New college graduates who grew up using AI tools are 'AI natives' with a massive advantage — they're already cracked on using these tools, while older graduates feel lost and lack agency.

(00:00:00) Okay, we are gathered here today in holy unity, brothers and sisters, to convene and discuss on this most holy day, the day the All-In podcast drops, many topics, AI data centers, China, justice, human dignity, Dari unwinding these SPVs hasn't been good for the Vatican. (00:00:27) We got in at 20 billion. (00:00:29) That was a fifty-bagger for us, so let's get started. (00:00:33) Jason, I'm pretty sure you believed you were the vicar of God before the encyclical, so this is nothing new for you. (00:00:39) We'll let your winners slide. (00:00:45) Rain Man David Sacks. (00:00:46) And it said we open sourced it to the fans and they've just gone crazy with it. (00:00:54) Love you, West. (00:00:57) The smoke has risen from Tramat's pool house and from the poker room. (00:01:04) He's staying in my pool house. (00:01:05) He's been there for the last three days. (00:01:07) It's been magnificent. (00:01:08) He didn't know. (00:01:09) You know what? (00:01:10) I understand where OJ was coming from. (00:01:13) You know, you put kiddo Kaleb in your house for long enough. (00:01:16) You just lose your shit at some point. (00:01:18) At some point, somebody's getting whacked. (00:01:21) All right, enough with the shenanigans. (00:01:23) But it's been great staying at the house because (00:01:26) There's actually, Chamantha is not aware of this, there's an iPad in the kitchen, and that's logged in to Uber Eats, DoorDash, Instacart, Amazon, Laura Piana. (00:01:35) Shut the fuck up. (00:01:36) Come on, stop. (00:01:37) No, there is. (00:01:37) It's literally every single service. (00:01:40) And I told the house manager, like, listen, any packages that come in the next 72 hours, right to the pool house, if it says JCow, right to the pool house. (00:01:50) So all these packages have been coming. (00:01:52) Then I relabeled them, gave them back, sent them to the ranch. (00:01:56) And now the house manager is sending that stuff to the ranch. (00:01:59) Laura Piano wants to know why my inseam went from 36 to 12. (00:02:02) Your waist size went from 32 to 36. (00:02:07) All right, welcome to the program, everybody. (00:02:10) David Sacks is here. (00:02:11) How are you doing, David? (00:02:12) I'm good. (00:02:13) Chamath Palihapitiya is back at the 80-90 office. (00:02:18) I was at the 80-90 office the last couple of days, and it's a vibe. (00:02:22) It's A vibe. (00:02:23) It's like a break culture going on. (00:02:25) If you're a bestie and you show up at that office, though, everybody there is a huge fan of the pod. (00:02:31) So I was like, it was like being royalty. (00:02:33) It's stopped by everybody. (00:02:35) Hey, I'm a developer here. (00:02:36) I'm a big fan of the show. (00:02:37) Thank you for giving it to Chamath. (00:02:39) We can't give it to him because he pays our mortgage and everything. (00:02:42) But every time you stick it to Chamath, we love it. (00:02:44) We're cheering for you in the secret Slack room. (00:02:48) There's a secret slack room. (00:02:51) There is. (00:02:51) There is definitely a secret slack room going on. (00:02:54) No, but it was great. (00:02:55) The vibes were awesome. (00:02:56) You're building a lot of software, a lot of young talent. (00:02:59) I don't want to say that where your secret source is, but there's a secret source of talent you have. (00:03:03) And man, those are some smart kids. (00:03:08) I'm happy to say it. (00:03:08) Look, when I was at Facebook, we became the most aggressive recruiter of Waterloo co-ops. (00:03:13) And so I went back to the well. (00:03:16) Yeah, we recruit more interns every quarter than we have full-time engineers, which we do on purpose because it puts a ton of pressure on the product actually being good. (00:03:27) We had 400 people apply this quarter for internships. (00:03:30) Wow. (00:03:30) It's very interesting with us sitting in for Friedberg, who's busy with some potato seed this week, doing great stuff at Ohalo. (00:03:40) The one, the only. (00:03:42) Bill Gurley is here. (00:03:44) He's been running down a dream. (00:03:45) If you haven't bought the book, get the book. (00:03:47) It's incredible. (00:03:48) And you're off book tour, so now you have time for us, yeah? (00:03:51) Yes. (00:03:52) And I, you know, I had told you, if you ever talk about the Pope, I'd love to hop on. (00:03:57) Yes, you were like. (00:03:59) Well, you know, Bill's an evangelical. (00:04:02) I'm A Catholic, so we do have some common ground here. (00:04:06) Do you, when's the last time you were at church, Bill? (00:04:08) Have you thought now that you're... (00:04:11) when J. (00:04:11) Cal gets sacrilegious, I got to come on there and make sure J. (00:04:18) Cal doesn't get out of line with the Pope. (00:04:21) Listen, the Pope is God's messenger on earth. (00:04:25) We should give him a base level of respect. (00:04:28) By the way, that's us imitating Bill Gurley. (00:04:30) It's not actually Bill Gurley. (00:04:31) Just for those of you listening, you think they were confused. (00:04:34) They were. (00:04:35) Literally they were confused. (00:04:37) We don't want to put words in your mouth. (00:04:39) But just a point of clarification. (00:04:40) But hey, everybody knows you were, you know, you handed the baton over at a benchmark after a very successful couple of decades. (00:04:50) in venture capital. (00:04:50) You wrote the book. (00:04:52) You've now got a nonprofit. (00:04:53) You're doing your own spin, I think, on maybe what Peter Thiel does with his fellowship. (00:04:59) You started your own Running Down a Dream Fellowship, I understand. (00:05:02) Yeah, it's targeted at a different demographic. (00:05:05) It's called rdad.org, runningdownadream.org. (00:05:08) There's a website we can give a link to. (00:05:11) We're going to do $5,000 grants to people who want to chase their dreams but need some help. (00:05:17) And so there is an application process like (00:05:20) Teal Fellows and other programs, and we've been out talking to those people. (00:05:24) But we're actually live. (00:05:26) We went live last week for the application. (00:05:28) So if you know people who have read the book and are inspired and need some help, have them apply. (00:05:34) Yeah, all right, folks. (00:05:35) Good for you, BG. (00:05:37) Yeah, it's great. (00:05:38) Me too. (00:05:39) Two other, so I did a TED Talk, which will come out soon, that's related to the book. (00:05:46) There's a professor in Miami that's built a course around the book, which I'm excited (00:05:50) about. (00:05:50) And he's doing it in a kind of an open source way so that other people can borrow that as well. (00:05:56) And so if there's anyone out there, I'd love to help them do that. (00:06:00) What's your take on all of this doomerism? (00:06:02) Like if you're a young person and you're in college or you're in high school, is this (00:06:07) Is this much ado about nothing, or how do you run down a dream in the face of something like that? (00:06:11) Yeah, well, I started the book before this happened, and I've been asked the question a lot, and it came up in the TED Talk. (00:06:19) I fear that a lot of people are in jobs, they actually don't care about that much. (00:06:25) And there's a Gallup poll that backs this up. (00:06:28) They came up with that word quiet quitters. (00:06:31) They're like 59% of the people they surveyed are kind of ambivalent about their job. (00:06:36) And when you're ambivalent about your (00:06:37) your job, you're not high agency, and so you don't lean in. (00:06:41) If you look at how Jason talks about how they implemented AI in all of his different working groups, you hear that enthusiasm and that high agency, and then you want to go try these things. (00:06:56) And I think the best way to protect yourself from AI is to be the most AI-enabled version of yourself you can be. (00:07:04) But if you're ambivalent about your job, you're probably not doing that. (00:07:07) And you could be a sitting duck. (00:07:10) So I think it's the mindset that's the problem. (00:07:13) I created an associate in training program for my firm, because we want to help people get into venture capital. (00:07:19) And we gave them a choice of assignments. (00:07:21) One of them was to write coverage of one of our portfolio companies that's breaking out, Micro One is the name of it. (00:07:26) And just give us like, hey, here's a competitive landscape. (00:07:28) Basically write a deal memo and coverage of that company. (00:07:31) And then we give them another option to vibe code a very specific project I've wanted to have for our venture firm for a long time on competitive intelligence. (00:07:41) And I would say, I think like maybe 80% of the students applying, and we had 400 or 500 people apply for six positions, (00:07:49) 80% of them did the vibe coding. (00:07:52) And I was shocked. (00:07:53) I thought it would be the exact opposite. (00:07:54) Anybody can write, anybody can throw shit in ChatGPT and get some output, but they actually built software. (00:08:00) And that's the scary thing. (00:08:02) The students who graduated, at least this is my perception, Chamath, the students who graduated like 5, 10 years ago before AI, (00:08:10) They're not AI first. (00:08:11) They feel lost in a drift. (00:08:13) They don't have agency. (00:08:14) But the group coming out of college right now that cheated their way through school using ChatGPT, doing their assignments, like using those tools, I'm joking, cheating, but I mean hacking. (00:08:24) I agree with that. (00:08:25) So they're totally cracked. (00:08:27) Yeah. (00:08:27) And they're just like, I know how to use these tools to get through my finals. (00:08:31) Gurley, I think you're saying something super important. (00:08:33) I said this last week, which is nobody asks the warehouse worker at Amazon whether they actually want that job. (00:08:40) And so to your point, job satisfaction isn't some external person judging your job to be valid and saying you must be able to have it. (00:08:48) I think it should be asking the person that does the job, do you like it and do you want to keep it? (00:08:52) Those are two very different questions. (00:08:54) And I think that all of this AI doom and gloom was a lot... (00:09:00) and too much, frankly, of the former and not enough of the latter. (00:09:03) And now this whole lie is kind of getting undone. (00:09:07) I think Sachs posted about it this week as well. (00:09:09) The Goldman Sachs CEO said it. (00:09:11) And now in this crazy twist of fate, now that we need to have trillion dollar IPOs, the entire Frontier Labs are all like, wow, it's going to be a bonanza of jumps. (00:09:19) Mark Cuban had a great quote. (00:09:21) He said, there are two types of people in the world, those that use AI to learn faster than they ever could before and those that use AI to avoid learning altogether. (00:09:31) And I think it's this notion of high agency or not. (00:09:34) That's pretty good. (00:09:35) Are you leaning in and using this stuff to be ever more powerful in what you try and accomplish? (00:09:41) Are you using it as a cheat code? (00:09:44) And if you're in the latter, yeah, you're probably at risk. (00:09:47) You get asked a lot about how to educate yourself if you're a parent of kids, so that you can put them on a path to launch and do well and chase their dreams. (00:09:57) You have a good answer for that question? (00:09:59) I mean, the second chapter of the book is all about lifetime learning. (00:10:03) And it's kind of a requirement that you're following your fascination because the lifetime learning comes for free if you're fascinated with something. (00:10:10) Like you just constantly soak up and devour new information. (00:10:15) And I do think that a lot of kids get exhausted because we've made high school and college such a grind that they think the learning ends the day that they walk out with their diploma. (00:10:26) And (00:10:27) as we all know, the best and brightest in all of our fields are on a constant learning journey. (00:10:33) And when something new comes out, they dive in and try and figure it out, right? (00:10:37) And so, and every single person in the book that we profiled has that kind of attitude about their craft, and every day. (00:10:47) And so I think the real test is if you're not proactively self-learning, then you're probably not tilting against something that you really adore and are fascinated by. (00:10:57) Max, you wanted to jump in there. (00:10:59) With respect to new college grads, I was going to say that I think the single most marketable skill in the economy right now has got to be proficiency in Claude. (00:11:08) If you're going into a firm right now and you're the only one who knows Claude, it would be like you're the only one who knows how to work a spreadsheet or word processor. (00:11:18) The advantage (00:11:19) would be enormous. (00:11:21) Now, I think that that's probably a short-term arbitrage because eventually everyone's going to have to figure out how to use these tools. (00:11:26) But as a young college graduate right now, you have such an advantage if you're an AI native, just knowing how to use these tools. (00:11:34) And this thought partially occurred to me when I saw what our producer, Nick, has been doing. (00:11:41) with using Claude for, he's been creating this daily briefing document. (00:11:46) We've been doing it for, we've been doing it for three months. (00:11:48) I just ran it for the first time. (00:11:53) Apparently I've never seen this. (00:11:54) Well, I just, you know, I thought it would just be AI slop and it would just kind of give me a roundup of news that I was getting in my ex feed anyway. (00:12:02) But actually the thing that was really impressive about it (00:12:06) was that it predicted topics that I would specifically be interested in based on my previous comments on the pod. (00:12:14) And also, it went back and looked at previous transcripts and what I had said, and then (00:12:22) had updates to those topics based on specific things I had said. (00:12:27) So again, it was highly, highly contextual. (00:12:30) But then I asked Nick, how do you generate that? (00:12:32) And he showed me the custom prompt that he designed for Claude and then the skills document. (00:12:38) And they were very long and detailed documents. (00:12:40) They weren't written in code, but they were very technical. (00:12:43) And I just realized looking at that, the average person is not going to be able to generate this. (00:12:46) I mean, this is why this idea that you're just going to be able to like throw AI into an organization and this is magically going to generate value is not true. (00:12:54) You have to know how to get value out of it. (00:12:57) I mean, maybe we could just show these documents on the screen. (00:12:59) Yeah, I mean, the interesting thing, Sax, is you can, you just have to ask your AI, you ask Claude or ChatGPT or whatever you're using. (00:13:10) Hey, I want you to make me a mega prompt and you like a mega pint. (00:13:15) Give me a mega prompt of you're a producer of a podcast. (00:13:19) These are the four characters on the podcast. (00:13:21) What would be a great prompt for me? (00:13:22) And it will actually suggest a prompt and then you can refine the prompt. (00:13:26) So you actually have a dialogue about a prompt as opposed to writing the prompt yourself. (00:13:31) And so I've started doing this and it is extraordinary. (00:13:34) Like. (00:13:35) Well, Nick, can you show on the screen to scroll through the training rules? (00:13:39) And then also there's the skills document that was written on how to be a producer for this podcast, which I thought was really impressive. (00:13:48) By the way, David, what you said, I think is true of almost every single job type. (00:13:53) Like it's not just tech or programming. (00:13:56) If you're in marketing, if you're in legal, if you're in accounting, like any role you might have at a firm, sales, if you're the most AI savvy person of all your peers, (00:14:09) You are golden. (00:14:10) Like golden are golden. (00:14:12) Like in your company. (00:14:12) You're 10x more valuable than the next person who's not, basically. (00:14:16) Yes, And I think that, I don't think it goes away because I think you learn how to get better at it over time. (00:14:24) So having an early advantage, I think, will extend for a while because you can learn more and more things you can accomplish. (00:14:32) Should we let producer Nick describe what we were just looking at? (00:14:35) Yeah, go ahead, produce. (00:14:36) Producer Nick, explain the process. (00:14:38) Yeah. (00:14:39) Once we got access to Claude Cowork and it had that like further expanded memory access, I thought it would be interesting to just start feeding every transcript into it and seeing if it could actually contextualize new stories that were coming out based on past things that you guys have said. (00:14:55) And I gave it like a general prompt of what I wanted. (00:14:57) And I said, how would you write a skills file or some training rules for this? (00:15:01) And it wrote all of it for me. (00:15:03) Yeah. (00:15:04) Oh, so you were less good than I thought. (00:15:05) Yeah. (00:15:08) No, it's a hack. (00:15:09) It's a hack. (00:15:09) You use AI to make the skills. (00:15:12) And you've been updating that over time, right? (00:15:14) As you've been iterating and learning. (00:15:15) Every single day, and every single day gets smarter and better. (00:15:18) The recursiveness of this is incredible. (00:15:20) So you need someone to manage that process, right? (00:15:22) Because the four besties are not going to do that. (00:15:25) So you need a producer of the show to do that. (00:15:26) This is why people will think, oh, this is going to wipe out all the jobs. (00:15:29) No, someone still has to supervise, iterate, validate, you know, all those kinds of things. (00:15:35) Yeah. (00:15:36) And it's. (00:15:39) It's really interesting. (00:15:40) The people who are coming into the workforce right now are super aware of this, and they're putting the tools to work, and it's much easier for them to get a job. (00:15:48) I mean, I literally looked at the top nine candidates for this associated training program I have, and we're going to do it every year. (00:15:56) Every summer we start it, we do it for a year. (00:15:57) We pay you to learn. (00:15:59) And (00:16:01) It was just extraordinary how you could tell immediately if the person had systems thinking sacks, like they understood the process of venture capital, that there was a structure to it. (00:16:12) You had to source deals, you had to make decisions on which ones to invest in, you had to do diligence, you had to double down on investments. (00:16:19) They just understood the process. (00:16:21) And then if you just talk to one of these LLMs, it will tell you what to do. (00:16:27) So you can say, I don't know what I'm doing. (00:16:29) What should I do next? (00:16:30) And then it actually tells you what to do next. (00:16:32) So for people who are intimidated about this and maybe think like, I'm already too far behind. (00:16:39) I encourage you to pop up Claude, go into co-work and say, what can I do to be better at my job and just start talking? (00:16:46) And literally, the more you talk and you can use voice, you know, text to voice, I use WhisperFlow is a really cool program for this, and I have a foot pedal to do it. (00:16:55) You just ramble and ramble and ramble and keep adding stuff. (00:16:58) You don't have to be structured. (00:17:00) It will build the structure around (00:17:02) The two or three paragraphs that you give it as instructions, that's the thing people are getting caught up on now is, Bill, they think they have to type, when in fact, if you just blather on and on, a skill I have a unique ability to do, you just blather on. (00:17:20) It's a superpower. (00:17:21) You blather on and the thing makes sense of it. (00:17:23) is unbelievable what the blather on prompt can get in terms of output. (00:17:28) Thanks for coming. (00:17:30) to my TED Talk. (00:17:33) All right, let's get started. (00:17:34) There's a lot to talk about. (00:17:35) And we got a big docket today. (00:17:38) We're going to start with the Pope. (00:17:40) The Pope is dope. (00:17:41) And the Pope, Leo, he's the 14th, released his first encyclical, encyclical on AI. (00:17:50) And it was long, 235 pages, over 42,000 words, (00:17:56) When did he write it, do you think? (00:17:58) When did he put that together? (00:18:00) Well, no, I think he used ChatGPT. (00:18:01) That's what it says here. (00:18:04) No, I mean, I'm guessing. (00:18:05) How long did it take for him to write this in between all of his other tasks? (00:18:09) I think it's a six-month process to do this, but I'm sure he had collaborators. (00:18:12) Bill, your book, I'm assuming, was 60 words. (00:18:14) He didn't write it. (00:18:16) I'm sure there was a team that wrote it, but Bill, your book's 60, 70,000 words, I'm guessing. (00:18:20) So this is almost a literal book, right? (00:18:24) In terms of (00:18:25) how long it is, and it's called Magnifica Humanitas, or Magnificent Humanity. (00:18:32) In it, he warns business leaders to safeguard humanity from AI. (00:18:36) His core argument is AI is not inherently evil, but technology is never neutral, and that technology takes on the characteristics, wait for it, of those who build, finance, and control it. (00:18:48) And I don't think he thinks super highly of that group of people, (00:18:52) The Pope called for regulation of AI companies. (00:18:55) Obviously, we're going to have that debate here. (00:18:58) Some of the things he called for, I think, are not very debatable. (00:19:03) And there's a lot of consensus around worker retrainment, safety for children and guardrails, a ban on autonomous weapons. (00:19:11) That's the Skynet rule. (00:19:12) Don't build terminators with your AI. (00:19:15) But he was joined by Anthropic co-founder Chris Ola. (00:19:19) I don't know how many co-founders there are of this company, but apparently there's dozens. (00:19:24) And Ola is not Catholic. (00:19:26) According to a Vanity Fair profile, he was raised evangelical, and now he's an atheist. (00:19:32) The folks at Amazon, Google, and Meta lobbied the Vatican on April 29th to soften the language in his missive, and he was not swayed. (00:19:41) His central question, Sacks, is will AI be used to concentrate power in the hands of a few, or will it serve everyone? (00:19:49) Something you brought up when you mentioned monopolies, duopolies, et cetera, 2 weeks ago on this very podcast. (00:19:57) What's your take on the Pope and his (00:20:00) interest and his missives on AI and promoting a bit of AI regulation. (00:20:07) Well, I very much agree with the Pope that the biggest risk of AI is a centralization of power and then it's misuse against us in some Orwellian way. (00:20:19) I think it's government that's going to do that, not necessarily an individual actor because it's governments that ultimately have the power. (00:20:27) So I do worry about the potential for AI to be used to surveil us, censor us, control us, as Orwell described in 1984. (00:20:37) So if that's where the Pope is going with this, I very much agree with him. (00:20:42) Maybe where we end up in different places is he thinks that government regulation is the way to prevent this. (00:20:48) And I would just say that we have to be careful not to empower government too much, because if you give government the power to (00:20:56) regulate or approve AI development, if you create, say, an FDA for AI, as many people are calling on, that will give government the power to approve models and therefore give notes to model developers. (00:21:11) And very soon, this definition of safety (00:21:15) will expand because the government always takes an expansive view of its powers. (00:21:19) And we saw this during the social media wars where the definition of trust and safety expanded to issues like psychological safety, microaggressions, disinformation, transphobia, and so on that, you know, again, these social media companies were told that they had to stamp out all of those threats to safety and it ended up becoming a censorship agenda. (00:21:43) So I get very worried about, you know, what (00:21:45) if some government agency can give notes to the model developers and they start telling the model developers that your definition of safety is not expansive enough, you have to, again, protect the public from disinformation or psychological harms. (00:22:00) So again, I think we just have to be careful not to aggrandize government because that's going to be the most likely culprit in terms of the centralization of power. (00:22:10) And (00:22:11) I know the Vatican likes Latin. (00:22:13) This is a problem of political philosophy that goes all the way back to Socrates. (00:22:17) It's called quis custodiet ipsos custodes, which is who will guard the guardians. (00:22:23) In other words, if we entrust a set of guardians to protect us from a bunch of threats, what's to stop them from becoming tyrannical and from becoming the new threat against us? (00:22:36) And I mean, this is the central dilemma of political power. (00:22:39) Who watches the watchers? (00:22:41) Yeah, who watches the watchers or who guards the guardians, meaning who's going to protect us against our guardians if they turn against us? (00:22:48) The genius of the American founding, by the way, is that it was a second order solution to this question. (00:22:53) The founders (00:22:55) of America very much understood this. (00:22:56) And what they came up with is we have to have the guardians guard against each other. (00:23:01) And so they came up with the idea of separation of powers. (00:23:04) We'd have separation of federal and state. (00:23:06) We'd have the three branches of the government, even within the legislative branch, it was a bicameral legislature. (00:23:12) So they divided up the powers in a way that hopefully the guardians would check against each other as opposed to becoming tyrannical against us. (00:23:22) And that is kind of my view on AI is that ultimately we have to have a solution of checks and balances. (00:23:29) If the AI market becomes monopolized and falls into the hands of one or two companies, I would use antitrust law very aggressively as a check and balance against their power. (00:23:40) Right now, we have a very competitive market. (00:23:43) We have 5 frontier labs competing very aggressively. (00:23:46) As long as the market is competitive, I would use that because I think competition generates the best outcomes. (00:23:52) It helps us win against China, but it also protects the population because these companies, if they get out of line, there's some competitor that can offer something better. (00:24:01) Consumers can opt out of it. (00:24:02) If they don't trust ChatGPT, they can use Anthropic, or if they don't trust Anthropic, they can go to Grok. (00:24:07) Bill, you had the number one rated (00:24:10) did talk at the All-In Summit in history, 2,851 miles. (00:24:16) You have been famously against regulatory capture. (00:24:20) In light of the Pope's comments of, hey, regulating, what do you think is common sense? (00:24:24) Because AI is everything. (00:24:26) AI can help people make bioweapons. (00:24:29) It can also help people get their term paper in or do, you know, be a better salesperson at, you know, Oracle. (00:24:36) Like we're talking about paper. (00:24:38) Like we're talking about oxygen here. (00:24:40) is like a fundamental horizontal technology. (00:24:43) So where do you think there is a case to regulating AI, if at all? (00:24:49) And where do you think, hey, yeah, free market will figure it out? (00:24:52) Well, I have two takes, one on the Pope and one on anthropic. (00:24:55) So your question is more about anthropic. (00:24:58) Let's go with the more powerful entity. (00:25:01) We'll go, okay, you want to go in reverse, the least powerful of the two. (00:25:05) So this Pope said, and I have to learn how to pronounce all these Latin words like you, (00:25:10) that this encyclical was mirrored after one done by Leo XIII in 1891. (00:25:17) And he invoked that. (00:25:19) He even said he chose the name because he's so enamored with Leo XIII. (00:25:24) Leo XIII encyclical warned that the Industrial Revolution was going to be bad for people. (00:25:29) So let me tell you what happened from 1891 till today. (00:25:34) The work week went from over 60 hours to 34 hours globally. (00:25:38) Real wages went up 8 to (00:25:40) 10X adjusted for inflation. (00:25:42) The medium worker now earns more than a doctor did in 1891. (00:25:47) Global GDP per capita went from 1,500 to 20K. (00:25:51) Child labor in the US went from 18% to 0. (00:25:54) Workplace deaths fell by 40X. (00:25:58) Life expectancy went up 60%. (00:26:00) And global poverty went from 75% of humanity to under 10%. (00:26:05) All those things happened because of technology, innovation, and capitalism, (00:26:10) which is exactly what Leo the 13th was warning against. (00:26:14) So he got it dead wrong. (00:26:16) He got the whole thing precisely wrong. (00:26:18) So it's an interesting thing to say you're borrowing from. (00:26:22) Yeah. (00:26:23) So now on to Anthropic. (00:26:26) Yeah, anthropic and just common sense around, do you think they're (00:26:31) How would you regulate and or protect against, maybe we'll broaden the term here, protect against nefarious uses of the technology? (00:26:39) Obviously, we all want children to be protected. (00:26:42) We want to have truth and honesty in terms of facts and all of us sharing some basic truths. (00:26:49) And we obviously don't want people using this technology for bioweapons and the Terminator scenario. (00:26:54) I have to tell you that Anthropic is a mystery to me. (00:26:58) I've never ever seen a company (00:27:01) that is both leading their field and the most negatively outspoken commenter on what they do. (00:27:09) I've just never seen it. (00:27:11) And my initial theory was the regulatory capture theory, that they just want to ensure there's regulation. (00:27:19) And quite frankly, I think they're very close to achieving that. (00:27:23) Like they have stirred up a frantic position, especially in America. (00:27:29) American consumers are definitely (00:27:31) afraid of AI. (00:27:33) I think I've talked to you guys in the past about, the book that Jonathan Heights written about social media. (00:27:39) And there's a whole bunch of state legislators that think we should have regulated social media. (00:27:45) And so now they're destined to want to get in front of it. (00:27:49) And we know that Anthropic's one of the most aggressive lobbying company startups of all time. (00:27:56) the amount of effort that they're putting in, the amount of money at a state-by-state basis. (00:28:01) So that was always my first theory, but then they just, they got so loud that I've literally in the past. (00:28:09) 30 days, read everything I can about Anthropic, and I've come up with a new theory. (00:28:13) This is my new breaking theory. (00:28:16) I call it the Dr. (00:28:17) Frankenstein theory. (00:28:20) You remember when Elon had that conversation with Larry Page, where Larry called him. (00:28:24) I was literally sitting next to him when he called him. (00:28:28) Explain the story real quick. (00:28:29) Well, we were at a birthday party and (00:28:34) Elon was like, listen, humanity needs to be protected from the stuff at DeepMind, because at DeepMind, they had an example of the AI having tried to break out, to jailbreak out of its computer and not be turned off and had some sentience or some inkling of sentience. (00:28:51) And he said, we have to protect the human species. (00:28:53) And he said, well, Larry said, well, what do you think (00:28:56) You got specious because you care about the human species over AI. (00:28:59) This is at least 15 years ago. (00:29:01) No, this is right before Elon co-founded OpenAI, right? (00:29:05) It was back in 2015 or something. (00:29:07) The actual story here is Elon and Google had backed (00:29:13) Demis and the team at DeepMind, when they were an independent company, then Elon was like, oh my God, Google's going to buy this. (00:29:20) And I remember having the conversation with Elon about this. (00:29:22) We have to figure out a way for DeepMind not to go to Google. (00:29:26) we have to block this somehow. (00:29:27) And he begged those folks to not sell to Google because Google was running the table on everything and he wanted this technology to be independent and he was on the board of the company. (00:29:37) And he also said this was his motivation to launch OpenAI as a non-profit. (00:29:41) And when Google got it, he just said, this technology is too powerful for any one person. (00:29:46) So once again, you got to give Elon a lot of credit. (00:29:48) He saw the writing on the wall. (00:29:50) If one person can, and he saw it 15, 20 years ago, and him and Sam Harris used to debate this over dinner, you know, (00:29:56) What happens if somebody controls this and they run away with it? (00:29:59) would be extremely dangerous. (00:30:01) It has to be available to all the people. (00:30:03) Essentially, the Pope's position. (00:30:04) It has to be in the service of humanity, not ruled by one person. (00:30:08) It's far too powerful. (00:30:09) So the reason I call this the Dr. (00:30:11) Frankenstein theory is the more I dig, I've met people who I dare say think it's their responsibility and they're excited about building a species that's superior to humans. (00:30:26) And I would just encourage people to read as much as they can about anthropic. (00:30:31) Chris Ola worked on this thing called the Constitution. (00:30:35) It's about 80 pages. (00:30:36) It's hard to get through, but I would encourage you to read it. (00:30:39) Amanda Askell, who is the chief philosopher, has started doing podcasts. (00:30:44) I would encourage you to listen to him and listen to her language. (00:30:47) And then Dario wrote this blog post called (00:30:51) The machines of loving grace. (00:30:53) Loving grace. (00:30:54) I read it. (00:30:55) And it was based on a poem. (00:30:57) And the poem is kind of weird. (00:30:59) We should put a link to the poem. (00:31:00) It's quite short. (00:31:02) But the last stanza of the poem says, I like to think of a cybernetic ecology where we are free of our labors and join back to nature, return to our mammal brothers and sisters. (00:31:15) I don't know what that means. (00:31:16) Like we're going to go live in the fields where the mammals live. (00:31:20) And then the (00:31:21) ticker and all watched over by machines of loving grace. (00:31:25) Sounds like overlord to me. (00:31:27) And then in Dario's post, he says, near the end, and it's very long, you read it, Jamal. (00:31:35) I mean, machines of loving grace is very long, but he's talking about in the future, what are humans going to do because he believes in the massive abundance and UBI and that we won't have to work. (00:31:45) I don't believe in any of those things, but he does. (00:31:48) And then he says, it could be a capitalist economy of (00:31:51) AI systems, which then give out resources to humans based on some secondary economy of what the AI systems think makes sense to reward in humans. (00:32:03) So that's envisioning a deity of sorts that's going to break ties and decide what humans do. (00:32:12) It's a computational reward function for humans. (00:32:15) It decides how much you're worth. (00:32:17) so I don't think they think they're writing software. (00:32:19) I think they're midwifing A deity here. (00:32:23) And I don't know which one I'm more afraid of, the regulatory capture or the second theory I call the Dr. (00:32:31) Frankenstein theory. (00:32:32) It's more scary to me, I think, the second thing. (00:32:35) These are delusions of grandeur. (00:32:37) Let's call it what it is. (00:32:38) They believe that they are so intelligent. (00:32:41) I know some of these folks, the Burning Man sort of offshoot of it, transhumanism. (00:32:45) They believe (00:32:47) that they're so powerful, these individuals, that they can create God. (00:32:52) And that by creating God, they are like this Prometheus kind of species. (00:32:57) It literally is... (00:32:59) the ultimate level of narcissism and delusion of grandeur to think you can create God. (00:33:05) And that then the God you create, like you're saying, Bill, is going to be so benevolent and perfect that you constructed the perfect God that will give you your pellet, will give you your little Skinnerian (00:33:18) I just would correct you. (00:33:19) I didn't say it. (00:33:19) Dario said. (00:33:20) Right, but no, but to your point of like just taking them at their word, they actually believe that they can create God and that they'll create a God so good that it's better than humanity. (00:33:29) Sacks, your thoughts. (00:33:30) Well, I guess the question then is, why are they pushing for the, let's call it red capture agenda where... (00:33:38) I know why. (00:33:39) Go ahead. (00:33:40) Jamal, go ahead. (00:33:41) that is very reductive game theory. (00:33:44) So if you want to be unexploitable, I think the best thing that you could do if you're trying to build a super god is have three or four entities in a room, close the door behind you, and then dominate those other three or four entities, and then you set the rules. (00:33:58) And because your counterparty is unable to track at the level of technical capability that you would have, (00:34:06) You create this massive asymmetry that allows you to exploit them. (00:34:09) That's just simple game theory optimization. (00:34:12) And you know what Bill said is so powerful. (00:34:14) I've read these things and it's laborious and it takes time, but every time they put these things out, just take the time to read it. (00:34:20) And what I have said before, Bill, I don't know your point of view on this, but I initially thought that this was mostly game theory. (00:34:29) that a lot of their reactions I thought were less rooted in their dogmatic beliefs and more rooted in a GTO approach to either raising capital or putting pressure on competitors. (00:34:42) Either way, both could be true, what your framing is and my framing, although mine's more tactical than yours, to be fair. (00:34:49) Because I've always thought that these moves make sense through that lens. (00:34:53) How do you absorb most of the capital? (00:34:56) How then do you make sure that you are in a position to disproportionately affect the rules? (00:35:02) And how do you create an oversight body that is less (00:35:08) capable and intellectually aware as you are about the actual details. (00:35:13) The referees don't understand the game, right, Shama? (00:35:15) If the refs don't understand the game, you'll run over the game. (00:35:19) By the way, by the way, one thing they have achieved by doing this is I think that if you polled the, let's just call it the intellectual elite, so everyone in the media and whatnot and the professors and all those, and they were to rank the different AI players by who they think, (00:35:38) is most caring, I think they'd probably put Anthropic first because they've been out with the doomerism talk. (00:35:44) And so it's given them a halo with the people that may matter for what they want to accomplish. (00:35:51) It's simultaneously creating a lot of trouble, like with the data centers and whatnot. (00:35:57) Like there's negative ramifications. (00:35:59) What you're saying is so important because on the one hand, they create empathy (00:36:04) And then they write these documents that expose what they think, and nobody actually connects the dots. (00:36:09) Yeah. (00:36:09) To steel me in their position for a second, I mean, I think probably the way they think about it is that they are creating something very powerful, something godlike, and therefore it needs to be safe. (00:36:20) And that they care the most about that out of everybody. (00:36:25) Nobody else takes this seriously. (00:36:26) Remember that Anthropic was basically a spin out of open AI, and they felt that (00:36:32) Sam and the company leadership weren't taking their point of view seriously enough. (00:36:36) It was the most woke portion of OpenAI. (00:36:39) Well, let's not say we're steel manning, so the most steel manning for seconds. (00:36:42) So they see the power of it. (00:36:44) They're the ones who are concerned about safety, and they care the most, and therefore they're in the best position to do that. (00:36:52) Now, I think the issue is just, you can see how this can lead to recapture, right? (00:36:57) Which is, (00:36:59) If you brand yourself as the safe AI company and then try to characterize everybody else as a reckless player and reckless AI needs to be stopped, you can see how this would basically further your monopolistic control over this industry. (00:37:16) And if you see AI through the lens that, you know, really, frankly, the Pope and I see it, which is (00:37:23) Centralization versus decentralization. (00:37:24) I do think that is one of the key lenses we should have on the technology is whether you want this to be a centralized or decentralized technology. (00:37:34) This way of viewing the world leads to more centralization. (00:37:38) And I think that's dangerous. (00:37:39) I mean, if AI is this very powerful technology, I think it needs to be decentralized so that (00:37:46) All of us can protect ourselves to some degree, right? (00:37:49) We need to be able to run, we need to be able to run the AI ourselves on our own hardware if we so choose, so we're not beholden to a single company that might be in bed with a deep state. (00:38:01) Let's say it very pointedly, if benefits. (00:38:04) and compensation and economic support were all of a sudden tied to some algorithmic decision. (00:38:11) This is a dystopian episode of Black Mirror that we're dealing with. (00:38:15) And to your point, Sachs, you want 100 or 1,000 or 100,000 versions of what that answer is so that there's actually a way to refute. (00:38:24) A singular answer, a singular answer to these kinds of questions, which is effectively what some folks would want, is incredibly dangerous. (00:38:32) And this is something that is in control, I think, of humanity. (00:38:37) I've been talking about AI sovereignty here for a bit. (00:38:41) just in terms of how much more cost effective it is and how you're not training other people's AIs with your knowledge and your insights. (00:38:49) This is why it's super important that open source, open source agents and local hardware be able to run these models and that consumers and companies learn how to roll their own language models, how to make a small language model, an SML, a VSML, a verticalized one, and run it on your Apple hardware, because Apple actually has taken a principled approach (00:39:11) Historically, to your sovereignty for your data, data sovereignty now is, yes, and now it's intelligent sovereignty. (00:39:18) The intelligent sovereignty is different than privacy. (00:39:20) Privacy is, oh, you can't see my photos, you can't... (00:39:24) peek into my notes app and what I wrote there in my journal. (00:39:27) Now, intelligence sovereignty is, you can't tell me what to think. (00:39:31) You can't use your AI to analyze my photos, to analyze my emails, to analyze my messages, and tell me how to interpret the world. (00:39:39) That's actually going to be the next key piece here. (00:39:41) This is why I think Apple is just... (00:39:43) the dark horse in this entire race. (00:39:45) If there is an open source product that can run on this hardware, the M5s, the, you know, 48 gigs, 128 gigs, the new Mac Studio coming out with supposedly a TB, that changes the whole game. (00:39:57) And this is so paradoxical, Bill,