Jun 7, 2025 ยท 2025 #22. Read the transcript grouped by speaker, inspect word-level timecodes, and optionally turn subtitles on for direct video playback
Edit labels for this show, save them in this browser, or download a JSON override for the production folder.
Transcript Playback
Abundance And You
Human Transcript
Timed transcript
Blocks are grouped by speaker for readability. Expand a block to inspect word-level timing.
Speaker 1
Hello, everybody. It's Saturday, June the 7th, 2025. A week ago, I did a show with my old friend Keith Teer, the publisher of That Was The Week, about what he called the abundance trap. It was a conversation from his newsletter last week. We are accelerating to abundance. And he and I talked about how inevitable this acceleration was and what exactly abundance should mean in the age of AI. And as it happens, history is repeating itself. A week later, and Keith's newsletter for this week is entitled Abundance Anew, and in some ways it's revisiting a number of the issues that we talked about last week. These are big issues, so I think it's actually rather healthy that we revisit them and talk them through again. Keith, is that fair? Why did you choose to write two weeks in a row about abundance?
Mainly because in the week since, a lot of other people have talked about abundance. One of them is a Financial Times writer who wrote a good piece on it, talking about abundance hype.
And I think in our conversation last week, we didn't clarify some things. We talked a lot about economics and politics and social policy, but I felt like we didn't do ourselves justice in really understanding all these things.
Well, you've had a week to recover. What have you learned this week, Keith, about abundance that we don't know last week? The cover of your newsletter this week is an eight-second video made from VO, Google's AI video of five people, two blonde women who look like twins in red tops, and three other characters in a... in a fork in the road. So obviously this is about our choice, which road we take on the AI trail.
Yeah, so the subtitle this week may be more interesting than the title. It says, everything is possible, but nothing is inevitable. Sounds like a movie. Wasn't there a movie called that? I don't believe there was, but I could be wrong. And what it speaks to is... disambiguating in the topic of abundance, disambiguating what is inevitable and what isn't.
When it comes to outcomes, well, remember, we talk about abundance and you. So as it pertains to the individual, nothing is inevitable. But as it pertains to abundance, something is inevitable. It is inevitable that the last several hundred years of human history where the working day has declined will continue to be the case at an accelerated rate driven by AI and robotics. So you could call that productivity, but let's just describe it in plain language. It's less people will be able to do more things due to automation.
I think I would accept that, but what I wouldn't accept is the degree of that change. So it's conceivable there'll be a 10, 20, 30 percent, optimists believe 80 or 90 percent. So whilst
productivity will probably increase, particularly obviously when it comes to mental work. There's nothing inevitable about the rate of change, is there?
No, the rate of change is impacted by science, execution against science, and also by government policy and all kinds of variables. So you can't predict the pace of change. I think you can predict that the pace is accelerating. And that going from where we are today with agents to agents that have physical form in the real world, robotics, seems to be something that's gonna happen within a period of time that we can understand in our minds and we'll probably live through. That is, just that alone is amazingly impactful on what human potential will be. And so I don't think we should underestimate it. I don't think we should overestimate it. But we should recognize that AI makes things possible that previously weren't. And the number of things that are possible literally today is a small number of things, mainly around automated coding or automated task management. It's going to get to be more and more in the near future.
Shall we call this Tia's Law as opposed to Moore's Law? I don't necessarily disagree with you. And I don't think it's a particularly controversial point that you're making there. So in your editorial, you'll talk about something called the promise and the problem of AI. What exactly is the promise? Is that what you've just described?
Well, I think the promise can be short-term or long-term. Broadly written, the promise is the reduction of human labor theoretically to zero. That's the promise. And enshrined within that is the belief that that would be a good thing. Of course, that is not necessarily something everyone agrees with. Some people think human beings are like animals. They need work in order to be disciplined. And they have this horror nightmare about individuals, you know, becoming couch potatoes because they have such a low assessment of humans, I think is a good thing. I think less work is good.
Well, yeah, I mean, I wouldn't agree with that. I think that there are some people who have a very high opinion of the potential of humans, which suggests that they're better off working because it brings out our humanness, whatever that is. But that's, again, a complicated issue. So I agree on the promise. The promise allows us to... to borrow some language from Marx, realize ourselves. So what's the problem, Keith?
The problem is that abundance used as a word that describes outputs, you know, the things produced, whether it's services or products or experiences, abundance described as outputs has no narrative within it as to who will benefit from those outputs. And so abundance could lead to a nightmare society close to a feudal system where a small number of people own everything, or it could lead to a utopia where everybody has anything they need. And which of those two futures happens, hence the fork in the road in the image, which of those two futures happens is dependent on human action. There's nothing built in that says the good scenario will happen versus the bad.
And this is obviously one of the themes in Criddle's piece in the FT about Silicon Valley's abundance of hype, overabundance about the promise of tech executives. And of course, tech executives are not disinterested observers of this. They have a huge amount invested. And this, to me at least, and we've talked about this before, this is the biggest issue of...
And that's going to be true in robotics. You know, the...
Words and timings
Andthat'sgoingtobetrueinrobotics.Youknow,the...
Speaker 3
the means of production, which is the stuff you need to make things, is going to be privately owned. Well, that already tells you you can't really be anti-capitalist.
No, I don't mean anti-capitalist in a political sense, but you actually have to embrace the fact that in the societies we live in, risk invested in outcomes is the source of those outcomes. And governments aren't going to do that. It's not the job of government to do that. And they have no intention of doing it. They don't want to do it. And populations wouldn't want their taxes spent doing that either. So you can rule out governments. Well, the only thing left is people. And people create companies to do that. And the successful ones become big. It seems to me that's the structure. Is that the promise or the problem, or both? Neither. It's the reality of how this abundance will be produced. And therefore, you're right, the proceeds will be owned by private companies unless something changes.
All right, so we're in agreement on that. So let's move on in terms of the editorial. You suggest that... I'm quoting you on the editorial, two distinct conversations are conflated. The first involves economics, and the second, what you call a more crucial conversation, concerns distribution. How does that play out in terms of AI? What is the economics of AI versus the distribution of AI?
Kind of like electricity, isn't it, in a sense? I'm not a historian of electricity, but I'm guessing at first the cost of electricity, both in production and access, was quite high, and it's dropped more and more over the last couple of hundred years.
Yeah, and I think the entire history of the division of labor is that. As tasks become automated, instead of having 100 people doing the task, you can have one. the entire history of the division of labor driven by innovation is a history of lower costs more outputs with individuals um less that's not entirely true i mean
adam smith's theory of economics which was built in the early history of capitalism and of the factory suggests that it was the division of labor that enabled the revolution in productivity
Yeah, but you can't have a division of labor unless you invent the seed drill, because otherwise everyone has to plant seeds. So they're correlated. It's neither one nor the other, it's both together. And so once you have this combination of technique driven by discovery and division of labor, the working day can decline from 16 hours down to modern norms of 40 hours a week or even less in some countries.
But it didn't always work out. I mean, you've got another section on lessons of history. I mean, one of the lessons of history is... History is complicated. I mean, in the 19th century, if your narrative was true, people's work days would have got less. But people were slaving away and literally quite often in factories for 16, 17 hours a day. They were working much harder than they ever did on the field.
Well, that's a snapshot of a moment in time. But I don't think you would claim that factories have increased the working day. Factories clearly have decreased the working day over the long haul. And, yeah, there are moments like wars where everything, you know, old rules are turned off.
Yeah, I mean, it's an interesting... I'm not sure. I mean, I'm not, again, enough of a historian. It would be interesting to compare the labour of a Shenzhen worker versus their labour when they were on the land. But anyway, let's go back to... this distinction between, I'm sorry, I keep on interrupting as I tend to do, the distinction between economics and distribution.
So distribution really is the word that describes how do these increased outputs produce more cheaply than ever before? How does the benefit of that, not necessarily the things themselves, but the wealth embodied in them, get distributed to everybody? So let's say you're out of work because teachers are no longer required because AI teachers are better than you.
So you have no income. So how do you personally pay for a house, probably a transport, food, that's distribution. How do the benefits of that get to you, especially if society has gone beyond labor as a requirement of life? And that's a good question at any stage. It already starts to be a good question now. Two of the teachers at my kid's high school retired this week. The implication was they were forced to. They're 60. And, you know, I think we're going to see in education in particular massive change over the next five, 10 years. How are those people going to live? Now, you know, the right wing point of view would be we don't care. but if it's everybody, suddenly you have to care.
Yeah, and it's not just teachers, especially when it comes to AI, it's engineers and lawyers and programmers and doctors and everybody else. So it sounds like, Keith, you're becoming more pessimistic. No, no. I mean, you've got this section on the lessons from history in terms of recent history of digital, for example, we were promised that the distribution of entertainment and social media would benefit everyone, and it clearly hasn't, or most people would argue it hasn't. So what lessons from history do we have to suggest that this outcome may actually be good?
Firstly, I dispute your lessons from history analysis. I mean, I think the very existence of Amazon Prime and Netflix and the huge numbers of people that subscribe to them and the fact that most people have several hours in their day when they can be entertained if they want to choose to use it that way, that's real. And that wouldn't have been the case, you know, even in my lifetime when I was born in 1954.
And so I don't really agree with your framing. That said, I think the amount of time in a day when you will have choice about what to do is going to grow. That's the real bottom line. Choice is, you know, it's a Margaret Thatcher and Ronald Reagan word in recent political history. They used it as a weaponized word against social welfare. But I think there's a real meaning of the word choice, which truly means how do you want to spend your time? And what, you know, which will resolve itself into ultimately what do you enjoy doing?
But you haven't answered, I mean, you brought up the problem which is the... Distribution. Well, you call it whatever you want, the gorilla in the room, that people aren't going to be working. They have to pay for rent. They have to pay for clothing. They have to pay to feed themselves. Where is this money going to come from?
Well, the debate, there's a lot of answers to that that are being discussed. The most recent is getting rid of the word basic from universal basic income. and calling it universal income, which is a concept that, for those who don't know, there's a book called For Us the Living that has this concept in it. It was written in 1939 and is based on some Austrian economic school theory, modern monetary theory, where you can give people things because they're so cheap to make, you don't have to, get money back when you give them to them. And so the core of it is the idea that production is so prolific and universal that everyone can have whatever they want.
Yeah, but if you want to take stuff from 1939, there's very little evidence that guaranteed minimum income, whatever, none of these things ever take off, even in the most minimal way. So why should we believe that they'll take off now?
I think you're being a little bit mischievous or entertaining, one of those two, because as you and I both know that Sam Altman's project, WorldCoin, now called world, they got rid of the coin. They just developed, remember that orb that you had to have a resume on? They've got rid of that now and they've improved on it to a much smaller device that's less expensive. And their goal is to have every human being on earth receive as a matter of course, their portion of the proceeds of AI in a crypto token, WorldCoin, monthly. And they've already got something close to 2 million people receiving it. And WorldCoin already trades on exchanges and has a value. So that's a practical... And you trust... Sam Altman with this?
It's not really under his control because it's automated from the moment you're in the system, you get your distribution. Now, it's not trouble-free. There's all kinds of detailed issues you could discuss about WorldCoin. But it does indicate that there are potentially practical steps to distribute wealth to everybody. that would allow them to have a high standard of living without working. if the society is wealthy enough to provide that. You can't pluck it out of nowhere. It has to come from the proceeds of automated production.
So this is what you mean by everything is possible. This is the age of abundance. So your exhibit A in the potential is Sam Altman's new scheme to give everyone in the world some sort of crypto coin.
And I don't think he's alone. I think there's a lot of different projects out there trying to figure this out. I heard Paul Ryan, the libertarian, interviewed this week, and he was asked about this. And he made the point that, and he's so anti-government that he's on his own, in a way, on the spectrum. He made the point that, yes, If there really is no need to work, government will have to deliver health, education, food, everything basically. So I think there is an intellectual understanding. It's a logical thing at the moment, it's not practical, it's not yet the time, but there's a logical understanding that the ultimate end game here will involve no work, but everyone has a good lifestyle.
Yeah, I have to admit that to me, this is just a fantasy. I mean, the reality of America, at least at the moment, is more and more inequality. The people who aren't working and who would like to work get less and less. So I just don't see any evidence of it. I mean, if your best case is some... Bizarre scheme by Sam Altman to give everyone in the world a proceed of the profit of AI. Have OpenAI and the other AI companies, have they bought into this? I think intellectually, many of them do.
I do think you're... What do you mean intellectually? Either they do or they don't. No, they do. I mean, look, Vinod Kostler invested in WorldCoin, for example, along with a bunch of other venture capitalists. And WorldCoin clearly isn't a profit-making enterprise. it's trying to solve a social problem.
I actually don't know if that's true, but there is no profit, I can tell you that. There's only costs. So I don't know where profit would come from. But your point about you see no evidence, isn't that a little bit like, you know, at the start of spring, all your plants look like they're dead and your lawn is, you know, not green. And you say, I see no evidence of life, but then spring comes and then summer comes and there's flowers and there's green. I mean,
but this comes back to the question of the title of your newsletter this week, everything is possible, but nothing is inevitable. There is an inevitability about the appearance of spring and summer and fall and winter, whereas these things aren't inevitable. You can't compare the seasons, Keith, with... Some of them are inevitable.
That's why I tried to separate out economics and politics.
Words and timings
That'swhyItriedtoseparateouteconomicsandpolitics.
Speaker 1
So what exactly is inevitable? The end of paid labor. Okay, so I take your point on that. I mean, that's such a long term. That's huge. Well, I actually don't. The end of paid labor, I think, is not inevitable. So you're suggesting that... No one will be paid for things they do in the future. That's inevitable. Nobody will be required to do work. No, but you didn't answer my question. You said that it was the end of paid labor. They may not have to, but they would choose to and want to.
No, they'll have to because there'll be no jobs. Why would I employ people if I can get software to do the things people do? I mean, I would never employ people. People are too expensive.
Well, we will see. It's all incredibly abstract, but also enormously important. It was inspired from a very good piece from Christina Criddle, Silicon Valley's Abundance of Hype Over Abundance. But the debate about abundance, as you note in the newsletter and in the pieces you cite, is very concrete too. You connect with a piece by Noah Smith, about the polemical significance of the concept of abundance on the Democratic side, and it's being triggered by the success of the Klein book on abundance. What's happening on that front? Leaving aside all these abstractions, why is the idea or the promise of abundance now important for the Democrats?
Well, I think the Democrats can't win unless the pie, if you will, is expanding. If the Democratic Party is the party of dividing up a small pie between people, that leads to all of the stuff we see today around immigration and contested access to assets and resources. For example, free healthcare. You can't imagine free healthcare unless the wealth of society is growing You can't imagine free education unless it's growing rapidly. So if the Democrats are the party of education, of healthcare, of minimum living standards, you need a growing pie. And that part of the democratic narrative has been missing. They've never talked about how to grow the pie. They only talk about how to distribute it. And that is a weakness because everyone's clever enough to realize unless the pie is growing, there's nothing to distribute. And that's going to take away from me. So it comes across as taking something away from people who are middle class instead of giving everybody something more. So I think it is a super critical debate for the Democrats as to how they grow the economy.
And not only grow the economy, but make government more efficient, as Smith notes in his piece. A lot of people who are commenting on this haven't even read the Klein-Thompson book. And it's about why government doesn't work and how Democrats need to somehow rethink or reinvent government. And I'm not sure how relevant AI could or should be in this.
Right. Well, AI is simply a tool that... that will help grow the pie. That's all it is. I mean, at that level, it's a human good. Even if it doesn't get distributed, a bigger pie is normally a good thing. But AI, it shouldn't be the centerpiece of the conversation. It should be the consequences of that increased production for social good.
Yeah, but again, you're sidestepping the question of how AI can make government. I mean, the problem with government in America is it can't do anything. I mean, lots of books have been written about that. Can't build high-speed rail. Can't build homes. So how can AI help with that?
I think that's a harder conversation. I mean, it comes down to... world coin versus the government. If you think about those two, world coin is a private initiative to distribute wealth. Government is a public initiative to manage.
I think They have the capability to do that, although they'd almost certainly do it as a government initiative. That'd be a government coin, if you will, which is what the US dollar roughly is. So they probably wouldn't get behind WorldCoin, but they may embrace the concepts behind WorldCoin, which is the distribution of wealth via a form of taxation, really. that takes assets from the AI-driven economy and makes those assets available to everyone for their life.
In other words, everything is possible, but nothing is inevitable. All very abstract. More concretely this week, a couple of interesting pieces of news on the AI front. One about 11 Labs debuting a conversational AI 2.0 voice, and another about Notebook LLM adding a new way to share our own notebooks publicly. Are you bullish both on 11 Labs and Notebook LLM?
They're both great. Notebook LM, of course, owned by Google. And whenever you and I don't do the show, I use it to create a show for us. And by the way, weirdly enough, those shows get way more listeners than we get. Probably because people are curious to see what it did.
Maybe it's because we're just really boring. It could be. I mean, Google should really buy 11 labs. I mean, my problem with Notebook LM, I'm sure they'll eventually change it, is that these... The AI, the conversational AIs, they sound the same. They're incredibly annoying. Yeah. They're unlistenable after a while. The first time you listen, you're amazed. Second time, you're half amazed. By about the fourth or fifth time, you're just irritated. Yeah. If Eleven Labs and Notebook LLM were actually... seamlessly integrated, you'd have an amazing product.
It's also uncertain. My interview of the week is with Chris Schroeder, who's a global venture investor. And he talked about drowning in black swans in what defines our age as its uncertainty. And it was an interesting conversation. And then I was particularly struck by the fact that someone of Barclays this week suggested that Alphabet's stock could fall as much as 25% in a Black Swan event. So I'm not sure if you saw the Black Swan interview with Schroeder. I did. But what do you make of a world of such radical uncertainty? And is it reflected not just in Alphabet, but in all the other big tech companies, particularly, I think, OpenAI? It's so uncertain.
Yeah, OpenAI is uncertain, but you'd bet on the upside. Google's uncertain. Well, you obviously bet on the upside. That goes without saying, doesn't it? Well, I was going to say Google is similarly uncertain, but you might bet on the downside. And that's to do with the dynamics of change. And to be honest, if you really want to drill into nothing is inevitable, one of the reasons for that is that global economics and politics... are counter-cyclical to the AI revolution. The AI revolution, one can mainly be optimistic about at least human capability. But the global economic and political overlay that is happening within is increasingly a fight between nations for who the winner is going to be. I hear conversations like the AI, you know, the USA can't lose the AI race, as if the AI race is a binary thing with only one winner, where clearly it is entirely likely that both China and the US will benefit from AI. And so you've got this highly contested global canvas, which is mainly driven by the rise of China and the pace at which China has risen and the therefore relative decline of the US and deindustrialization and all those trends leading to internecine fighting, internecine, I think I should say, fighting. where even Americans are fighting immigrants.
And I talked about this a little bit last week. This is my biggest concern with all these conversations about abundance is that Maybe there is a certain kind of abundance you're describing which AI will enable, but it will create new scarcity. So in the 2020s, it doesn't seem to me entirely uncoincidental that land and territory and accessibility to countries is increasingly scarce at the same time as we have this supposedly abundant technology. They're connected. Coming back to Google, there was a really good piece in The Times about what if Google just broke itself up, a tech insider makes the case, which actually suggested that everyone, Google shareholders and Google employees, be better off if Google actually split up, which goes back to your point about the upside of the downside of Google in a Black Swan, as a Black Swan event.
Yeah. Yeah, devil in the detail there about what they break up. But intuitively, you know for sure that if Google broke out Gemini and Notebook LM into a pure AI company, it would be worth a lot.
And you've caught, of course, Waymo, which is also in its own way an AI company. And they made Demis Hassabis, the head of DeepMind, the CEO of the new company.
Yeah. There was another article this week that said the opposite, which is Google isn't worth. The pieces of Google are not worth the current value of Google. Therefore, it would be a decline. And that's because the value of Google is largely driven by its search engine and its advertising business.
Where we pontificate. Finally, your post of the week is by David Friedberg, which explains AI's abundant future. More time for family, friends and ledger. I haven't actually, I have to admit, I haven't.
here's what really annoys me about all this and yeah i'm sure that you will call me a reactionary and all the rest of it but what's particularly annoying is friedberg or friedberg however you pronounce his name is an example a classic example of a successful silicon valley type And this in Silicon Valley is defined, and you know this better than I do, by how intensively people work and how bad they are socially and what an absence usually of leisure and private life they have, often out of choice. So it's always particularly rich to hear from, and I use that word, I guess, with a degree of irony, It's particularly rich to hear from Silicon Valley billionaires and multi-billionaires about the abundant nature of AI's future because there'll be more time for family, friends and leisure. I mean, I don't know enough about Friedberg to discuss whether or not he's exhibit A in this, but is there some truth to that?
So Friedberg is a scientist. He's a venture capitalist stroke scientist. His main work is figuring out how to genetically modify plants in order to be better. whatever the goal of the plant is, to provide food or shade or whatever, you know, trees and so on. So he's a pretty sober kind of a guy. And he is wealthy, for sure. But the reason I put this in is because it was a very articulate... and very short summary of what the potential is. And I think it's good if these powerful, wealthy people in the valley are making social points about the future they want. I like that because it means they're not just talking about how much money they can make. They're talking about what the impact can be on the world. Vinod Khosla does that a lot.
In a week where... And we haven't talked about Musk, and I'm pleased we have, and about his bizarre breakdown of his close friend, Donald Trump. These people, and again, I know it's hard to generalize, and I'm not saying it's true of everyone, but there is, you know this as well as I do, you're a Silicon Valley guy much more than I am, that the social skills of a lot of these people are relatively minimal. I mean, Zuckerberg, for example, comes to mind. So they often talk about social and the value of friends and talking to one another, but they're not skilled in it themselves.
I think sometimes that might be what you glean from their public persona, Andrew. And I used to live in College Terrace here in Palo Alto where Zuckerberg lived before he went and bought his big compound. And we trick-or-treated him one year. And he used to walk around the neighborhood, by the way, and talk to you. He's fine. He's perfectly normal and nice, actually. So I think the public persona gets in the way of who they really are. And that's always true. You know, whenever you become a little bit famous, it kind of warps your public persona a lot.
Yeah, I'm sure that's true. I'm sure he's not quite as... evil, and I'm sure the same is true of Musk and Andreessen and all the rest of it. But they seem to be sort of socially troubled. But coming back to the Friedberg point, again, it comes back to this abundance and scarcity. One of the defining qualities of our age is its antisocial nature. We have all these books like The Age of anxiety which suggests that people are more and more isolated can blame on social media on technology on the changes of our politics and culture so again it doesn't seem it seems to me sort of eerily coincidental that we have this new technology that's come along that is supposed to solve all our big problems of loneliness and isolation and lack of meaning in our life. But the reality is that some of this technology at least is causing it, is part of the reasons why we're falling into this trap. And there's very little evidence that it's actually going to make it better. If anything, it'll probably make it worse. Well,
I think there's an interesting conversation maybe for another week there, which is to do with can a society that is increasingly individualized also be more social than a society that was more dependent on communities? I would say that economic progress tends to create individuals because you don't need help. Everything you need Is there available to you? When I was growing up, the neighbors were really important. Sometimes you ran out of money, you needed food, and they had some, or babysitting, all kinds of stuff. Increasingly with economic progress, people become independent and they like anonymity. They don't really want their neighbors in their life as much as we had them in the past. And they see that as a plus, that anonymity. Imagine if your neighbor's knocking on the door three times a day. And so I don't believe individualism is antisocial. In fact, I think what it does is it increases the space where you choose to be social. And that is, you know, on your terms. So especially in a world where AI is doing all the work, yet everybody can have food, That would be the most social thing I could imagine because everybody's connected to the core of an economic engine that can make their life good. But it isn't social in the narrow sense of you're making friends and hanging out with people the whole time.
in other words keith everything is possible but nothing is inevitable which sets us up for next week i hope you're going to do another abundance post though who knows it depends what's in my newsletter Well, we will see. But good conversation, important subjects, which is certainly only going to intensify. Much to discuss, Keith. Have a good week. And we will talk again, abundance or otherwise, next week. Thanks so much. Thank you, Andrew. Bye, everyone.