Tyler Cowen: Creativity, Talent & AI

It’s a special episode today. We recently celebrated the 200th episode of the Future Work/Life podcast, which coincided with the same milestone for my guest’s show, "Conversations with Tyler."

Tyler Cowen is an economist, columnist, blogger and podcast. He’s a professor of economics at George Mason University, founder of the Marginal Revolution blog (which he’s been writing daily for 20 years), and a columnist for Bloomberg.

Tyler and I delve into various topics in this episode, reflecting his diverse interests and expertise, including his recent book GOAT, which explores the greatest economists of all time, and his previous work on talent in collaboration with venture capitalist Daniel Gross. Our conversation also ventures into unexpected areas like football and dating!

TALKING POINTS:

• Writing and podcasting: Tyler reflects on his journey as a podcaster and writer and how these roles have influenced his thinking.

• GOAT: Tyler shares why he published his new book in GPT and what we can learn from asking questions like an economist.

• Music and creativity: I ask Tyler about his passion for music and its influence on his creative endeavours.

• Future of work: Tyler's views on the impact of AI on jobs and productivity, especially in knowledge work.

• Remote work trends: Remote & hybrid work's impact on productivity, culture, and talent development.

• The role of AI in identifying talent: The potential and limitations of AI in talent acquisition.

TRANSCRIPT:

Ollie: Hello, and welcome to the future work life podcast. My name is Ollie Henderson and my guest today is Tyler Cowen. Very recently, I celebrated the 200th episode of this podcast and appropriately enough, Tyler also recently celebrated that milestone.

And today's episode, I've got to say is one of my favorites ever. Tyler is a professor of economics at George Mason University in Virginia in the US. He's the host of the long running economics blog, Marginal Revolution and a Bloomberg columnist. Now I got to know Tyler's work through his called conversations with Tyler, and he interviews people from all sorts of backgrounds, whether it's sport, people like Martina Navratilova and Kareem Abdul Jabbar, technology and entrepreneurship, people like Peter Thiel and Mark Zuckerberg, or indeed music. He recently had Rick Rubin, the legendary producer on his show. And that's one of the areas we discussed today.

So as you'd expect for someone with such breadth, we talk about a range of topics, many of which of course, focus on work jobs, and particularly AI being an economics professor, we also talk about that subject and specifically GOAT, which is Tyler's most recent book about the greatest economists of all time, and the first to be published in GPT, he explains why he wrote the book and why it's relevant even to us non economists. We also talk about talent, which was the previous book Tyler wrote with venture capitalist Daniel Gross.

Along the way we also talk about some unexpected subjects like football and dating. If you want to find out more about Tyler, you'll find links in the show notes.

In which I'm going to be writing more about the themes I talk about with Tyler today over the coming weeks.

Now if you're having conversations in your company about the effect of AI on jobs and productivity, you can also get in touch with me via the link in the show notes and I can come in and talk to your team. That's enough from me.

Here's my conversation with Tyler Cowen.

Ollie: Well, Tyler, it's a pleasure to have you on the show today. I've just tipped over 200 episodes of this podcast and I believe you've reached the same landmark recently.

So first of all, Congratulations 200 episodes in, what have you learned makes a good podcast interview?

Tyler: Just doing it, right? If you just wake up and do it. The main thing I've learned is I should have been doing more of them. Not that I think they're all good, but people can pick and choose. And ideally you should do a podcast episode every day and, you know, let the market sort it out.

Let your AI. Listen to them all and tell you which ones you'll like.

Ollie: Yeah. It's all those. Have you ever tried doing that? I've tried twice doing what I have called a pod storm, in tribute to weet storms and just trying to do a podcast every day over a period of time. So, in December, I did four weeks every working day. I put out a short one, actually five to seven minutes.

And it's surprisingly difficult to keep the consistency at that rate.

Tyler: I’ve not tried but in my experience, everyone wants April and October. So parts of October feel a bit like that, except they're an hour, and you have a lot of prep.

Ollie: Well, I appreciate you squeezing me in on December the 29th.

I've listened to a lot of your podcasts, and I actually listened back to the very first one you did with Peter. Thiel and I don't think it actually, it didn't sound like it was supposed to be a podcast per se. I mean, it sounded like a conversation in front of an audience.

How's your approach to interviews changed since that first episode with Peter Thiel?

Tyler: I prepare a lot more now. I'm doing Peter again and it will be released as a Conversations with Tyler podcast and we'll revisit some of those questions, but I had no idea it was going to be a podcast So it was having Peter into George mMason and people would get to hear him and meet him And this was before he was super well known also before he was mega controversial It was like there were people who didn't know who Peter Thiel was at that time.

So I thought this will be nice and then I was shocked when he said he would come And then we just kept on doing them and there you go.

Ollie: I was running a business for 10 years up until January, 2020. And then I. I left that company, my business partner bought me out, which actually was great timing just before COVID actually in hindsight. And, I didn't really know what I was going to do at that point. So I started writing. That was the first time I really started writing regularly. I started blogging and that was in February, 2020. And I was just exploring my relationship with work. So as I said to you before, we started recording three young kids and I'd burnt out several times, probably spent a bit too much time at work. Sometimes, I just felt like I wasn't doing enough at home. I thought I'd just. Rethink how I did it. So I started writing about that and then it was around the time COVID hit and lots of people were interested in it and they signed up to my newsletter, which I started seeing after, and I had enough people saying, you should do a podcast. I did what you just suggested, which just, I just did it, , recorded a few. I got the bug for it really just enjoyed meeting really interesting people asking the questions. I wanted to ask them , and you know, that's the spirit. That's usually the advice I give by the way if people say should I start a podcast?

I say well, is there anybody you'd like to ask interesting questions to in which case? Yes.

Tyler: Strong agree?

Makes it better. You become too mannered and you're like trying to engage with them as if that's a meaningful notion.

Ollie: And actually it's funny because I think when I've, when I've not got it right, it's because I've almost been thinking too much about what I think other people want to hear. So the people listening, you know, what, what do I think other people want to hear? And it's far better when I just write a bunch of questions down or questions come to me, which I'm just intrigued by a curiosity.

It's a classic, you think, right? , I'm interested you, how, whether your podcasts improved your writing.

Tyler: Not that I know of. So I've learned a lot. And then that knowledge has improved my writing. If you count that, yes. But I don't think doing a podcast has improved my writing. It may have made it more conversational. Which could be plus or minus.

Ollie: Yeah. How about being an economist? Did that help your evolution into a podcaster? Yeah.

Tyler: Why do you do it this way? What are you optimizing?

Know, what are your trade offs? So I find that very useful, yes.

Ollie: Yeah, I, I did a little in the spirit of your book, which I'm sure we can talk about a bit later. I've been getting chat GPT to help me prepare for this conversation. And, , I asked it to estimate how many words you've written on marginal revolution, which your blog, a blog that you've been writing for 20 years, it reckons about 6.

5 million words. So, Clearly for you, it's a habit, which probably by this point, pretty difficult to break. It would probably seem strange to you if you didn't write every day, but what would you say, to people who do struggle to write?

, is it something that everybody can do? Or is that one of those things that's particular to individuals?

Tyler: I think it's particular. So if someone is already at the point of asking me how to write, My best advice is often just stop, don't be trying to write.

You know, it's not that you have to be a writer. Writers on average are not paid well. The few who are paid well often just love writing and always wanted to be writers.

Who's making you write? Now maybe you need to write memos at work, then that's a different consideration. But that's not usually the problem people have who can't bring themselves to write. It's writing for themselves. And they have this weird idea in their head.

Tyler: Do something better.

Ollie: Yeah, do you write for yourself?

Tyler: Yes, absolutely. And marginal revolution, I'm not too much into the audience. It's working out problems, things I'm thinking about.

Ollie: Yeah, said before, I hadn't really written very much in years apart from in emails and I have found. Well, it has certainly helped me clarify my thinking about things, but it's, it's also I found it quite the reflective part of the process of writing very useful as well. Actually, I think in just this speed of life and as life sort of flies by, I find it both useful to sit down and take that time to reflect, but also looking back, it's easier to identify moments in time by. The writing that I've done, you know, and admittedly I've only been doing it for four years. , but it's interesting when I look at the way I was thinking and, you know, not that much has changed in my points of view or opinions about things, but I can kind of pin, I can remember my, that moment in time and it's actually improved my memory.

Tyler: Perhaps four months ago, I would have thought that Iran would take on a more aggressive posture in the Middle East than has turned out to be the case. So I'm modestly more optimistic about that part of the world.

I know what is happening is terrible, and it's a terrible price to pay. But so far, let's hope these words hold, what we're seeing about the underlying equilibrium is the countries in the region mostly don't want to fight.

And I thought on, say, October 7th, that it had a much bigger chance of turning into something much larger.

And so far, we're not seeing that. So that's been a very recent update I've made. But, you know, if you write on a lot of issues every day, you're making updates every day. So feel free to pick an issue.

And I'll tell you what I think.

Countries went better than I was expecting. That's an update. I'm not sure which broader update to make from that, but I know I need to make a bunch of updates from that one.

Ollie: Bringing it back to a subject I talk on this podcast and about jobs and AI's effect on jobs. I mean, what's your position historically been on it's perhaps say five year effect on. Knowledge work, let's be relatively specific. What's your thinking being? And so has that shifted at all in the past 12 months?

Tyler: Most issues, I revise. But on that issue, I feel my earlier views are being confirmed, which is that there'll be a few fields. Where job replacement is very rapid.

Be an obvious example. Most areas, people will respond quite slowly, and there'll be a lot of time to adjust.

Sides. And I think there's no reason why we can't stay at full employment, but in the longer run, there'll be more people working in the energy sector. As carpenters, as gardeners, as greeters at restaurants, and many jobs will go away, so, but I think we'll stay at full employment.

Ollie: Yeah, who stands to gain the most from AI, do you think? I mean, what jobs will be created, for example?

Tyler: There's a small number of people who take a lot of initiative and who can manage projects very well, and they now have a near infinite supply of what you might call research assistants, and they'll just be able to juggle a lot more balls and have all this free quote unquote labor, and they will earn phenomenal sums of money.

And then people who do things like smart carpentry. So we're going to build much more. There'll be more planning, more projects, more programming, more of everything. So if you're a physical builder, but not just brute force labor, you actually think how to put things together in a way that's not easily automated.

I think those people, they're not going to become truly rich, but it will be common for say a very good carpenter to earn 170 K a year.

Ollie: You mean aided by augmented by AI, the people working physical jobs, but in some way. , their, their work is, is supported by AI. How do you envisage that?

But then the actual work has to be done, and it will be complex rather than just lifting heavy things, and if you call those people generally carpenters, they're in for great times.

It's not an easy job.

Ollie: No, I, they're the sorts of jobs which I admire because I'm just completely useless at them. , I can barely put up a shelf.

Tyler: Here, like I couldn't put together our ping pong table.

We had immigrants, so called unskilled immigrants, come over and do it. And I just think there's going to be a lot more of that.

Ollie: Yeah, absolutely. Move any of the pictures that you see behind me and there'll be about five holes where I try to get it right the first time and failed.

So what we've seen interesting over the past few years, and I've seen this myself, so I've been running companies, , and at changing the way that I've built teams.

So, whereas before with tech talent we might've hired locally, gradually you began outsourceing part of that. And certainly over the past few years, as teams have become more distributed for many organizations, outsourcing skilled labor, you know, knowledge work to different parts of the world has clearly increased. Do you see AI reducing that increase in outsourcing to emerging economies? And I wonder what the positive and negative implications might be if that's the case.

Tyler: I'm not sure of the net effect, but AI smarts are like outsourced smarts. They just happen to be in your lap, so to speak.

My intuition is that there'll just be a lot more quite concrete projects. We'll have a much higher demand for energy. So the notion of people getting together in the same physical space to build something will make a major comeback.

In your country, if your regulators allow it, we have more federalism. I'm sure it will happen here.

So that will diminish in percentage terms work from home because you'll have to be there on the lot.

Ollie: Yeah.

Tyler: That's my intuition. I don't think that's close to established or proven.

Ollie: I'd like to come back to this regulation question actually a bit later on because, , there's been some pretty big cases recently in which the UK regulator has played quite a big role in obstructing, , acquisition. Certainly the Adobe and Figma one is the most, the most recent. Let me, let me save that one though.

Let me just ask a question about, Keynes. John Maynard Keynes , because you've. Recently written a book, and it's appropriate, we're talking about it alongside AI because you, you wrote this alongside chat PT and it's called goats. It talks about the, the determining whether you can identify the best, the greatest economist of all time. But it reminded me in the context of this conversation about keynes's prediction of a 15 hour work week. I mean, could AI be a factor that pushes us towards that? Because right now it doesn't. Appear likely that we're going to get close. Four day weeks, about as close as it appears that we're likely to become, , to be getting to anytime soon.

Tyler: Who want 20 to 25 hour work weeks may have that as an option. 15 seems like quite a low number. To me. There is a fixed cost of building a team, starting with the company, developing relationships. So Keynes was wrong. And I don't think most people want as little as, say, 20 hours of work a week.

So he'll, he'll mostly be wrong, but there's plenty of slackers now, right? In both of our countries, they get by different ways, family support, spousal, partner, maybe black or gray market. They sell some pot, who knows what they do. , that's already a thing. But it's not that fun.

I don't think those people are envied so much.

Ollie: Yeah. Well, what's your take on universal basic income because it often comes into the conversation around, around now when we're talking about this future in which, , you know, perhaps we have requirements for less paid work, the idea that, that there's a basic income prepared to every individual. I mean, do you see that as a attractive proposition, realistic proposition?

Tyler: That's another view revision I've had, so I used to entertain the idea. But witnessing life under COVID, people really want to get out and work. Now, the quality of jobs is much too low, but I think we can improve that, and that's a more fruitful path than giving everyone money not to work.

Ollie: I'm not interested about your journey in writing the book. Perhaps you'd explain in the relationship with track GBT in, in the context of goats before I ask you a few questions about it, how did, how did you decide to do it? What did it actually look like in practice and what are your reflections on it?

Post publication.

Tyler: Sure, let me just tell all the listeners, my book is free and online. If you just google Tyler Cowen Goat. You get right to it. The site is econgoat. Ai. The entire book is written by me.

Not a single word by AI. Not even edited by AI. But what I did was I took the book and loaded it into an app so you can interrogate the app about the book in any way you want.

If you want a summary of Chapter 4, you don't want to read Chapter 4, you can do that. You can ask AI to criticize the book. You can ask AI, what was the most interesting point in Chapter 3? You can really ask it anything and the service gives you that. So I thought this is an innovation in book publishing when I wrote the book none of this existed It was not in the back of my mind And then one day I woke up and it's like hey I've got to put this, you know, in AI and have it be what is, in fact, the first full length book published in GPT 4.

Ollie: Yeah.

Tyler: Pele is Adam Smith, right?

Tyler: You know, it's funny you bring this up. Im about to record a podcast in which we compare the greatest basketball players to the GOAT economists. Which I know much better than, you know, football for you, soccer for us. And we're going to pursue that line of approach.

Soccer, I could only tell you a few people, period.

Ollie: Well, I mentioned Keynes before. Apparently Lionel Messi is the Keynes of football. It says with his revolutionary ideas during the Great Depression, Keynes changed how governments interact with the economy, much like Messi's impact in modern football through his unique playing style and numerous records.

So there we go. Great, great minds. Although I've never, never been sure about that great minds quote. I think I'm not sure if it was you or someone else I heard talk about the idea of great minds think alike being a, any other inadequate representation of why we sometimes have similar thoughts to others.

Tyler: Talen said it. Great minds always, or usually disagree.

Ollie: That's it. That's who said it.

Tyler: I'd like to see the data on that one.

Ollie: Yeah, exactly. There we go. , so yeah, so there we go. That was my little addition. And actually, there's a good example about how interactivity can actually help people connect with something which wouldn't otherwise be possible for them, like my 11 year old, but I'm interested for us non economists. Why should we care about the debate or a debate about who the greatest economist is?

Tyler: It's a way of understanding economics, a way of understanding our history, a way of learning how to judge talent. Like, how good are we at assessing these people? What are their strengths and weaknesses? So I view it as another laboratory for testing talent judging abilities. And as you may know, my previous book with Daniel Gross, it's just called Talent.

It's about judging talent. This is like an applied case study.

Ollie: Yeah why is it not possible to clearly identify one single goat?

Tyler: Excellence is multidimensional. We're comparing people over, you know, basically two centuries worth of time, so the standards evolve. It's interesting, if you ask GPT 4 repeatedly, who does Tyler say is goat, each time it gives you a different answer. It'll say Smith, it'll say Keynes, it'll say Mill. And that's reflecting the wisdom of GPT 4.

It's not a malfunction. I can tell you my favorite candidate is Mel. I think the simplest winner is Adam Smith, if only because he was first. But I don't think he's the obviously correct choice.

Right?

Tyler: None of these people were, and that, to me, is quite interesting.

If you want to be in goat contention, you have to do more than just economics.

Ollie:. Well, just to bring it back to the football players or soccer players. I mean, everybody tries to debate all the time. Who's better right now between Messi and Ronaldo and it's impossible because they have different skill sets, different characteristics, character. , so to do it across generations is impossible.

So it's a, it's, it's good to have the debate.

Tyler: Who does GPT say is better, Messi or Ronaldo?

Ollie: That I don’t know.

Tyler: You've never tried?

Ollie: I've never tried. I'll ask my, , my 11 year old, definitely would have tried that. I mean, there are literally whole YouTube channels about this question. I've watched many, recreations of Messi versus Ronaldo done in different ways, but I've sadly not asked GPT, but I will straight after this call.

Now, you're a prolific writer, but I know you're, you're a massive music fan. You're a prolific music listener. Do you combine the two? When you write, do you listen to music or do you write in silence?

Tyler: Usually I write in silence. The music doesn't bother me, but I mostly stop hearing it.

And that I do absorb.

Ollie: And what sort of music? I mean, can you, somebody can read while there's lyrics in the music? Or is that a distraction?

Tyler: It's a distraction.

Instance. , but you'd typically classical music. Now, some of that has lyrics, but not in a way that's distracting me.

If there's a Bach motet or choral piece by William Byrd, it does not get in my way because I'm not really listening to the words.

It's not like Strawberry Fields forever, right?

Ollie: Yeah.

Tyler: A lot of jazz, classical music, some world music, and rock and roll, I save for the car.

Ollie: Nice.

I'm interested in whether you learned something from writing the book you've just written, but something which isn't to do with either economics or AI, do you typically learn something about yourself or about other subjects when you're writing a book about a specific, specific topic?

Tyler: I've learned I always like to be doing new things. I'm very pleased by that, and I want to keep on doing new and different things with the manuscripts I write.

Information we have to deal with.

And the current version, it's a little slow. In some ways, it's a bit clunky. But in less than two years, it will be much, much better. And you might just wake up in the morning, not look at Twitter, and just say, you know, GPT, tell me what I should know.

Ollie: Yeah.

Tyler: And you might get images, you might get text, you might get spoken, you'll be able to choose.

And it will just handle that for you. And that's not far away.

Ollie: It is remarkable when you think about new technologies. There's many different ways I suppose you can judge the impact of a new technology, but for me personally, I suppose it, I could, the easy judge is how much time do I spend with it? And I think I spend a worryingly, a, a large amount of time with GPT nowadays.

I mean, I'm, I'm just starting a new company at the moment and. I work with teams based in different locations around the world. So in my previous business, I worked in close proximity with someone. I saw him in the office all the time. I use GPT like a co founder. I mean, I'm asking them to pull apart. I've got a lot of ideas that I've got and propose things.

And like you say, you know, they're never perfect. And you know, they're, they're not necessarily particularly well articulated sometimes, but it's really bloody useful, you know, just to have somebody there to constantly throw ideas at. And I, I don't actually know if you can, it can tell you how much time you spend on it, but I suspect it's, you know, it's going to be at least an hour or two a day already a year in, it's remarkable really.

Tyler: If I'm not traveling, I use it every day, and sometimes when I travel. But I find the distribution is evolving bimodally. There's plenty of people who try it and then never use it at all. And then there's people like the two of us, and not that much in between.

So I worry about those other people. I guess they must worry about us.

Tyler: But I would double down my bets.

Ollie:

I saw my brother last week and he's got the voice activated GPT on his phone and he was showing the kids the conversations he was having it while we're having our family Christmas meal last Saturday.

So there we go. We're already, voice is already becoming an important part of that interaction as well.

Tyler: I think AI girlfriends and boyfriends, on average, will be good for us. That it will draw people out of their shells a bit, give them practice, make them feel a bit safer.

Some people will get addicted in very bad ways, but I'm expecting it to be a net plus.

Ollie: Is that a business? Idea potentially that the essentially it's a date trainer So that when you go out in the real world, you've already learned some of those social skills That can help you interact better.

Tyler: Yeah, people are doing this already. It's not just potentially. There's demos out there, at least a dozen startups trying to do this. Obviously, they mostly won't make it, but some of them will.

Ollie: Interesting. I will look that up. I mean not i'm happily married, but you know We can always improve can't we?

Tyler: Well, the AI could teach you how to be nicer to your wife.

Tyler: So if men and women have different ways of communicating, it's a safe space where you can practice whatever methods of communicating your partner might want from you more without having to offend them all the time. And that may be useful as well.

It's interesting that probably most people don't want to practice that though, right?

Ollie: No, yeah, exactly. That's uncomfortable, isn't it? Well, here's the thing, right? If GPT could speak to my wife and say, look, what do you really want from Molly? You know, what are the things that, maybe not tick in the boxes right now.

Maybe that's, a net positive for both of us. But, , I think she might find that a little weird and maybe she's right to find that a little weird. We need to soften that somehow before it's palatable for most normal people.

Tyler: But your children won't find it weird, or your grandchildren.

Ollie: Yes, exactly. , it sounds like, I mean, I know clearly you\'re someone who just relishes learning. Do you learn something new every day?

I could say. I would feel like a total failure. I hope I'm learning something every moment. Every minute, every, at least every 15 minutes at the worst. That's a hard pace, but I think that should be the aspiration. When I walk the dog, I don't learn that much.

I did it first, about the dog, but that's asymptoted.

So I don't actually like walking the dog for that reason. Though I like the dog, I prefer to let the dog out in the back. But we have coyotes, so that's a problem too.

Ollie: Okay. Would you listen to music when you walk the dog or do you leave yourself alone? In silence?

Tyler: I don't like buds and pods. I think it's the wrong way to listen to music, and one should listen at some distance.

So I don't listen when I walk the dog. And furthermore, you're in the suburbs. You want to be able to hear cars or other things. I don't know if it's dangerous, but I don't somehow philosophically like walling myself off from the sounds of the world.

I don't do it.

Tyler: Oh, of course. The last concert I saw was about ten days ago in Buenos Aires. There was a performance of Handel's Messiah at the Teatro Colón, and that was incredible. , my wife and I, we have, like, tickets for a whole bunch of concerts coming up. And, , we go all the time. It depends who's in town.

So I never go just for the sake of it. , but we're most likely to see classical music.

Ollie: Right. Interesting. I know you, , on Rick Rubin's podcast recently.

And I’m sharing a lot about what's happening in my home at the moment, but it's been the holidays. So we're spending a lot of time with me and the family, but I had your DJ mix on in the kitchen yesterday while I was making dinner. , the second half of it went down better than the first half with my wife and kids.

Tyler: That I can understand.

Ollie: Although I enjoyed it. I enjoyed it. There was a much. I learned actually some new artists, some new music. It was really great. What's your relationship with music and, and learning? I mean, do you consider learning and searching for new music part of that process as well?

Tyler: No, it's always new. Maybe, on average, I buy a CD every two days. I don't prefer streaming. I still like some kind of physical product. The house is overflowing with stuff. And there's always, there's a lot new coming out, especially end of the year lists.

I'll buy heavily from those and just sample things.

So like the new Lana Del Rey, , release this year is excellent. I've been listening to that lately. Been listening to a lot of Renaissance music. , all sorts of odds and ends. I've been making some playlists for Rick Rubin.

That he will use for a project of his. The last one I made was a playlist of transcriptions only.

And a lot of transcriptions. I knew I had to find what, what's the best version of that on streaming.

Ollie: Nice.

Tyler: Like. What instruments are good for transcription? So clearly piano can do a lot. Guitar can do a lot. Harp and accordion are somewhat underrated for transcriptions.

They're imperfect, but you, you can do some things with them.

Ollie: Interesting. It's funny. I go on the album of the year list actually. If you've done, if you've been on that website album of the year, which breaks it down into different genres and it is always useful at this time of year because you have a pretty, you know, by this point there are, there's enough of a sample size to know that top 10 is going to be pretty good, you know, and, , I, but particularly over the holidays, I tend to go back to the old favourites, just some safe ground and break, breaking up the, relentless Christmas music on the radio. So I have to just have some stuff which can, allow the family to settle into that as well.

I think you said on the Rubin podcast, you get the Beatles radio on. Occasionally. And that's always a safe bet as well.

Tyler: That's right. Do you have satellite radio in the UK?

Ollie: It's not the same. It's not the same really. I mean, you can listen to all the American satellite radio shows easily enough. , but yeah, it's not, it's not quite the same as in the States, classic gold was the first station which just played all the old hits.

And now that's segmented into different types of gold, you know, the sixties or, you know, you know, bands and these types of things, but it's not quite as specific as satellite radio.

Tyler: Interesting.

Ollie: Yeah.

Tyler: And I guess you drive less.

Being an American who lives in suburbia, I'm driving all the time.

Ollie: Yeah. I've still got turntables and I've got records in different rooms and I do listen, I've got my Sonos set up in different parts of the house. But when I'm out and about the only way is with headphones on really, cause, I do. Sadiq Khan our man's made it very difficult for us to drive in London.

Tyler: Yes.

Ollie: Well, interesting to talk to you about music. I do have a few questions about talent and jobs though, if you don't mind. So, formal credentials, more or less important in the work of the near future.

Tyler: They're becoming less important, and I think AI will accelerate this. So the mere ability to manipulate words and ideas is where large language models are strong. And so often that's what credentials are measuring.

Credentials that maybe measure your determination or your abilities to build things or manage other people.

But current credentials are, are right now overvalued and due for a pretty steep tumble.

Ollie: And related to that, I mean, to what extent is self reported career and personality data useful for identifying talent?

Tyler: It's better than skeptics think. So if you ask people, are you extroverted, are you introverted, and so on, the answers are not terrible. I'm not sure we'll ever have anything that much better short of taping people at great length and applying machine learning.

And what's their determination, those are really tough to measure in any case.

So, progress on those is going to be slow.

Ollie: Relating AI to talent. I mean, I work in a world in which it's popular to try to apply AI to any problem. So for example, talent acquisition. AI is just, it's part of the value proposition for almost every business that I'm seeing right now. But I mean, what are the limitations when using AI to identify talent?

Because, you know, again, perhaps we must be overblowing some of the potential for them, certainly in the short term about how useful AI is in actually identifying whether someone is the right person to join our company.

Tyler: I don't think it's that useful at the moment. Now, if it's a mechanical exercise, like who can read through 2, 000 resumes and sort out who's finished college and who hasn't.

AI is awesome at that, the mechanical side of it. But the judgment side, I don't think it's useful yet. Or if it is, it's something proprietary and secret that I haven't seen.

But for kind of rote labor, it's just way, way quicker, and it does it immediately.

Ollie: Can appreciating what's uneven about someone help understand whether they\'re a fit for your team?

Tyler: Well, say you're choosing a co founder, as I know you've done at least once. You have to be honest and ask, why does this person want me?

Ollie:.

Tyler: And if you don't have a good answer to that question, I start to get worried. It means there's things about them you don't understand. And the chance that those things are bad rather than good is pretty high.

And if your answer is just, well, I'm the best person in the world, and the co founder is the second best person in the world, that's great when it's true. But again, you can't count on that. If you have a complex answer involving their strengths and weaknesses, and maybe why they want to live in London, and why you're the right match for them, Then you can start to feel pretty happy that you understand the match.

And the, the honest answer is going to be uneven in almost every case. So if you can only think of good reasons why they want to work with you, you're missing the boat. Start worrying.

Ollie: What about, you know, with limitations, so everyone's got their limitations or they've got their strong points and they've got their weaker points. In your experience, I mean, do you just become more aware of these as you get older? Or do people in general become more aware of those and more open to revealing them?

So I think about myself personally, I think when I was 25, I just had huge blind spots about those areas that I was weaker at. Whereas if you were to ask me now, I would just tell you straight out what they\'re. And I think actually in many cases in that example you gave there, it does help people understand far quicker and far easier whether I\'m the right person to work with them.

Does that just come with age or just are some people better at revealing that than others?

Tyler: I think for people high in openness it comes with age. But I also observe another skewed part of the distribution where people, even at pretty young ages, they just ossify.

And they have their fixed stories. They're not going to budge from them. They still might be quite productive like oh i'm a great policeman, right?

But they're not self critical about how they might get better. But maybe they\'re courageous and experienced whatever and that's that so It cuts both ways age matters But some people it makes their narratives more closed.

Ollie: Have you pushed yourself beyond your comfort zone? I look at the guests that come on your podcast, and I know you\'re someone who has breadth, I suppose, in terms of your, your knowledge, but are you purposefully pushing yourself out of your comfort zone in the guests that you bring onto the podcast and the things you write about?

Tyler: I try to pick people.

Ollie: Yeah, I'm actually reading a book at the moment, which somebody recommended to me called the comfort crisis. , and it's a study in how we have through modern technology and, you know, just through improvements in our standards of living. Become a society, which is incredibly comfortable. And actually it's to our detriment.

And he literally goes to Alaska and that's how he starts the story by putting himself out of his comfort zone. So, you know, that, that scenario you laid out is, is a, is a real thing, which someone's written a book about.

Tyler: But I see more and more people they use the extreme event as some kind of placebo.

In terms of what you would voluntarily arrange, I don't see much of people going outside their comfort zones.

Ollie: Let's talk about podcasting. I mean, if I were to weigh too deeply into economics, I suppose I would be going out of my comfort zone. You know, I'm I and we rewind three years or so when I started this podcast. I think I'd have been petrified. You know, having a conversation with you, Tyler, honestly, because I think, you know, you're, this is, you're, you write about subjects which I don't have a deep understanding, but I can now have the conversation because I've built the skills it takes to interview someone on a podcast. But I suppose where I'm trying to push myself is to delve into some of those areas in which I maybe don't have as much knowledge or much experience, but I'm trying to understand more. And I think that's, In some way, what I'm doing personally, and I get the sense that you're perhaps trying to do that with your questions as well.

It must be that you\'re pushing yourself to the limit and trying to extract insights and interesting conversation from those that you interview.

Tyler: But I think people are rarely at their limits. I feel I could do more. I hope to do more. , I don't ever feel right now that I'm at my limits. And that's my fault.

Ollie: Yeah. All right. I mean, one other thing I think I've heard you say, which I'd love to just mention as well is the value, not just of reading opinions that you don't necessarily agree with, but actually writing out contrary points of view.

Better when you have to state them. So steelmanning, it's called sometimes. And most people are bad at that.

Ollie: And do you, you try to do that regularly? That's just part of, again, your thought process in trying to, , improve your learning.

Tyler: That's right, and if you end up disagreeing with them, you'll have a much better sense of why. And sometimes you get converted.

Ollie: Yeah. One more question just related to work. Cause, , you know, we're nearing the end of our conversation, but, , Nick Bloom, Stanford professor, , remote work expert has just been on the podcast and he's been studying hybrid and remote work extensively and his conclusions in short, I mean, it goes in a lot more detail, but in short, , the hybrid work has no. Discernible impact on productivity and there's a roughly 10 percent reduction in productivity for fully remote work, but that's largely offset by cost savings for a lot of businesses because they don't have to have a fixed location. So I find that interesting, but what I'm more interested in is what we're missing from this by just thinking about productivity as it relates to our business strategy and decisions about. Workforce and particularly leading and developing talent.

Tyler: I like Nick and his work a great deal, but I disagree with him on that point. One reason of having face to face work is you can spot who amongst your interns, say, will someday be your director, or COO, or whatever. And you don't get that remote. And that can be a 10 to 15 year process, or maybe longer. And they imbibe more of the culture.

You realize they're the ones to invest in. They get promoted more rapidly. So it's for the very promising 23 year old, 15 years later, where I think you're taking the big productivity hit. Now, it's fair to say we're not sure. We can't measure that yet. But that's my view. And that's why I don't agree with him.

Ollie: Yeah, it's tricky when I've, I obviously see it from different sides. I've spent most of my working life and certainly around my last company. Almost exclusively working in an office. I mean, it was hybrid in a sense, I suppose, because we had flexible work, but the last few years, for obvious reasons, my working life has been completely shifted and now I spent, I'd say 90 percent of my time working with people in other countries, let alone in other places in the UK.

So I just am forced to do it. And it's definitely changed the way I connect with others, and it's definitely changed the interactions that I have. And I suppose it's more difficult to spot whether they have that thing, which is sometimes difficult to define. It's a feeling that you get from spending time with people every day.

But I must be honest, I've not landed fully on one side or the other. , so I find it interesting to see that data, , certainly at scale. It's an interesting, it's interesting to see as companies make decisions about return to office or not.

Tyler: That they can develop face to face.

Ollie: Yeah, very early in this conversation, we talked about regulation.

I do want to just go back to that quickly. , does over regulation threaten the UK's role as an innovation center, as a location for innovative businesses?

Tyler: Yes, especially near Cambridge. You need to build a lot more. You can do it. The land is there. Now, near Cambridge is pretty, but it's not that you're, say, destroying Stonehenge or the Lake District. A lot more startups want to be there, a lot more businesses. There are various parts of your country where it seems clear to me it's way too hard to build, and I hope if Labour wins, they remedy that.

The Tories seem to have too many older voters. I don't see the prospects there, even though Tory elites, when you speak to them, they understand this problem very well.

Ollie: What about Brexit? Do you think that's going to prove to be a smart decision? Long term?

Tyler: You know, when it happened, I was against it. I was concerned. I still think it was a mistake. But with the passage of some time, I've increasingly seen it as inevitable. And if you read History of the Reformation, Henry VIII, and all that, it just was never going to work out. And I've made my peace with Brexit.

By being different from Europe. And I don't think you've done that yet. You've started a bit with AI, regulating AI less than the EU is likely to be doing. So you need to make the best of it. And I don't think it's going to be undone. So I'm unhappy with both sides of the Brexit debate.

Tyler: Just different than them.

You're so different.

Tyler: Referenda are a bad idea, by the way. It was a big mistake how it was done. All the Brexit means Brexit. How do you see it through? No one ever knew what the deal actually had to look like. The Ireland's like a huge mess.

, many things were badly handled, I think, no matter what your point of view.

Ollie: Yeah, well, we've had a series of short term snap decisions over the past, I don't know, five to 10 years as, , the government sought to cling on to power, which often usually happens. It happened with labor at the end as well, and it's not leading to, , great prospects for us in the short to medium term.

New year.

Tyler: Optimistic about England, I have to say. But not like in the next year.

Ollie: Yeah, exactly. So one last question for you. I'm not suggesting anything by this question, by the way. But what does retirement mean to you and how do you see it changing in the U. S. And UK over the next say 20, 25 years?

Tyler: Well, to me personally, it's a, you know, synonymous with death. That will come.

But I don't want to retire before then.

Jobs goes up , people are wealthier, but they retire they tend to die more quickly They lose their social networks like what's the upside?

So I hope the notion of retirement can be partly ushered into a gentle retirement That said there are people with health problems disabilities who really cannot work in rewarding ways And retirement in some manner will always be with us.

Ollie: Yeah. Well, Tyler, I really appreciate your time. Thank you for speaking with me today. Anything you want to leave us with before we wrap up?

Tyler: Well on the future of England. I often visit and everyone’s doom and gloom But you look at your assets, your best cities, your best universities, your cultural heritage, the english language Everything from the Beatles to James Bond to Monty Python to whatever Uh, I think there are a few parts of the world that are going to do better over the medium term Not the short term

Ollie: , well, I'm, I'm finishing this podcast with a spring in my step after that. So thank you.

Ollie: Thanks Tyler.

So thanks very much to Tyler for joining me again. Thanks for listening. You'll find links to Tyler's podcast, books and blog in the show notes. And I'll see you again next week.