:18 MO GAWDAT
[01:47:13] Should I chill or should I, freak  out? Right? And accordingly,
[01:47:18] you keep flip flopping and and and that  constant indecision is what stresses you,
[01:47:23] right? So when you feel worried, turn  it into either fear or safety or a sense
[01:47:29] of safety. So tell yourself, am I going to  make up my mind? Am I going to lose my job?
[01:47:36] 01:47:29:18 - 01:47:50:00 MO GAWDAT
[01:47:36] So I now need to go look for another job and and  go down that path? Or am I going to actually keep
[01:47:41] my job so I need to double down and get the  next promotion right. So so this is worry.
[01:47:48] Panic is not a question of, the threat.  It's a question of how soon is the threat.
[01:47:56] 01:47:50:05 - 01:48:12:16 MO GAWDAT
[01:47:56] It's a question of time. We panic when the  threat is imminent. Right. So if if you have
[01:48:01] a presentation in a month's time, you don't panic  about it. Right? But when it's tomorrow and you're
[01:48:06] not ready, you start to panic, right? And so  when you panic, don't treat it as a threat. Don't
[01:48:13] treat the threat. Because if you're out of time,  treating the threat makes you panic more right.
[01:48:19] 01:48:12:23 - 01:48:33:27 MO GAWDAT
[01:48:19] When you're when you feel a panic, treat time.  Try to give yourself more time. Call the person
[01:48:25] and say, can we make it 3 p.m. instead of 1 p.m.?  Can we make it next week? You know, find a friend
[01:48:30] that can help you, give you more time by doing  some of the tasks. Empty your agenda and don't,
[01:48:35] you know, drop the things that you don't need to  do tomorrow so that you're preparing and so on.
[01:48:40] 01:48:33:29 - 01:48:56:08 MO GAWDAT
[01:48:40] Right. And then finally, anxiety, the  top of all pandemics of our world today
[01:48:45] is not about the threat either. Anxiety  is about my capability of dealing with
[01:48:50] this act. Right. So if I if I'm if I feel  that there is something threatening in the
[01:48:56] future and I feel that I'm not prepared  to handle it, I feel anxious. Right.
[01:49:03] 01:48:56:14 - 01:49:16:25 MO GAWDAT
[01:49:03] And so if you treat it like fear  and attempt to deal with the threat,
[01:49:07] you discover your inability. So it reinforces  your anxiety. And that cycle continues. Right?
[01:49:14] When you feel anxious, work on your skills.  Don't work on the threat, okay? You know,
[01:49:18] find someone to teach you that bit that you  don't understand. Learn it on on YouTube.
[01:49:23] 01:49:16:25 - 01:49:40:02 MO GAWDAT
[01:49:23] Find someone that you can partner with that  can take the bits that you don't know and so
[01:49:26] on and so forth. So. So what am I trying to  say? I'm trying to say that even though we're
[01:49:31] surrounded with stressors, life is never  going to stop stressing you. The truth,
[01:49:36] which is quite interesting, is it's a choice. It's  a choice for you to limit some of those stresses,
[01:49:42] and it's a choice for you how you deal  with those stresses by developing circuits.
[01:49:46] 01:49:40:04 - 01:49:50:11 MO GAWDAT
[01:49:46] Right. And if you know, the more you invest  in those things, knowing that stress is not
[01:49:51] your natural state, the more it becomes an  easier task because you develop those skills.
[01:49:56] 01:49:50:14 - 01:50:15:01 GEOFF NIELSON
[01:49:56] I wanted to talk about one specific scenario  that I think is probably fairly common with
[01:50:02] people these days, and maybe, maybe  you've experienced it somewhere along
[01:50:04] the way at Google. And I think you  can probably see it whether you're,
[01:50:08] you know, a junior employee or even a leader,  which is that certainly more even since the
[01:50:15] pandemic. I think this anxiety and people  and and blow this up if you don't like it.
[01:50:21] 01:50:15:07 - 01:50:56:07 GEOFF NIELSON
[01:50:21] But this anxiety that we feel is we've we've  ended up in this world where either our boss
[01:50:29] or our organization is the tiger bow.  So, so the way based on our workloads,
[01:50:37] if we're knowledge worker, is based on everybody  pushing us harder. All these tasks coming down
[01:50:42] the pipeline and coming that coming down  the pipeline in a way that's unpredictable,
[01:50:47] makes you just feel like you're always in  the cage with the tiger because there's
[01:50:51] anticipatory anxiety, because you're in  these organizations that are disorganized
[01:50:57] enough that you can't predict what your  day or your week is going to look like.
[01:51:02] 01:50:56:09 - 01:51:10:19 GEOFF NIELSON
[01:51:03] And that triggers this, this cycle of stress.  How what tactics or what approaches would
[01:51:11] you recommend people take if they find  themselves in the situation like that?
[01:51:17] 01:51:10:21 - 01:51:37:11 MO GAWDAT
[01:51:17] It depends on how. So by the way, that's  true. Sometimes the bosses, the tiger for
[01:51:22] sure. Sometimes an email is the tiger. But is it  true? Like is it, does it have to be that way?
[01:51:31] You know, so it depends on where you are in the  organization. And, you know, in my junior years,
[01:51:36] I used to never start working any day until I had  a things to do list next to me on my desk, right?
[01:51:44] 01:51:37:18 - 01:52:01:03 MO GAWDAT
[01:51:44] With times allocated to it. Right. That actually  clearly showed that I wasn't a lazy person,
[01:51:52] that I was doing the absolute best I can to do  as many tasks I as I can. I prioritize them.
[01:51:58] They normally were only a subset of  all of the tasks available to me,
[01:52:01] and then someone would pop up, and say, Mo,  seriously, I need you to do that review.
[01:52:07] 01:52:01:03 - 01:52:19:00 MO GAWDAT
[01:52:07] It's really important the customer is waiting. I  brought that up, but whatever. And so on. Right.
[01:52:13] And my response in a very common way would be, oh,
[01:52:16] I would love to do it. But we need to remove one  of those. Okay. If you want to remove this one,
[01:52:23] talk to that person. If you want to  remove this one, talk to my boss.
[01:52:25] 01:52:19:00 - 01:52:51:11 MO GAWDAT
[01:52:25] If you want to remove this one, you know  and so on. And it's not that I'm lazy,
[01:52:31] it's CDC. Those people expect those things  from me. So would you kindly just do that
[01:52:37] task so that I can prioritize my work? I'm here  to help. Right. So if you're if you're a junior
[01:52:43] in the organization, being on top of your, on  of of on your tasks because when you're junior,
[01:52:50] you're at Task Clercq, being on top of your tasks  really helps you midway in the organization.
[01:52:58] 01:52:51:11 - 01:53:20:14 MO GAWDAT
[01:52:58] So, you know, if you're in management  or junior leadership or, you know,
[01:53:03] not the top leader if you want. Okay. Sure.  You, you you need to shift the the focus of
[01:53:13] your boss from tasks to objectives.  Right? So I remember vividly one of
[01:53:21] my favorite bosses of all time was my first  boss at Google, who was very harsh, right?
[01:53:27] 01:53:20:18 - 01:53:48:24 MO GAWDAT
[01:53:27] Harsh in terms of he wanted us to to, to  to to to thrive, really. And and, you know,
[01:53:36] I, I did things differently. I have some brain  defects. Some areas of my brain are missing. And
[01:53:43] so there are tasks that I'm not good at, but there  are tasks that I'm better than others. And I'm in
[01:53:49] one of those management meetings, you know, one  of my peers said, why doesn't region four do that?
[01:53:55] 01:53:48:24 - 01:54:13:12 MO GAWDAT
[01:53:55] Why is more not doing this like you're asking  it from us? Okay. And and my boss was about to
[01:54:02] pounce on me. And I responded quickly. And  I said, because I'm growing 29% and you're
[01:54:07] growing too. Is that a good reason? Okay. And so  we had this interesting organized conversation.
[01:54:14] And then I basically told my boss, look,  please let me do things the way I want.
[01:54:20] 01:54:13:14 - 01:54:14:16
[01:54:20] GEOFF NIELSON Back off.
[01:54:21] 01:54:14:18 - 01:54:40:21 MO GAWDAT
[01:54:21] I, I'm really doing well here. If you  if you force me to do them differently,
[01:54:26] I'm going to fail because it's solves my skill  set, right. The day I failed doing the my way
[01:54:32] fired me. Right. And get someone who can do it  your way. So funny. Funny. The next morning,
[01:54:39] we are standing in the international sales  conference, or the next week or something
[01:54:43] where we have, you know, basically  8000 Googlers in the in the audience.
[01:54:47] 01:54:40:21 - 01:55:03:17 MO GAWDAT
[01:54:47] And, and, you know, someone asks and says also why  why is region for not doing this this way. And,
[01:54:54] and you know, my boss responds and I quote,  he goes like, well, I have no idea how small
[01:55:00] does what he does, but when he stops doing  it, I'm going to fire. Okay. So my response
[01:55:05] in the audience is I put my hand in the air  and say, yeah, that's exactly what I want.
[01:55:10] 01:55:03:19 - 01:55:31:26 MO GAWDAT
[01:55:10] I want the freedom to perform the way  I perform. While that comes with the
[01:55:15] responsibility of delivering to the  company as the company wants. Right.
[01:55:19] If you're the top guy. Seriously. Chill.  Right. So, so I had I hosted I'm normally in,
[01:55:29] in my approach you know, and at Google  X for example, my business team would
[01:55:33] come in and and talk about you know we  have this pipeline of 16 opportunities.
[01:55:38] 01:55:31:26 - 01:55:51:06 MO GAWDAT
[01:55:38] This is this, this is that. And then  after opportunity number three I go like,
[01:55:42] that's it. I don't need to know more. These three  are enough. Right. And they go like, no, no. But
[01:55:47] the others are interested. And I'm like, look,  if you focus on 16 you're not going to be able to
[01:55:52] serve them properly. I think you should go to the  other 13 and tell them we'll work on those later.
[01:55:58] 01:55:51:09 - 01:56:12:20 MO GAWDAT
[01:55:58] Right. Focus on the three, close them  and then let's talk again. Anyway,
[01:56:01] they they wouldn't. But that was  my style until I met. I hosted a,
[01:56:06] a fortune 500 CEO at Google X and had a wonderful  conversation talking about things and, you know,
[01:56:15] running out of time. I said, you know, you  know what? You need to come back another time.
[01:56:19] 01:56:12:20 - 01:56:33:11 MO GAWDAT
[01:56:19] I really want to show you this is very  interesting. And he said, why another time? I
[01:56:24] have time. I was like, oh, that's an interesting  CEO. You're not that busy. And he says, no,
[01:56:28] I work four hours a day. And I said, what?  He said, I work four hours a day. And I said,
[01:56:34] how? And he said, look, any meeting that's  less than an hour is too personal for me.
[01:56:40] 01:56:33:11 - 01:57:00:27 MO GAWDAT
[01:56:40] So that's why I don't attend. Right. Any meeting  that starts and five minutes in, they're not well
[01:56:47] prepared. I leave, okay? And because, and I have  I take only four meetings a day because more than
[01:56:56] that means that there are way too many strategic  problems in the company. Right? If a company is
[01:57:01] running well, more than four strategic decisions  a days means you're changing too much, right?
[01:57:08] 01:57:01:03 - 01:57:28:13 MO GAWDAT
[01:57:08] So basically, he said. And then in the remaining  four hours, I walk around the corridors and hug
[01:57:13] everyone. What a strategy, right? And and once  again, remember, the difference between leadership
[01:57:21] and management is that management is whipping  everyone. To try and squeeze 1% more. Leadership
[01:57:28] is hugging everyone, and so many people, as they  go through the ranks, fail to recognize that.
[01:57:35] 01:57:28:13 - 01:57:52:10 MO GAWDAT
[01:57:35] They fail to recognize that I really  don't need to with anyone anymore. I'm.
[01:57:40] I've hired senior VP's who are some of the most  intelligent people in the world reporting to me,
[01:57:47] so I might as well let them be senior VP.  Right? And so again, it depends on which part
[01:57:53] of the organization you are. It all starts with an  acknowledgment that I'm not here to suffer, okay?
[01:57:59] 01:57:52:10 - 01:58:03:14 MO GAWDAT
[01:57:59] I'm here to perform. And performance  doesn't necessarily like the guy on
[01:58:04] the cover of fortune magazine, you know,  doesn't necessarily come from stress.
[01:58:10] 01:58:03:16 - 01:58:39:18 GEOFF NIELSON
[01:58:10] Yeah. Thank you for that. That's that was,  a really, really excellent answer. And I
[01:58:17] love the way you broke that out. And, it it  really resonated with me. And I hope that it,
[01:58:24] that it resonated with a few people listening as  well, and. Yeah. Yeah, I mean, the, the comments
[01:58:29] about chilling out, you know, certainly it feels  like we've extrapolated too far this idea of,
[01:58:39] you know, line work of I'm only as productive  as the number of hours I put in the day,
[01:58:43] all the way up to a CEO of whatever organization.
[01:58:46] 01:58:39:21 - 01:58:55:28 GEOFF NIELSON
[01:58:46] And being able to break free of that and  saying, no, actually, less is more. And,
[01:58:51] you know, there's a quote somewhere about  strategy is choosing what not to do.
[01:58:55] I can't attribute it properly off  the top of my head. But but but I
[01:58:58] love that. And I think it's such an  important message for for leaders.
[01:59:03] 01:58:56:00 - 01:59:23:16 MO GAWDAT
[01:59:03] Yeah. It's so true. It is so true that 80% of what  you do makes you advance 5% more. And, you know,
[01:59:13] again, it's a bit like consumerism and capitalism.  Really. Do I really need that 5%? Like, you know,
[01:59:21] if if I work my backside off this year, the money  that I can make might help me buy a fancy car.
[01:59:31] 01:59:23:19 - 01:59:58:06 MO GAWDAT
[01:59:31] Should I trade my life one full year for a fancy  car? It doesn't sound very wise to me. Honestly.
[01:59:41] And and I truly and honestly believe that most  people, when they look back at their life,
[01:59:48] they just realize that they've invested  their heartbeats in the wrong things,
[01:59:53] right? I mean, in a very interesting way.  Remember, even today, I, I sit on many,
[01:59:59] many boards and I, you know, I advise  many governments and leaders and so on.
[02:00:05] 01:59:58:08 - 02:00:33:22 MO GAWDAT
[02:00:05] It's not because of my heartbeats, do you  understand? Is that I don't sell time. This
[02:00:11] is really interesting that most people who really  figure it out understand that if you really invest
[02:00:19] in something that you're good at and become  noticeably better than the average person at it,
[02:00:28] you can probably live a very comfortable life.  Just, you know, sharing what it is that you know,
[02:00:36] about that thing and it and that, by  the way, applies to employment as well.
[02:00:40] 02:00:33:24 - 02:01:07:21 MO GAWDAT
[02:00:40] We used to have distinguished engineers, okay.  Distinguished engineers really didn't code much
[02:00:47] at all. Most of the time they didn't even code  right. But they had that incredible skill that by
[02:00:55] them sharing half an hour with a junior engineer,  that junior engineer becomes twice as productive,
[02:01:01] solves a problem that could have taken him  six days. Right. And and really, you know,
[02:01:07] you really need to reflect on your life and  say, am I still behaving as that freshman?
[02:01:15] 02:01:07:21 - 02:01:20:00 MO GAWDAT
[02:01:15] Just that just came out of college,  right? Just putting more of this
[02:01:19] in my life every day and thinking  that I'm becoming a senior leader.
[02:01:27] 02:01:20:02 - 02:01:50:27 GEOFF NIELSON
[02:01:27] Yeah. Wow. Well, and it it sounds like  there's so much room for reflecting on what,
[02:01:35] what are you really good at, for one. And and  what is actually going to have that impact
[02:01:40] and move the needle 100% versus 5%, 100%.  And and having I'll call it the courage,
[02:01:46] I guess, to let go of all the other  things and and getting rid of the
[02:01:50] mindset of just more as more and every  incremental 1%, you know, is worth it.
[02:01:58] 02:01:50:29 - 02:02:17:08 MO GAWDAT
[02:01:58] My wonderful ex-wife, at a point in time,  I reported that let's not mention names,
[02:02:04] but one of my peers was the funniest human  being, the loveliest human being alive.
[02:02:10] Right? So still one of my best friends today.  And he worked. He was good at what he did,
[02:02:18] but he was a party animal. Like he  would take the boss every other evening.
[02:02:24] 02:02:17:08 - 02:02:36:16 MO GAWDAT
[02:02:24] They would go left their heads off. Right.  And. And you can't help it. The boss loved him,
[02:02:29] right? He's very lovable. I love him, okay.  So one day I went back to my wife and I said,
[02:02:35] baby, I really think I should be  more of a wine and dine kind of
[02:02:39] person. I'm a businessman. I'm supposed  to take the boss and the clients out.
[02:02:43] 02:02:36:18 - 02:02:56:10 MO GAWDAT
[02:02:43] And so some evenings I'll be late for  dinner or, you know, I won't. I won't
[02:02:48] join for dinner. And she looked at me. You know,  that's what a good wife should do. And she said,
[02:02:54] of course, maybe we should do it.  We'll do whatever you think is right,
[02:02:57] but you're going to be mediocre at it  at best. I said, so what do you mean?
[02:03:03] 02:02:56:10 - 02:03:16:01 MO GAWDAT
[02:03:03] And she said, this is really not you.  You're a you're a thinker and a philosopher,
[02:03:08] and you know what? Client wants to go out and  talk about the, you know, the, the ailing,
[02:03:14] you hear human, you know, fortune as a  result of capitalism. That's not. No.
[02:03:20] Nobody wants that. You know,  you're friend is good at it.
[02:03:23] 02:03:16:03 - 02:03:33:27 MO GAWDAT
[02:03:23] Okay. You might as well just come home. You know,  I never really came home early at the time. I,
[02:03:29] you know, come home at 8 p.m., relax a little.  You know, sleep well, go out the next morning
[02:03:35] and keep growing your business better than  everyone else, right? It's a choice. Yeah.
[02:03:41] 02:03:33:29 - 02:03:52:20 GEOFF NIELSON
[02:03:41] Yeah. No, I think that's, I think that's  very, very well said. There was one more
[02:03:49] thing we didn't talk about that I  did want to talk to you about today.
[02:03:52] And now especially that we're this  deep into the conversation that. Yeah,
[02:03:56] I like to pretend no one is listening at this  point anymore, so we can talk about whatever.
[02:03:59] 02:03:52:21 - 02:04:12:10 GEOFF NIELSON
[02:04:00] That's good. Yeah. You know, we we talking  about snake oil salesmen, and all the hype for,
[02:04:06] you know, 1,000,001 different things that  we absolutely have to have or learn about
[02:04:10] or buy. What's at the top of your bullshit  list right now? What are the things you're
[02:04:16] hearing about that people are talking  about, or hawking that you're saying?
[02:04:19] 02:04:12:10 - 02:04:23:25 GEOFF NIELSON
[02:04:19] You know what, this is bullshit. You know,
[02:04:22] if you're if you're investing in this  either financially or in terms of attention,
[02:04:27] you're wasting your time. It's not going  to pan out the way people are saying.
[02:04:31] 02:04:23:28 - 02:04:32:29 MO GAWDAT
[02:04:31] That's such an interesting question.  I do not know the answer to that. I
[02:04:35] actually waste none of my time to look  at bullshit. It's quite interesting.
[02:04:40] 02:04:33:01 - 02:04:35:06 GEOFF NIELSON
[02:04:40] That's fantastic of you.
[02:04:42] 02:04:35:09 - 02:05:02:06 MO GAWDAT
[02:04:42] Yeah, I, I was shocked by this question. I will  tell you, though, even if it's not bullshit,
[02:04:48] we're probably going to get a dotcom bubble  style thing, right? So in the current world where
[02:04:56] things are moving so fast, you're bound to make  mistakes, right? You know, if you're an investor,
[02:05:02] you're bound to invest in a company that has  all of the promising, you know, elements to it.
[02:05:09] 02:05:02:09 - 02:05:30:16 MO GAWDAT
[02:05:09] Correct. Founders. Good idea. Good technology,
[02:05:12] whatever. And then maybe someone else  beats them to it. Or maybe, you know,
[02:05:18] we don't know. It is such a fast paced world. And,  you know, with someone like Trump at the helm,
[02:05:24] you have absolutely no idea what will  happen tomorrow. So so, you know,
[02:05:29] it's actually it. You should probably expect  that 60% of your choices will be wrong, right?
[02:05:38] 02:05:30:18 - 02:06:02:21 MO GAWDAT
[02:05:38] And even if they're right, he's going to  do something stupid and and they're going
[02:05:41] to fail anyway. Right. And so so when you really  think about it, I wouldn't say I have a portfolio
[02:05:52] approach, but I would probably say invest in  industries, not companies. You know, if, if,
[02:06:00] if and if you're a startup founder yourself or if  you're a, a business yourself, invest in segments.
[02:06:10] 02:06:02:23 - 02:06:40:15 MO GAWDAT
[02:06:10] Not ideas. So basically, tell yourself I'm  going to be the absolute best at customer
[02:06:16] service and then invest in every part of that  segment or tell yourself, I'm going to be,
[02:06:21] you know, leading in efficiencies. Right.  And and so on. And and you can then add
[02:06:29] segments but if you try multiple approaches to  increasing your efficiency and multiple vendors
[02:06:36] and multiple ideas and you know, some will fail  and some will succeed, it's such a fast paced,
[02:06:43] you know, market that you're bound to make  some wrong mistakes that are some mistakes.
[02:06:48] 02:06:40:15 - 02:07:13:06 MO GAWDAT
[02:06:48] And, and I think making mistakes is actually much  less harmful than not deciding at all. Right. So
[02:06:56] so, you know, if, if you're going to be in call  center improvements, find the top five players,
[02:07:03] split your call center into five little  units and try each of them. Right.
[02:07:08] And and and believe it or not, as four of  them fail and you find out the one that works,
[02:07:14] you know, you can scale that in no  time at all and and benefit everyone.
[02:07:19] 02:07:13:09 - 02:07:51:14 MO GAWDAT
[02:07:20] Having said that, there is a lot of hype,
[02:07:25] and a lot of what actually matters is not  really hyped. Okay. It's quite interesting. I,
[02:07:33] I believe that, of course, reasoning and math  for AI has absolutely been the breakthrough. It's
[02:07:41] not, I don't think I don't think AI is fabulous.  It will be the core of everything that we do,
[02:07:47] and it's probably going to be an interesting  part of our demise because as we open up to
[02:07:52] agents as a CIA fisher, criminal intelligence,  as I call it, will find so many entry doors.
[02:07:59] 02:07:51:17 - 02:08:19:05 MO GAWDAT
[02:07:59] But, but the real breakthroughs has been  reasoning and mathematics. I mean, I,
[02:08:05] I used to say that my AGI, when it comes  to, linguistic intelligence, happened, in
[02:08:11] 2024. Right. But I could still beat them in math.  Good luck. Now, I'm. I'm nothing. And, you know,
[02:08:21] very few of my friends can beat them in math now,  you know, very few of my geeky friends, I've.
[02:08:26] 02:08:19:05 - 02:08:43:05 MO GAWDAT
[02:08:26] I was wiped out in 20, 20 and 23. In terms  of coding. Right. Some of my friends are
[02:08:35] still better coders than they are, but they'll be  wiped out in a year, for sure. And these, I think,
[02:08:43] are the true breakthroughs. These are the ones  that will make a massive difference. So if we.
[02:08:50] 02:08:43:05 - 02:09:00:06 GEOFF NIELSON
[02:08:50] Get, you know, deep reasoning which which you've  said before, we're probably less than a year away
[02:08:56] from if we get to this next level of reasoning of  math, of understanding what's what, what does that
[02:09:02] unlock, what what doors are open, or what are  the implications from AI being able to do that?
[02:09:08] 02:09:00:12 - 02:09:28:12 MO GAWDAT
[02:09:08] But both. It's always a singularity.  You're going to get some people that will
[02:09:11] use deep reasoning to, to hack the stock  market. And you're going to get people,
[02:09:17] that will use deep reasoning to invent  something amazing. Right. And, and,
[02:09:23] and both it's not one or the other.  Both would happen at the same time. My,
[02:09:29] my hope is that humanity will respond to the  hackers, by saying, hey, let's work together.
[02:09:36] 02:09:28:15 - 02:10:02:20 MO GAWDAT
[02:09:36] But, you know, there is no denying that there  are incredible breakthroughs in terms of our
[02:09:41] understanding of things because of the level  of intelligence that we now have access to.
[02:09:46] It's refreshing. It's refreshing, radiant. And  I say, I say that with, with a very, childlike,
[02:09:55] happiness. Because with age, I, I sort of started  to feel that I'm slowing down a little, like, you
[02:10:04] know, I still am a very reasonable mathematician,  but it takes me longer, which is really weird.
[02:10:10] 02:10:02:20 - 02:10:24:27 MO GAWDAT
[02:10:10] I hate it, okay. Takes me longer to do the  math. Maybe I'm not using it as often. Or
[02:10:16] maybe I'm just slowing down. And now suddenly,  you give me this new boost where I just need to
[02:10:22] know how to state the problem and someone will  do the math for me. And it's just incredible,
[02:10:26] right? You know, I, I just need to state the  problem and someone will do the research for me.
[02:10:32] 02:10:24:27 - 02:10:57:15 MO GAWDAT
[02:10:32] It's just so empowering. And and and  and it's, you know, when it comes to
[02:10:38] reasoning. Just think about this One of the top  limitations of humanity was multi disciplinary
[02:10:48] reasoning. Meaning there is a certain point  at which, for me to be a meaningful physicist,
[02:10:58] I need to so deeply specialize that I have no  space left in my head for chemistry or biology.
[02:11:05] 02:10:57:18 - 02:11:26:21 MO GAWDAT
[02:11:05] Right. And that's the truth of me and every,
[02:11:08] every scientists I've ever worked with.  You really, it's becoming so complex that
[02:11:14] you have to specialize. Right? And so your  reasoning when you solve complex problems
[02:11:21] is limited to your own capability. And if  you want to bring other specialists in,
[02:11:27] it's limited to the ridiculous bandwidths of,  of information communication that humans have.
[02:11:34] 02:11:26:23 - 02:12:12:15 MO GAWDAT
[02:11:34] Right. Imagine if I can if I can reason across  disciplines next year with that efficiency.
[02:11:42] Right. Imagine if I can allow artificial  intelligence to look at climate change,
[02:11:49] not just as a recycling and manufacturing  problem, but also as a physics problem that
[02:11:55] includes a bit of biology, a bit of, I don't know,  astrology. Right. And and basically, maybe we
[02:12:02] end up finding that if we took a certain bacteria  from Earth and sent it to space in a certain way,
[02:12:09] at a certain speed, in a certain angle, and  then brought it back and it fell on a palm tree,
[02:12:15] you know, it would, you know, consume  more of the CO2 in that in the world,
[02:12:20] 02:12:12:17 - 02:12:17:16 MO GAWDAT
[02:12:20] I don't know. Right. But that's the  promise of that is just incredible.
[02:12:25] 02:12:17:18 - 02:12:44:09 GEOFF NIELSON
[02:12:25] Yeah, yeah. And that's, I was thinking about  it earlier, much earlier in our conversation
[02:12:30] when you were talking about synthetic data.  Because, you know, for me, if you asked me,
[02:12:36] Jeff, what's the fastest way to start coming up  with scientific breakthroughs? It would be point
[02:12:41] AI at cross-disciplinary, you know, papers or  or pieces of literature or finding and saying,
[02:12:47] take all the physics papers here, take all the  biology papers here and cross-reference them.
[02:12:52] 02:12:44:09 - 02:13:05:24 GEOFF NIELSON
[02:12:52] Just all of them. Just do it and see what insights  you come up with, you know, and it doesn't have to
[02:12:57] be, you know, just two fields. You can do it with  every field and the amount you could unlock that
[02:13:02] no human could ever do so quickly. It's really  easy, at least for me, to imagine a world that
[02:13:08] completely transforms, you know, technology  and science in a very short amount of time.
[02:13:13] 02:13:05:24 - 02:13:06:17
[02:13:13] GEOFF NIELSON Yeah.
[02:13:14] 02:13:06:19 - 02:13:10:07
[02:13:20] MO GAWDAT Totally.
[02:13:20] 02:13:10:10 - 02:13:33:26 GEOFF NIELSON
[02:13:20] Yeah. My, I know we've had a very long  and at least for me, extremely interesting
[02:13:26] conversation. Thank you. I want I wanted to  say, you know, a huge, huge thank you for,
[02:13:31] for making the time and for sharing your  insights. There were so many things I wanted
[02:13:36] to talk with you about today, and I feel like  we covered just, a silly amount of ground, but.
[02:13:41] 02:13:33:26 - 02:13:51:02 GEOFF NIELSON
[02:13:41] But, everything to me still ties.  It ties together as we think about,
[02:13:47] you know, what's coming next for us, what  it means for people, what it means for the
[02:13:50] world. Like we we went up to, you know, the  level of, you know, the earth and the climate
[02:13:55] and nation states. We were down to the level  of, you know, us as individuals and purpose.
[02:13:59] 02:13:51:02 - 02:14:00:21 GEOFF NIELSON
[02:13:59] So I really appreciate it. I learned a  ton. I'm walking out of this, you know,
[02:14:04] room with a lot to think about. So I really  appreciate you sharing your insights. I, I.
[02:14:08] 02:14:00:21 - 02:14:23:06 MO GAWDAT
[02:14:08] I, I really enjoy it. I'm very, very grateful  for the time. I'm very grateful for the way
[02:14:13] you handled it and the questions you asked. I, you  know, I should again, maybe just close by saying,
[02:14:19] please don't take any of what I said as true. Just  take it as an interesting direction to consider.
[02:14:26] It's the, you know, the best of my analysis,  but it could absolutely be complete garbage.
[02:14:31] 02:14:23:06 - 02:14:43:16 MO GAWDAT
[02:14:31] So, you know, nobody knows. The future is  very arrogant to predict. That's when anyone
[02:14:35] knows. But yeah, I'm really grateful. And  I think it's by this moment it's just you
[02:14:40] and I in the podcast, everyone else left.  So, if anyone's still here, tell us. And,
[02:14:48] Yeah, I'm. I'm really grateful  for the opportunity. Thank you.