Skip to content

The Peter McCormack Show AI Business Career

A Futurist's Warning: "Work Is About to Break" | Amy Webb

[Amy Webb](amywebb.io), the quantitative futurist who runs [Future Today Strategy Group](ftsg.com) and teaches strategic foresight at [NYU Stern](www.stern.nyu.edu), sits down with [Peter McCormack](www.youtube.com/channel/UCzrWKkFIRS0kjZf7x24GdGg) in New York for roughly ninety four minutes to argue that work itself is about to break. Her thesis is not that a particular job vanishes in thirty six months. It is that the operating system of society is being rewritten underneath us, because the technologies arriving now, [AI](en.wikipedia.org/wiki/Artifici

Published Jul 9, 2026 1:34:21 video 46 min read Added Jul 11, 2026 Open on YouTube →

At a glance

Amy Webb, the quantitative futurist who runs Future Today Strategy Group and teaches strategic foresight at NYU Stern, sits down with Peter McCormack in New York for roughly ninety four minutes to argue that work itself is about to break. Her thesis is not that a particular job vanishes in thirty six months. It is that the operating system of society is being rewritten underneath us, because the technologies arriving now, AI and biology, self improve and evolve in ways no one can predict, and for the first time you could see GDP rise while employment falls, an economy that thrives and has no use for your labor.

McCormack pushes and prods rather than nods along, folding in his own six week sprint building software with Claude despite not being able to code, a live demo of a voice assistant he built two days earlier, and a running argument about debt, regulation, and where power actually sits. Webb answers with data and with concrete forecasts: lights out factories, a collapsing professional services pyramid, a "contribution credit" to replace both taxes and welfare, China as a fast follower treating the United States as a free research lab, and a warning that everyone alive now belongs to what she calls Gen T, the transition generation. What follows rebuilds the whole conversation in order, with the numbers, names, analogies, and asides kept in place and attributed to whoever said them.

Figure 1. The spine of Webb's warning. Each step is meant to sound mundane on its own; the alarm comes from stringing them together. The break is not mass firing on a single Tuesday, it is the quiet decoupling of wealth from human work, which historically have always moved together.

A humanity scale reordering

McCormack opens by naming the feeling: we seem to be moving into a new era, a reorganization of the operating system for society. Are we looking at a humanity scale reordering? Webb's answer is yes, and she thinks we can see it already. The nexus of power is shifting. In the United States she describes a battle royale pitting the financial sector against the tech sector against the government sector. In years past it would have been clear who won. This time no one is sure, because each side is exerting enough strength that the others have to give and take in a way we have not seen before.

She adds a caution about prediction that becomes a theme. Ten years ago the big consulting firms all published studies saying truck drivers would lose their jobs to self driving cars, and named the year it would happen. Webb remembers reading them and thinking no model bore that out. What actually happened is the opposite of what people expected: it is white collar workers whose jobs are fundamentally shifting, not drivers.

Then she reframes the anxiety itself. We each carry mental models of how the world works, some from observation, some from science fiction, and reality tends to arrive different from the model. That gap is what produces the feeling of lost control and lost agency. Some of what is shifting is real, she says, and some of it is just perception.

Show us receipts, and the AI assistant that sounded coked up

McCormack tests her with what he has been building. He mentions a UK program, Question Time, where politicians dodge audience questions, then reveals he built a website called Show Us Receipts that grades politicians' tweets. It ingests every tweet by a politician, fact checks each against primary sources, and labels it true, mostly true, misleading, false, or unverifiable. He built it in three days and has not launched it yet because it still makes mistakes. He can feed in a show transcript and live fact check people while they are on air.

Webb points out he just answered his own question about how the world is changing. Back in the 2016 election, the zenith of Twitter functioning correctly in her view, people were injecting misinformation with botnets, complicated networks of automated accounts. Organizations like Pew, nonprofits, and news outlets tried to brute force fact checking with humans. McCormack cranked out in seventy two hours what teams of employed humans used to do by hand.

Then McCormack says: should we do something really cool live? He phones his own AI assistant on the desk. A voice answers, briefs him that he has sponsor invoices to chase for Real Bedford and an ad question sitting with a colleague, and confirms his only calendar item is an interview with Amy at eleven. Webb jokes that the digital system sounded coked up at the top, all sped up. McCormack explains the arc: six weeks ago he discovered Claude, and with no ability to code, HTML 4 being the last thing he programmed, he has built what he calls three years of work in six weeks. He got there by taking instructions from Claude. He hates the term vibe coding; Webb agrees it sounds stupid. To him the profound point is that anybody can now play around and get a usable website out the other end, and that is a fundamental shift. He calls it the printing press for technology, and connects it to the anxiety of listeners whose jobs are disappearing: Coinbase cut about 17% of staff, Square cut 4,000 people. AI has become an accelerant people do not know what to do with.

AI is not one technology

Webb clicks back before going forward. The printing press, the Gutenberg moment, let a few people set type and mass produce leaflets, pamphlets, and books far faster than a scribe. That shifted the course of humanity because it flooded the market with printed material, which galvanized education, literacy, and religion. The press became a platform for further innovation, a general purpose technology.

AI, she stresses, is not a technology. It is an umbrella term for a whole bunch of different technologies. In 1956, not far from where they are sitting, up in Dartmouth, New Hampshire, a couple of research scientists invited their friends to campus for a two month summer meeting to figure out what a machine that can think might mean. The idea was not original to them. Charles Babbage, who was English, and Ada Lovelace, who did most of the work, had built the first tabulation machine, the first computer, decades before. The Dartmouth meeting produced no AI powered computer, but it seeded public discourse, then investment, then some overexuberance, then technology that failed to materialize in the 1980s. That is how we get to now: AI is not a thing, it is many things, spanning the physical world and visual information and text.

Here is the key difference she wants to mark. The printing press could not learn on its own. The internet, HTML 4, HTML 5, all powerful, do not self improve either. The technologies that exist now, whether AI or the biology she says is increasingly linked to it, self improve and evolve in ways we cannot predict in advance. If you are sick of hearing about AI, and Webb says she is too, the reason to keep paying attention is that self improvement piece.

Thinkers and think nots

So which jobs go, how soon, and what does it mean for you? Webb turns the demo back on McCormack. His six week sprint was made possible by the technology but mainly by him. He has had an unusual, polymath career, which means a lifetime of experience and knowledge, so that when he sits down with AI he already knows what to do. The AI is a supplement that accelerates him rather than replacing him.

She invokes a professor at MIT who talks about thinkers and think nots in the age of AI. Thinkers are people like McCormack, who spent a life learning and experiencing, so the AI layer supplements them. The people in jeopardy are those who have either decided AI will replace them no matter what and so stopped learning and growing, or who got started with the technology too young, before accumulating the experience and knowledge that lets you use it well.

Both of them have sixteen year olds. Webb describes an unusual household: a highly powered server rack in the basement, their own internal setup, a dedicated network node for her daughter, and what she calls a great Chinese firewall built just for her. No phone, no social media. The result is a kid with a vast amount of experience for her age, who sits down with AI and does not ask it to write her essay. She tinkers, she builds with it. That is not true of a lot of her friends who had unfettered access to phones and social media, who just want to get the task done so they can go back to TikTok. Those, Webb says, are the think nots who will have problems in the future.

So after thirty seven minutes, she says, to finally answer the jobs question: the right conversation is not which job in which industry is gone in thirty six months. The better question is who is more vulnerable, and it does not just come down to the specific job. The first question is whether you are a thinker or a think not. Thinkers will be fine, because they are agile and have accumulated skills across a lifetime. (This segment is bracketed by a read for Iren, the show's lead sponsor, an AI cloud that builds data centers and GPU infrastructure on renewable energy.)

How a think not becomes a thinker, and the coal miners who learned to code

If a listener does not want to be a think not, how do they get out? Webb's test is only half a joke: if you are not really listening but just dumping the transcript into ChatGPT to grab a couple of takeaways, you are the think not. Anyone who makes it through an episode asking probing questions is by default a thinker. The path forward is agility and continual learning.

Then she attacks the standard remedy. In the United States we talk about upskilling and reskilling people, which she calls condescending. If you are fifty and have a lifetime of experience, why discount all your skills? Her side quest is the coal country story. Across Virginia and West Virginia the mines had been closing for two centuries, but in one concentrated West Virginia area the closures hit hard, and the mine had been the town's central employer for generations. Around 2008 to 2010, the start of the iPhone era, apps were the new thing, so someone had the bright idea to take laid off miners and teach them to code, HTML 4 or JavaScript, and make apps. These are people who spent their lives operating heavy machinery, working in confined spaces, moving their bodies, doing first aid, and now they were put at a desk and promised high paying jobs at places like Salesforce if only they finished the training. Never mind that they were competing against a marketplace of people thirty years younger who had only ever been digital natives.

The point, she stresses, is not that a miner cannot learn to code. Of course any of them could; it is not that hard. The failure was not acknowledging the skills they had spent lifetimes building. Who else might use those skills? A German heavy industry or auto manufacturer could have been invited to put a plant in West Virginia, with a trained workforce ready. Better yet: phones are popular, that means more demand for rare earth metals, the United States has some, and there were almost no miners who could do that work. That is strategic foresight: not marveling at the new technology but planning for what it implies, going into the future to decide where you want to be and backwards engineering the path. Instead the local government paid a startup to sit miners at desks, the programs collapsed within two years, and the towns were no better off. She calls it an abdication of responsibility, because thinking about our futures is part of the job of elected officials and business leaders alike.

McCormack flags the obvious flaw: the word she used was government. He never holds much confidence in government run programs. Webb agrees that could be its own podcast, but holds her line that strategic foresight is part of the responsibility of people in office.

China is having a better jobs conversation

McCormack raises the competing visions. On the All In podcast he heard that AI has created new jobs, an increased demand for programmers even though they no longer need to code by hand. Against that is the fear factor, and the people who lose jobs will look to government and ask what it will do for them.

Webb says notice who is not having that conversation. In the United States and the UK and much of the world we talk non stop about which jobs go away, split between a utopian pole where thousands of new jobs bloom because that is what has always happened, and the doomers, who get louder because doom gets clicks. She calls herself a pragmatist: just acknowledge we are in a transition, and stuff is going to change. She will not give a jobs prediction, because there is no way to use math to do it right now, too many variables in play.

Who is doing a better job of the conversation? China. Chinese coworkers talk about AI too, but their worry is not the AI, it is that they are not learning fast enough to stay competitive. People there are not at the two poles of will my job vanish versus I will get universal basic income and go read the classics. They ask the harder, more nuanced question: what do I need to learn to be a little better, a little more clever, a little more competitive.

She grants she will not generalize 1.4 billion people, but the orientation differs. China's five year plans, once dismissed by scholars as bold proclamations that never happened, are now being executed well under Xi Jinping. The current plan is technically about AI and jobs but mostly about infrastructure: heavy investment in electricity and power lines, internet access for every family, and an education system oriented toward a world where you use these tools alongside your own thinking. Her summary: China will fast follow. It treats the United States as a free R&D lab, watching the Valley and the hyperscalers create all the debt and all the cool tools, then building its own version and spending its money to make sure everybody has access. Playing that forward, Webb's job in strategic foresight is to use data to answer three questions: where is the world going, where will value be created, and how will we participate. Europe she reads as heavy on regulation, England a question mark, and China the most coordinated effort at answering those three questions in a way that benefits everybody. Not the best decisions across the board, she says, there are all kinds of problems there, just a very different approach, which is why people there are less freaked out about jobs.

Question	United States and much of the West	China, as Webb describes it
Mood about jobs	talk non stop about which jobs vanish; split between utopians and doomers	worried they are not learning fast enough to stay competitive
What the plan funds	the frontier: models, hyperscalers, and a great deal of debt	electricity, power lines, internet for every family, retooled education
Role in the race	the free R&D lab, paying to invent the tools first	the fast follower, copying the tools and paying for access
Worker orientation	will my job go, or will I get universal basic income	what do I need to learn to be a little more clever

Figure 2. Webb's contrast between the two orientations. She is careful not to call China's decisions better across the board, only more coordinated at answering where value goes and how a population participates. The Western debate, she argues, gives a better dopamine hit and kicks the can down the road.

What happens when labor stops creating the wealth

McCormack anchors it to the debt based economy. We live under beyond comprehension levels of debt, and Webb has argued wealth and prosperity may not come only from labor. Elon Musk has floated "universal high income"; Webb doubts even Musk knows what that means. If value gets created elsewhere and not from labor, what happens to a hierarchy that runs on what we earn and then buy, the car, the house?

Webb sets status aside for later and takes the economy first. Go back to Adam Smith and the pin factory: put skilled people in one place, let each specialize in one piece of making a thing, and you scale both the output and the prosperity. That division of labor is the foundation of the modern economy, a lot of labor going into productivity for a good or service someone else buys. The problem now is that tools automate some of that productivity. Until recently automation lived in narrow applications, robots doing highly skilled technical manual work in a car factory. The shift coming is broader: automated systems that supplement people in some cases, and in others mean we do not need as many people, or any.

This is where her firm's research on convergences comes in, the macro forces of change, uncertainties, and trends combining into something net new. One convergence they discovered is lights out industrialism: factories that run twenty four hours a day literally with the lights out, as productive as a modern factory, powered by robots that need no light, no air conditioning, none of the things modern factories were built to give the people at their center. A lights out factory can be a dark box that produces everything you need. And if human capital, which has always undergirded economic structures, is no longer the driving force behind productivity and therefore wealth, then you can get the thing that normally cannot happen: GDP going up and unemployment going up together, in a wealthy, prosperous community. The owners of the lights out factories, the owners of the new mechanism for labor, do very well; others do not. It is a new divide, a new concentration of wealth, and a future where the economy thrives but has no use for you and your labor.

AI is collapsing the pyramid

It is not just the pin factory, Webb says, it is the law firm. In many countries, especially the United States, the firm is a pyramid: a mass of very young people at the bottom logging billable hours on research and manual work, leading up to a handful of partners at the top, and the deal is you climb the ladder over time. That shape, she says, is starting to look more like a narrow spike, and pretty soon by necessity more like a dot. McCormack lands the line: AI is collapsing the pyramid. Webb agrees, and says it is happening across professional services.

Then the number that makes McCormack's face drop. Last year in the United States roughly 36,000 new lawyers entered the marketplace. Only about 300 people skilled at rare earth mining did. Go five years forward: if we are still minting that many lawyers a year on top of the lawyers we already have, what happens to all the people who should have been contributing productively to the economy? Do they still have the means to do it? This is why it is a transition, she stresses. There is no light switch; the jobs do not all vanish tomorrow. But we have to think differently, and think of ourselves as part of a great transition.

Figure 3. The mismatch Webb uses to argue the economy is training people for the pyramid that AI is collapsing while starving the work it will actually need. The miners' bar is a sliver on purpose: about one hundred and twenty times more new lawyers than new rare earth miners.

We are Gen T, the transition generation

McCormack is a Gen X cusper; so is Webb. Their producer is Gen Z, their daughters are Gen Z or Gen alpha. It does not matter, Webb says, because all of us are Gen T, the transition generation. For the next few decades everyone alive is transitioning from the economic structures, social order, and world order we were used to, to whatever comes next. This is the hardest part to grasp, because it means there is no single policy, no single universal basic income, no single fix, since the challenge keeps unfolding.

On status and brands, Webb thinks about it out loud, doing the scenario work she would normally do with her team. Take the weight loss drugs, the GLP medications. Being thin used to be genuinely hard, partly genetic, so it was a status marker. Now there is a way to reprogram yourself to be thin, so thin is no longer unobtainable, which prompts a botched Avatar unobtanium joke between them. If thin is easy, what becomes the next scarce thing? We are always on a quest for what others cannot have. Today wealth buys a designer bag; tomorrow the coveted thing may be cognitive, knowing the prompt engineering to build a digital assistant you can call into. The status symbols of the future may be less about objects and more about cognitive things some people can do and others cannot.

McCormack is more interested in the hierarchy built through the financial system. In a world of more abundance where people do not have to work and government distributes something to live on, how do we distribute housing, how do people choose where to live if we can build anything? Webb asks whether he actually thinks maximal abundance and wealth redistribution will happen. He says no, that is only what he hopes. He thinks there is always an elite group with wealth or power, maintained through debt and inflation, and that being close to the government money printer means you benefit more. So his real question is how a debt based economy keeps its order when labor no longer contributes to it.

Tokens, metering, and a different kind of tax

Webb points out a debt based system some AI users already live inside: the lack of tokens. Ask an AI enough, or start coding, and you hit the limits of the context window; the systems can only parse and remember so much, and it takes energy and compute, so people run out. On the free tier of Claude you get only a few questions before you are cut off. McCormack has had a very expensive month; he reckons he spent about $8,000. It feels like doing all this for free instead of hiring an agency, Webb says, but you are very much being metered, and that is a different form of a tax.

He thought of it as a utility. Webb pushes back: your electricity is one way distribution, they send power and you pay, and they are not learning about you in the process. Every AI system McCormack used, Perplexity for fact checking, ElevenLabs for voice, benefited from him, probably more than he benefited from them. He jokes he should have charged them, which is exactly her point.

The contribution credit

That opens Webb's proposal. She cannot think of a moment in any history where a community stayed stable without being productive. Productivity stabilizes and balances a community; when people lack purpose and are not productive members of society, things go poorly, over and over. If you want to test it, she says, play Civilization. Universal basic income, when tested, has not worked in the long term. Norway found oil, built a sovereign wealth fund, and distributes some of it, but nobody decided to stop working; they kept doing what they did with extra resources to play with. UBI only works when people still contribute to their local communities.

Her alternative is a contribution credit, with two big components among many. First, back pay for your work training AI. Webb has written four books; her books sit in a corpus she says was roughly 150,000 books, hoovered up with no call about royalties. McCormack's public output has been ingested the same way. (He interjects a read for Plaud, the Note Pro recorder he uses to brief his producer after long interviews.) It is more than the corpus of human knowledge, she says. Have you ever driven a car and used GPS? Then the data from that drive went back to the company to refine its systems. All of us contribute in some way, some more than others, and we should get back pay for that work.

Second, forward pay for the invisible work that holds communities together. Teachers in the United States effectively work for free. Webb spends time helping her aging father, who has Parkinson's, and her husband's aging parents. She volunteers as an assistant scout master for one of the first all girl Scout troops in the country, mentoring about twenty girls. This bottom up work stitches communities together, and a lot of it is going away. In her system, once a company is making measurable profit on these tools, a small percentage is set aside into a fund, small enough not to shock the markets or block the companies trying to IPO. Productive members of society earn credits, and the credits let them draw from that fund. It rewards productivity rather than paying people to do nothing, and she frames it as an economic and moral solution that is neither a tax, which she thinks is bad, nor regulation of AI, which she thinks should not be done.

McCormack likes it: if the busywork of commuting and moving data between screens gets automated, people could do the neglected work, the environment, community groups, helping neighbors, and that could be a new economy. Webb resists the word utopia. It is a market based approach, because at the end of the day money drives decisions, and we should lean into that rather than wave our hands. If we know purpose matters, that some jobs go and some appear, that people get left out of the transition, and that invisible work keeps communities healthy, then what to do as AI emerges is fairly obvious.

Figure 4. Webb's proposed mechanism, offered as an alternative to both a tax and AI regulation. The design goal is to keep rewarding productivity through the transition, since she argues no community stays stable without it, while acknowledging that a shrinking group of hyperscalers is concentrating the wealth.

Should AI be regulated

McCormack wants the regulation thread. He notes export controls on Fable, the Anthropic model, which he says were unworkable for Anthropic, and a limited release of ChatGPT 5.6 that he reads as OpenAI preempting export controls. Part of the argument is not handing tools to people who might hack important systems; he is more suspicious it is gatekeeping access, whether from fear of what people build to expose government, a desire for a head start, or fear that China reverse engineers it for DeepSeek style models. He woke up one morning to find he no longer had access to Fable while doing incredible things with it. Webb notes he was not alone: an AI law firm reportedly sued the government for cutting off its Fable access.

On China she is blunt: from her vantage China is about six months behind, give or take depending on development cycles. Refusing to release models because you worry China will catch up is silly. Then she reminds him that tech companies have a long history of making proclamations and backfilling with facts. The 1956 Dartmouth crowd was hyper networked, friends with people in Hollywood and business. Some of 2001: A Space Odyssey grew out of that circle. By the late 1960s and early 1970s figures like Marvin Minsky had become celebrity technologists making insane proclamations, machines that would translate English to Russian and back at the height of the Cold War, when a mainframe was larger than the room they are sitting in and had nowhere near the needed compute. The promise and peril were so spectacular that an entire media cycle ran on it. The hype cycle is not a recent phenomenon.

Her modern example: in 2022 OpenAI put out a press release, she thinks February, about a new thing called ChatGPT, possibly 3.5, that was so powerful they could not release it to the public. It was not the first time a company said its creation was too powerful and terrifying to ship, and, McCormack adds, it is great marketing. What these companies are exceptional at, Webb says, is marketing and accumulating debt, which does not necessarily match what is happening in the real world. Are some systems genuinely capable of terrifying things? Yes. But give her a fork and she can stab someone in the neck with it. She is not being glib; we do not know how these systems evolve, but we have to stop the breathless speculation and get pragmatic about what is actually being built and why it matters.

What a futurist actually does

McCormack says it feels like we are living in the future, everything we were promised starting to happen, minus interplanetary travel, and he watched the "funeral for the trend report," Webb's retirement of her long running emerging tech trends report. Has AI collapsed all her timelines?

She explains the work. Colloquially she is a futurist; in real terms the job is boring, she is like an accountant, quantitative. There are people who call themselves futurists and just make things up. Her background is game theory and economics, taking in different data and figuring out outcomes. She had a stint as a foreign correspondent for Newsweek and the Wall Street Journal about twenty five years ago, doing early computer assisted reporting, which sounds funny now because everyone uses a computer. She landed in strategic foresight. At the Journal she could not just talk to a few people and write; she wrote a pitch, got interrogated and edited, and every claim needed a verified piece of data. She does the same now but looks ahead instead of back: report the hell out of the present, then use modeling to figure out plausible future outcomes. That is different from scanning headlines and going with the gut, and different from speculative foresight, which is science fiction and has its own role. Most of what she does is advising a chunk of the Fortune 100 and government agencies around the world, so she has to be as right as possible while constantly updating her priors. She wishes more people would go into it and actually get trained, because it is not giving speeches or telling tall tales, it is sitting at a computer running models and talking to colleagues, and it has never been more important than now.

AI and biology

What excites her about the future is the intersection of AI and biology. She admits, with a wince, to being a bit of a longevity enthusiast, Bryan Johnson world, though she is more interested in how people approach it and in extending her own health span. Several things changed in the last five years. GLP drugs help people lose weight, and a large body of research suggests they also improve some factors that drive aging, so you get older without getting old. That matters to her personally: her father's Parkinson's is horrific and debilitating, cognitively he is fine but he has largely lost the ability to speak and move. Synthetic biology now lets us treat biological code somewhat like computer code, what she calls write access to life. Instead of a single Parkinson's medication that sort of helps and must be constantly monitored and adjusted, a treatment targeted to him specifically could keep him from being locked into his body. Her mother died young of neuroendocrine cancer, cancer that affects the cells in ways hard to locate; better diagnosis or a protein based therapy might have changed that. What she is optimistic about is AI linked to other fields unlocking optionality we never had. (A sponsor read for Leen, Bitcoin backed loans, sits in here.)

McCormack recalls a man whose dog had cancer and who used AI to decode its genome and get a specialized drug. Webb wants to stay pragmatic: we are not curing cancer tomorrow. She asks everyone to unhook from the timeline question, especially investors who want to know exactly when. The honest answer is faster than before, but the exact month and year are unknowable because there are too many variables. Are there promising signals? All over the place. Japan is far ahead on genetics and AI. There are finally options for climate change; it is unlikely we reverse the added atmospheric energy driving temperature swings, so we may instead mitigate outcomes. You can now combine AI and biology to re engineer almond trees, which need a lot of water, so they need a quarter as much and can grow in other conditions. These excite her because agriculture has barely advanced in fourteen thousand years, and even in health and medicine we have been stuck in places. AI is the accelerant because it lets us simulate and learn faster than the squishy computers in our heads allow.

Convergence, divergence, and ambient compute

There is a lot happening in physical AI, intelligence that uses the real world. Webb predicts a literal explosion of wearable devices over the next twenty four to thirty six months, glasses, wristbands, rings, and weird combinations that will not last. McCormack mentions the Midjourney body scan announcement, which he thought was a new MRI; Webb clarifies it is a scan, she thinks ultrasound, and warns of the 1990s mall CAT scans that only produced false positives. Allbirds is no longer making shoes but AI devices; Midjourney, already passed by Nano Banana, is getting into hardware.

She uses her firm's framing of convergences and divergences. Back in 1997 she carried a heavy backpack: a fifteen pound Toshiba laptop that doubled as a self defense weapon, a wireless sniffer, a MiniDisc player kept for its sound quality, an early GPS device. By 2007, just ten years later, the first iPhone converged all of it into one device with more compute. Now, twenty years on, the things layering on top are diverging again: ChatGPT and Claude, Snap Spectacles and Meta glasses, bike computers and wrist wearables scattered across the body. The next convergence, she says, is not a single device but ambient compute, all of these things funneling into one system that makes sense of the data and hands back the insight. McCormack calls it his cocaine guy, the sped up assistant, and she says that is exactly the example.

1997A backpack full of single purpose gadgets: a fifteen pound Toshiba laptop, a wireless sniffer, a MiniDisc player, an early GPS unit. Each device did one thing.
2007The first iPhone converges all of it into one device with more compute than the whole bag combined.
2020sThe stack diverges again: ChatGPT and Claude, Meta glasses and Snap Spectacles, rings, bands, and bike computers spread back across the body.
NextAmbient compute: not a new device but one system that funnels every stream together and hands the insight back. McCormack's "cocaine guy" assistant is an early taste.

Figure 5. Webb's convergence and divergence cycle, the concrete version of her point that novelty has become the new normal. The lesson is to watch the pattern, convergence then divergence then convergence, rather than fixate on any one gadget.

Novelty as the new normal, and where power sits

That morning McCormack saw Meta announce brain to query, using AI to decode brains and translate thinking into words, like Neuralink without an implant. Webb says this is not new. About twenty years ago at the University of Washington, one researcher thought "type the letter F" and, across campus, moved another person's finger to type it, with video still online. We keep getting stuck in the moment, treating everything we see as new and novel, which means novelty has become the new normal. What he saw on Twitter is not the end product; there is a long continuum behind and ahead, and keeping that perspective matters when we ask whether something is okay and who will accumulate more power.

On power, if he had asked her twenty or even ten years ago, when she wrote The Big Nine, she would have said the tech companies. Her frame then was a three sided prisoner's dilemma among the finance sector, the tech sector, and government, each thinking it had the most power while actually balanced precariously, and better off collaborating. That has changed. Tech has some power, but in the United States the current administration has amassed a great deal, which she attributes less to the administration itself than to the Project 2025 group who had the vision and the discipline to execute, with the Trump administration as the beneficiary taking the credit. The finance sector has lost leverage because everyone is willing to throw capital; she notes the thing OpenAI has made most of is debt, and it wants more, and Musk is making a lot of debt too. So in the three sided game it is really tech companies versus government, at least here, with a flavor of it in the EU, which has a strong and heavy handed regulatory arm. A hyperscaler with a supply chain is vulnerable to an administration that can cut off access, which is the Apple problem: components from China, the administration threatening to cut things off, which she thought stupid in the long run given how much Apple contributes to the economy.

England, the swing, and the temple roof

Webb asks McCormack about England. Power there, he says, sits almost entirely with the state, which wants more, but it is subservient to the asset managers of the world, the BlackRock types; when Keir Starmer got into power his first meeting was with Larry Fink. There is heavy regulatory lobbying and a cutting out of the little guy. He is skeptical of "Brenter," a re entry to the EU after Brexit, calling it a Hail Mary for a leftist government trying to hold power. His read of UK politics: normally a Conservative government builds wealth, an imbalance grows, a Labour government comes in to redistribute, a fine pendulum. What they actually got was a center left Conservative government, high tax and high surveillance with little to distribute, then a Labour government arriving amid low productivity, low growth, stagflation, and heavy debt, squeezing living standards. Starmer is out, he says, replaced by Andy Burnham, who will likely go harder.

On AI in the UK: DeepMind is based there but was acquired by Google. McCormack thinks the UK has little innovation, and if you are smart and ambitious you leave for the United States, the place of opportunity, because the UK is buried in regulation. He read of an AI investment in a town like Barnsley that might have been in the hundreds of thousands, at most five million, and thought, what are we doing. He expects a reaction, a swing to a right wing government, something more like the Trump administration eventually.

Webb calls the swing itself the challenge. Part of why it keeps happening is that leaders across countries have not established a concrete long term vision everyone can get behind; the long term plans that do exist leave out chunks of the population. Project 2025, she notes, has a very clear vision that alienates half the country, but it has an organized group who learned to execute a plan built over a long period. What Britain has is resistance against a vision rather than a shared one.

Then the temple. Webb spent years living in Japan and China, and her friend Taka is a Buddhist monk running a temple in Kyoto, the twenty sixth generation to do so. One day he pointed at a new roof made from a particular wood from a particular region, and the reason they could build it was that two hundred years ago a family member had planted and tended the tree, knowing someone in the future would need it to repair the roof. A tiny example, she says, but analogous to what we lack: agreement on what we desire for the future. His family desired that the temple keep being a place for spiritual growth and community; they understood the world would change but knew they would need a roof. Most of our plans, and most of what we argue about, is the how, not the what.

McCormack would love a long term UK plan, but they have cycled through about seven prime ministers in ten years, and everything is short term. People do not want to deal with the debt, so they keep borrowing or taxing to fund public services the country cannot afford. He wants a long term plan, and thinks it may get worse before it gets better; the NHS always runs out of money, and if the country were a company it would be trading insolvent. Rising taxes push people out, a vicious loop like the Argentine brain drain, with young people unable to get on the housing ladder. His fixes: strip out the mountains of regulation (his own coffee shop is drowning in it, plus energy prices and minimum wages), take a libertarian approach with tax incentives for investors, reduce taxes, and shrink a state that has become a giant, closer to what Adam Smith would say.

The trap of left and right, and the fan car

Webb tells her own political journey. In ninth grade, about fifteen, she read The Communist Manifesto and declared herself a communist, until her parents told her never to say that outside the family. Two years later someone handed her Ayn Rand's The Fountainhead and she flipped to the other side. She was a Democrat for five seconds, then made some money and became a Republican. At fifty one she wants no labels, mainly because each one, including libertarianism, connects to structures she does not think work going forward. She wants to stop framing it as regulation versus no regulation, a binary, and find something that is not regulation.

McCormack calls himself a trajectory libertarian: everything is too big, too much government, regulation, debt, and money printing, working against what we want as a society, so we need to reverse a little. His favorite book is The Law by Frédéric Bastiat, which describes two periods a society moves through, productivity and wealth creation, then plunder, where plunderers fight for control of the state to plunder the productive class and the society contracts. He thinks the UK is in the plunder phase and wants back to productivity. Left and right, he says, is a trap of the mind.

Webb agrees they are spiritually aligned but says the point is structural. It is America's 250th anniversary that week, and the laws and structures made sense mostly two hundred and fifty years ago, in a world with no global trade as it exists today, no drones that assassinate a person, warfare being militias with muskets on a Virginia farm rather than unmanned drones entering Moscow. Some of it does not make sense now. But the answer, from her point of view, is not deregulate or regulate more; it is what could exist that is not regulation but has the desired impact. That is a fundamental reperception, another concept from her firm: taking the same inputs that already exist and seeing them differently, exploring the white space rather than the obvious. We need to do something about AI so the best decisions are made in the public interest, but it cannot be regulation.

Her illustration is Formula 1. In 1978, the year McCormack was born, a car showed up at a race with a giant fan bolted on. The team said it cooled the hot engine and driver. The driver was Niki Lauda, one of the best ever, and with the fan he accelerated through corners, unusual, and won handily. The fan also acted like a vacuum cleaner, sucking the car to the tarmac so there was no drag. A week later they banned giant fans, but his win stood. Her lesson: in Formula 1 people constantly build crazy new technology because no rule says they cannot, and the rule comes after the fact. That is how emerging technology emerges. To write a regulation you would need answers to exactly how the thing will evolve and look over time, which we cannot have. It does not mean do nothing, but regulation as we have always thought of it does not work; we need something else, something that enables competition and investors rather than stifling them. They agree they have hit the point where both must leave and travel, and that the current structures do not make sense for the world we are about to live in.

McCormack floats a London part two in the fall. His parting hope is a world where technology does so much for us that we need it less and go back into the real world to savor it. Webb says she can grant that hope today: all he has to do is not use his technology. He laughs that he needs it sometimes. It is the show's final episode of the trip back to London.

Key takeaways

Webb's core claim is structural, not a job by job forecast: the technologies arriving now self improve, which lets wealth decouple from human labor, so you could see GDP and unemployment rise together for the first time, an economy that thrives without needing you.
The right question is not which job vanishes in thirty six months but who is more vulnerable, which starts with whether you are a thinker, who uses AI as a supplement to a lifetime of skill, or a think not, who stopped learning or started too young.
Reskilling programs that ignore existing skills fail. The West Virginia coal miners taught to code collapsed within two years; strategic foresight would have matched their real skills to heavy industry or rare earth mining.
China is having a calmer, more nuanced jobs conversation, treating the United States as a free research lab and spending on infrastructure and access while it fast follows, roughly six months behind.
Lights out industrialism, factories running dark and unmanned, is the concrete mechanism by which human capital stops driving productivity and wealth concentrates in whoever owns the new means of production.
AI is collapsing the professional services pyramid; the United States minted about 36,000 new lawyers last year and only about 300 new rare earth miners, training people for a shape that is disappearing.
Everyone alive is Gen T, the transition generation, so there is no single policy fix. Webb's own proposal is a contribution credit that pays back for training AI and forward for invisible community work, funded by a small slice of measured AI profit and drawn down as earned credits.
Webb is optimistic about AI linked to biology, longevity, synthetic biology as write access to life, water thrifty crops, but insists on unhooking from the exact when, and argues the answer to AI is not regulation as we have known it but a reperception of what could work.

Chapters

0:00:00 The Economy May Not Need You 0:01:13 Humanity Scale Reordering 0:04:25 Show Us Receipts 0:07:10 Peter's AI Assistant 0:08:00 Three Years In Six Weeks 0:08:46 AI Is Not One Technology 0:12:00 Self Improving Technology 0:13:30 Thinkers And Think Nots 0:16:44 How To Become A Thinker 0:21:23 Failed Transition Planning 0:23:58 China's AI Strategy 0:28:31 What Happens Without Labour? 0:30:56 Lights Out Industrialism 0:33:47 AI Collapsing The Pyramid 0:35:12 We Are Gen T 0:41:39 Contribution Credit 0:52:43 Should AI Be Regulated? 0:58:52 What Futurists Actually Do 1:03:54 AI And Biology 1:30:16 Regulation Can't Keep Up

Notable quotes

0:00:00 "We could be looking at a future where the economy is thriving but has no use for you and your labor and what your contributions can be." (Webb)
0:00:38 "You could have GDP going up and unemployment going up for the first time together." (Webb)
0:00:42 "The current structures don't make sense for the world we're about to live in." (Webb)
0:08:40 "This is essentially the printing press for technology." (McCormack)
0:11:20 "AI is not a thing. It's a whole bunch of different things." (Webb)
0:13:30 "Those are the think nots who are going to have problems in the future." (Webb)
0:22:20 "If anybody can predict the future for you, it's me. And I am not going to give you a prediction for jobs, because there's no way to use math to do it right now. There's too many variables in play." (Webb)
0:25:40 "They look at the United States as a free R and D lab." (Webb, on China)
0:31:10 "Imagine factories that can operate for 24 hours a day, literally with the lights out." (Webb)
0:33:47 "AI is collapsing the pyramid." (McCormack)
0:34:20 "We graduated I think 36,000 new lawyers. Only 300 people skilled at rare earth mining did." (Webb)
0:35:12 "We are all the transition generation." (Webb)
0:40:00 "I reckon I spent about $8,000. I've had a very expensive month." (McCormack)
0:50:00 "Money is what drives decisions. It just is." (Webb)
0:57:40 "We have to stop with the breathless speculation and start to be much more pragmatic about what's actually being built and why it matters." (Webb)
0:59:20 "Colloquially I am known as a futurist. What that actually means is the job is pretty boring. I'm like an accountant." (Webb)
1:04:40 "We kind of have write access to life." (Webb, on synthetic biology)
1:06:20 "Agriculture hasn't really advanced in 14,000 years." (Webb)
1:12:00 "Novelty has become the new normal everywhere." (Webb)

Resources mentioned

Amy Webb, quantitative futurist, CEO of Future Today Strategy Group, professor of strategic foresight at NYU Stern, and author of The Big Nine.
The Peter McCormack Show, the channel hosting this interview, and Real Bedford, McCormack's football club.
Claude and Anthropic, including the Fable model referenced in the export controls discussion.
OpenAI and ChatGPT, and DeepSeek, the Chinese models Webb references.
Perplexity and ElevenLabs, tools McCormack used to build his fact checker and voice assistant.
The Dartmouth workshop of 1956, with Charles Babbage, Ada Lovelace, and Marvin Minsky; the printing press and Johannes Gutenberg.
Adam Smith and the division of labor, the foundation of the pin factory argument.
China's five year plans under Xi Jinping, and the All In podcast, the source of the new jobs claim.
Layoffs cited: Coinbase and Square; employers named: Salesforce and the rare earth mining shortfall.
Universal basic income, the Norwegian sovereign wealth fund, and the game Civilization, all used to argue that productivity stabilizes communities.
GLP weight loss drugs, synthetic biology, Parkinson's disease, neuroendocrine cancer, agriculture, and Bryan Johnson world, from the AI and biology segment.
Hardware and interfaces: Midjourney, Nano Banana, Allbirds, Snap Spectacles, Meta glasses, Neuralink, ambient computing, the iPhone, MiniDisc, and the University of Washington brain to brain experiment.
Power and politics: The Big Nine, Project 2025, Donald Trump, Apple, the EU AI Act, BlackRock, Larry Fink, Keir Starmer, Andy Burnham, Brexit, DeepMind, Google, Barnsley, and the NHS.
Ideas and books: The Communist Manifesto, Ayn Rand's The Fountainhead, Frédéric Bastiat's The Law, libertarianism, the US 250th anniversary, and Kyoto.
Formula 1, Niki Lauda, and the Brabham fan car, Webb's parable for how technology outruns rules; plus 2001: A Space Odyssey and Avatar.
Sponsors read during the show: Iren (AI cloud), Plaud (the Note Pro recorder), and Leen (Bitcoin backed loans).

Where it stands

This is a forecast and a proposal, not a report of settled facts, and it is worth marking which parts are which. The hard anchors check out. The layoffs Webb and McCormack cite are real events, the 1956 Dartmouth workshop with Babbage and Lovelace as forebears is history, China's five year plans did shift from aspiration to execution under Xi, and the Formula 1 fan car is a documented episode: the Brabham BT46B won the 1978 Swedish Grand Prix with Niki Lauda driving and was withdrawn shortly after, the win standing, close to Webb's telling. One small correction: the longevity figure she gestures at is Bryan Johnson, not Brian.

The load bearing claims, though, are projections. Lights out industrialism at scale, GDP and unemployment rising together as a durable pattern, the professional services pyramid collapsing to a dot, an explosion of wearables inside thirty six months, and China being exactly six months behind are Webb's modeled forecasts, and she is candid that she will not put dates on them because the variables are too many. Her convergence and reperception methods are proprietary to her firm, so an outside reader cannot fully audit how the forecasts are derived. The contribution credit is her policy idea, not enacted anywhere, and its hardest questions, how profit is measured, how credits are earned without gaming, whether it truly beats a tax, are acknowledged but not resolved on air. Her own evidence that universal basic income underperforms is contested in the wider literature, where results are mixed rather than uniformly negative. And rhetorical flourishes like agriculture not advancing in fourteen thousand years compress a lot of real progress, from the green revolution to modern genetics, into a single line. The value of the conversation is less any single number than its central provocation: if wealth can keep growing while human labor stops being needed, the institutions we built around jobs, taxes, welfare, and status may be measuring the wrong thing, and the useful work is deciding what we actually want before we backwards engineer the path to it.

Full transcript

We could be looking at a future um where the economy is thriving but has no use for you and your labor and what your contributions can be. Are we looking at a humanity scale reordering? I think the answer is yes. And I think we're starting to see that already >> and it feels like AI has just become this accelerant that people don't know what to do with. The key difference with the technology that exists now, whether that is AI or related technologies like biotechnology, which it's like biology and I kind of linked together, these self-improve and they can evolve in ways that we cannot predict in advance. There's a professor at MIT that talks about in the age of AI thinkers and think nots. I don't think the right conversation is specifically which job in which industry will be gone 36 months from now. I think the better question is who are the people that are more vulnerable? You could have GDP going up and unemployment going up for the first time together. >> We are all the transition generation. The current structures don't make sense for the world we're about to live in. >> All right, Amy, I I've been so excited to talk to you. We seem to be uh moving into a new era where we're potentially looking at a reorganization of the operational system for society. >> Mhm. Yeah. Um so first of all, thank you for including me on your US uh tour. Um happy to be with you in New York. >> I really am legit. I've been looking forward to this as well. >> Um yeah. So are we looking at a humanity scale reordering? Um I think the answer is yes. And I think we're starting to see that already. Um, we're seeing the nexus of power shifting a little bit. Uh, and and that depends on the country, but I think here in the United States, there's definitely been this sort of battle royale pitting the financial sector against the tech sector against the uh government sector. Um, and typically it would have been years past like clear who the winner was. And this time, you know, I'm not I'm not sure what's going to happen because, uh, because everybody seems to be exerting enough strength that the others are having to give and take a little bit in a way we haven't seen before. But also um if you go back maybe 10 years there were a lot of like all the big consulting firms had their studies out saying you know truck drivers are all going to lose their jobs because of self-driving cars and this is the you know this is the year that it will happen and I remember seeing those studies thinking like did you just hire a bunch of analysts out of like none of this makes any sense there's no model that bear this out. Um, you know, that being said, it's the white collar people whose jobs are fundamentally shifting, which is not what people had thought. Anyhow, this is all to say, what is a world model to begin with? So, if we stop and think for a moment, why are we feeling this sensation of change and profound uncertainty? It's because we all live with these mental models for sometimes a really long time. And some of that comes from what we've observed. Some of it comes from science fiction. Um, and our expectations for what reality is going to be wind up being pretty different from what reality actually is. And that gap is what makes us have these feelings of anxiety and like we don't have any more control and we don't have any agency which is not the case but it's this this sense of like everything is changing and I and I I can't do anything about it. Um which which all ties back to that question about you know the operating system of humanity and and all of these things that are shifting. Yes, a lot is but some of it is also just our perception I think. Yeah, they we get a lot of emails in to the show and talk to a lot of people about AI. There was also there's also a regular program in the UK called Question Time where they have four or five politicians and the audience asks questions and they didn't >> do they answer Wait, I got now I got to ask you questions. Do they actually answer the questions or is it just gobbly talking points? >> Well, it's interesting you should say that because uh I just built a website called Show Us Receipts which grades tweets. >> Awesome. >> Yeah. So, this is the power of AI. I built this in three days. It sucks in every tweet by a politician. It checks, fact checks them against primary sources and it graves them as true, mostly true, misleading, false, or unverifiable. I haven't put it live yet because there's some mistakes in there. So, I'm fine tuning it. >> But with this question time program, I was like, well, I can take the transcript of the show and I can put that in as well and live fact check them while people are watching. >> Okay. So, you just answered the question that you asked me. How is the world changing? So if we go back in this country to the 2020 election and the 2016 election, um 2016 was sort of the zenith of of Twitter functioning correctly, I would say. >> And um people were still trying to inject misinformation. Um botn nets, which are complicated networks of automated systems that are responding to content and posting their own content. Um that was the beginning of maybe we should be fact-checking this. And there are organizations in the United States like Pew and there were a handful of nonprofits, some news organizations trying to brute force what you are talking about with humans. >> So there were a lot of humans that were employed to do exactly what you just cranked out in what sounds like 72 hours. So, so that is a and and it's you know once you have it fine-tuned and launch it, it's going to be a really good service assuming that the you know the the labeling is accurate. >> So that's a so but here's why this is this matters because a obviously this has says something about jobs but B this is almost like a real time thing that will exist. You will be holding politicians more accountable >> which is great >> which is awesome. >> Yeah. But the the amazing thing with this and the six weeks I've had is because uh I can't code. I know HTML 4 was the last thing I programmed. >> Oh, that's a that's a while ago. >> But I just get instructions from Claude. >> I was just going to ask, did you use Claude to do this? >> Yes, I use Clawude. >> Yeah. So, it's an easy you and whatever. I hate the like term vibe coding because I think it's just one of these terms for right now that's going to sound stupid in a couple years. >> I think it sounds stupid now. >> I think it 100% sounds stupid. >> Software engineer. >> Thank you. Um, but the but the fact that anybody can just play around and have an output be a usable website is a fundamental shift. And >> should we do something really cool live? I wasn't planning to do this. >> This might not work. >> Maybe. I've already I should not say yes before I like see what we're doing. >> Let's see what we do. Let's see if this works. I wasn't planning this. >> Hi Peter. One moment. >> Is that your voice? >> No. Two things pressing. Peter, you've got sponsor invoices to chase for real Bedford. Applauded advert question sitting with Alec. Kurt copied in. What would you like to tackle? That's me. Everything all right? Did I sound a bit off? Anyway, the two things pressing this afternoon. You've got sponsor invoices. >> Calm down. Calm down. >> Uh, what do I have on today? Do I have any meetings or anything today? Any interviews? >> Just the one thing, an interview with Amy Weber at 11. That's all I've got in your calendar today. Anything else? >> No, that's fine. Let let me give you the background to that cuz I only built that two days ago. >> Okay. Did you also did you give a little cocaine because that's what it sounded like. Sound like your digital system was all coked up at the beginning. >> Yeah. Something went well. So we have we I'm still fine-tuning but there's been a journey to this which has got me to the point where I'm so excited to talk to you because it's been a profound six weeks. >> Six weeks ago I discovered Claude. I've built three years of work in six weeks with no ability to code >> and uh and fundamentally changed the way I work. And and so the profound moment I've got got to and why I wanted to talk to you is essentially this is the printing press for technology and when I think of our listeners that there are there are jobs that have been wiped out. Coinbase got rid of um what was it 17% of their staff. Square got rid of 4,000 people >> and and that anxiety people have about will my job exist? How do I keep my job? What happens if my job goes? That's the anxiety people are dealing with. I see all the opportunity but I recognize the anxiety and it feels like AI has just become this accelerant that people don't know what to do with. >> Let me let's click back a little bit before we go forward. Um so first of all you mentioned the printing press and this is this um >> you know influential moment in time when if you go back to Guten the Gutenberg press um >> the you know at that moment in time you would have had one or more probably several people literally putting t you know tiles together um uh into a press so that you could uh more much more quickly than a human scribe would have been able to do uh mass produce mass being small scale at that stage uh leaflets and pamphlets and and books and you know whatever it might have been. Um and that did fundamentally shift the course of humanity because it meant that um there was much more supply in the marketplace of printed materials and that galvanized education in new ways and religion and the places where people would learn how to read. So in a way that printing press became a platform for additional innovation. It was in a sense a generalpurpose technology. >> Um artificial intelligence first of all if everybody listening is sick of hearing about artificial intelligence so am I. You are not alone. Um and we're going to keep talking about it. But the reason I bring this up is because AI, first of all is not a technology. It's an umbrella term for a whole bunch of different technologies. And in 1956, when there was not too far away from where we are right now up in Dartmouth, um, New Hampshire, uh, there was a meeting over the summer for 2 months, um, a couple of research scientists invited all their buddies, uh, to campus and said, "We've got this idea about a machine that can think." Now, this wasn't their initial original idea. Others had already posited this many decades before. in fact going all the way back to uh Charles Babage who's English and Adal Love Lace who was doing most of the work um had created the first tabulation machine so it was like the first computer. So people have been thinking about this for a while but they like got these guys together and said let's really figure out what it might mean for a machine that can think. um we didn't have an AI powered computer after that but we had a lot of stuff in the public discourse which then led to investment and uh to some degree some overexuberance um and technology that failed to materialize in the 80s but but that's how we get to now AI is not a thing it's a whole bunch of different things um that involve the physical world and visual information and exactly what you just discussed um the printing. So AI is a general purpose technology and we're going to get to jobs in a second, but I just want to highlight the key difference between the printing press as an inflection point as a general purpose technology and AI. The difference is the printing press couldn't learn on its own. So it couldn't self-improve. and the internet HTML 4 HTML 5 very powerful stuff but it also doesn't learn on its own or self-improve the key difference with the technology that exists now whether that is AI or related technologies like biotechnology which it's like biology and I kind of linked together these self-improve and they can evolve in ways that we cannot predict in advance so if you are sick about hearing about artificial intelligence because you keep feeling like you're hearing the same stuff over and over again. The reason it's good to be paying attention is because of that piece, the self-improvement piece. Now, jobs, which jobs are going away, how soon are they going away? What does it mean for you? So what you just described that you built your your past six weeks of a journey of experimentation and learning and getting to the point where you called your digital assistant and that assistant just read out for you everything you need to know. Um it was made possible because of the technology but mainly because of you. So but this is the part that people forget. So like Peter you have had this unbelievable career. You are a polymath. you are somebody who is well I did you did some research on me I did some research on you before before I showed up but you've done all of these different things which means you have a lifetime of experience and knowledge so that when you sit down to figure out what am I going to do with AI you already know >> yes >> um there's a professor at MIT that talks think nots the thinkers are people like you so you have spent your life learning and experiencing and now the AI layer is a supplement. Okay, which means that it doesn't obviate you, it is accelerating you. But there are a lot of people out there who have in a way decided um either AI is just going to replace them no matter what. And so they've stopped learning and they've stopped trying to grow or they are getting started with this technology too young and they haven't had the experience to low to to learn and to gain new knowledge. Therefore in a few years those are the people who will actually be in jeopardy. We both have 16-year-olds. >> Um, my daughter, we had, we live in an unusual situation because we have like a, you know, highly powered server rack in our basement and we basically built our own internal bra. We're super nerds. So, >> no surprise. >> Yeah. So, like my my kid had her own dedicated network node, but we you know, she thought she was out there exploring everything and it we basically built a great Chinese firewall just for her. Um, but she uh shouldn't have a phone and no social media or anything else. So, this kid now has a vast amount of experience in her young 16 years and she's she's learned all of these things so that when she sits down to use AI, it is actually not write my essay for me. It's none of that stuff. It's more clever like what what you're trying to do. She's tinkering with it. >> She's building >> she's building with it. That is not true of a lot of her friends. >> Yes. who uh had unfettered access to their phones and to social media, whatever else. I mean, there there are correlations here and they're just trying to use it to like I just want to get this stuff done so I can go back to screwing around on TikTok. >> Those are the thinkotss who are are going to have problems in the future. So at last after 37 minutes to answer the question that you did about jobs, >> I don't think the right conversation is specifically which job in which industry will be gone 36 months from now. I think the better question is who are the people that are more vulnerable and it doesn't just have to do with the specific job they have. The first question is is this a thinker or a think not kind of person. The thinker people are going to be fine because they're going to be agile and they will have accumulated those skills throughout life. >> This show is brought to you by my lead sponsor, Iron, the AI cloud for the next big thing. Iron builds and operates next generation data centers and delivers cuttingedge GPU infrastructure all powered by renewable energy. Now, if you need access to scalable GPU clusters or are simply curious about who is powering the future of AI, check out iron.com to learn more, which is irre.com. >> So, how does a think not become a thinker? If they're listening to this, they're like, well, I don't want to be in the thinkot category. How do I get there? Well, if the person listening is actually not listening but just downloading your transcripts and dumping them into chat GPT to get the couple of things they need to know, um, those are your thinkotss. I think everybody listening probably is a thinker because what you're doing here is asking challenging, probing questions. So, by default, if you make it through an episode, >> you're you're a thinker. So, I think people here are fine. if you want to play this forward um you have to be agile. So, you know, being in continual learning mode, especially in this country, we talk about um upskilling people or reskilling people, which I think is I think that's like a that's a really condescending thing to say to somebody to reskill them. If you're 50 years old and you've had a lifetime of experience doing something, why would we discount all of the skills that you have? Can I can I take a quick side quest and tell I mean it's going to happen anyways, but can I quick story? >> It's not going to be quick. I'll try to make it quick. >> Um, in the so in this country, we had a lot of coal mines throughout the the east coast. So, Virginia, West Virginia, and um coal just kind of went out of fashion over the past 200 years, right? But there's still some operational mines in in both places. In one concentrated area in West Virginia, um just the mines started shutting down. And for a long time, like that mine, that was the town that was the central employer. Generations of people had been, you know, from the same family had been working and that like that's it. And when the mines closed, there's no other industries, nothing else to do. So I think it was 2008, 2010, something like that. Uh we were at the beginning of the iPhone era and not too long after people started talking about apps, right? Apps were the new thing. So, somebody had the bright idea to take all of these miners who are now out of jobs and reskill them or upskill them and teach all these minors how to code. Okay. So, we're we're talking about somebody who has spent his entire life operating heavy machinery, going into very confined spaces, having to physically move their body around, you know, first aid, like all of these accumulated skills, and now we're going to put them at a desk and they're going to learn HTML 4 or whatever it might be, right? JavaScript, uh, and they're going to make apps. Now, this is nothing against the coal miners. Could any coal miner learn how to code? Of course, it's not that hard. This is about not acknowledging the skills that they had spent their entire lives building and building upon for those in generational families. Who else might make use of those skills? Um, instead of thinking through that, the local government paid some startup a bunch of money to come into town and have all these minors sit at desks. And these people, by the way, were were promised very highpaying jobs at places like Salesforce, right? If only they get through this this training program. Forget the fact that you're competing against a marketplace of people literally 30 years younger than them, okay? who have only ever been digital natives. Why didn't the local community instead have the foresight? So this this is where the strategic foresight comes into play. It's not enough to marvel at the new technology. You have to plan for what it implies. >> If it was me, I would have said, yes, there are these apps and mobile phones. More to the point. Mines are going to close, but we actually still need those skill sets. So they could have called a German company in let's say auto manufacturing or heavy industrial or something and said look come to West Virginia put up a plant here we have highly trained workers who can continue to do the work you know that that you need to have done or better yet you know these phones seem to be pretty popular. I bet we're going to need a lot more of them. That probably means we need more rare earth metals. We've got some of that here in the United States. We don't have any minors right now who can really do that. Why don't we see if this group of people can also do that type of work? >> But there was a very obvious flaw in all of that is that the one word you used was government. >> Well, sure. And I'm sure we'll talk about Yeah. But but this is the point. The point is um coming as a shock to literally nobody. uh these amazing programs to teach the coal miners how to code all collapsed within two years and then these towns were no better off than they could have been. And that irritates me because government is charged with thinking about our futures. I mean that is part of the part of the gig. >> Um and to me that was an abdication of responsibility. I'm not saying that government should have come in with a very heavy hand and sort of put everybody into a government pro, you know, none of that stuff. What I am saying is local governments, city governments, you know, federal, whatever. The part of the job is let's go into the future and figure out where it is we all want to be and then backwards engineer that way. Backwards engineer that um given what we know to be true. So it's strategic foresight that is part of the responsibility of people in elected office as well as business leaders and everybody else. But >> well I never hold too much confidence with governmentr run programs but that could be a whole podcast on its own. But there's these kind of um competing ideas of what what AI is going to mean for the future. There is um I mean I listened to the allin podcast recently and they were talking about certain areas where the AI has created new jobs. There's like an increased demand for programmers even though they don't need to code anymore. Uh but there's also the the fear factor of people worrying about jobs going and if jobs do go it will be the people losing their jobs who will be looking to the government to say what are you going to do for me >> and so >> yeah you know who's not having that conversation so so you're right you're absolutely right in the United States in England and a lot of places around the world we are we are talking non-stop about jobs >> which jobs are going away depending on who you talk to it's either some kind of polyiana you know ultra utopian all the all these new job, thousands of new jobs will bloom, right? >> Uh because that is what has always happened. And if you talk to the doomers who are increasingly loud because it gets more clicks on the internets. Um you know, all jobs are going away and there'll be economic collapse. Uh look, I'm a pragmatist. Yes. >> Which is to say, just let's all just acknowledge that we're in a transition. >> Stuff's going to change. >> Stuff's going to change. That's right. I don't have an Look, if anybody can predict the future for you, it's me. I am I am the best I think at this um and I am not going to give you a prediction for jobs because there's no way to use math to do it right now. There's too many variables in play. Let me tell you who's doing a better job of having the jobs conversation. China. >> Okay. >> So if you go to China and you talk to people there the convers they are also talking about jobs in a very different way. So co-workers will talk about AI and their concern about AI. Their concern is not the AI. It is that they are not learning enough quickly enough so that they can continue to be competitive. People in China are not all at these sort of opposite poles wondering if their jobs are going to be gone or when they're going to go start learning the classics because they're all going to get universal basic income, right? It's not that. They're thinking much more specifically and in a nuanced way. What do I need to learn how to do a little bit better? How can I be a little bit more clever? How can I be more competitive? It's a very different orientation. It's the harder thing to do. The easier thing to do is to sit back and get yourself all, you know, riled up. Um, from a dopamine point of like, you know, like that's going to give you like a better dopamine hit. Um, but it's also just going to sort of kick the can down the road. You're still gonna have to deal with this. >> Well, is that So you are you basically saying China has more thinkers and less think not. >> I think that China is a very large place. So I would hate to generalize 1.4 billion people. >> Um, but I think the orientation there is a little different. So uh there is so China has these five-year plans that the that the government creates and then um you know to some degree they they get executed and if you go back 20 years um a lot of scholars would say these and policy makers these 20-year plans were these bold proclamations and they they never like they were silly nothing ever happened but that has changed under Xi Jinping >> uh and they're actually execcut ing now quite well on these plans and the current 5-year plan is technically about AI and jobs but it's mostly about infrastructure. So this is another key difference that people don't really know a lot about. So China is heavily investing in electricity >> like power lines. Um they are working hard to make sure that every family has access to the internet. Um they're also making sure that the education system is oriented a little bit more toward a world in which you have these tools that you can use along with the thinking work that you're already doing which is a way of saying China is going to fast follow. They look at the United States as a free R&D lab. So they're looking at what's happening out in uh in the valley uh where our large AI companies are based and hyperscalers are based. China's like you you guys go out and you know create all the debt you want and uh and develop all the cool tools. We we'll we'll let you do that and then we'll take whatever it is that you've learned and then we'll make our own thing uh and we're going to spend all of our money to make sure that everybody has access. So if you play that forward. So again my I work in strategic foresight. My job is to figure out using data. Where is the world going? Where will value be created? And how will we participate? If I look at what we're doing here, what is happening in Europe? England's kind of over here in a question mark zone from my point of view. And and China seems to be the more coordinated effort toward answering those three questions in a way that benefits everybody. And look, I'm not saying China's making all the the best decision. There's all kinds of problems there. I'm just saying the approach is quite different. So the so people are not as freaked out there about jobs as they are here. >> So when you think uh and and again I always refer it back to the people who might be listening thinking about this and what they're considering with their their lives and not me because I know what I'm doing. Um, but if they're trying to prepare for the future and think about this, >> this seems to be that uh because we live in a debt based system, right? We live in a debt based economy and you've talked about that >> we we we live in a heavily debt based economy right now. My god, >> a beyond comprehension levels of debt. But if you start to think about the future, you've talked about how uh wealth and prosperity might not just come from labor. Mhm. >> U Elon Musk has talked about universal high income. Um I don't know what that means or if it's a reality. >> I don't I don't know that he knows what that means. You know, >> but it but if if the value is created elsewhere and it doesn't come from labor, have you thought much about what that means for the hierarchy in society? Because a lot of a lot of our status and hierarchy comes from what we earn and then what we buy from it, the car we drive, the house we build. But if there's a fundamental shift to the operating system of society, how do how do we even square that? How do we even consider what that will mean? >> Yeah. Um let me come back to the question about how we present ourselves and status last. Let me first address the what does the economy look like going forward. Um, so if you go back to Adam Smith and the pin factory, um, and for those of you who do not hang out over the weekend, you know, with a cup of coffee thinking about Adam Smith, um, the concept was if you, uh, have a bunch of skilled people in one place and each person is gets very very good, specializes in sort of one piece of making a thing, um, you can you can scale the things that are made as well as the prosperity and everybody kind of flourishes. is afterwards. Um, it's a lot of labor going into productivity for a good or a ser service that somebody else then might buy. This is the foundation of the modern economy. One of the problems we have right now is that tools automate some of that productivity, some of that work. And it's not the same as a car factory where you can, you know, use robots to make most of the cars because up until recently a lot of that automation, let's say robots have been used in fairly narrow applications um involving manual work and highly skilled let's say technical manual work. The shift that's coming is that we're going to have automated systems that will supplement to some degree, but there are cases where, you know, um we just don't need as many people making the things or in some cases we don't need people at all. So, if I go back to China for just a moment, um we we at my firm have been doing a lot of original research around convergences. Yes. >> Which is the um macro forces of change and uncertainties and and trends sort of combining such that the output is something net new and one of the convergences that we have discovered using this new methodology is uh lights out industrialism. Um, so imagine factories that can operate for 24 hours a day, literally with the lights out. Um, that can be just as productive as a modern factory. Um, >> what that and and that makes sense because you've got robots powering the entire thing. They don't need lights. They don't, you know, modern factories were designed all the way back to Adam Smith with people in mind. People at the center, >> you need lights. You need air conditioning or something. You know, you need all of these things. Well, a lights out factory could just be a box literally dark um that is able to produce everything that you need. And if that's true and we already see these things in existence, then that means that um it's human capital which has always undergurtded the economic structures that we live in is is no longer the driving force behind productivity and therefore wealth. Which means that you could have GDP increasing. GDP is not the only important measure of the health of an economy, but it's a big one. But you could have GDP going up and unemployment going up for the first time together in a wealthy pro prosperous community. Normally those would be at odds. M >> um and that means that the people who own the lights out factories, you know, who who own the mech the new mechanism for labor will do quite well and it means others will not. So it creates a new type of divide. It creates new concentrations of wealth. It also means that we could be looking at a future um where the economy is thriving but has no use for you and your labor and what your contributions can be. So I think that that is concerning because it's not just the pin factory, it's also the law firm. In a lot of countries, law firms, especially here, um the business model is sort of a pyramid where you have a ton of very young people at the bottom with a lot of billable hours doing tons of manual work, research, similar manual style work that all sort of lead up to a handful of people at the top. Um, and the idea is that over time, you know, you sort of climb that that ladder up the pyramid. Um, this shape is starting to look more like this. And pretty soon, by necessity, it's it's kind of going to be that. And then maybe it's just going to be like a dot. >> AI is collapsing the pyramid. >> It is. And that's come that's happening for professional services. >> So, we're we're starting to see this happening in a lot of different places. So, again, that begs the question, how does that reorder society? Because um just as an example, last year in the United States, we graduated I think 36,000 new lawyers entered the marketplace, >> only 300 people skilled at rare earth mining did. Okay. So if we go forward, Yeah, I know. The look on your face right now is exactly right. So if we go 5 years into the future, >> do you know if we're graduating that many new lawyers every year, we already have a lot of lawyers. what happens to to the people who should have been contributing in a productive way to the economy? Do they have the means to do that anymore? So, we're going to have for a while that that's why this is a transition. >> There's not a light switch. All the jobs aren't going away tomorrow. Okay? >> We have to think differently. It is where the nuance comes in. >> We have to think about ourselves as part of this great transition. So, I think you're probably a millennial. I'm a Gen X. >> I am Gen X. >> Okay. So that's >> we're cuspers. So we're we're the cool people. Our daughters are I think Jen He's laughing. Are you a millennial? >> He's a Gen Z. >> You're a Gen Z. I I think my daughters I think our daughters are Gen alphas. Although my my kid tells me all the time she's Gen Z. >> Uh it doesn't matter because all of us are Gen T. >> We are all the transition generation. So for the next few decades, we are all going to be transitioning uh from the economic structures, the social order, the world order that we were used to to the thing that comes next. So this is the hard the hardest part for people to wrap their heads around because it it means that there's no single policy. There's no single universal basic income. There's no single thing that's going to solve the challenge because this challenge is going to be unfolding. Now what does that mean for brands and for status and for everything else? >> Um that's a really great question. Um I bet you if we go into the future to some degree status will be found in in other ways. So, I'll give you one example. So, all of the weight loss drugs that are suddenly popular and and accessible and available, what used to be unobtainable, being thin, that was a really hard thing to do. >> I'm I'm aware. >> Uh, well, it's genetics, right? Sometimes it's like you could be disciplined and calorie restrict and drink and not drink alcohol and do all the best things, but reality you you're our biology is sort of pre-programmed and you can you know and then that changes as you age. >> I would like that excuse but it's beer and cheeseburgers. >> Well, okay. But but for some people but the point is now there is a way to reprogram yourself which is partially how these GLPs work. Um, so that you can be thin. So thin is no longer unobtam. Unobtamian. Unobtanium. That's not even a I was trying to make a Marvel reference and I screwed up. >> Hold on. Wasn't that um uh Avatar? That was the That was what they were mining. >> Was that what they were actually mining thing? Okay, that's a joke we have inside of our family. But you know what I Okay, so it's uh if you can now be thin, >> what's the next thing? you can't be toned and you know in shape and whatever. We're always going to be on a quest for things that other people can't have. >> Yes. >> So today if you have enough wealth you can buy a designer bag, but at some point that may not be as desired. What may be desired is I know how to do all of the prompt engineering that you did to make your cool digital assistant that you can call into to get your you know what I mean? So the status symbols may be more cognitive. I don't know yet. I haven't really thought this through kind of >> scenario. I'm actually doing the scenario stuff that I would normally do with my team just out loud ad nauseium. But I wonder if the status symbols of the future have more to do with um just cognitive things that that people can some people can do and others can't. >> Yeah. Well, I I think I'm I'm more interested in the hierarchy that's built through the financial system >> in that if we move to a world of more abundance >> um >> and people don't have to work and perhaps there is a program that the government distributes money so you have something to live on. Uh, how do how do we how do we distribute housing? How do people choose where they live if we can build anything? >> Well, can I ask you a question? >> Please do. >> Do you think that's gonna So, do you what you just described >> um maximal abundance uh the government's going to redistribute the wealth and like and everybody >> No, I don't think it's going to happen. >> Well, I mean, >> some people think it's going to happen. What do you >> That's what I hope. >> What do you What do you think? I think we always have a world where there are is a certain elite group of people who are wealthy or have power and there are uh controls and mechanisms for maintaining that usually through debt >> uh debt and inflation >> and uh if you're close enough to the government money printer you tend to benefit from a bit more that I tend to think about that so actually I think the real question I'm asking you is in a debt based economy where labor is uh no longer uh contributing to the economy how how do how does the government maintain that uh that cast system the order system >> which is based on on debt and >> the growth of the debt >> well because >> so that's interesting um you know it's interesting there is a debt based system that people exist that that we some people who are using artificial intelligence have already found themselves in the middle of and that is lack of tokens >> right so I don't want to get overly technical here but if you use AI systems you have to have enough uh you have to have enough of a mech mechanism to operate what's called a inside of a context window. So basically like once you start coding things or asking a lot of questions, the AI systems can only remember so much. That's um or they can only parse so much. So you can only dump so much stuff into it. Just takes a lot of energy and compute to make it work. Um which means some people run out. So, if you're on the free version of Claude, for example, you've probably found that after a certain amount of time, you you only get a couple questions, right? >> 24 hours. I've had a very expensive month. >> I spent about I reckon I spent about $8,000. >> I was I was going to ask you. Yeah. So, but that's a good point. >> So, it feels like we're doing all of this stuff for free. We don't have to hire an agency. We can do But it's not, >> right? Um you are very much being metered >> and it is a different form of attacks if you stop and think about it for a moment. >> Yeah. Okay. Uh >> well, I thought of it more like a utility. >> Well, I think a lot of people would like for us to think about these systems as a utility, but in reality, we're not being metered quite, you know, again, your local electricity service, that's one way distribution. They're sending power to your flat. Yeah. And you're paying for it. They are not learning about you in the process. So I would argue that the more like all of this work you did to create these cool systems and tools um you were not the only one who benefited there. every AI system that you used whether that was perplexity for the factchecking or you know 11 for the voice recogn I don't know what you used but whatever you use right okay so whatever you did >> they all benefited from you probably more than you benefited from them >> I should have charged them >> well that's kind of my point one of the option opportunities in front of us is something that I call a contribution credit >> okay >> because and I'll explain what that is in a moment but let me set this up by saying um you know I cannot think of a moment in modern in any history regardless of how far you go back where a community was stable without being productive. Productivity it turns out is a wonderful way to stabilize and balance a community. If people don't have a purpose and if they are not being productive members of society, things tend to go poorly. Like we we've seen that over and over and over again. And if you want to just see if that maybe if there's a way to game it, play Civ. Do you ever play video games like Civilization? >> Uh not. But I know it. I know it. >> Yeah. >> I conquer. >> Okay. So like but there's plenty of stra like things don't go well, right? If if people feel like they don't have a sense of purpose. >> So universal basic income when it has been tested, it has not worked in the long term. There are a couple places. So like Norway found its oil reserves kind of you know recently. Um so there's a sovereign wealth fund and some of that gets distributed but Germany didn't like decide or not Germany Norway didn't decide like nobody's working again. Uh everybody kind of continued with what they were doing but now they just have additional resources to play with. Um, there have been a couple of places that have tried UBI, but ultimately it only works out when people are still contributing to their local communities in some way. So, uh, my plan that I have been trying to get people in this country at least excited about is a contribution credit which has two key components. There are many more, but sort of the two big ones are um back pay for your work training AI and forward pay for the invisible work that happens every day in our communities. I've written four books. I've been I'm a public person, so I've been videoed a lot. You I'm just I'm my stuff is out there and all of it has been hoovered up. >> Yeah. and you've had the same thing happen to you um by AI systems. >> All right, check this out. This is Plaude. Now, one of the hardest parts of doing long form interviews is what happens after we stop recording. I could be sat here for 3 hours talking about AI, politics, economics, and all that civilizational stuff. And then immediately afterwards, I need to provide a brief to my producer Connor. He wants to know what the title's going to be, the thumbnail, what clips matter, what's going to be the open and hook. And that normally means waiting for a transcript, digging through the notes and trying to remember what the strongest moments were when it happened. It never really happens like this. Usually a couple of days later, Connor is chasing me and I can't remember what we spoke about. So when Plaude reached out and they said they had a solution, I was interested. So I've been using this. This is the Plaude Note Pro. I just literally leave it here on the desk during an interview. And once we're done, I instantly have access to searchable text from the conversation. So instead of relying on my memory after a three-hour show, I can immediately pull quotes, identify themes, and send a proper brief over to Connor. And honestly, some weeks I'm doing three to four long form interviews. So Plaude has become incredibly useful. But it's not just for interviews. We're planning shows in the car. There's post show discussions and sometimes just random ideas after recording. All those conversations we don't normally capture. So look, if you're thinking of using Plaude, obviously follow local laws and get consent when recording conversations. If you're a journalist or a podcaster, I think Plaude is something you're going to like. So, if you want to find out more, please head over to plauda.ai/mccormac for 20% off. That is plaud.ai/mccormac. And plaude is spelled p l aud. Did you Eric Weinstein talked about this very interestingly. He put out a profound tweet that sat with me for a long time. >> I I haven't seen it. What was it? >> Yeah. So, I mean, I'm I'm going to try and remember it cuz it was a while back, but I contacted him afterwards to talk about it because he said, "These AI systems are sucking up all the uh uh the history of everything we've learned, built >> um the entire history of humanity and monetizing it for a group, a small group of people." Correct. >> And uh and in that process, >> we will be taking jobs away and >> and we need to fight back against this. We need to have a discussion against this. But I don't the fight back I don't agree with because anytime you show up for a fight >> some you're going to get a fight on the other end. So that's actually not what I'm proposing. >> Yeah. No, but I'm just more interested in the point that like I hadn't thought about it is that essentially they're just sucking up the entire knowledge of human history. >> Sure. But it's more than that. So I've all of my books there's a database that if you've written books you can check to see they've the the main corpus was about 150,000 books I think. So, my books are in it. Um, at any rate, so, but nobody like called me first to see if if that was cool or if I what my royalty was going to be. But let me ask you a question, Peter. Have you ever driven a car? >> Okay. And when you drive that car, have you ever used, let's say, GPS? >> Okay. So, guess who else? So, if you've ever driven a car and used GPS, this is another one of these things where you're not the only one benefiting. what's happening during that drive is being sent back to the company so they can continue to refine and fine-tune. Point being all of us in some way you know we are all contributing um some of us more than others so in my case just because of the volume of cognitive knowledge work same with you but in some way we all have I should get back pay for that work there's also forward pay so let's talk about teachers teachers in the United States effectively work for free they're they're volunteers they don't get paid very much at all >> um but I have a aging father and my husband has aging parents. I hope they never listen to this. This is going to pop up on my father-in-law's feed and I'm going to get an earful for having called him aged. But >> does he listen to British conservative? >> You know what? He might. >> We'll find out. I'll let you know next month. Um anyways, so uh we spend a lot of time helping them. And my dad's got Parkinson's so like there's a lot of doctors, a lot of stuff. Um I volunteer as a scout ma assistant scout master um for scouts in the United States. We have the boy scouts. It's now just scouts. >> Um I we have one of the first all grew all girl scout troops in the country >> and I mentor I think we've got 20 girls on our troop now. Spent a lot of time mentoring those kids and teaching them and helping them learn and they spend a lot of time camping and all this stuff. A lot of us are doing invisible work that stitches our communities together. I would argue that um this is not government top down work. This is the more bottomup stuff that connects us to each other. Okay. So we are seeing a lot of that going away for various reasons. Some of which has to do with AI, some of which has to do with just other stuff. So in the contribution credit system, what we're doing is acknowledging exactly what you said. M >> small group of people increasingly and you know those people are hyperscalers increasingly uh concentrated wealth um government hapless you know either trying to siphon some of that off or not and here in this country every four years somebody changes their mind and that's you know so let's just acknowledge the situation that we're in rather than it being a tax which I think is not good or rather than trying to regul ulate AI which I think should never should not be done and we can talk more if you want about why >> I definitely want to >> um what this does it's an economic and moral solution we uh once a company is making measurable um profit on these tools uh then a small percentage of that is put into a fund so such that it doesn't shock the markets we've got a lot companies that are trying to IPO now. So this doesn't impinge their ability to do that. We're just taking a small amount and putting that aside. And then if you are a productive member of society, you earn credits and the credits allow you to take from that fund. >> So economy really >> it's a different approach. It's there's and yes, I know there's like a zillion different potential ways to game all of this. I have thought through it. Um >> but the general idea is more of a fluid economy. >> Correct. The general and the idea came from um right so so this is more of a earning credits system that doesn't immediately so it's not just giving people money for not doing stuff. It's rewarding productivity. M >> the the jobs conversation is the first conversation to have, but ultimately we're going to be having a conversation about productivity because that impacts how people behave toward each other. So what we're doing is creating the conditions so that productivity continues to be rewarded in a during the transition when jobs are changing. >> It is quite interesting if it if it can work in the way that you're talking about. We have at the moment people are just very busy uh spending maybe time commuting to work spending time in an office with a screen moving data one thing to another. If that is all being handled and automated and these people can go and do the things that maybe are being neglected in society which is the environment we live in um you know giving time to community groups and uh to like a lot a lot of people want to work with their local community and and help people and if that can be a new economy that's exactly what I'm talking about. I mean that that does take us a little bit more towards that kind of the utopia that AI can deliver. >> Well, look, I don't believe in such a thing as a utopia, but what I'm more saying is this is a market-based approach because at the end of the day, what do people I look everybody can be >> altruistic and good and everything else, but it's still money. >> Ultimately, and that's we should acknowledge it, you know, money is what drives decisions. It just is. M >> um so let's lean into what we know is the driver of of decisions and figure out a way to address that. >> So it should not be a top- down regulation. We shouldn't just wave our hands and say well the you know let fate do what it will and let's plan effectively and if we know what is important is people feeling like they have a sense of purpose and we know some jobs will go away and some new jobs will happen but they're going to be people left out of that transition and we know that it's all of this invisible work that historically has made our communities healthy then to me it's pretty obvious what we should be doing as AI um emerges. >> I like it. I do want to touch therefore on the regulation because you mentioned that and we have seen >> export controls on fable um which obviously were unworkable for anthropic uh and we've also seen this release of uh GPT chat GPT 5.6 I think would limited release which is them >> preempting the export controls. Um part of the argument is that uh they didn't want to give tools to people to maybe you know uh hack into important systems or also really it would to me I'm a little bit more suspicious is uh it's gatekeeping access to technology >> whether it's fear that what people will do to like build things like I've built to expose the government or just cuz they will just want to get a head start or they fear that China will reverse engineer some of it for their own deep sync models. I mean I don't know the answer. I was just very frustrated. I woke up one morning and I no longer had access to Fable and I was doing some incredible things with it. >> Yeah. So, you were not the only one. There's a law firm. I forget which. There there was a lawsuit. Um there's a company suing the government for cutting off its access to Fable. And it turns out it was a an AI law firm. >> Uh so, so you're not the only one upset about these things. Let's um let's address the China question first. my point. Look, I'm no I'm not an expert in any of this. I just, you know, pay attention to to some stuff. I would never call myself an AI expert or a geopolitical expert. Um, but I do research for a living. And, you know, >> when it comes to China, from my point of view, China's about six months behind us. >> That's about it. >> And that changes obviously depending on where everybody is in their development cycles. Um I I think if we are not releasing models because we're worried China is going to catch up um then that is being silly. Uh so that's the first thing. Um I think we forget that technology companies have a pretty long history of making proclamations and announcements and trying to backfill with facts later on. even in AI. So, you know, I I will send this to you when you get back. Yes, >> cuz you will enjoy this. >> There were some hilar there were some ridiculous articles. So, 1956, those guys >> Dartmouth University, you know, they they come up with all these crazy ideas. Um, not too long after, they're also now the same group of people. They're super networky, so they they know people in Hollywood and they know people in business and all over the place. Everybody's kind of getting a little bit excited about these things. If you've ever seen uh 2001 Space Odyssey, do you can't remember how? >> Okay. that was actually based that some of that came from this meeting because the people who worked on all that stuff were friends. By the I want to say 19 early 1970s maybe late60s early '7s there were that group of Marvin Minsky. So some some of those people were out talking to the media about the coming age of artificial intelligence and they were they're their the spotlight on them was getting brighter and brighter and brighter and they became celebrity, you know, celebrity technologists in their own right. They were making insane proclamations about machines that would simultaneously translate English into Russian and Russian into English. Remember the height of the Cold War. Now, if you go back in time and look at the size of a computer and the amount of compute in it, you know, the late 1960s, like a mainframe was larger than the room that we are currently sitting in and had nowhere near the compute that would have been required to do what they were talking about. that the the promise um and the perils of of artificial intelligence were so spectacular that there was an entire media cycle that lasted for quite a while where this is all anybody talked about. That hype cycle is not a recent phenomenon. In 2022, uh we've all forgotten this now, but OpenAI sent out a press release. I think it was February. They had this new thing called chat GPT. I think this one was 3.5 maybe. And it was so powerful that they could not they could not release it to the public. There's a press release where they say this. Okay. So, this is not the first time we're hearing from these companies that the thing they've built is so powerful it is terrifying. >> It's great marketing. It was methos when they came out and said, "Of course it is. It's we're just going to release it to a select few companies to begin with." So, um, the point that I'm making is the thing that these companies are exceptional at is marketing and accumulating debt. That doesn't necessarily translate to what's happening in the real world. Now, are some of these systems incredibly capable and powerful of doing some terrifying things? Yes. But also, if you give me a fork, you know, a fork can be a deadly weapon if you stab somebody in the neck with it. Mhm. >> So, I'm not trying to be glib. >> You know, obviously we don't know how some of these systems are going to evolve over time, but I think we have to stop with the breathless speculation and start to be much more pragmatic about what's actually being built and why it matters. >> It it's one of the most interesting parts of all of this. I said to you before we started um and I said it at the very start is it does feel like we are now living in the future. Yeah, everything we were promised is starting starting to kind of happen. We've been promised interplanetary travel. I think we're some way off on that. But we do have access to these incredible technologies. Uh and I also I watched the uh the funeral for the trend report. Um has has it has AI collapsed all the timelines for you now? And is it very is it become more of a difficult job? >> You mean the work that I do? >> Should I tell everybody what work I do? cuz I know I've got a strange so colloquially I am known as a futurist. >> in in real world terminology uh what that actually means is the job is pretty boring. It's I'm like an accountant. >> It's not always well can we be fair and and and have the quantitive part in it because there are people who call themselves futurists and they just make stuff up. I think this is going >> Yes, there's a lot of that >> data. >> There is. So my I started off uh so my economic background is game theory and economics >> and a lot of what that I mean you know this but like a lot of economics is you're taking in a bunch of different data and you're trying to figure out outcomes. Um I had a temporary professional side quest as a journalist. So I was a foreign correspondent. I used to write for Newsweek. This is like 25. This is a billion years ago, 25 years ago. Uh I was at Newsweek and I was at the Wall Street Journal and a lot of what I was doing was uh I was covering technology, merging technology. Um but I was using early days computational systems and trying to like find stuff. So I was doing something called computer assisted reporting, which is hilarious if I say that it does that doesn't exist now because everybody's using a computer, but back then that was kind of a new thing. Um, I wound up in this field, strategic foresight, but all of these jobs are related. If you're a really good, like when I was at the journal, I couldn't just talk to a few people and then like write the story. I had to write a pitch. That pitch often times got edited and I got interrogated. This is for a daily story. before I could sit down to do it, you know, everything that I wrote had to have a verified piece of data or just you would be able to show your work, right? Um, I like to think that what I do now is what I learned how to do with economics and game theory and journalism, but rather than looking backwards, I'm looking ahead. So, can I report the hell out of the present, use modeling to figure out plausible outcomes in the future? That's it. That is very different from scanning a couple of headlines and saying, "Well, you know, my gut tells me or or getting super excited about a technology and just then speculating about what the future might look like." There is such a thing as speculative foresight, which is sci-fi. um it's not just I mean there's you know and and there is a a role for that but when it comes to making decisions and most of what I do is advising you know we we advise a you know chunk of the Fortune 100 and government agencies around the world we have to be right we we have to be as um as right as we can knowing that you kind of have to update your priors all the time >> I expect you're busy at the moment >> we're quite busy Um, so it is it is an important field. I honestly wish more people would go into it. It's, you know, if you're somebody who's like, "Yeah, I'm a futurist." Awesome. Please go get trained and learn how to actually do the work, which is most of the time sitting in front of a computer just simulating and and running models and talking to her colleagues to figure out, you know, how to sort of uh adjust things. Um it is that it is is not giving speeches or you know telling tall tales. It's actually really important work. Really important work. It's never been more important than right now. >> And so what what does excite you then right now? What >> what excites me? >> Yeah. What on the in the future? >> Um well I'm particularly interested in the intersection. Well, I'm I'm kind of a um God, I'm going to regret saying this out loud. I'm like a longevity maxer. Okay. >> Not maxer, but I'm a deeply interested in longevity. >> Um, >> Brian Johnson world. >> But, uh, I'm here. Let me say I'm, uh, I'm more interested in how people are are approaching this, if that makes sense. >> So, and I, um, I mean, I do a ton of research, so I'm also trying to extend my health span, >> but I we're also living through this moment in time where it's possible to finally do that. I mean, there's a bunch that has changed over the past five years. So, if we go back to GLPS for a moment, um, you know, one of the things that GLPs do is they help people lose weight. There's also a giant body of research that's that's either out or going to be out soon. You know, they also help reverse aging in some ways or they improve the factors that lead to aging so that you get older without feeling or getting old. That's really interesting to me. Not just because I'm selfish and I want to, you know, I want to >> live forever. >> Live forever and like have my face not melt. It's not that. You know, my my dad's got Parkinson's. Um, and that is a horrific, debilitating condition. Um, cognitively he's fine, but he's pretty much lost his ability to speak and to to move. Um, synthetic biology is the kind, you know, there are ways now to sort of treat biological code similarly to how we treat computer code, meaning we kind of have right access to life. So rather than a single Parkinson's medication that kind of sort of helps him, but he's got to constantly be monitored and they have to adjust it every now and then. if there was something for him specifically uh or a better type of treatment um you know then he wouldn't he wouldn't be spending this part of his life sort of locked into his body which just sucks. Uh my mom died early of uh when she was very young of neuroendocrine cancer which is a fancy way of saying cancer that affects the cells but we can't quite figure out how or where. >> Yeah. um if there had been better diagnosis, something more targeted, a new type of therapy leveraging proteins, there's just this what I'm optimistic about is artificial intelligence when you link it with other things starts to unlock uh totally other fields and give us optionality we just haven't had before. >> All right, let's talk to you about my sponsor, Leen. Now, if you're borrowing against your Bitcoin, there's one thing that matters more than anything else. Is it actually safe? Well, Leen have just dropped their lowest rates ever. But more importantly, they haven't changed the thing that matters most. Your Bitcoin is never lent out. So, they're not chasing yield. There's no hidden risk and there's no rehypothecation. Just collateralized loans done properly. Now, they've done over $10 billion in loans. And they've done this through bull markets, through bare markets, and everything in between. And they've done it without ever losing any client assets. So, now you get lower rates, which means the bigger the loan, the lower the rate. and full transparency before you apply. There's no monthly payments or early repayment penalties. And they give you tools to stay in control from auto topups to LTV alerts. So, you're not choosing between a good rate and safety anymore because with Leen you get both. Now, you can check out your rate using the calculator at leen.io/a. I read a story and I I don't have the full details on memory but about a guy who had a dog and his dog had some form of cancer and he used AI to decode his uh genome of some kind and then they >> it was the DNA spiral. >> Yeah. And they got a specialized drug for him for the dog and I was like that's fascinating. >> Yeah. Now look I want to be very pragmatic here. We're not curing cancer tomorrow so it's not that. >> But we're in the timeline. >> I don't know. So this is the thing with Everybody's got to stop asking. Not I mean you can it's your job to ask but like I'd like for us to unhook from the timeline piece of this especially >> investors because they like they want to know when. Um the answer is faster than before. But exactly what month and what year? I don't know yet because there's too many variables in play. >> But are there promising signals? >> Oh my god. Yes. All over the place. Um, Japan is so far ahead of everybody else on genetics and artificial intelligence and um, doing all kinds of interesting things. There's we finally have options for climate change. It is hot as hell right now in New York City. It's going to get worse uh, over the next couple of days because we've got a heat dome. uh it's unlikely that we're going to be able to reverse the things that created um more energy in the atmosphere which is leading to these you know swings in in temperature that we're seeing. Uh so if that's true maybe we figure out ways to mitigate the outcomes of that. So you can now combine AI and biology to create almond trees that don't require almond trees require a ton of water. you can re-engineer them so they require a quarter of the amount of water, you know, and and can grow in other types of conditions. These are the types of things that I get very excited about because agriculture hasn't really advanced in 14,000 years. And while we have miracles in health and medicine, we've also kind of been stuck for a little while in in a lot of places. And that's the AI accelerant on things. Well, right, because it gives us the ability to simulate and to learn faster than we would have before because we're somewhat limited by the squishy computers inside of our heads. >> Um, there's a whole bunch of uh interesting things happening in the realm of physical AI. So, this is artificial intelligence that makes use of the real world in some way. So there are wearable devices you're you are about to see over the next probably 24 months to 36 months a literal explosion of wearable devices coming to market. There were a bunch of recently made announcements about glasses um but different wristbands, different rings, um weird combinations of wearables that probably won't last very long. Uh but basically this creates the conditions for people being able to do more better stuff. >> Experimentation. Sure. I mean there's been two things recently that have just blown my mind. Uh did you see the midjourney announcement for the body scan? I was like what is this? This is interesting. Or is it >> Sorry. Yes. Um you want to explain to people what that is? >> Well, so my understanding it was a new way of doing MRI scans. >> Well, it is a way of doing scans and I think it's ultrasound not uh but >> um yeah. So, there was a thing in the 90s where you could get a a CAT scan in the shopping mall and and you there was a promise to like scan you for every single possible thing and all that did was lead to false positives. >> Um, allirds is no longer making shoes. They're making AI stuff. Midjourney has already been passed by by Nano Banana and so they're now getting into devices. >> But this is so we talk about convergences a lot in our company. We also talk about divergences. So, what we're going to see if you So, very quickly, if you go back to like 1997, >> um, I was carrying a lot in my backpack. You probably were, too. >> I had a very heavy laptop, Toshiba laptop. >> It's like 15 pounds, doubled as a self-defense weapon if I needed it. >> Um, I had a Canary W. I had a wireless sniffer because, you know, I had a miniis player because it had wonderful sound quality. at all this I had a early days GPS device had all this stuff by 2007 so that is just 10 years I had the first iPhone and that first iPhone took all of these disperate things and put it into one device that had more compute and the ability to do all of these things >> um today so we're about 20 years afterwards um that one device uh the the things that layer on top of it are now diverging again. So, we've got Chat GPT and we've got Claude. We have Snaps Spectacles if you're somebody wearing them or Meta's glasses. >> Um, you know, I've got I'm a endurance cyclist so I've got bike computer and wrist. I've got all this Yeah. all this stuff all over the place. >> Um, what will happen the the next convergence is not, you know, a single device. It's ambient compute. So we will have all of these things sort of funneling into a singular system of some kind that takes all of those data and helps us make sense of it and then gives us access back to the insights and whatever on the other end. >> My my cocaine guy. >> Yeah, exactly. That's exactly right. That's a good example. >> So this is why in the moment with so much change happening all the time Yeah. It can it can be really hard to see the bigger picture of what's happening, but that's what I really want everybody to try to focus a little bit of attention on. >> Yeah. I mean, so this morning I saw the announcement from Meta, and I only briefly read it on Twitter about the brain to query. >> Are you still on Twitter? >> Uh yeah, I still use it. It's good for me. >> What's what's going on over there? I haven't I haven't uh been to Twitter and I'm an OG, one of the original Twitter users. And >> so it very quickly changes your uh your algorithm. You watch you watch post of a fight. All your fights are post people fighting. Mine's just all AI now. And I love it because I just I see tools, I see prompts, I see skills, but I see things being built. So I I useful to you. >> Yeah, I I get that. And then a small dose of um British politics, which um is a car crash. And so the AI stuff's great, but this morning meta announced brain towerty. They've been using AI to decode brains. >> Uh that sounds like a good thing to do. Yeah. And I was like, "Wow, that's I mean, again, that's the AI being the accelerant." And it's just like these are the It's >> Wait, is this doing what? How are they doing the brain to >> I I just saw the tweet. So, all I saw I saw >> Well, we know it's you you need to use your new I've got the receipts system on that tweet to make sure it's a real thing. >> But it but it's scanning brain uh uh brain scans to translate what people are thinking into words. Okay. And so it's essentially what it's like neural link but but without an implant. Okay, >> which is kind of cool if they can do that. But it's more this accelerant that seems to be happening everywhere. I I I almost have AI exhaustion in that there's so much you can do. I can't even keep up. >> Well, let me first say and again I'm happy you brought that up because we keep getting stuck in the moment thinking that everything that we are seeing is new and novel. >> Which means that novelty has become the new normal everywhere. So what you're describing is actually not new technology. About 20 years ago at the University of Washington, um one person they there were two researchers on opposite ends of the campus in you know Seattle or I think it's Seattle. Um anyhow somebody sent thought you know type the letter F and on the other side of campus and there's a video you can see online of this. They it moved that person's finger and and they typed. The reason I bring this up is because sometimes it can feel like every new thing is the first time this thing has happened and we forget that there's been a continuum this entire time. >> Um, and that's important because when we start to ask questions about is this okay? >> Should we be doing this? What are we giving up? Who is going to accumulate more power? >> Yeah. I think it's important again to just keep the broader context in mind because what you just saw on Twitter is not the end use or the end product. We've got a ways behind us and a ways to go >> and it's just we got to keep the perspective I think. >> Yeah. I mean where where should we be worried in terms of control of the power? I mean is it going to be the state? Is it going to be these nine corporations? Well, that's a great question and I think if you had asked me that question maybe 20 years ago, I would have or even 10 years ago when I wrote the big nine, I would have said the tech companies, you know, like there's this three-sided prisoners dilemma sort of pitting the finance finance bros and the tech sector and the government and everybody thinks they have the most power when in actuality they're balanced, you know, precipitously and if they only collaborated everything would be better. That's changed. Um the tech companies do have some power, but at least in this country um the the Trump administration has amassed uh and it's as I don't want to give the Trump administration credit for this because it's actually not the Trump administration. It was a group of people who worked on um project 2025, right? That's what it was called. Um it was a small group of people who worked on a policy and before that had the vision to execute you know what was what did they want the country to be like and how do they execute against it this is just the administration that is the um >> beneficiary >> the current modality of that plan >> um and and I think they're taking credit for a lot of what was done but it I don't think it was actually this administration I think it was that group of people at any rate I think the power has shifted And I think the financial sector, look, look at how much money you could argue that OpenAI has made many things, but the thing they've made the most of is debt, >> right? >> And they want more of it. >> And they want more of it. And Elon Musk is, you know, making a lot of debt. I mean, they're trying to issue like so, you know, the the finance sector no longer has the leverage that it used to because everybody's willing to throw capital. So that means in this three-sided, you know, prisoners dilemma, it's really the technology companies versus government, at least in this country. Um, and I think there's a flavor of that in the EU because the EU actually has a fairly strong and heavy-handed regulatory arm. >> Um, and it just so it just depends on where you are in the world, how it all shakes out. I think that the it's again, it depends on where you are. the tech companies are working hard. So if you're a hyperscaler with products that require a supply chain, then that makes you a little vulnerable with this administration because they can use regulatory structures to cut off access to things. So that's with Apple, right? >> So Apple products come from, you know, the components come from China. Um, the Trump administration threatens to cut things off. That creates enormous problems for Apple, puts Apple in a terrible position. Our economy, like Apple contributes an enormous amount to our economy. So, the government taking them on, I thought was kind of a stupid thing to do in the long run. Uh, but it does change the dynamics. You know, England, I mean, I I think we all kind of understand Europe. What do you think about England? I haven't really thought about that before >> in terms of >> well in terms of the balance of power. I mean is it >> I mean we the power is in well the power is entirely with the state and and they're trying to get more power but they are a little bit subservient to the um the asset managers of the world the black rocks and such. So when karma got into power first meeting he had was Larry Frink came in and so uh there is a lot of regulatory uh lobbying. Okay. Um there's a lot of cutting out of the little guy. >> Does this You're going to There's There's talk of uh Brener, is that what you're calling it? Instead of Brexit, re-entering the uh >> I mean, it's I just don't I think it's a Hail Mary for a leftist government who want to maintain power because they've we we took on a um a leftwing government at a time when there wasn't much to give out. So traditionally in the UK what happens you have a conservative government we build a bit of wealth there's a uh an imbalance in society you then tend to get maybe a Labor government come in and there's a bit more redistribution and it's a it's a fine pendulum what we actually ended having was quite a center I would argue center left conservative government >> high tax high surveillance and there wasn't much to distribute and by the time people were fed up with them they went >> um the Labor government arrived at a time when there was low productivity low growth way stagation w stagnation a lot of debt and they tried to do what Labor traditionally does and and it's just squeezed everything out, the living standards. So, we're in a really precarious position. Um Karma's obviously out now. He's been replaced by Andy Burnham who's probably going to become prime minister, but he's going to go even harder. So, >> does AI what happens? So, what about AI and technology? Does that come into play? Cuz >> I mean >> I mean Deep Mind is based in UK, but >> Yeah. Um Yeah. I'm not sure if weren't they acquired by Google? They are, but they're based in Yeah. for me, we just I think broadly, we don't have much innovation in the UK, and if you're smart, ambitious, you come out here. >> This is this is the place of opportunity. We have so much so much regulation in every area. I read about a an an investment of in a I think it was a town called Barnsley into AI and the number was something like it might have been in the hundreds of thousands. If anything, at most it was like 5 million. Okay. I was like, "Huh, what are we doing here?" The UK is not in a good position where um if you're if you're going to say you're brighter smart, you should leave and go somewhere else. We are not pro- innovation at the moment. >> It it will change because there will be a reaction as there always is and we will swing to a right-wing government and something more like the Trump administration is going to come at some point. >> Well, but this is a wonderful point you're making. It's the swing that poses the challenge. >> Yes. And I think part of that from my point of view, part of the swing keeps happening because our leaders in all of the our various countries have not established a concrete long-term vision that everybody can get behind. >> They can't though. The problem is is the news cycle and the demands from the public um are so short term that you can't sell a long-term plan. People are >> I don't know. I I think you can, but it has to be it has to be something where everybody feels like they have a role to play. The long-term plans that have been sold here leave out good chunks of the population. So, look, the the project um plan has a very clear vision of the future, but it alienates like half the country. The difference is there's a group of people organized behind it who have learned how to execute and this plan didn't get hatched overnight. It's the product of a long long period of of thinking and planning. Um and but what we have here is just resistance against it. Um you know there's I'll tell a story about my friend Taka who's a I spent a long time living in Japan and China. Um and my friend Taka is a monk. He's the in in Japan in Cotto uh Buddhist. He runs a temple. He is the 26th generation >> Wow. monk in in running this temple. Um I was hanging out with him one day. He's like super fun. So we're hanging out and he's pointing to the roof. I was like, "What's new?" And he goes, "That." And I said, "You got a new roof? Come on, Taco." Like, and he's like, "We were able to get a new roof." And this wood is a particular wood that comes from a blah blah blah. Particular region, all this other stuff. And the reason they were able to do that was because 200 years ago, one of his family members planted the tree that took care of the tree that they knew at some point in the future somebody would need to use to prepare to repair the the roof. Now, this is a tiny little example of a small beautiful temple, you know, in Kyoto, but I think it's analogous to what we lack. M >> what we lack is um can what what is the thing that we desire for the future? Um what he his family desired for the future was that this temple is continues to be a place for spiritual growth and community service and all the things that people need. We understand that things may change in the world but we know we're going to need a roof. That's what I'm talking about that sort of we we need to align on and regardless of your opinions on the economy like how we get there is different. >> Most of our plans and most of what people argue about is the how, not the what. >> See, I would love us to have a long-term plan in the UK. We've gone through, I think, something like seven prime ministers in about 10 years. >> Yeah. You're you're cycling through on uh >> soccer team. The guy said the soccer can't kick me afterwards. I mean football team. Um but but it's but there's this everything's so shortterm at the moment. Everyone's very demanding and it's very difficult because it's very obvious that people don't want to deal with the debt issue and so they keep borrowing more money or they keep trying to tax more to solve the demand that the public has for public services which we can't afford. I would love a long-term plan. It might have to get a little bit worse >> before we get there. I'm really >> It seems pretty bad that didn't your national NHS like run out of money or >> I mean it always runs out. I mean we are I mean if we were a company we would be trading ins solvents. I mean our debt levels are crazy and the problem is is they keep increasing the taxes then people are leaving. It's this vicious loop that Argentina went through where they had the brain drain in the country. We we have a brain drain here. Um you know people connorate age they can't get on the housing ladder. There's lots of economic problems. I think they can all be fixed. I speak to >> what would be your like what's a what's a one easy maybe not easy but what's your sort of like here's where we start to fix? >> Oh gosh. I don't think it could be one, but I can give you a couple. Firstly, we we have to get rid of the mountains of regulation that exists in our I mean, one of our businesses is a coffee shop, and the amount of regulation we have for a coffee shop is crazy. And you throw in the energy prices, the minimum wages. You cannot make money out of this business. We do it for the love of coffee. >> But there's a lot of people out there who have a coffee shop. They need to. So, but I would uh deregulate. I mean, I'd take a pretty libertarian approach. I would deregulate. I would um uh I would make it a welcome environment for investors. So I'd have tax incentives for investors. Uh I would reduce taxes and I would reduce the size of the state itself. The state itself is a is a giant at the moment. But it all libertarian ideas. It would be you know it would be what trying to think of a good example. It would be somebody like um probably back to what Adam Smith would say. >> It's funny. So I when I was in high school, I must have been ninth grade, so probably 15, I read the Communist Manifesto. >> And I was like, I'm a communist now. This all makes sense to me. Like I get it. Uh and then my parents were like, literally never tell that to anybody outside of Harvest family. Um two years later, uh somebody gave me a copy of Ein Ran's Like the Fountain. >> Fountain had Yes. My brain is starting to glitch. Um, anyhow, I read that and I was like, well, clearly every you I'm on the other side of things now. Um, I have I worked really hard for a long time to find a group of people. It's like to fit in. So, for 5 seconds I was a Democrat. Like I'm Yeah, that makes sense. And then I started making some money and I'm like I'm a Republican now. >> here's where I've landed. >> Yes. I'm 51 years old and nobody's singular system like I don't need any labels anymore. >> Um what I would and mainly because each one of those labels including libertarianism connects back to structures that I don't think work anymore going forward. So I would rather like we we frame the conversations as regulation, no regulation in a binary. >> I think what we need is something not regulation if that makes sense. >> Well, so the way I explain to people when I I say I'm a libertarian, I'm really uh a trajectory libertarian. I just I look everything at the moment is a bit too big. >> Uh there's too much government, too much regulation, too much debt, too much printing of money. It's clearly working against uh what we're trying to achieve as a society. Um, we need to we need to reverse this a little bit. It's the um, have you ever ever read I talk about this on the show all the time, but it's my favorite book. It's The Law by Frederick Bastier. >> No, I have not. >> It's a fantastic book, but he he basically said, "Look, there's two periods that society goes through. There's one of uh, productivity and wealth creation. And then there's a second period of plunder. In and in the period of plunder, you have the uh, plunderers who fight for control of the state to plunder from the productive class and you you essentially contract. That's where we are at the moment in the UK. So me as a a trajectory libertarian, I think we're in the plunder phase. I'd like us to get back to the productivity phase. So what will lead to more productivity? I think left and right is a trap. >> It's a trap of the mind. It's a trap. >> Look, I I think spiritually we're aligned. But the point is structurally >> we we are living in a it's it's America's 250th anniversary. >> Yes. this week as we are talking and the way that we created laws and government and the initial structures for this country made sense mostly >> 250 years ago >> during a time in which there was no such thing as global trade that not in the way that it exists today. We did not have drones that could in an unmanned way find and assassinate a person. Like, you know, the way they were thinking about warfare was militias like a guy with a musket, you know, a farm in Virginia, not um unmanned drones entering Moscow to try to take down a political official. It's, you know, >> um or automated like automated systems into all of the things that make up modern society. um regulation and at least in this country the way that we create rules the way our checks and balance system all of these things made sense 250 years ago mostly I think some of it doesn't make sense now however the answer to that is dereg is not from my point of view deregulate or more regulate it is what could exist that is not regulation that has the desired impact >> and that represents a fundamental reperception um in my field. So one of the things we talk about um a lot my colleagues and I that I work with is something called reperception. It's taking all the same inputs that had already existed but just seeing them in a different way exploring whites space versus the stuff that's obvious. I think this moment demands because we need to do something about AI so that the best possible decisions are made for the public interest. But it cannot be regulation. >> Um, you want me to explain why really quickly? >> Please do. Please. >> So, we were talking about football. >> And F1 earlier. Let me tell you a quick hilarious maybe not F1 story that illustrates this point. So 1978, >> the year I was born, >> the year you were born, >> there was a race and all the car companies show up at the race. These are fancy Formula 1 people. And one of the cars has a giant fan bolted on it and everybody's looking at this car and they're like, "What's what's this fan? Why do you have a giant fan bolted to the car?" And the team is like, "The engine gets so hot. The driver gets hot, the engine gets hot. this is just to cool everything down. So they're like, "Okay, well, looks kind of wacky, but so they start driving and the person driving that car is Nikki Laa, who was one of the best >> drivers ever." >> So he would have been pretty fast anyways. But on this particular day, at this particular race, with a giant fan bolted to the back of his car, he is going really fast, especially around the corners. And as he's cornering, he's he's actually accelerating beyond the other drivers, which is a very unusual thing to do. So, he handily wins this race. And uh afterwards, everybody's like, "Well, we want to use these giant fans now." Because in addition to cooling off the engine, that giant fan was also acting like a vacuum cleaner and it was basically sucking the car to the tarmac. So that as Nikki's driving around, he's he's losing no speed. There's no drag. There's no problem. Uh a week later, they banned the use of giant fans on the cars, but his win still stood. All right. What does this tell us about the current and future era of artificial intelligence in Formula 1? They're still doing like that all the time. So like they're they're people are constantly coming up with crazy new technologies because there's no rule saying you can't. So they build it and then maybe somebody finds out and then then then there's a rule after the fact. That is what happens with emerging technology. Uh it emerges right and exactly how something is going to look I don't know. Um and if you were to try to write a regulation or a set of rules you would have to have answers to exactly how is it going to evolve and look over time. We can't do that. So this is a really good it doesn't mean we shouldn't do something but regulation the way we've always thought of regulation doesn't work. We need something else >> and we need to not stifle competition. >> That's right. So we need to enable competition. We need to enable investors and I mean we could spend many many more hours talking about the I I have thought concretely through how to do this but the point is the current structures don't make sense This is um we've we've kind of hit a point where we've both got to leave and travel. >> Well, I'll be in London this fall. Maybe we'll have a part two. >> Yeah, I would love it. I mean, I've got so much more I wanted to talk about. I didn't I didn't even use one of my questions. I the opening question this uh this is everything I hoped and better. And I've got so much more I want to I could do this for hours with you. I really appreciate it. I know I know you're busy and it was a real pleasure. And if you're in London, we'd love to have you in our studio. There's so much more I want to ask you about and I just Yeah, I'm really grateful you did this for us. Thank you so much. >> Well, I love to be on here. Thank you. >> I did have one other question, but it's something I hope I'm hoping we head towards a world where technology does so much for us that we need it less and we just go back into the real world and let's savor it for next time. But that's my hope. >> You know what? I can grant that hope today cuz all you have to do is not use your technology. >> I know, but I need it sometimes. Um Amy, thank you so much. Thank you. I appreciate you. >> Thank you. >> And thank you to everyone for listening. That's our final uh show on this trip back to London. See you soon.