AMA: career advice given AGI, how I research ft. Sholto & Trenton

Dwarkesh Patel 0:00 Today, this is going to be an Ask Me Anything episode. I'm joined with my friends, Trenton Brickin and Sholto Douglas. You guys do some AI stuff, right? (laughs)

Trenton Bricken 0:08 Yeah. A dabble. (laughs)

Dwarkesh Patel 0:12 (laughs) Um, they're researchers at Anthropic. Other news, I have a book launching today, it's called The Scaling Era.

Trenton Bricken 0:17 Yeah.

Dwarkesh Patel 0:17 Um, I hope one of the questions ends up being, why you should buy this book. (laughs)

Trenton Bricken 0:23 (laughs)

Dwarkesh Patel 0:23 But we can kill two birds with one stone. But, um, okay, let's just get at it. What's the first question that we gotta answer?

Trenton Bricken 0:28 Take us away.

Sholto Douglas 0:29 So, I wanna ask the flyball question that I heard before of, why should ordinary people care about this book? Like, like, why should my mom buy and read the book?

Dwarkesh Patel 0:38 Yeah. First, let me tell you about the book, what it is. So, you know, these last few years, I've been interviewing AI lab CEOs, researchers, people like you guys obviously, but also scholars from all kinds of different fields, economists, philosophers. And they've been addressing I think what are basically the gnarliest, most interesting, most important questions we've ever had to ask ourselves. Like, what is the fundamental nature of intelligence? What will happen when we have billions of extra workers? How do we eco- how do we model out the economics of that? Um, how do we think about an intelligence that is greater than the rest of humanity combined? Is that even a coherent concept? And so, what- what I'm super delighted with is that with Stripe Press, we made this book where we compiled and curated the best, most insightful snippets across all these interviews. And you can read Dario addressing, why does scaling work? And then on the sa- next page is Demis explaining, um, DeepMind's plans for whether they're gonna go with the RL route and how much of the AlphaZero stuff will play into the next generation of LLMs. And on the next pages is of course, you guys going through the technical details of how these models work. Um, and then there's so many different fields that are implicated. I mean, I feel like AI is one of the most multi-disciplinary fields that one can imagine because there's no field, no domain of human knowledge that is not relevant to understanding what a future society of different kinds of beings will look like. Um, there's, you know, you're gonna have like Carl Schulman talk about how the scaling hypothesis shows up in primate brain scaling from chimpanzees to humans. On the next page might be an economist trying to argue with like Tyler Cowen, explaining why he doesn't expect explosive economic growth, and why the bottlenecks will eat all that up. Um, anyways, so that's why your mom should buy this book. It just like, it is the distillation of all these different fields of human knowledge applied to the most important questions that humanity is facing right now.

Sholto Douglas 2:40 I- I do like how the book is sliced up by different topics-

Dwarkesh Patel 2:43 Yes.

Sholto Douglas 2:43 ... and across interviews.

Dwarkesh Patel 2:45 Yeah. Yeah.

Sholto Douglas 2:45 So, it- it does seem like a nice way to listen to all of the interviews-

Dwarkesh Patel 2:49 Yeah.

Sholto Douglas 2:50 ... in- in one digestible way.

Dwarkesh Patel 2:52 Yeah. There's, um, two interviews I've done that haven't been released publicly before that are in the book.

Sholto Douglas 2:58 Mm-hmm.

Dwarkesh Patel 2:58 So, one was that Jared Kaplan, um, uh, who's one of your co-founders, and this is another example where it's like th- he's like a physicist and he's explaining scaling from this like very mathematical perspective about data manifolds. Um, uh, and then on the same next page, you have like, a totally different perspective. It's like Goren talking about, you know, what- why- why- why can't we just have like distilled int- like, why did general intelligence actually evolve in the first place? What is the actual evolutionary purpose of it? Um, and it's like, page by page, right, you can just get addresses. Even for me, I mean, like the person who's been on the other end of these conversations, it was actually really cool to, um, like read it and just be like, "Oh, I actually like, I, now I realize how these insights connect to each other."

Sholto Douglas 3:36 Yeah. The, the only other thing that stood out to me as well is the introduction section-

Dwarkesh Patel 3:39 The only thing that stood out to you? (laughs)

Trenton Bricken 3:42 (laughs)

Sholto Douglas 3:42 Like, like-

Dwarkesh Patel 3:42 Shh.

Trenton Bricken 3:42 (laughs)

Sholto Douglas 3:42 Yeah, yeah, that was, that was really the only thing that was noteworthy.

Dwarkesh Patel 3:45 (laughs)

Sholto Douglas 3:45 Um, I just mean in ter- it stood out in accessibility.

Dwarkesh Patel 3:49 Yeah.

Sholto Douglas 3:50 Is the introduction section and the diagrams for like, all the different inputs that-

Dwarkesh Patel 3:55 Yeah.

Sholto Douglas 3:55 ... enable you to train a machine learning model.

Dwarkesh Patel 3:58 Yeah.

Sholto Douglas 3:58 Stripe Press books are also just beautiful-

Dwarkesh Patel 4:00 Yes.

Sholto Douglas 4:00 ... and have these nice like, side captions for explaining-

Dwarkesh Patel 4:02 Yep. Yep. Yep. Yep.

Sholto Douglas 4:02 ... what parameters are, what a model is, these sorts of things.

Dwarkesh Patel 4:05 Yeah. Actually, when we did our episode together, a bunch of people, I don't know if you saw this, independently made these like, blog posts and Anki cards and shit-

Sholto Douglas 4:15 Yeah. Yeah.

Dwarkesh Patel 4:15 ... where they were like, explaining the con- 'cause we just kind of passed over some things. Um, and hopefully we've given a similar treatment to every single interview I've done, where you can read a very technical interview with a lab CEO or something, or an engineer or researcher. And then the side is like, here are, here- here is like more context, here's more definitions, here's more commentary. Um, and I, yeah, feel like it elevated the conversations.

Trenton Bricken 4:37 So, in other words, my parents will finally understand what I do for a job.

Dwarkesh Patel 4:40 (laughs)

Trenton Bricken 4:41 (laughs) Well, they do. They go, "Very well."

Dwarkesh Patel 4:43 Yeah, yeah, yeah.

Sholto Douglas 4:44 (laughs)

Dwarkesh Patel 4:44 Maybe my parents will also ƒ (1:00:00) .

Trenton Bricken 4:47 (laughs)

Sholto Douglas 4:47 (laughs)

Trenton Bricken 4:47 (laughs)

Sholto Douglas 4:48 All I need to know is that my name's in a book.

Dwarkesh Patel 4:50 Yeah.

Trenton Bricken 4:50 Yes. (laughs)

Dwarkesh Patel 4:51 (laughs)

Trenton Bricken 4:51 You're a co-author. (laughs)

Sholto Douglas 4:53 They're like, "Cool." (laughs)

Dwarkesh Patel 4:55 (laughs)

Trenton Bricken 4:55 Should we get into the AMA questions?

Dwarkesh Patel 4:56 Let's

Dwarkesh Patel 4:57 do it. Let's do it.

Trenton Bricken 4:57 All right. Uh, so Brian Krav asks, "The issue you raised with Dario, and occasionally tweet about relating to models not making connections across disparate topics, some sort of combinatorial attention challenge. What are your thoughts on that now? Do you solve it with scale, thinking models or something else?"

Dwarkesh Patel 5:11 By the way, so the issue is, one of the questions I asked Dario is, look, these models have all of human knowledge memorized. And you would think if a human had this much stuff memorized, um, and they're moderately intelligent, they couldn't be making all these connections between different fields. And there are examples of humans doing this, by the way. Um, there's f- Donald Swann or something like this. This guy noticed that the- the way, what happens to a brain after magnesium deficiency is exactly, uh, the kinds of, I don't know, structure you see during a migraine. So, then he's like, "Okay, you take magnesium supplements and we're gonna cure a bunch of migraines." And it worked. And there's many other examples of things like this where you just like notice two different connections between pieces of knowledge. W- why, if these LLMs are intelligent, are they not able to use this unique advantage they have to make these kinds of discoveries?Um, I feel a little shy, like, me giving answers on AI shit- (laughs)

Narrator 6:03 (laughs)

Dwarkesh Patel 6:03 ... with you guys here. But, uh, uh, so actually, Scott Alexander addressed this question in one of his AMA threads, and he's like, "Look, humans also don't have this kind of logical omniscience," right? It's... He used the example of, in language, if you really thought about, like, why are two words connected? And it's like, oh, I understand, like, why, why rhyme has the same etymology as this other word, if you really thought about... But you just, like, don't think about it, right? There's, like, this combinatorial explosion. Um, I don't know if that addresses the fact that human, we know humans can do this, right? Like, the, humans have in fact done this, and I don't know of a single example of LLMs ever having done it. Um, actually, yeah, what, what is your answer to this?

Trenton Bricken 6:38 I think my answer at the moment is that the sort of pre-training objective doesn't necessarily... Like, it imbues you with this, like, nice, flexible, general knowledge about the world, but doesn't necessarily imbue you with the, like the skill of making, like, con- novel connections or, like, research.

Dwarkesh Patel 6:51 Yeah.

Trenton Bricken 6:51 The, the kinds of things that, uh, people are trained to do through PhD programs and through, like, sort of the process of exploring and interacting with the world. Um, and so I, I think, like, at a minimum, you need significant RL in at least similar things to, to be able to, like, approach, like, making novel discoveries. And so I would like to see some early evidence of this as we start to build models that are d- sort of interacting with one, trying to make scientific discoveries-

Dwarkesh Patel 7:17 Yeah.

Trenton Bricken 7:17 ... and, like, and sort of, like, modeling the behaviors that we expect of people in these positions-

Dwarkesh Patel 7:21 Yeah.

Trenton Bricken 7:21 ... because I don't actually think we've, we've done that in a, in like a s- meaningful or scaled way as a, as a field, so to speak.

Dwarkesh Patel 7:27 Mm-hmm.

Sholto Douglas 7:27 Yeah. R- riffing off that with respect to RL, I wonder if models currently just aren't good at knowing what memories they should be storing.

Dwarkesh Patel 7:35 Mm.

Sholto Douglas 7:36 Like, most of their training is just predicting the next word on the internet and remembering very specific facts from that. But if you were to teach me something new right now, I'm very aware of my own memory limitations, and so I would try to construct some summary-

Dwarkesh Patel 7:51 Mm-hmm.

Sholto Douglas 7:51 ... that would stick, and models cur- currently don't have the opportunity to do that. Memory scaffolding in general is just very-

Dwarkesh Patel 7:58 Hm.

Sholto Douglas 7:59 ... primitive right now. I-

Trenton Bricken 8:00 Right, like, Claude plays Pokemon, right?

Dwarkesh Patel 8:01 Exactly, yeah.

Sholto Douglas 8:02 Yeah. Or, like, someone worked on it, it was awesome, it got far, but, uh, another excited Anthropic employee then, like, iterated on the memory scaffold and was able-

Dwarkesh Patel 8:11 Hm.

Sholto Douglas 8:11 ... to, like, very quickly improve on it.

Dwarkesh Patel 8:13 Interesting.

Sholto Douglas 8:13 Um, so yeah, that's one. I do also just wonder if models are idiot savants. Um, the best analogy might be to Kim Peek. So Kim Peek, uh, was born without a corpus callosum, if I recall correctly. Um, each hemisphere of his brain operated quite independently. Uh, he could read a page of a book. So he'd open a book, there'd be two pages visible. Each eye would read one of the pages.

Dwarkesh Patel 8:37 Huh.

Sholto Douglas 8:38 Uh, and he had, like, a perfect encyclopedic memory of, like, everything he'd ever read. Um, but at the same time, he had other debilitations.

Dwarkesh Patel 8:46 Yeah.

Sholto Douglas 8:46 Functioning socially, these sorts of things. Um, and it's just kind of amazing how good LLMs are at very niche topics, but can totally fail at other ones still.

Dwarkesh Patel 8:56 I, I really wanna double-click on this thing of why there's this trade-off between memorization... Like, yeah, why does cutting it off... Like, apparently it's sort of, it's connected to this debilitation, but why can't we... Like, Wikitext is not that, it's like five megabytes of information. The human brain can store much more, so why does the human brain just not want us to lear- memorize these kinds of things, um, and is actively pruning? Uh, and, yeah, I don't know.

Sholto Douglas 9:19 Yeah.

Dwarkesh Patel 9:19 Uh, but w- we don't have to do it right now. We'll do a (laughs) we'll do a separate episode.

Sholto Douglas 9:22 Yeah. Just, the, the one thing I'll say on that is, like, there is another case study of someone with a perfect memory.

Dwarkesh Patel 9:26 Yeah.

Sholto Douglas 9:26 So they never forgot anything.

Dwarkesh Patel 9:27 Yeah.

Sholto Douglas 9:27 But their memory was too debilitating.

Dwarkesh Patel 9:29 Hm.

Sholto Douglas 9:30 Um, it'd be like your context window for the transformer is, like-

Dwarkesh Patel 9:35 Hm.

Sholto Douglas 9:35 ... trillions of tokens.

Dwarkesh Patel 9:36 Yeah.

Sholto Douglas 9:37 And then you spend all your time attending to past things, and, like-

Dwarkesh Patel 9:39 Right.

Sholto Douglas 9:40 ... are too trapped in the details-

Dwarkesh Patel 9:42 Yeah.

Sholto Douglas 9:42 ... to extract any meaningful generalizable insights from it.

Dwarkesh Patel 9:45 Yeah, yeah. Uh, Terrence Deacon, whose book you recommended, had this interesting insight about how we learn best when we're children, but we forget literally everything that happened to us when we were children, right? We have total amnesia.

Sholto Douglas 9:55 Hm.

Dwarkesh Patel 9:55 And adults have this in-between where we don't remember exact details, but we, um, we can still, like, learn in a pretty decent way. And then LLMs are on the opposite end of this gradient, where they'll get e- the exact phrasing of Wikitext down, but they won't be able to generalize in these very, uh, obvious ways.

Trenton Bricken 10:11 Yeah, a little bit like Gwern's theory, optimizer theory, no?

Dwarkesh Patel 10:13 Yeah, yeah, yeah.

Trenton Bricken 10:14 Yeah.

Dwarkesh Patel 10:14 I think I probably got it from that (laughs) .

Sholto Douglas 10:16 (laughs) Yeah, yeah, yeah.

Trenton Bricken 10:17 Oh, yeah.

Sholto Douglas 10:17 Gwern, Gwern has definitely had a big influence on, on all this for me as well.

Dwarkesh Patel 10:20 That's right, yeah, yeah, yeah. I mean, but what I, I feel like what's underappreciated, um, on the podcast is, like, we have this, like, group chat, and we also just, like, meet up a lot in person.

Trenton Bricken 10:28 Yeah.

Dwarkesh Patel 10:28 And just all of the offer from the podcast just comes from you and a couple other people just, uh, uh, feeding me their, like, ideas and nudges and whatever, and then I can just use that as an intuition pump during the conversation.

Sholto Douglas 10:40 Y- yeah, you're not the only one. (laughs)

Dwarkesh Patel 10:41 (laughs) What do you mean?

Sholto Douglas 10:43 Oh, like, I benefit immensely from-

Dwarkesh Patel 10:44 Oh, right, yeah.

Sholto Douglas 10:45 ... just hearing what everyone else has to say.

Dwarkesh Patel 10:46 Yeah, yeah, yeah.

Sholto Douglas 10:48 And then, yeah. It's all, it's all regurgitation in one way or another.

Dwarkesh Patel 10:49 Yeah, yeah, yeah, yeah. Yeah.

Trenton Bricken 10:51 Another question?

Dwarkesh Patel 10:51 Yes.

Sholto Douglas 10:52 Uh, MaybeRabid, MaybeRabidMonkey, um-

Dwarkesh Patel 10:58 (laughs)

Sholto Douglas 10:58 ... asks, "Imagine you have a 17-year-old brother/nephew just starting college. What would you recommend he study given your AGI timelines?"

Dwarkesh Patel 11:08 I don't know. Become a podcaster? (laughs) I feel like that job's still gonna be around. It's funny because I, I studied computer science, and in retrospec... I mean, at the time, it was like, you could have become a software engineer or something. Instead, you became a podcaster. It's, like, kind of an irresponsible career move, but in retrospect, it's like... (laughs)

Trenton Bricken 11:23 It's great. (laughs)

Dwarkesh Patel 11:23 It kinda worked out. Just as these guys are getting automated. (laughs)

Trenton Bricken 11:27 I get asked this question all the time.

Dwarkesh Patel 11:28 Yes. Okay, go ahead.

Trenton Bricken 11:29 Um, and one answer that I like to give is that you should think about the next couple of years as increasing your individual leverage by, like, a huge factor every year.

Dwarkesh Patel 11:37 Mm-hmm.

Trenton Bricken 11:37 So, or, you know, already software engineers will come up and say, "You know, I'm two times faster," or, "In new languages, I'm five times faster than I was last year." Um, I expect that trend line to continue, basically-

Dwarkesh Patel 11:46 Right.

Trenton Bricken 11:46 ... as you sort of go from this model of, well, I'm working with some model that's assisting me on my computer, and it's like, like, basically a pairing session, to I am managing a small team, uh, through to I'm managing, like, a division or a company-

Dwarkesh Patel 12:00 Yeah.

Trenton Bricken 12:00 ... basically, that is, like, targeting a task. Um, and so I think that deep technical knowledge in fields will still matter-

Dwarkesh Patel 12:08 Yeah.

Trenton Bricken 12:08 ... in four years.

Dwarkesh Patel 12:09 Yeah.

Trenton Bricken 12:09 Like, it absolutely will-

Dwarkesh Patel 12:10 Yes.

Trenton Bricken 12:11 ... because you will be in the position of managing dozens... Or, like, like, your sort of, your, your individual management bandwidth-... will be maxed out by trying to, to manage, like, teams of AIs-

Dwarkesh Patel 12:21 Yeah, yeah.

Trenton Bricken 12:21 ... in this kind of thing. And maybe AIs... You know, maybe we end, end up in, like, a truly, like, singularity world, where you have AIs managing AIs, and this kind of stuff. But I think in a very wide part of the, like, possibility spectrum-

Dwarkesh Patel 12:31 Yes.

Trenton Bricken 12:31 ... you are managing enormous, like, vastly more resources-

Dwarkesh Patel 12:35 Yeah.

Trenton Bricken 12:35 ... than an individual could command today.

Dwarkesh Patel 12:37 Yeah, yeah.

Trenton Bricken 12:38 Um, and you should be able to solve so many more things with that.

Dwarkesh Patel 12:40 That's right. And, and I think, like, I, I would emphasize that this is not just cope. Like, it genuinely is a case that these models lack the kind of long-term coherence which is, like, absolutely necessary for making a successful company or-

Trenton Bricken 12:53 Right.

Dwarkesh Patel 12:53 Like, just, like, getting a fucking office is, like, kind of complicated, right?

Trenton Bricken 12:55 Yeah.

Dwarkesh Patel 12:55 It's like you gotta deal with all these, um... And so, uh, uh, you can just imagine that for sector after sect-... Like, the economy is really big.

Trenton Bricken 13:02 And really complex.

Dwarkesh Patel 13:03 Exactly. And so, especially if it's... I mean, I don't know the details, but I'd assume if it, if it's, like, a data sparse, uh, thing where you kind of just, like, gotta know what's actually, what is the context of what's happening in the sector or something, I, I, I feel like you'd be in a good position.

Trenton Bricken 13:15 Yeah.

Dwarkesh Patel 13:15 Maybe the other thought I have is that it's, it's really hard to, like, plan your career in general.

Trenton Bricken 13:20 Mm-hmm.

Dwarkesh Patel 13:21 And I don't know what advice that implies, because I remember being super frustrated. I mean, I was in college, and the reason I was doing the podcast was to figure out what it is I want to do. Um, it wa- it wasn't the podcast itself. And I would go on, like, 80,000 Hours or whatever career advice, and I, it, it just, like, in retrospect, was all kind of mostly useless, and just, like, just try doing things. I mean, especially with AI, we just, like, don't know what it's gonna... Like, it's so hard to forecast what kind of transformations there will be, so try things, do things. I mean, it's such banal, vague advice, but I, I am quite skeptical of career advice in general.

Trenton Bricken 13:54 Well, may- maybe the, like, piece of career advice that I'm not skeptical of-

Dwarkesh Patel 13:57 Yeah.

Trenton Bricken 13:57 ... is put yourself close to the frontier-

Dwarkesh Patel 13:58 Mm-hmm.

Trenton Bricken 13:59 ... because you have a much better vantage point.

Dwarkesh Patel 14:00 That's right.

Trenton Bricken 14:01 Right? Like, you can study deep technical things, uh, whether it's computer science or biology or, like, and get to the point where you can see what the issues are-

Dwarkesh Patel 14:10 Yeah.

Trenton Bricken 14:10 ... because it's actually, it's, like, r- remarkably obvious at the frontier what the problems are.

Dwarkesh Patel 14:13 That's right.

Trenton Bricken 14:13 And it's very difficult to see-

Dwarkesh Patel 14:14 But actually, do you think there is an opportunity? 'Cause one of the people, th- one of the things people bring up is the, um, maybe the people who are advanced in their career and have all this tacit knowledge will be in a position to be accelerated by AI, but you, you guys four years ago or two years ago, when you were getting discovered or something, that kind of thing, where you have a GitHub open issue and you try to solve it, is that just, like, that's done and so the onboarding is much harder?

Trenton Bricken 14:38 Uh, it's still what we look for in hiring.

Dwarkesh Patel 14:40 Yeah.

Trenton Bricken 14:40 So, (laughs) you know?

Dwarkesh Patel 14:41 Yeah.

Trenton Bricken 14:42 (laughs)

Sholto Douglas 14:42 I'm in favor of the learn fundamentals, gain useful mental models-

Dwarkesh Patel 14:47 Yeah.

Sholto Douglas 14:48 ... but it feels like everything should be done in an AI-native way or, like, top-down instead of your bottom-up learning. So first of all, learn things more efficiently by using the AI models, and then just know where their capabilities are and aren't.

Dwarkesh Patel 15:01 Yeah.

Sholto Douglas 15:01 Um, and I would be worried and skeptical about any subject which prioritizes rote memorization-

Dwarkesh Patel 15:06 Mm-hmm.

Sholto Douglas 15:06 ... of lots of facts or information-

Dwarkesh Patel 15:08 Mm-hmm. Yeah, yeah, yeah.

Sholto Douglas 15:08 ... um, instead of ways of thinking.

Dwarkesh Patel 15:10 Yeah.

Sholto Douglas 15:11 Um, but if you're always using the AI tools to help you, then you'll naturally just have a good sense for the things that it is and isn't good at.

Dwarkesh Patel 15:17 That's right. Yeah, yeah, yeah. Next one.

Trenton Bricken 15:20 What is your strategy, method, or criteria for choosing guests?

Dwarkesh Patel 15:24 The most important thing is, do I wanna spend one to two weeks reading every single thing you have ever written, every single interview you've ever recorded, talking to a bunch of other people about your research? Um, 'cause I get asked by people who are, like, quite influential often to be like, "Would you have me on your podcast?" Um, and more often than not, I say no, for two reasons. One is just, like... Like, wh- okay, you're influential or something, and just, like, wh- you're, it's not fundamentally that interesting as an interview prospect. N- not from, like, uh, I don't think about the hour that I'll spend with you. I think about, like, the two weeks, 'cause it's g- this is my life, right? The research is my life, and I, and I wanna have fun while doing it. Um, so just, like, is this gonna be an interesting two weeks to spend? Is it gonna help me with my future research or something? And the other is, big guests don't really matter that much in the, in the... Like, if you just look at what are the most popular episodes or, uh, what, in the long run, helps the podcast grow, um, by far, my most popular guest is Sarah Payne. And she, before I interviewed her, was just, like, a scholar, um, who was not publicly well- well-known at all, and I just found her books quite interesting. Same goes with... So my, my most popular guests are Sarah Payne, and then Sarah Payne, Sarah Payne, Sarah Payne, 'cause I did a lecture series with her.

Trenton Bricken 16:35 (laughs)

Dwarkesh Patel 16:35 Um-

Sholto Douglas 16:36 (laughs) That's awesome.

Dwarkesh Patel 16:37 ... and by the way, from a viewer-minute adjusted basis, I host a Sarah Payne podcast where I occasionally talk about AI.

Trenton Bricken 16:44 (laughs)

Dwarkesh Patel 16:44 (laughs)

Sholto Douglas 16:44 (laughs)

Dwarkesh Patel 16:45 Um, and then it's David Reich, who is a geneticist of ancient DNA. I mean, he- he's, like, somewhat well-known, but... He had a best-selling book, but he's not, like, he's not Satya Nadella or Mark Zuckerberg, who are the next people on the list. And then again, I think, at, like, pretty soon, it's, like, you guys or Leopold or something, and then you get to the last CEOs or something. So, big names just don't matter that much in the, like, for what I'm actually trying to do. And so it's just like, well... And it's also really hard to predict who's gonna be the David Reich or-

Trenton Bricken 17:09 Mm-hmm.

Dwarkesh Patel 17:09 ... uh, Sarah Payne, so just, like, have fun. Who- talk to whoever you want to spend time researching, and it's a pretty good proxy for what will actually be popular.

Trenton Bricken 17:18 What

Trenton Bricken 17:19 was the specific moment, if there was one, that you realized you, that producing your podcast was a viable long-term strategy?

Dwarkesh Patel 17:26 I think when I was shopping around ad spots for a Mark Zuckerberg episode, and which were s- which are, you know, like, uh, and now when I look back on it, it's like, not in retrospect that mind-blowing, but at the time, I'm like, "Oh, I could actually hire an editor" (laughs)

Trenton Bricken 17:42 (laughs)

Dwarkesh Patel 17:42 ... "full-time, or maybe more editors than one." Um, and from there, like, turning into a real business. Um, that's when I... 'Cause before I just, like, didn't... People would tell me, like, "Oh, these other podcasts are making whatever, whatever amount of money," and I'd be like, "How?" You know? (laughs)

Trenton Bricken 17:57 (laughs)

Dwarkesh Patel 17:57 Um, I have this running joke with one of my friends that (laughs) ... I don't know if you've seen me do this, but every time I encounter, like, a young person who's like, "What should I do with my life?" I'm like, "You gotta start a blog. You gotta be the Matt Levine of AI. You can do this."

Trenton Bricken 18:09 (laughs)

Dwarkesh Patel 18:09 That's like a-

Trenton Bricken 18:09 Mm-hmm.

Dwarkesh Patel 18:09 ... a totally empty niche. You could, like, it's i- you, you can... um, and I have this running joke with them where they're like, "You're like a country bumpkin who's, like, won the lottery." And you're like-

Trenton Bricken 18:17 (laughs)

Dwarkesh Patel 18:17 (laughs) You go up to everything, of course, and like-

Sholto Douglas 18:19 (laughs)

Trenton Bricken 18:19 (laughs)

Dwarkesh Patel 18:19 ... "Guys, a scratch pad."

Trenton Bricken 18:20 You have to love it.

Dwarkesh Patel 18:20 "Get the scratch pad." Yeah, yeah. (laughs)

Sholto Douglas 18:22 (laughs)

Trenton Bricken 18:22 (laughs)

Dwarkesh Patel 18:22 Um...

Sholto Douglas 18:24 I, I, I, I do wanna press on that a bit more 'cause your immediate answer to the, uh, 17-year-old was to start a podcast.

Dwarkesh Patel 18:30 Yeah.

Sholto Douglas 18:30 So, like, what, what niches are there? What sort of things would you be excited to see-

Dwarkesh Patel 18:34 Yeah.

Sholto Douglas 18:35 ... in, like, new blogs podcasts?

Dwarkesh Patel 18:36 I mean, I wonder if you guys think this too, but I think this, like, Matt Levine of AI, um-

Trenton Bricken 18:41 Yeah, absolutely.

Dwarkesh Patel 18:42 ... is, is like a totally open niche as, as far as I can tell. And I-Apologize to those who are trying to fill it in. (laughs)

Sholto Douglas 18:48 (laughs)

Trenton Bricken 18:48 (laughs)

Dwarkesh Patel 18:48 I was like, "Oh, I'm aware of at least one person right now doing this," but, um, the other thing I really emphasize is, it is really hard to do this based on other people's advice, or to say, like, "Here's a niche I'm f- f-" Um, like, I, at least I'm trying not to fill, like, a specific niche. If you think about any sort of successful new media thing out there, there's, it has two things which are true. It's, like, often not just geared towards one particular topic or interest, and two, it's, the most important thing is that it is propelled by a single person's vision. Um, it's not like a collective or whatever. And so I would just really emphasize... Sorry, the thing I really want to emphasize is, it, it can be done. Two, you can make a lot of money at it, which is not the most important thing probably for the kind of person who would succeed at it, but still, it's just worth knowing that this is a viable career. Um, three, that, I, yeah, that basically you're gonna feel like shit in the beginning where it's, like, all your early stuff is gonna kinda suck. Um, maybe some of it will get appreciated, but it seems like bad advice to say still stick through it in case you actually are terrible, because some people are terrible, but, (laughs) um, in case you are not, like, just do it, right? Like, what is, what is the three months of blogging on the side really gonna cost you? And people just don't actually seriously do the thing for long enough to actually get evidence or get the sort of RL feedback on like, oh, this is how you do it, this is how you frame an argument, this is how you make a compelling thing that people will want to read or watch.

Trenton Bricken 20:16 Bloging is definitely underrated.

Dwarkesh Patel 20:18 Yeah.

Trenton Bricken 20:18 I think, like, most of us have probably-

Dwarkesh Patel 20:20 So you both had blogs which were relevant. I don't know if they're-

Trenton Bricken 20:22 Yeah.

Dwarkesh Patel 20:22 ... actually relevant to getting-

Trenton Bricken 20:23 They weren't that... They were, like, somewhat relevant.

Dwarkesh Patel 20:25 Yeah.

Trenton Bricken 20:26 Um, but I think it was more so that we have all read almost all the blogs that do an-

Dwarkesh Patel 20:31 That's right.

Trenton Bricken 20:31 ... in-depth, like, treatises on AI.

Dwarkesh Patel 20:33 That's right.

Trenton Bricken 20:33 Um, like, if you write something that is high-quality, it is almost, like, invariably going to be, like, shared around Twitter and read.

Dwarkesh Patel 20:40 Oh, th- this is so underappreciated.

Trenton Bricken 20:41 Yeah.

Dwarkesh Patel 20:42 So, two pieces of evidence. Um, I was talking to a very famous blogger you would know, and I was asking him, "How often do you discover a new, like, undiscovered blogger?" And he was like, "Eh, it happens very rarely, like maybe once a year." Um, and then I ask him, "How long after you discover him or her does the rest of the world discover them?" And he's like, "Maybe a week."

Trenton Bricken 21:04 Interesting.

Dwarkesh Patel 21:04 W- and what that suggests is, like, it's actually really efficient. Like... Oh, so, sh-

Trenton Bricken 21:08 Yeah.

Dwarkesh Patel 21:09 I, I, I have some more takes. (laughs)

Trenton Bricken 21:12 (laughs)

Sholto Douglas 21:12 (laughs)

Trenton Bricken 21:12 Let's hear them.

Sholto Douglas 21:12 This is the AMA.

Dwarkesh Patel 21:13 So, uh, I, I believe that slow compounding growth in media is kinda fake. Um, like, Leopold's situational awareness, it's not like he was building up an audience for a long time, for years or something. It just, like, it was really good-

Trenton Bricken 21:26 Mm-hmm.

Dwarkesh Patel 21:26 ... um, disagree or agree with it. And if it's good enough, it, literally everybody who matters, and I mean that literally, will read it. I mean, I think it's, like, hard 'cause, like, zero shot, something like that, but the fundamental thing to emphasize is, you're, the, the compounding growth, at least for me, has been I feel like I've gotten better.

Trenton Bricken 21:43 Mm-hmm.

Dwarkesh Patel 21:43 And it's not so much that somehow the three years of having 1,000 followers was somehow, like, a compounding... You know, I don't think it was that. I think it was just like, I, it took a while to get better.

Trenton Bricken 21:53 Mm-hmm. Yeah. Suddenly, when Leopold posts that, like, the next day, it's almost like you can picture it, like, being s- almost stapled, not, it wasn't, but it was stapled to the wall, so to speak, on Twitter, like-

Dwarkesh Patel 22:02 That's right. Yeah.

Trenton Bricken 22:02 ... you know, everyone was talking about it. You went to any event for the following week, every single person-

Dwarkesh Patel 22:07 That's right.

Trenton Bricken 22:07 ... in the entire city was talking about that essay.

Dwarkesh Patel 22:09 Yeah, yeah.

Trenton Bricken 22:10 Um, it was like Renaissance Florence or whatever. (laughs)

Dwarkesh Patel 22:11 That's right, that's right, that's right, yeah. The world is small.

Sholto Douglas 22:14 Yeah, what, what would you say was your first big success? I, I'm trying to think back to when I first found-

Dwarkesh Patel 22:18 Yeah.

Sholto Douglas 22:18 ... your podcast. I distinctly remember you had your blog post on, uh, the Annus Mirabilis.

Dwarkesh Patel 22:24 Mm-hmm.

Sholto Douglas 22:24 And Jeff Bezos retweeted it, I think.

Dwarkesh Patel 22:26 Yeah, yeah.

Sholto Douglas 22:27 I, I'm trying to remember if you, if, if it was even before that or not, but, but yeah, what, I'm curious, your answer.

Dwarkesh Patel 22:31 I feel like that was it.

Sholto Douglas 22:32 Okay, yeah, yeah.

Dwarkesh Patel 22:33 Yeah. I, I mean, it wasn't something where I'm like, it was some big insight that deserved to blow up like that. It was just taking some shots on goal that were all, like, whatever, insight porn-y, and then one of them, um, I guess caught the right guy's attention and, (laughs) um, yeah. But I think that was it.

Trenton Bricken 22:48 Yeah. That's something else which is underappreciated which is that a piece of writing doesn't need to have a fundamentally new insight so much-

Dwarkesh Patel 22:53 Yeah.

Trenton Bricken 22:53 ... as give people a way to express cleanly-

Dwarkesh Patel 22:55 Mm-hmm.

Trenton Bricken 22:55 ... a set of ideas that they are already, like, aware of in, in a, like, sort of broader way.

Dwarkesh Patel 23:00 Yeah.

Trenton Bricken 23:00 Um, and if it's really crisp and, like, articulate, then-

Dwarkesh Patel 23:04 Yeah.

Trenton Bricken 23:04 ... even still that's very valuable.

Dwarkesh Patel 23:05 And, sorry, the one thing I should emphasize, which I think is maybe the most important thing to the feedback loop, it's not the compounding growth of the audience. Um, I don't even think it's the compoundin-, like, it's me getting more shots on goal in terms of doing the podcast. I actually don't think you improve that much by just doing the same thing again and again, uh, if there's, like, no reward signal. It just-

Trenton Bricken 23:24 Mm-hmm.

Dwarkesh Patel 23:24 ... you'll keep doing whatever you were doing before. I genuinely think the most important thing has been the podcast is good enough that it merits me getting to meet people like you guys. Um, then I become friends with people like you. You guys teach me stuff. I produce more good podcasts, so hopefully slightly better. That helps me meet people in other fields. They teach me more things. Um, like with the China thing recently, I wrote this, like, blog post about a couple stories about things that happened in China, and that alone has, like, netted me an amazing China network-

Trenton Bricken 23:54 Mm-hmm.

Dwarkesh Patel 23:54 ... in the matter of, like-

Trenton Bricken 23:55 Mm-hmm.

Dwarkesh Patel 23:55 ... one blog post, right? And so hopefully, uh, uh, if I do an episode on China, it will be better as a result. Um, and hopefully that happens across field after field. And so just getting to meet people like you is actually the main sort of, uh, flywheel.

Trenton Bricken 24:08 Interesting. So, move to San Francisco?

Dwarkesh Patel 24:10 (laughs) Yes. If you're trying to do AI, yeah.

Trenton Bricken 24:11 Yeah. Next questions. Shall we do...

Sholto Douglas 24:16 I mean, can we do a-

Dwarkesh Patel 24:18 Oh, very important question.

Trenton Bricken 24:20 (laughs)

Dwarkesh Patel 24:20 Yeah. (laughs)

Trenton Bricken 24:20 From, uh, Jacked Paajeet.

Dwarkesh Patel 24:22 (laughs)

Sholto Douglas 24:22 (laughs)

Trenton Bricken 24:23 How much can you bench?

Dwarkesh Patel 24:24 Um, uh-

Sholto Douglas 24:26 You can't lie 'cause we want to know the answer. (laughs)

Trenton Bricken 24:29 (laughs)

Dwarkesh Patel 24:29 (laughs) Um, at one point I did bench 225 for four. Um, now I think-

Trenton Bricken 24:34 Okay.

Dwarkesh Patel 24:35 ... I'm probably, like, 20, uh, 20 pounds lighter than that or something.

Trenton Bricken 24:38 Okay.

Dwarkesh Patel 24:38 Um, uh, the reason you guys are asking me this-

Sholto Douglas 24:41 (laughs)

Dwarkesh Patel 24:42 (laughs)

Trenton Bricken 24:42 (laughs)

Dwarkesh Patel 24:42 (laughs) ... is because I've gotten lifting with both of you. And I remember Trent and I were doing, like, pull-ups and, uh, bench. And then he'd be like, "Bench," and he'd, like, throw on another plate or something. And then, like, instead of pull-ups, he'd, like, be cranking out these muscle ups.

Sholto Douglas 24:57 (laughs)

Dwarkesh Patel 24:57 (laughs)

Trenton Bricken 24:57 (laughs) It's all technique. Just making sure. (laughs)

Dwarkesh Patel 25:01 (laughs) Um, yeah, so they both bench more than me. (laughs)

Trenton Bricken 25:05 (laughs)

Dwarkesh Patel 25:05 But I try my best. (laughs)

Trenton Bricken 25:07 A- ask again in, uh, six months. Yeah.

Dwarkesh Patel 25:09 Yeah. Yeah.

Trenton Bricken 25:09 Yeah, exactly.

Dwarkesh Patel 25:11 (laughs) Um,

Dwarkesh Patel 25:12 what's your favorite history book? There's a wall of them behind you.

Trenton Bricken 25:15 Mm-hmm.

Dwarkesh Patel 25:17 Oh, uh, I mean, obviously the, uh, Caro LBJ biographies.

Trenton Bricken 25:20 Oh, okay, yeah.

Dwarkesh Patel 25:21 Yeah.

Trenton Bricken 25:21 The, the-

Dwarkesh Patel 25:22 Sorry, the one, main thing I took away from those books is LBJ had this quote that he would tell his debate sc- he, in his early 20s, he taught debate to these, like, poor Mexican students in, um, in Texas. And he, he used to tell them, "If you do everything, you'll win." I think this is an underrated quote, so that's the main thing I took away. And you see it through his entire career, where there's a reasonable amount of effort, which, you know, goes by, like, 20/80. You do the 20 to get the 80% of the effect, and then if you go beyond that to get, like, "Oh, no, I'm not just gonna do the 20%, I'm gonna just, like, do the whole thing." And there's a level even beyond that, which is like, like this is, like, just, like, an unreasonable use of time. This is hav- gonna have no ultimate impact, and still try doing that. Um, uh, yeah.

Trenton Bricken 26:05 You, you've shared on Twitter, uh, using Anki.

Dwarkesh Patel 26:08 Yeah, yeah.

Trenton Bricken 26:08 And even, like, a Claude integration.

Dwarkesh Patel 26:10 Yeah.

Trenton Bricken 26:10 Uh, do you do book clubs? Do you use Goodreads? And what are you reading right now?

Dwarkesh Patel 26:16 I don't use... I don't have, uh, book clubs. I do, the, um... But the Space Bar edition has just genuinely been a huge, like, uplift in my ability to learn. Mostly because it, it's not even the long-term impact over years, though I think that is part of it and I do regret all the episodes I did without using Space Bar because all the insights have just sort of faded away.

Trenton Bricken 26:37 Mm-hmm.

Dwarkesh Patel 26:38 Um, the main thing is, if you're studying a complicated subject, for, at least for me, it's been super helpful to, um, consolidate. So it's like y- if you don't do it, you feel like a general, where you're like, "I'm gonna wage a campaign against this country," and then you, like, climb one hill, uh, and then the next day you gotta retreat, and then you climb the same hill. Um, there might be a sort of more kosher analogy (laughs) -

Trenton Bricken 27:02 (laughs)

Dwarkesh Patel 27:02 ... uh, sorry. And the other question was what am I reading right now?

Trenton Bricken 27:05 Yeah.

Dwarkesh Patel 27:06 Oh, uh, my friend Alvaro de Menard, author of Fantastic Anachronism. Can I, can I just pull it up, actually? It's right here.

Trenton Bricken 27:12 Yeah.

Dwarkesh Patel 27:13 He... I hope he's, I hope he's okay with me sharing this, but he wrote, um, he made, like, 100 copies of this translation he did of his favorite Greek poet. And they're like... Yeah, uh, Cavafy. Hopefully I didn't mispronounce it.

Trenton Bricken 27:28 Huh.

Dwarkesh Patel 27:28 Um, that one has a good inscription for Grin because that's his coffee, but... (laughs)

Trenton Bricken 27:32 (laughs)

Dwarkesh Patel 27:32 Um, but it's, it's super delightful, and that's what I've been reading recently. Um-

Trenton Bricken 27:36 Any insights from it so far?

Dwarkesh Patel 27:38 Poets will hate this framing. I feel like, uh, poetry is like, it's like TikTok, where... (laughs)

Trenton Bricken 27:49 (laughs) Oh, boy.

Dwarkesh Patel 27:49 Um, where it's like you get this, like, quick vibe of a certain thing, and then you, like, uh, huh, and then like swipe, and then you get the next vibe, swipe. Alvaro (laughs) , I'm sorry.

Trenton Bricken 28:02 (laughs)

Dwarkesh Patel 28:02 (laughs)

Trenton Bricken 28:02 No, that's interesting. How do you go about learning new things or preparing for an episode? Like, what does that... You mentioned the one to two-week period where you're, like, deep diving on the person. What does that actually look like?

Dwarkesh Patel 28:11 Um, it's very much the obvious thing, like you read their books, you... If they have papers, you read the papers. If they're, have colleagues, you try to talk to them to better understand the field. I will also mention that I, all I have to do is ask them questions, and I do think it's a much... I think it's much harder to, like, learn a field to be a practitioner than just learn enough to ask interesting questions. But, um, yeah, for that, like, it's very much the obvious thing you'd expect.

Trenton Bricken 28:37 Based Carl Sagan asks-

Dwarkesh Patel 28:40 (laughs)

Trenton Bricken 28:40 ... uh, "What are your long-term goals and ambitions?"

Dwarkesh Patel 28:45 Yeah, the, uh, AGI kind of just makes the prospect of a long term, like, harder to articulate, right? Um, you know, the Peter Thiel quote about, "What is your ten-year plan? And why can't you do it in six months?"

Trenton Bricken 28:56 Mm-hmm.

Dwarkesh Patel 28:56 Like, it's especially salient (laughs) -

Trenton Bricken 28:58 (laughs)

Dwarkesh Patel 28:59 ... given timelines. Um, for the foreseeable future, grow the podcast and do more episodes, maybe more writing, but, um, yeah. So we'll see what happens after, like, ten years or something. The world might be different enough. It, it, yeah, so basically podcast for now.

Trenton Bricken 29:17 Something you've spoken to me about, uh, and particularly when you were, when you were trying to hire people for the podcast, was what you wanted to achieve with the podcast.

Dwarkesh Patel 29:24 Mm-hmm.

Trenton Bricken 29:24 Like, w- w- what, in what way do you want the podcast to, like, shape the world, so to speak? Do you have any thoughts on that? Or, uh, 'cause I remember you talking like, "I really want people to actually understand AI, and, like, how this might change their lives," or, like, you know, how, like, what we could be doing now to shape the world such that it ends up better.

Dwarkesh Patel 29:39 Yeah. Um, I don't know. So I have, uh, I have contradictory views on this. On the one end, I do, I do know that important decisions are being made right now in AI.

Trenton Bricken 29:54 Mm-hmm.

Dwarkesh Patel 29:54 And I do think we're, uh, g- I mean, riffing on what we were saying about situational awareness-

Trenton Bricken 29:59 Mm-hmm.

Dwarkesh Patel 29:59 ... if you do something really good, it has a very high probability of one-shotting the relevant person.

Trenton Bricken 30:04 Mm-hmm.

Dwarkesh Patel 30:04 You know, people are r- just generally reasonable. Like, you make a good argument, it, it'll go places. The, on the other hand, I just think it's very hard to know what should be done. It, like, you gotta have the very correct world model, and then you gotta know how in that world model the action you're taking is gonna have the effect you anticipate. And even in the last week, I've changed my mind on some pretty fundamental things about, um, what I think about the possibility of an intelligence explosion or transformative AI as a result of talking to the Epoch folks.

Trenton Bricken 30:32 Yep.

Dwarkesh Patel 30:33 Basically, the TL;DR is I want the podcast to just be an epistemic tool for now until I... 'Cause I think it's just very easy to be wrong, and so just having a background level of understanding of the relevant arguments is the highest priority.

Trenton Bricken 30:46 Makes sense.

Dwarkesh Patel 30:46 Yeah. What's your sense? What should I be doing?

Trenton Bricken 30:49 I mean, I think the podcast is awesome, um, and a lot more people should listen to it. And there are a lot more guests I'd be excited for you to interview.

Dwarkesh Patel 30:56 Nice.

Trenton Bricken 30:56 So-

Dwarkesh Patel 30:56 So, glad to be on your recs.

Trenton Bricken 30:57 ... seems like a pretty good-

Dwarkesh Patel 30:58 Yeah.

Trenton Bricken 30:58 ... seems like a pretty good answer for now.

Dwarkesh Patel 30:59 Yeah, I think-

Trenton Bricken 31:00 Making sure that, like, there is a great debate of ideas-

Dwarkesh Patel 31:04 Yeah.

Trenton Bricken 31:04 ... on not, not just AI, but on, like, other fields-

Dwarkesh Patel 31:06 Yeah.

Trenton Bricken 31:06 ... and, and everything is, I, incredibly high leverage-

Dwarkesh Patel 31:08 Yeah.

Trenton Bricken 31:08 ... in value.

Dwarkesh Patel 31:09 Yeah, yeah.

Trenton Bricken 31:10 How do you groom your beard? It's majestic.

Dwarkesh Patel 31:12 (laughs) Um, I, I don't know what to say, just (laughs) genetics. (laughs)

Sholto Douglas 31:19 (laughs)

Trenton Bricken 31:19 (laughs)

Dwarkesh Patel 31:19 I do trim it, but-

Trenton Bricken 31:21 No beard oil?

Dwarkesh Patel 31:22 I, sometimes I do beard oil.

Trenton Bricken 31:23 Oh, okay.

Dwarkesh Patel 31:23 So, yeah.

Sholto Douglas 31:23 How often?

Dwarkesh Patel 31:24 Uh, o- once every couple of days.

Sholto Douglas 31:26 Okay.

Trenton Bricken 31:26 That's not sometimes. (laughs)

Dwarkesh Patel 31:28 (laughs)

Sholto Douglas 31:28 (laughs)

Trenton Bricken 31:28 That's pretty often.

Dwarkesh Patel 31:29 And, uh, do you have sh- different shampoo for your head and your beard? No.

Sholto Douglas 31:32 What kind of shampoo do you use?

Dwarkesh Patel 31:33 (laughs) Anti-dandruff. (laughs)

Sholto Douglas 31:36 (laughs)

Trenton Bricken 31:36 (laughs)

Dwarkesh Patel 31:36 Do you condition it?

Sholto Douglas 31:38 Um, yeah.

Dwarkesh Patel 31:40 How often do you shampoo it?

Trenton Bricken 31:42 (laughs)

Sholto Douglas 31:42 (laughs)

Dwarkesh Patel 31:43 Who pushed you up to this?

Sholto Douglas 31:44 We're, we're, we're, we're-

Trenton Bricken 31:44 Big shampoo. (laughs)

Sholto Douglas 31:44 ... we're giving people the answers that they want. (laughs)

Trenton Bricken 31:48 Big beard oil.

Dwarkesh Patel 31:50 (laughs)

Sholto Douglas 31:50 (laughs) Yeah, you can sell some ad slots to different, uh, shampoo companies-

Dwarkesh Patel 31:54 (laughs)

Sholto Douglas 31:54 ... and we can edit it.

Trenton Bricken 31:54 Maybe we solve an ad slot, who knows?

Dwarkesh Patel 31:57 (laughs)

Sholto Douglas 31:57 (laughs)

Dwarkesh Patel 31:57 Sorry, you had this idea of merch. Do you wanna explain this (laughs) , this t-shirt idea?

Sholto Douglas 32:01 Yes.

Trenton Bricken 32:01 Oh, yeah.

Sholto Douglas 32:01 Yeah, yeah. So, people should react to this.

Dwarkesh Patel 32:03 Uh-huh.

Sholto Douglas 32:03 Someone should make it happen. D- Dwarkesh wants merch, but he doesn't wanna admit that he wants it-

Dwarkesh Patel 32:07 (laughs)

Sholto Douglas 32:08 ... uh, or he doesn't wanna make it himself 'cause that seems tacky.

Dwarkesh Patel 32:10 Yeah.

Sholto Douglas 32:11 So, I really want a plain white tee with just Dwarkesh's beard in the center of it.

Dwarkesh Patel 32:16 (laughs)

Sholto Douglas 32:16 That's it, nothing else.

Dwarkesh Patel 32:17 But you were saying it should be, like, have a different texture than the rest of the shirt.

Sholto Douglas 32:20 Oh, so, so then I was really ripping off it-

Dwarkesh Patel 32:23 (laughs)

Sholto Douglas 32:23 ... where maybe, like, a limited edition set can have some of your beard hair-

Dwarkesh Patel 32:27 Okay.

Sholto Douglas 32:27 ... actually sewn into the shirt.

Dwarkesh Patel 32:30 (laughs)

Trenton Bricken 32:30 (laughs)

Sholto Douglas 32:30 (laughs)

Trenton Bricken 32:30 All right.

Sholto Douglas 32:32 That'd be pretty cool. I would, I would, I would pay. I would pay for that.

Dwarkesh Patel 32:35 Okay.

Sholto Douglas 32:35 (laughs)

Trenton Bricken 32:35 (laughs) How much?

Dwarkesh Patel 32:36 I've, I've got, like, patches all over my beard. (laughs)

Sholto Douglas 32:39 (laughs)

Trenton Bricken 32:39 (laughs) Depends on how much hair, if it's, like, one is, like, in there somewhere versus, like, the whole thing.

Dwarkesh Patel 32:44 (laughs)

Trenton Bricken 32:44 Like, do I have to dry clean it? Can I wash it as, like, on the delicate setting? Like... (laughs)

Dwarkesh Patel 32:49 (laughs)

Sholto Douglas 32:49 (laughs)

Trenton Bricken 32:52 But, but really, I, I think you should get merch. If, if you wanna grow the podcast, which apparently you do, then-

Dwarkesh Patel 32:56 You, you, you think beard-

Trenton Bricken 32:57 ... this is one way to do that.

Dwarkesh Patel 32:58 ... beard hair in the future is necessary? (laughs)

Trenton Bricken 33:00 Oh, yeah. Oh, yeah.

Dwarkesh Patel 33:00 Yeah, yeah, yeah.

Trenton Bricken 33:01 (laughs)

Sholto Douglas 33:01 (laughs)

Trenton Bricken 33:03 Which historical figure would be best suited to run a frontier AI lab?

Dwarkesh Patel 33:06 This is definitely a question for you guys.

Sholto Douglas 33:08 Oh. No, I mean, I'm curious what your take is first.

Dwarkesh Patel 33:11 Yeah.

Sholto Douglas 33:11 You've spoken to more of the heads of AI labs than, than I have.

Dwarkesh Patel 33:14 I, I was gonna say, LBJ? Sorry, is the question who would be best at running an AI lab or would be best for the world, or?

Trenton Bricken 33:22 Let's say-

Sholto Douglas 33:23 Yeah, what's, what outcome do you want? (laughs)

Dwarkesh Patel 33:24 (laughs)

Trenton Bricken 33:24 Yeah, what outcome do you want?

Dwarkesh Patel 33:26 'Cause I imagine, it seems like what the best lab, AI lab CEOs succeed at is raising money, building up hype, um, setting a coherent vision. I don't know how much it matters for the CEO themselves to have good research taste or something. But it seems like it, their role is more as a sort of emissary to the rest of the world. And I feel like LBJ would be pretty good at this. Like, just getting the right concessions, making projects move along, um, coordinating among different groups to... Maybe r- oh, Robin Moses.

Sholto Douglas 34:00 Yeah.

Dwarkesh Patel 34:00 Again, not necessarily best for the world, but just in terms of, like, making shit happen.

Trenton Bricken 34:04 Yeah. I mean, I think best for the world is a pretty important precondition.

Dwarkesh Patel 34:07 Oh, right.

Trenton Bricken 34:07 Um...

Dwarkesh Patel 34:08 Uh, there's a Lord Ackwood quote of, "Great people are very rarely good people."

Trenton Bricken 34:13 Hmm.

Dwarkesh Patel 34:14 So, it's hard to think of a great person in history who are like, I feel like they'd really move the ball forward and also I trust their moral judgment.

Trenton Bricken 34:20 We're lucky in many senses with, like, the set today, right?

Dwarkesh Patel 34:23 That's right.

Trenton Bricken 34:23 Like, the set of people today are both, like, they try and care a lot about the moral side as well as, uh-

Dwarkesh Patel 34:29 Yeah.

Trenton Bricken 34:30 ... just sort of drive the labs forward.

Dwarkesh Patel 34:31 This is also why-

Trenton Bricken 34:32 Uh-

Dwarkesh Patel 34:32 ... I'm skeptical of big, grand schemes, like nationalization or some public-private partnership, or just generally shaking up the landscape too much, because I do think we've, we're in, like, one of the better... I mean, the sort of, like, the difficulty of whether it's alignment or whether it's some kind of deployment, um, safety risks, that is just, like, the nature of the universe is gonna make that some level of difficulty. But the human factors, in a lot of the counterfactual universes, I feel like we don't end up with people... Like, we could even be in a universe where they don't even pay lip service.

Trenton Bricken 35:05 Yeah.

Dwarkesh Patel 35:05 This is like, not an idea that anybody had.

Trenton Bricken 35:08 Yeah.

Dwarkesh Patel 35:08 They could have an ASI takeover. Um, I think to think we live in a pretty good counterfactual universe.

Trenton Bricken 35:12 Yeah, we got a good-

Dwarkesh Patel 35:13 All things considered.

Trenton Bricken 35:13 ... good set of game players on the board.

Dwarkesh Patel 35:14 That's right. That's right.

Trenton Bricken 35:16 How are you preparing for fast timelines?

Dwarkesh Patel 35:19 If there's fast timelines, then there will be this six-month period in which the most important decisions in human history are being made. And I feel like having an AI podcast during that time might be useful. Um, that's basically the plan.

Sholto Douglas 35:35 Have you made any shorter term decisions, um, with regards to, like, spending or health or anything else?

Dwarkesh Patel 35:43 After I interviewed Zuckerberg, my business bank balance was negative 23 cents. Uh, when the ad hi- money hit, I immediately reinvested it into NVIDIA. So, that is the... (laughs)

Sholto Douglas 35:56 (laughs)

Dwarkesh Patel 35:56 Um, sorry, but you were asking from a sort of altruistic perspective?

Sholto Douglas 36:00 No, no, just in general, like, have you, have you changed the way you live at all because of your AGI timelines?

Dwarkesh Patel 36:06 I never looked into getting a Roth IRA. (laughs)

Trenton Bricken 36:10 He brought us Fiji water before. (laughs)

Sholto Douglas 36:13 (laughs)

Dwarkesh Patel 36:13 (laughs)

Sholto Douglas 36:13 Which was in plastic bottles, so...

Trenton Bricken 36:14 Dwarkesh has changed. (laughs)

Sholto Douglas 36:16 (laughs)

Dwarkesh Patel 36:16 (laughs) Well, have you guys changed your lifestyle as a result?

Trenton Bricken 36:21 Not really, no. I, I just, like, work all the time.

Dwarkesh Patel 36:24 (laughs)

Sholto Douglas 36:25 (laughs)

Dwarkesh Patel 36:25 But you were doing that anyways, or would you not?

Trenton Bricken 36:26 Yeah. (laughs)

Dwarkesh Patel 36:26 Ah...

Trenton Bricken 36:30 I would probably be going very intensely at whatever, like, thing I'd picked to devote myself to.

Dwarkesh Patel 36:34 Yeah.

Trenton Bricken 36:34 Yeah.

Dwarkesh Patel 36:35 How about you?

Sholto Douglas 36:35 I canceled my 401K contributions earlier.

Dwarkesh Patel 36:37 (laughs) Oh, really?

Sholto Douglas 36:38 Yeah. Yeah, that, that felt like a, a more serious one. Um, it's, it's hard for me to imagine a world in which I'm, like, s- have all this money that's just sitting in this account and waiting till I'm 60 and things look so different then.

Dwarkesh Patel 36:52 But that, you could be like a trillionaire with your, uh, marginal 401K contribution.

Sholto Douglas 36:56 I, I guess, but you also can't invest it in, like, specific things.

Dwarkesh Patel 37:00 That's right.

Sholto Douglas 37:00 And, um, I don't know. I might change my mind in the future and, and, and can restart it. Um, and I've been contributing for a few years now.

Dwarkesh Patel 37:08 On, on, on a more serious note, one thing I have been thinking about is-How could you use this money to an altruistic end? And basically, if there's somebody who's up and coming, eh, uh, in the field that I know, which is, like, making content, could I use money to support them? And I'm of two minds on this. One, there are people who did this for me.

Trenton Bricken 37:27 Mm.

Dwarkesh Patel 37:27 And it was kind of actually responsible for me continuing to do the podcast-

Trenton Bricken 37:30 Mm.

Dwarkesh Patel 37:30 ... when it just, like, did not make sense as a ... there were, like, a hun- a couple hundred people listening or something. Um, esp- uh, wanna, uh, shout out Anil Varanasi-

Trenton Bricken 37:38 Mm.

Dwarkesh Patel 37:38 ... for doing this. And also Leopold, actually, for the-

Trenton Bricken 37:41 Mm.

Dwarkesh Patel 37:41 ... foundation he was previously running. On the other hand, it's n- it's like, I feel like I would n- it's the thing about what's, that blogger was saying, that the good ones you actually do notice. It's like, it's hard to find a hidden talent. Maybe I'm totally wrong about this. But I'd feel like if you, if I put up a sort of grant application, I give you money if you're trying to make a blog, I'm actually not sure about how well that would work.

Trenton Bricken 38:04 There's different things you could do, though. Like, there's, I'll give you money to move to San Francisco for two months. Uh, and, and, like, you know, sort of meet people and, and, like-

Dwarkesh Patel 38:12 Mm-hmm.

Trenton Bricken 38:12 ... sort of get more, like, context and taste and, like-

Dwarkesh Patel 38:14 That's right. That's right.

Trenton Bricken 38:14 ... feedback on what you're doing. And, like, uh, it's not so much about the money or time, it's, like, it's putting them in an environment where they can more rapidly grow. Like, that's something that one could do. Um, I mean, you also, I think you do that, like, quite proactively in terms of you, like, deliberately introduce people that you think will be interesting to each other and this kinda stuff, so.

Dwarkesh Patel 38:31 Yeah.

Trenton Bricken 38:31 Yeah.

Dwarkesh Patel 38:32 No, I mean, that's very fair. And I, obviously, I've benefited a ton from moving to San Francisco.

Trenton Bricken 38:36 Yes.

Dwarkesh Patel 38:36 And it's unlikely that I would be doing the podcast, at least on AI-

Trenton Bricken 38:39 Yes.

Dwarkesh Patel 38:39 ... to the degree I am if I wasn't here. So maybe it's a mistake to judge people based on the quality of their content as it exists now, and just throw money at them. N- not throw money, but give them enough money to move to SF, to get caught up in this intellectual man- milieu, and then, um, maybe do something interesting as a result.

Trenton Bricken 38:56 Yeah.

Sholto Douglas 38:56 Yeah. The, the thing that most readily comes to mind is the MATS program for-

Dwarkesh Patel 38:59 Mm-hmm.

Sholto Douglas 38:59 ... AI research. And this seems like it's just been incredibly successful at giving people the time, the funding-

Dwarkesh Patel 39:06 Yep.

Sholto Douglas 39:07 ... and the j- like, social status justification, uh, to do AI safety relevant research-

Dwarkesh Patel 39:15 Right.

Sholto Douglas 39:15 ... with mentors.

Dwarkesh Patel 39:16 Oh, and you-

Sholto Douglas 39:16 Um-

Dwarkesh Patel 39:16 ... you have a similar program. Uh-

Sholto Douglas 39:17 We have the Anthropic Fellows Program.

Dwarkesh Patel 39:19 That's right, yeah.

Sholto Douglas 39:19 Yeah.

Dwarkesh Patel 39:19 And then what, what is your, um ... I know you're probably selecting for a slightly different thing, but I assume it's gonna be power law dominated. And have you noticed a pattern among the, even if it's the MA- MATS fellows or your fellows who is just like, "Th- this was m- this made the whole thing worth it."

Sholto Douglas 39:33 Yeah.

Dwarkesh Patel 39:33 What characteristic helps them?

Sholto Douglas 39:34 I mean, there have been multiple people who Anthropic and other labs have hired out of this program.

Dwarkesh Patel 39:38 Yeah, yeah.

Sholto Douglas 39:39 So, I, I, I think the return on investment for it has been massive.

Dwarkesh Patel 39:42 Mm-hmm.

Sholto Douglas 39:42 Um, and yeah, apparently the fellows, I think there are 20 of them, are like really good.

Dwarkesh Patel 39:46 Oh, what is the trick to making it work well, or finding that one person?

Sholto Douglas 39:49 I think it's gotten much better with time, where the early fellows, some of them did good work and got good jobs. And so then now later fellows, like, the quality bar has just risen and risen and risen, and there are even better mentors now than before.

Dwarkesh Patel 40:04 Got it. Mm-hmm.

Sholto Douglas 40:04 Uh, so it's this really cool flywheel effect. Um, but originally, it was just people who, like, didn't have the funding or time to, like, make a name for themselves or, or do ambitious work. So, it was kind of like giving them that niche to do it.

Dwarkesh Patel 40:20 Right. Right, right, right.

Sholto Douglas 40:20 Uh, seems really key.

Dwarkesh Patel 40:21 Yeah.

Trenton Bricken 40:22 You can do other things. There doesn't have to be money. You know, like, you could, uh, put out ideas for things you'd be really interested in reading-

Dwarkesh Patel 40:28 That's right.

Trenton Bricken 40:28 ... or, like, promoting, right?

Dwarkesh Patel 40:29 Yeah, yeah, yeah. Um, there's something coming there.

Trenton Bricken 40:32 Okay, there we go.

Dwarkesh Patel 40:33 So, if th- this episode, hopefully will launch Tuesday at the same time as the book, by the way, which you can get at stripe.press/scaling. But on Wednesday, which is the day after, hopefully there's something useful for you here.

Trenton Bricken 40:45 Okay, exciting.

Dwarkesh Patel 40:46 Cool.

Sholto Douglas 40:46 Yeah.

Trenton Bricken 40:47 Any other questions we wanna ask?

Dwarkesh Patel 40:50 The, um, the thing I have takes on, which I rarely get asked about, is distribution.

Sholto Douglas 40:54 Distribution of AI?

Dwarkesh Patel 40:55 No, no, sorry. (laughs) Like MrBeast-style d- distribution.

Sholto Douglas 40:58 Oh, yeah.

Dwarkesh Patel 41:00 Um.

Sholto Douglas 41:00 Yeah, yeah, yeah.

Dwarkesh Patel 41:00 Um, where people, I think, rightly focus on the content. Um, and if that's not up to snuff, I think you ju- you won't succeed. But to the extent that somebody is trying to do similar things, the thing they consistently underrate is putting the ... yeah, the, putting the time into getting distribution right. Um, I just take, like, random takes about, for example, the most successful thing for my podcast in terms of growth has been YouTube Shorts.

Trenton Bricken 41:30 Mm-hmm.

Dwarkesh Patel 41:30 It's a thing you would never have predicted beforehand. And, you know, they're, like, responsible for, like, basically m- at least half the growth of the podcast or something. Uh-

Trenton Bricken 41:41 I mean, I'd buy that.

Dwarkesh Patel 41:42 Yeah.

Trenton Bricken 41:42 Why wouldn't you have predicted it? Like, I mean, like, I mean, I guess there's the contrast of, like, the long form deep content-

Dwarkesh Patel 41:48 Mm-hmm.

Trenton Bricken 41:48 ... and, like, YouTube Shorts and stuff. But I do definitely think they're, they're good hooks.

Dwarkesh Patel 41:52 That's right.

Trenton Bricken 41:52 Good content.

Dwarkesh Patel 41:52 Yeah, yeah.

Trenton Bricken 41:53 Yeah.

Dwarkesh Patel 41:53 And I have, like, takes on how to write tweets and stuff. The main intuition being-

Trenton Bricken 41:56 Yeah.

Dwarkesh Patel 41:56 ... like, write like you're writing to a group chat.

Trenton Bricken 41:58 Yeah.

Dwarkesh Patel 41:58 Um-

Trenton Bricken 41:59 Yeah.

Dwarkesh Patel 41:59 ... to a group chat of your friends rather than this, like, formal whatever. I don't know, I, I, I could ... but just, like, these sort of, like ...

Sholto Douglas 42:06 Yeah, no, I mean, what el- what else comes to mind here?

Trenton Bricken 42:08 Well, maybe it's interesting, the difference between, like, TikTok and YouTube Shorts.

Dwarkesh Patel 42:11 Oh yeah, we've never cracked TikTok.

Trenton Bricken 42:12 Yeah.

Sholto Douglas 42:13 Why not? Like, you've tried?

Dwarkesh Patel 42:15 Yeah. I mean, we try-

Trenton Bricken 42:16 Tried, have you done everything? (laughs)

Dwarkesh Patel 42:17 (laughs) No, I have not done everything.

Sholto Douglas 42:18 If you read these poems.

Trenton Bricken 42:20 Yeah. (laughs)

Dwarkesh Patel 42:23 (laughs)

Sholto Douglas 42:23 (laughs)

Trenton Bricken 42:23 (laughs)

Sholto Douglas 42:23 Maybe you're, like, in a bubble bath with, like, some beard shampoo on. (laughs)

Trenton Bricken 42:27 (laughs) Reading poems? That'd be an incredible video. I bet you that would go viral. (laughs) You have to do that now. (laughs)

Sholto Douglas 42:35 (laughs) Manifest.

Dwarkesh Patel 42:36 (laughs) Reading a poem, uncross your legs.

Trenton Bricken 42:38 Yeah. (laughs)

Sholto Douglas 42:39 (laughs) Last one, last, last episode it was the interpretability challenge-

Trenton Bricken 42:41 (laughs)

Sholto Douglas 42:41 ... now it's George Mesh in a bubble bath. (laughs)

Dwarkesh Patel 42:44 I gotta sell the book somehow, you know? (laughs)

Sholto Douglas 42:47 (laughs) Wait, literally do it like Margot Romney-

Trenton Bricken 42:49 Yeah, exactly. Explaining the CDI as it's

Narrator 42:51 (laughs)

Dwarkesh Patel 42:51 Yeah, yeah, yeah.

Trenton Bricken 42:52 ... shot.

Sholto Douglas 42:52 Yeah. (laughs)

Trenton Bricken 42:52 So why is-

Dwarkesh Patel 42:53 Scaling. (laughs)

Sholto Douglas 42:58 (laughs)

Trenton Bricken 42:58 (laughs)

Dwarkesh Patel 42:58 Um, and that's how you crack distribution. (laughs)

Trenton Bricken 43:00 And that's how you crack distribution. (laughs)

Sholto Douglas 43:03 (laughs) Oh. But yeah, no, like, when we did our episode, it launched and you were sharing interesting tidbits about how it was doing and the thumbnail you wanted-

Dwarkesh Patel 43:14 Yeah.

Sholto Douglas 43:14 ... to use and the title, and-I think I even asked you, like, to share more details, 'cause y- it seemed interesting and cool-

Dwarkesh Patel 43:21 Yeah.

Sholto Douglas 43:22 ... and subtle things. Um, but it seemed like you also kinda just hated it.

Dwarkesh Patel 43:25 (laughs)

Sholto Douglas 43:25 Like playing this game-

Dwarkesh Patel 43:26 Oh, yeah.

Sholto Douglas 43:26 ... of like really having to optimize all these knobs. So like-

Dwarkesh Patel 43:29 I, what I, what I realized, I mean, talent is everything, so having, I'm, I'm really lucky to have, um, uh, three to four editors who I'm just, uh, like incredibly prou- proud, uh, to work with. I don't know how to hire more of them. Like they're just so good and self-directed. So honestly, I don't have the tips to, how to correct that. I hired those guys. So one of them was a farmer in Argentina. One of them was a freshman master's student in Sri Lanka. One of them, uh, was a former editor for one of MrBeast's channels. The other is a director in Czechoslovakia who makes I- these AI animations that you've seen in, um, the Notes on China, and he's working on more essays like that.

Sholto Douglas 44:06 Mm-hmm.

Dwarkesh Patel 44:06 So, um, I don't know how to replicate that catch again. (laughs)

Sholto Douglas 44:10 God, that's a pretty widely cost net, I'm gonna be honest.

Dwarkesh Patel 44:13 Yeah. Uh, but they're all, like, goddamn, they're so good.

Sholto Douglas 44:16 And this was just through your challenges and just tweeting about.

Dwarkesh Patel 44:18 That's right. Yeah.

Sholto Douglas 44:19 Yeah.

Dwarkesh Patel 44:19 That's right. I, I had a competition to make clips in my podcast. I rounded up a couple of them this way. Um, yeah. It's hard, it's hard to replicate because I've tried. Um-

Sholto Douglas 44:27 Yeah. Why do you think this worked so well with the video editors? 'Cause you tried a similar approach with your, uh, chief of staff.

Dwarkesh Patel 44:34 Yeah. The, the difference is, with the video editor, I think there is this arbitrage opportunity where there are people... It, it is fundamentally a sort of, are you willing to work hard and obsess about getting better over time, um, uh, which all of them go above and beyond on. But you can just find people in other countries who are, like, um... And it's not even about the wages. Like I've 10X'd their salaries or something like that. Um, it's, uh, just about getting somebody who is really data oriented. And there is this global arbitrage there. Whereas with the general manager... By the way, so the person I ended up hiring, and who I'm super excited to work with, is your childhood best friend.

Sholto Douglas 45:13 Yeah.

Dwarkesh Patel 45:13 Uh, Max Herranz.

Sholto Douglas 45:14 Max is so great.

Dwarkesh Patel 45:15 And he would have plenty of other opportunities. There's not this, like, weird arbitrage where, you know, you, you find some farmer in Argentina. (laughs) Um-

Sholto Douglas 45:22 But you know, it, it, it is striking that you were looking for a while-

Dwarkesh Patel 45:25 That's right.

Sholto Douglas 45:26 ... and then-

Dwarkesh Patel 45:26 Yeah, for like a year.

Sholto Douglas 45:26 ... just kinda mentioned offhand that Max was looking for something new.

Dwarkesh Patel 45:30 I genuinely... Uh, this is, this is gonna be, like, a total, um, uh, 12-year-old learns about the world kinda question, but I genuinely don't know how big companies hire. 'Cause I was trying to find this person for a year, and I'm really gl- uh, glad about the person I ended up hiring. But it was just like, if I needed to hire 100 people for a company, let alone 1,000 people, I just, like, do not know how to find people like this at scale.

Sholto Douglas 45:53 Yeah. I mean, I think this is, like, the number one issue that startup CEOs have. Hiring.

Dwarkesh Patel 45:57 Yeah.

Sholto Douglas 45:57 Like it's just relentlessly the number one...

Dwarkesh Patel 45:59 Yeah. And, and the thing I was stunned with is how it didn't seem like my platform helped that much.

Sholto Douglas 46:04 Mm-hmm.

Dwarkesh Patel 46:04 I got, like, close to 1,000 applications across the different rounds of publicizing it that I did. And a lot of, I think, really cool people applied, but the person that I ended up hiring was somebody who was just a reference, um, uh, you know, like a mutual friend kinda thing. Um, and a couple of other top contenders were also this way.

Sholto Douglas 46:24 Mm-hmm.

Dwarkesh Patel 46:24 So it's, it's weird. Like the best people in the world don't wanna apply, at least to things like this.

Sholto Douglas 46:29 Yeah.

Dwarkesh Patel 46:30 Um, and you just gotta seek them out. Even if you see- think you have a public platform or something.

Sholto Douglas 46:33 Yeah. Yeah, I mean, the, the job might just be so out of distribution-

Dwarkesh Patel 46:36 Mm-hmm. Yeah.

Sholto Douglas 46:37 ... from anything else that people would do.

Dwarkesh Patel 46:39 Maybe. That's right, yeah, yeah.

Sholto Douglas 46:41 Um, um, so ............................

Dwarkesh Patel 46:42 I did do writing.

Sholto Douglas 46:43 Ray asks, uh, "How do you make it on Substack as a newbie writer?"

Dwarkesh Patel 46:47 Okay. I think, um, if you're starting from scratch, there's two useful hacks. One is podcasting, because you don't need to have some super original new take. You can just interview people who do, and you can leverage their platform.

Sholto Douglas 47:01 Mm-hmm.

Dwarkesh Patel 47:01 And two is writing book reviews.

Sholto Douglas 47:03 Mm-hmm.

Dwarkesh Patel 47:04 Again, because you have something to react to rather than having to come up with a unique world view of your own. There's so many other things, and it's h- really hard to give advice in advance. Just try things. But, um, those, I think, are just, like, good, um, cold starts.

Sholto Douglas 47:16 The book reviews is a good suggestion. I actually use, like, Gwern's book reviews-

Dwarkesh Patel 47:19 That's right.

Sholto Douglas 47:19 ... as a way to recommend books to people.

Dwarkesh Patel 47:21 By the way, this is a totally under-supplied thing.

Sholto Douglas 47:23 Yeah.

Dwarkesh Patel 47:23 Because I...

Sholto Douglas 47:24 Yeah.

Dwarkesh Patel 47:24 If anybody has book reviews... Jason Furman is this economist-

Sholto Douglas 47:26 Yeah.

Dwarkesh Patel 47:26 ... who has, like, a thousand, you know, Good Reads reviews. And I, like, I probably have visited his Good Reads on a thou- a hundred independent visits.

Sholto Douglas 47:34 Wow.

Dwarkesh Patel 47:35 Same with the Gwern book reviews or something, right? So book reviews are a sort of very under-supplied thing, uh, if you're looking to get started making some kind of content.

Sholto Douglas 47:43 Hmm. I like that.

Dwarkesh Patel 47:44 Yeah. Cool. Thank you guys so much for doing this.

Sholto Douglas 47:47 Yeah, this was fun.

Dwarkesh Patel 47:48 We're, um-

Sholto Douglas 47:48 This was fun.

Dwarkesh Patel 47:48 Uh, uh, we'll turn the tables on you again pretty soon.

Sholto Douglas 47:51 Yeah.

Dwarkesh Patel 47:51 But, uh-

Sholto Douglas 47:52 How does it feel being in the hot seat?

Dwarkesh Patel 47:54 It, it, it's nice. (laughs)

Sholto Douglas 47:56 (laughs)

Dwarkesh Patel 47:59 Nobody ever asks me questions. (laughs)

Sholto Douglas 48:02 (laughs) Nobody ever asks how is Dwarkesh? (laughs)

Dwarkesh Patel 48:07 (laughs)

Sholto Douglas 48:08 Cool.

Dwarkesh Patel 48:08 Yeah, yeah, super excited for the book launch.

Sholto Douglas 48:10 Thank you. Thank you. Your website's awesome, by the way.

Dwarkesh Patel 48:11 I appreciate it.

Sholto Douglas 48:12 Oh yeah, the website's sick.

Dwarkesh Patel 48:13 Yes. Yeah. Uh, strife.press/scaling.

Sholto Douglas 48:18 Yeah.

Dwarkesh Patel 48:18 Um, cool.

Sholto Douglas 48:19 Cool.

Dwarkesh Patel 48:19 Thanks, guys.

Sholto Douglas 48:20 See you later. Thanks.

Dwarkesh Patel 48:21 Okay, I hope you enjoyed that episode. (instrumental music) So as we talked about, my new book is out. It's released with Stripe Press. It's called The Scaling Era, and it compiles the main insights across these last few years of doing these AI interviews. And I'm super pleased with how it turned out. It really elevates the conversations and adds necessary context. And just seeing them all together even reminded me of many of the most interesting segments and insights that I had myself forgotten. So I hope you check it out. Go to the link in the description below to buy it. Separately, I have a clips channel now on YouTube. Um, people keep complaining about the fact that I put clips and the main video on the same channel, so request granted. There is a new clips channel, but please do subscribe to it, um, so we can get it kickstarted. And while you're at it, also make sure to subscribe to the main channel. Um, other than that, just sh- honestly, the most helpful thing you can do is share the podcast. If you enjoyed it, just send it on Twitter, put it in your group chats, share it with whoever else who you think might enjoy it. Um, that's the most helpful thing. If you wanna learn more about advertising on future episodes, go to dwarkesh.com/advertise. Okay, see you on the next one.