AI, Agents & the New Age of Progress: Work, Poverty & the Future
Nov 23, 2025 ยท 2025 #44. 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
AI, Agents & the New Age of Progress: Work, Poverty & the Future
Human Transcript
Timed transcript
Blocks are grouped by speaker for readability. Expand a block to inspect word-level timing.
Speaker 3
Hello, everybody. It's Sunday, the 7th of December, 2025. We're back on This Week in Technology with my friend Keith Teer, the author of That Was the Week newsletter. Last week, Keith argued that the AI race is a myth, that there aren't going to be a single winner, that there are a number of winners in different categories. And I think that was, I don't always agree with everything that Keith says, but I think it was a very wise observation. This week, he's pursuing that theme again. He's talking about something called the Great Compression, which in some ways is an extension of what we talked about last week, but not just about an AI race, a life race, a tech race, a political race. about whether there are going to be any winners or a winner-take-all society in broad terms. Keith, is that a fair description of this debate over what you call the Great Compression?
It is. The Great Compression really is when a small number of outcomes is most likely. And therefore, attention, money, effort focuses in on those small number of likely outcomes. You know, the best way of imagining it is in the venture capital space. Most money is going into a very small number of funds, and most money from funds is going into a very small number of companies.
Yeah, and the way you've put it in a very interesting way in the newsletter this week is venture capital is no longer venture capital, or maybe it's something different. Maybe it's just a semantic issue, but it's certainly profoundly different from how we used to think about venture capital.
Yeah, and this comes from the underlying articles in the newsletter, by the way, which are a fantastic collection this week. They all suggest this narrative. And it's that venture capital is becoming much more like industrial policy capital. where government and investors are all converging around a single set of themes and a small number of companies. And that is a winner takes all. You know, last week I said there is no winner. There's going to be a lot of winners. This week suggests that there's going to be a few winners. Not a lot, but a few.
I mean, 20%. When I was reading the editorial this week, I thought of the great Italian turn of the 20th century sociologist, Pareto, who came up with his rule, his 80-20 rule. He was a very controversial, but I think accurate observer of society. Is this really Pareto's law for the AI age, Keith, an 80-20 rule?
Well, that's why I think This great compression is so interesting because it's not just about tech. It's not just about Silicon Valley. It's not just about venture capital. It's about all our lives. It's impacting us on every front, particularly on the career front. I was intrigued by your argument that our career, at least according to Keith Teer, is becoming what you call a high stakes tightrope. What does that mean, Keith?
Well, it means that, and this is particularly focused on professionals who bill by the hour, it implies that most of the specialist skills that those professionals can bill for are going to be automated. And therefore, to be a professional, focus on one of those skills is highly risky at this point in time. And what you're gonna end up with is a small number of very clever generalists Like you and I. Hopefully. I mean, hopefully, Keith. The answer should have been yes, absolutely. Absolutely, yes. And then, you know, each generalist will have an army of software-based specialists that can carry things out.
It's not physical people, isn't it? It's not physical people, which is, again, the great compression of... in the world of jobs is about less jobs and more people with leisure time.
So just coming back to this word, compression, what does it literally mean? You talk about this thing called the compression of capital and work, the compression of platforms and reality. I mean, it's kind of interesting for a PowerPoint slide or a TED talk, but I'm not entirely sure what it means.
I think it means the reduction of choice due to the likelihood of events happening. So Netflix acquiring Warner Brothers this week is one of the- Maybe if the American
president whose name I will not mention approves of it.
Words and timings
presidentwhosenameIwillnotmentionapprovesofit.
Speaker 2
Yeah, it seems very likely that, I mean, Warner Brothers clearly want to sell. There was only two buyers. either one of the two buys would have represented a compression of ownership, at least, over media. And Netflix won. I don't know that it would be better from anyone's point of view if Paramount had won. It seems like roughly the same outcome.
Keith, you and I, one of our long-running jokes is over a certain graduate student you secretly, I think, rather admire, Lena Khan. Doesn't this all just prove Lena Khan's version of the world?
is it? It's a terrible thing. It's going against history in a direction that will result in less productivity, less wealth. So she thinks that she's doing a good thing, but actually, from a historical point of view, she's King Canute trying to hold back the tide. Or Queen Canute, I think she'd probably be. wouldn't you?
Who knows? These days could be either. Oh, Keith, now you're getting into trouble. You may have to cut that bit out of this conversation. So with your Marxist background and understanding of Hegel and the nature of history, All this is inevitable. Why write about it? I mean, in your editorial, you suggest it might not work out that well. And you've often talked about human agency and shaping or reshaping the Great Compression. So there's nothing inevitable about this, is there? And if there isn't anything inevitable, then why shouldn't we admire someone like Lena Kahn, who's trying to re-architect it, at least in her mind, in the interests of human beings?
Well, the first thing to understand is the great compression is not an end point. It's a moment in time. So there is an open question, what comes after? But taking it as a moment in time, I believe it's effectively progressive, to use a word. It's progressive and a precondition for abundance that this great compression happens. And therefore, stopping it is the opposite of progressive, which is conservative.
I take your point. You and I have disagreed in some ways on this, but probably more in agreement than disagreement. But you do end on a rather... concerning note. You say, and I'm quoting you here in the editorial, while this concentration of resources may be a rational response, it also concentrates risk by linking retirement savings and our collective economic stability to a handful of highly correlated high-stakes trades We are betting our collective future that the winners of this curve will be benevolent and that the system can survive the compression required to crown them. Those are not certain assumptions about the benevolence of the winners.
So that's definitely happening. Which is right. You know, you can't imagine a future in which Elon Musk isn't more powerful than he is today and more wealthy than he is today at the moment. Same with Sam Altman. I mean, of course, there can be missteps.
Yeah, and our interview of the week, actually, is with two other VCs, Dave McClure and Amman Vergy, who... Tell us the truth, at least, about Elon Musk in their view and Sam Altman, presenting them both as incredibly smart and competitive individuals who want to even more concentrate their wealth now.
Yeah. And so the progressive point is that in order for there to be ever enough wealth to raise everyone up, the success of this stage of capitalism is a precondition. Capitalism is accelerating wealth production right now by productivity gains and investment, and probably the most in America, because it's the most capitalist of all the economies, but China's certainly doing the same in a different structure. And the wealth that that produces, which is simply measured by outputs from inputs, economically speaking, is growing and is gonna grow even faster.
Exactly. And so the minute you move your brain to a political domain, which says what kind of policies are appropriate, there's really two answers that people have. The first is to try to stop it, which is the Lena Kahn answer. To be fair to Lena, I'm not sure she wants to stop it.
I mean, if she was on the show, I don't want to put words into Lena's mouth, but she would talk about monopoly power. rather than just concentration. And she would make a legal argument, but go on.
And she'd be wrong because there's many more than one players that are likely winners. It's highly competitive, as we've seen in the last few weeks, with the baton of who's the leader in AI passing almost every day. So there's no monopoly here. You know, if there was collusion, there might be an oligopoly, but there is no collusion. They hate each other. Yeah, that is for sure.
I think that's one of the other things that came out of my conversation with McClure and Virgie is that they're, both of them, members of the PayPal mafia, so they've worked closely with Altman and Musk, and they do underline the fact that these guys don't like each other.
And then the second way is the Ezra Klein approach, which is abundance and progress are the same word. Capitalism for abundance and policy for progress as a duopoly, if you will. And therefore, you need to have a dialogue. with the Elon Musk, the Sam Altmans about what the future of the whole world looks like. By the way, I've made Musk's interview, the interview of the week this week. And during that interview, he made the point that he doesn't believe nation states will be relevant in the future, which is a point we often make You don't agree. Yeah, I don't make it. And so you need a dialogue about the policy and societal implications of success.
And I would actually, you made Musk your interview of the week. I don't want to get sucked into our Musk conversation. We'll have to agree to profoundly disagree on Musk and not mention that word anymore. But my interview of the week actually would have been a very interesting interview of Ezra Klein on his own show by another New York Times staff member where Klein was being interviewed about 2025. So if you're interested, this is to our audience and also to you, Keith, if you're interested in the Klein argument about abundance, it's well worth listening to as well as reading the book.
Yeah, but that is the kind of, you know, the road forks on that question. Well, it's a political question. Well, it's economic and political. I think step one I am now 100% pro-capitalist when it comes to growth, and I can accept concentration of wealth and power as going along with that. And I see that as the journey you have to go on to get to the outcome, which is enough wealth to uplift our wealth.
Okay, we've had this conversation. Let's talk more concretely, Keith, because I think it's worth just to get sucked into it. abstract visions of the future. Let's talk about the Great Compression in more detail. You describe it as a concentration of wealth and power. You think it's an essential precondition or a station on the way to abundance. I'm not so sure, but that's another question. How does this manifest itself in media history? You have an interesting section in this week's editorial on media history. You mentioned the Netflix paramount deal that's in the works. How would this play out in history in terms of the concentration of wealth and power? I mean, are you talking about guys like the Ellison family or others? I mean, you don't get really, You don't get really wealthy these days by being in media. There are other media ownership now. It's for Bezos and others or the Salesforce guy, Benioff, who bought Time. It's the equivalent of aircraft or yachts, isn't it? How do you lose a big fortune or how do you get a big fortune by starting with a large fortune and investing in media companies?
I apologize for my bad, badly articulated joke. Keith has clarified that. But it's true, isn't it? I mean, you don't make any, any entrepreneur who has the ambition to make billions of dollars is not going to go into media these days.
Well, look, I think the currency of media is human attention, time. And there is a battle for human time between the big screen and the small screen. And on the small screen, there's a battle between AI and web search as the discovery mechanism. And so the great compression here is the shrinking number of interfaces we humans will use to engage with media. And therefore, our time and attention will be channeled through less and less channels.
And for OpenAI. I mean, the red alert that Sam Altman announced this week, one of the pieces I read suggested that by concentrating on the software, he's kind of giving up more on the business model and focusing less on potential advertising models for OpenAI. So it's interesting.
Well, yeah, in a way, that's a counter trend to the Great Compression. What happened this week is In response to Google success with Gemini 3, Altman put out a memo to his team saying, code red, we have to focus back in on the quality of the Frontier model, ChatGPT model, and take our attention away from some of the products we were developing, one of which is advertising. And so that code red is a little tap on the shoulder saying, well, wait a minute, there's quite a few players here. And the idea of OpenAI as the leader, which it still is with roughly 70% of traffic from AI chat,
Last week it was 80, used to throw around the 80%. When it gets down to 60, it'll become more interesting. Let me ask you one question, Keith, slightly off topic. You've been the CEO of lots of tech companies. This idea of Sam Altman sort of behaving as if he's driving the Star Trek enterprise and calling this code red. Is it just infantile? Is there something in it? I mean, would you ever issue? You run lots of companies. You still do run one. Would you ever issue something to all the people who work for you saying code red? It just sounds like the language of a four-year-old.
Well, I suspect that language was filtered through his internal PR team before it. Well, even more. What adult would filter anything through a PR team? But look, I do think leadership is about making ambiguous decisions and being decisive about them. Clearly, that is an ambiguous decision because Google is not ahead. And so it probably, truthfully... No,
but my question wasn't about, I mean, obviously...
Words and timings
butmyquestionwasn'tabout,Imean,obviously...
Speaker 2
No, no, I'm going to answer your question. I'm going to answer your question. But I'm saying the truth is he's called this code red. You know, I did a similar thing at SignalRank in the last few weeks. Did you do a code red, Keith? We didn't call it Code Red, but we made a very decisive decision about what our next 12 months is all about. that was certainly not the only possible decision we could make. But it's important to make one decision and get everyone behind it. And it shapes your future. So running a startup is not unlike a journey with air traffic control.
Maybe that's why I'm such a terrible startup guy. Everything I've ever done has failed because I don't have enough code reds. Maybe I should include a code red in Keenon.
Well, it involves thinking ahead. and then reversing from the future to a current set of things you should do. So the precondition for leadership is having a view of the near future and also the more distant future and aligning your day-to-day work to that.
I take that point, but why not, and maybe this is why I've been such a failure as a CEO, why not just issue an email to everyone who worked for you saying, look, this is really important. rather than code red. It sounds like, again, four-year-old boys playing together.
But your disagreement with it is a bit four-year-old-ish as well, because it's just a word. Who cares? Well, then don't use it. Language matters, doesn't it? I think you wouldn't. Yeah, but it's very effective for him. I don't know if you've noticed, but the whole media seized on that, especially the information. And the last week has all been about OpenAI focusing back in on its core problem.
Yeah, and we've talked about it. Exactly. So you're right, I'm wrong. We've talked about it too much already. Let's talk more about this battle for reality's front door. Is it a battle between Google and Google? Altman's company and Microsoft, or is it a broader battle? Who is involved in this? I mean, this is the ultimate battle for wealth and power in the 21st century, isn't it?
Yeah, I think it's a broader battle. For example, this week's stories are a lot about the browser becoming more of an AI interface than a browser. And that's one of the front door.
And one of the other headlines also, which I found really interesting, is about this very sharp restructuring of Apple's senior executives. Their top lawyer has left. They replaced their AI person. So no company is secure in this world in terms of what you say is controlling reality's front door. Historically, Apple have one realities front door for the last what 10 15 years through our iphones but that doesn't seem to be guaranteed in the future either right look meta also acquired a
company that does a pendant that listens to everything in your day this week the front door almost for sure is going to end up being uh a microphone that captures things around you, and earphones that speak to you, and a database of history that you can refer back to. It probably isn't a keyboard and a web browser at all.
Yeah, and it's so hard. And of course, five years ago, Mark Zuckerberg made, at the time, it seemed relatively prescient, at least from some people's point of view, that the front door to reality would be 3D. And he renamed Facebook Meta. And this week marks a very clear redirection of Facebook from Meta and virtual reality to AI. So all these companies are having to face this. One of the interesting things about the the Facebook redirection, two things strike me about it. Firstly, that he's pretty open about the failure of meta, virtual reality. So on the one hand, it reflects, I guess, a weakness. On the other hand, he's very open about recognizing that and redirecting the company. What do you make of the meta announcements this week?
Well, look, I think it's always been his strength. to be honest with himself about where the risk is. No code red for Zuckerberg. Well, he does. He doesn't say code red, but he acts as if there's one.
Yeah. So acquiring Instagram back in the day and before that, WhatsApp or was it the other way around? I always forget. That was the other way around. I think they acquired Instagram first. Yeah.
Well, what they did know is the risk of being a web company was too much and they had to become a mobile company. They definitely knew that. They didn't know they would be successful, but they had to make those plays. And today, You know, no one believes that 3D face masks are the future interface. It's possible that glasses could be that are see-through, but they're not good enough yet to play that role. The pendant, which is a microphone and has Bluetooth so it can speak to you, is another take. That seems a bit more promising, you know, logically where humans end up, which is they can hear things and and the AI can speak to them. But I think even beyond that, once most people have a robot, that robot is gonna be the microphone, the speakers, the memory. And so you'll actually have a physical companion that you can talk to about everything. And so the front door to knowledge, when knowledge is more and more contained in a database, in quotes, not an actual database, but a knowledge base, let's call it. And the interface to that knowledge base is varied, but includes vision, voice, hearing. That's the most likely end point.
Yeah, and it's interesting, coming back to your battle for reality's front door, that Altman's been forced through the market to focus, to double down on, I guess, his software, his AI software. We haven't heard much about this joint venture he's doing, or I think he was doing with Johnny Ives, of course, who invented at least the look and feel of the iPhone with Steve Jobs. Does that mean that OpenAI might have given up on the battle for reality's front door in terms of a physical product?
No. Actually, Lauren Powell Jobs, Steve Jobs' wife, interviewed the two of them on stage last week. And it was a bit of a love fest, as it always is with the job. He's got that English... English disease of charming everyone around him. Like you and I, Keith. Exactly. Especially you. And he even used the phrase that he was in love with Sam Altman. And they're definitely still working on something.
I bet Code Red ran off in Sam Altman's head when he said that. Right. Is Johnny Ive gay? I have no idea. I mean, Altman is. So maybe that there may be an item in the future. Who knows? I don't know. I don't care. And good luck to me with years. Well said. So coming back to our compression argument, Keith, let's talk. I try to I don't always like allowing you to talk about VCs because that's your day job. But This week, it is interesting. You note that venture capital is no longer venture capital. And early stage rounds, you note, are now made up of billions of dollars. Does even the idea of a stage of a series A or B or C or late stage round, does that even make sense in our age of great compression?
It doesn't make sense for the most prolific AI companies. They seem to be stage independent in terms of check size and stage. You know, you have people raising more than a billion dollars when they haven't built anything yet because of who they are, like the former CTO of OpenAI, for example, or Ilya is another example. So you only mentioned his first name because we could pronounce that. The second name is what? And so for AI companies, no, but for the rest of venture, there's still stages. And if you, I have the data, but I'll describe you that the top 5% of funding rounds early stage are massively higher than let's say the 75th percentile of the median. So there is a separation, but there still is a normal venture market in other segments.
So what should ordinary, I mean, most of us, for better or worse, aren't VCs. We're not Johnny Ive. We're not Sam Altman. What should we all be doing in this age of great compression? Where should we still be going to law school? I mean, you've noted that the billable hour era is coming to an end. White collar work is being squeezed. What should people be doing? Should they be just spending their time on an anthropic or open ai or google gemini and learning
how to manage armies virtual armies i think there's going to be three kinds of people the first kind will ask the question if billable hours are going to zero can't i build a law firm with no people and make a lot of money And those will be the innovators, the entrepreneurial types who see an opportunity in the law.
Yeah. And in that world, you've got lower costs. Therefore, you've got expanding margins, that is, say, profit margins. And so a normal profit margin in a law firm is probably as low as 20%. It might go to 80% to 90% more like a software company. And then the second group, will be in denial and will keep doing what they're doing which is running a law firm and paying lots of salaries and they will eventually um not be viable and then the third group uh are the employees of of those leaders who will be insecure i'm
wondering what comes next you got three kids young we both got kids in their 20s um in terms of the tightrope for the professional. I know one of your sons is a software person. Maybe one of the others is more entrepreneurial and same with mine. What advice would you give young people in this age of careers for better or worse, being a high stakes tightrope? We're all tightrope walkers. Do we need to literally become them? Learn how to, not just to walk on a tightrope, but learn that, When we fall off, we've got to get back on and do it again.
Well, at least for the near future, everybody knows that AI needs to be led by humans. Nobody's going to build autonomous AI that has no human oversight. And the human plays the role of the generalist and the guide, almost like a product manager is in a business as opposed to an engineer or a product leader. And I think my advice would be to become a competent AI driver in whatever your domain is, to be the air traffic control, to be the person who defines the endpoint that you're trying to achieve, and the person who then runs AI to achieve that endpoint. That will be very valuable as Korea, almost in any domain, whether it's legal or engineering, almost any. And then I think physical AI is still some way away. So if you have a job that's physical, even driving or flying airplanes or, you know.
Yeah, anything physical. But I think societally, we have to embrace the idea of a reduced working week and consider that to be progress and fight for it. and then discuss what is the standard of life that people can expect to maintain as the working week reduces as a good conversation, as a progressive conversation.
Yeah, I'm haunted. A couple of weeks ago, I think I may have mentioned it on the show. I was in Washington, D.C. There was a reunion of my graduate class, and most of the people there were professional academics. And they were all, as professional academics, by definition, they're deeply conservative, reactionary, maybe not politically, but in every other sense. And one of them said to me, oh, things are terrible. He teaches at UC Berkeley, so a very, very prominent university. He said, things are really bad. The AI is ruining everything. I had a student who I told the kids, and I'm sure this is true of all academics and probably school teachers as well. I told them that they have to do the reading. And a woman came to me, this is what he said, and he was shocked by this. This was his argument to show the crisis. He said, a woman came to me and said, I'm not going to do the reading. Why would I do that? And on the one hand, I'm a writer and I love reading books, so I take his point. But on the other hand, you can understand this woman's position. Why would you read a book? to download information, which would take several hours, when you could do it in 30 seconds. Is that a fair sort of way of conceptualizing what is actually happening?
Well, no, because it isn't just a library anymore. It's a library with a brain that can summarize anything and teach you anything. So she obviously is on the right side of history. And I think what happens is reading becomes pleasurable, not as part of your education, but as part of you serving your human need. And education will become much better, much easier to bring everyone up to a high level.
Yeah, I mean, if you're an author, then you shouldn't be thinking of writing books which are where their value is in their information. It has to be in the experiential quality in terms of art or some other way of getting value from reading the book. Because if all the value of the book is purely an information, you as an author, you no longer have any value or your book doesn't have any value anyway.
Yeah, I think there's a correlation between value and content in this way. if the book focuses on a contested part of human knowledge, for example, currently there's a raging debate among physicists about the source of consciousness, where there's no established answers. And probably never be one.
Right. And there's another raging debate about was there a universe before the Big Bang due to the James Webb Telescope? These are still interesting to read because it's contested knowledge and you're trying to puzzle things through. But where it's established knowledge, there's way less reason to be reading.
Then we get into the area of vaccines and autism, which some people say has contested, others aren't. Let's end, Keith, with again a return to politics. I'm still not convinced by your argument that We've all got to just lie back and think of England, so to speak, in terms of history and accept that this is all inevitable. What is the role of politics and political parties in our age of this Great Compression, where everything is changing so fast, where everything we took for granted? Marx famously put it in his manifesto in the 19th century, where all that is physical is... How did he put it in the manifesto? Did...
argument was in terms of the industrial revolution and its reshaping of the industrial, the agricultural world. Now this AI age is doing the same when it comes to the industrial world. Yeah. Well, what is the point of politics or are you just basically saying there is no point to politics anymore?
So, The idea that all that is real turns to dust is really another way of saying that we are reinventing the entire human division of labor on the basis of new techniques that we've invented. And therefore, all your assumptions about continuity need to be put on one side, and you need to be more asking the question about what is next.
he wrote in the manifesto, all that is solid melts into air, all that is holy is profaned, and man is at last compelled to face with sober senses his real conditions of life and his relations with his kind. In other words, he realizes at least in early Marx's species being, I'm not sure whether he would agree with that as an older man, but This idea of all that being is solid melting into air seems to be happening today in every sense in terms of our lives, our politics, our economics, everything.
Yeah. Remember, he wrote that in 1848 at the time when almost every European country was having a revolution. So it was tangible to him that everything was changing. And for us, everything is changing again. Politics, the role of politics is another way of saying the role of human beings in shaping the future. Politics is where we discuss preferences and then seek to make them happen. And so the absence of that discussion, and I will say, it's very hard to be optimistic about either of the political parties having really grasped the discussion that would be required today to design the future based on what's happening. But the bar is high because the outcome for humanity is on the table. And so politics is where you discuss that. Political parties are formed, opinions are shaped. Populist movements can arise, usually will arise if there's enough enthusiasm.
Yeah, and they can be reactive or they can be progressive, populist movements. And often both at the same time. Can be both at the same time. So in some ways, you know, this is an exciting moment because human beings are no longer passive. they are actually wanting to take some control over their future.