Ben Horowitz: RSI, Crypto as AI Money, & Classified Physics

Moonshots host Peter Diamandis speaks with Ben Horowitz, cofounder and general partner at a16z, alongside regular cohosts Salim Ismail, Dave Blundin, and Dr. Alexander Wissner-Gross, about whether AI can or should be paused, what happened when Horowitz told a Biden administration official that regulating AI means regulating math, why crypto is the natural money for AI agents, and why the gap between AI capability and societal adoption may be wider than people think. This episode originally aired on Peter Diamandis's Moonshots podcast.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  A large number of departures from XAI, from the founding team.

• 0:04Speaker 01

  It wasn't clear to me whether they were fired or whether they left, you know, because they all leave on good terms.

• 0:08Ben Horowitz?

  I don't know the answer to that question. I will say that it's...

• 0:14Speaker 03

  Recursive self-improvement, RSI, is the real trigger for the singularity, and it happened a while ago. We're exiting the industrial age permanently, as we're talking.

• 0:23Ben Horowitz?

  We're obviously going into a new world. Like with the industrial revolution, I think it's scary at times to think about.

• 0:31Speaker 04

  I think we have 150,000 people per day dying on Earth, and I think AI is probably the best chance we have at stopping that.

• 0:40Ben Horowitz?

  Whoever is building the AI has a lot of control about how society is going to work. So I do think there's real danger along these lines of attempting to pause it.

• 0:49Speaker 00

  When are we going to have discovery by an AI of something as significant as relativity on its own?

• 0:56Speaker 04

  I don't think it's the next 12 months. I think it's...

• 1:01Ben Horowitz?

  Now that's a moonshot, ladies and gentlemen.

• 1:05Speaker 05

  Ben Horowitz argued directly to the Biden administration officials that regulating AI means regulating math. Their response? "We did that in the 40s with nuclear physics, and some of it is still classified today." Horowitz sees the real danger not in AI moving too fast, but in US regulations slowing progress enough that China ends up leading how AI reshapes society. The Biden administration's final AI chip export controls required government approval for GPU sales to most of the world. This conversation, previously aired on Peter Diamandis's Moonshots podcast, covers recursive self-improvement, why crypto is the natural money for AI agents, and what it would mean for Apple to own the local AI hardware strategy. Peter Diamandis speaks with Ben Horowitz, co-founder and general partner at A16Z,

• 1:50Speaker 05

  alongside co-host Salim Ismail, Dave Blunden, and Dr. Alexander Wissner-Gross.

• 1:59Speaker 00

  So everybody, welcome to Moonshots, another episode of WTF here with my Moonshot mates, DB2, AWG, Mr. EXO, and a friend of the pod, someone who's been with us before, the amazing Ben Horowitz of Andreessen Horowitz. As I like to say every week, welcome to the number one podcast in AI and exponential tech. Our job here is getting you future-ready. And it is an insane week. We've actually recorded two podcasts this week, just because the speed is over the top. And we're going to be recording again in another four days. I mean, Ben, it's like, it's my goodness, right? Thank God, my AI avatar is getting really good.

• 2:41Ben Horowitz?

  Yeah.

• 2:43Speaker 00

  Let's open up with top stories in voices, video, XAI, multis.

• 2:51Speaker 00

  All right, first one. Here we go. We're starting to see a little bit of Doomer conversations coming. AI disruption will soon hit sooner than most expect. This is something that's been making the rounds for Matt Schumer, CEO of OtherSide AI. Ben, have you seen this article?

• 3:10Ben Horowitz?

  Yeah, yeah, of course.

• 3:11Speaker 00

  Of course.

• 3:14Speaker 00

  Thoughts? So, I mean, is this what we already know? I mean, we're about to hit recursive self-improvement. Once that hits, all of our curves go out the window, everything accelerates. Everything we've been preparing for is being redefining you time scales. What do you think of it?

• 3:33Ben Horowitz?

  I think it's, I think the AI timeline is somewhat unpredictable, but kind of certainly more predictable than what he's talking about, which is societal change. I think that...

• 3:51Ben Horowitz?

  You know, it tends, I would be like very surprised just in seeing how even companies in Silicon Valley have changed so far if companies like outside of that sphere, you know, just completely change everything they did in one to five years. Like, I think that's a little aggressive for societal change. You know, look, we're obviously going into a new world. And, you know, like with the Industrial Revolution, I think it's scary at times to think about. But, you know, like, there are going to be, he kind of, I feel

• 4:35Ben Horowitz?

  like, highlighted all the negative changes and not the positive ones. And I feel like there's way more positive change coming than than negative change at a much more rapid rate. So, I don't know, like, I thought it was a little aggressive.

• 4:52Speaker 00

  I mean, I'm just trying to understand why it made the rounds, since this is not a brand new conversation, but I got it sent to me by everybody. Dave or Alex?

• 5:01Ben Horowitz?

  I think it, I think it really has to do with OpenClaw and, you know, kind of the new coding models. So, people in Silicon Valley are talking about AI differently since then, because it is kind of different. And I think that that's a trigger for this one being so viral.

• 5:22Speaker 01

  Yeah, I completely agree. I think a lot of things that we've been saying for, you know, up to a year now on the pod are suddenly resonating a lot more because of, you know, OpenClaw, but also a bunch of other eye-opening nano banana-type things where, you know, denial a year ago was very easy. Denial today is much, much harder in the face of like what's right in front of you. But also, the flavor of this rollout has changed a lot in my mind recently, because, you know, this was board meeting week for me. You know, three back-to-back mega board meetings, 1,100 people affected. And what they were thinking of before was, well, will AI be able to do what I do and replace me? No way. Now they're like, oh, wait, AI is easily going to make me three times more productive.

• 6:06Speaker 01

  Okay, well, that's the same thing, right? In terms of the headcount you need to get a job done, that's effectively the same thing. And like, oh, okay, I didn't think of it that way. Now you're exactly right. So the hope is that these companies will grow into it and can keep current headcount and expand 3X. But if you don't expand 3X, you're still looking at, you know, a two-thirds reduction in headcount to get the same job done. So it effectively is a huge amount of displacement because, you know, the big banks, insurance companies are not going to triple their size in the timeframe.

• 6:38Ben Horowitz?

  Yeah, I also think they're not going to go to total efficiency very fast. Like, I mean, I could be wrong, but like, I've dealt with these guys. They've had plenty of opportunities to be more efficient. And we'll see, but we'll see, we'll see.

• 6:54Speaker 03

  I have a couple of thoughts about that article.

• 6:56Speaker 01

  Yeah, please, go ahead.

• 6:57Speaker 03

  One, I thought it was like a summary of the podcast for the last eight months. That's all we've been talking about is stuff's going to change. It seemed a little dramatic, as Ben put it, and I don't agree with the timelines, but definitely there's something coming. I think it's just a, the reason it's making the rounds is just got the zeitgeist of, of what's exactly happening that we need to track right now, which is we're in a singularity of multiple types. So that's what I got.

• 7:27Ben Horowitz?

  By the way, it was also like super well-written, like a really compelling.

• 7:33Speaker 04

  AI-written.

• 7:34Ben Horowitz?

  Yeah, yeah, yeah.

• 7:36Speaker 04

  I'll chime in here and just say, while I enjoy writing about the singularity in general, obviously I'm doing it almost every day at this point. I found it completely unremarkable. Maybe I'm just too, too deep in the weeds of writing about more pressing advances every day, but the, the sort of style where you talk about advances that, by the way, are moving even more quickly than I think described. In the essay and comparing it back to the COVID pandemic, which I think is a relatable touch point for a lot of people. Like, something big is about to happen. Let's be really millennialist. And, you know, if you read my essay, it's the ultimate viral hook. Read my essay to know exactly what's about to happen and how to survive the next five minutes of your life. It's a natural viral moment,

• 8:20Speaker 04

  but I don't think the information value was especially high compared to others.

• 8:23Speaker 00

  I agree completely.

• 8:25Speaker 03

  Exactly what I said, except Alex said it more eloquently.

• 8:28Speaker 00

  Yeah. We have a couple of articles here on C. Dance 2.0 out of ByteDance.

• 8:36Speaker 03

  Man, is that thing good?

• 8:37Speaker 00

  Oh, my God, it's amazing. It's, you know, it's going to change everything. Let me play this particular video first. The response here is, you know, and you've said this before, Alex, Hollywood is cooked. Right, so this is a video clip with a one-line prompt. So, you know, Tom and Brad fighting karate on a rooftop. And it generates what 10-second clips right now. Alex, what are your thoughts?

• 9:08Speaker 04

  I think at the risk of sounding, again, ho-hum, unremarkable, we saw copyright infringement at the scale already, or alleged, I should say, with earlier video models. And we saw the industry response and we saw settlements that eventually deals were struck to handle it. I think people are at this point, again, maybe I'm just sounding overly jaded with some of these advances, but I've seen remarkable advances in video models. I tend to think that people are so easily awed by video models that are able to show celebrity faces and scenes that they recognize, that maybe they over-index on the underlying quality of the models. I think world models that are interactive in real time

• 9:53Speaker 04

  are profoundly more interesting than video models. I think this is just 10 different copyright infringement lawsuits waiting to happen.

• 10:03Speaker 00

  But I still was wowed. I'm still one of those people that said.

• 10:06Ben Horowitz?

  Yeah, that was very.

• 10:07Speaker 00

  That was amazing.

• 10:09Ben Horowitz?

  So I would say on this one, the two, the two videos that I watched that were like, like where the entertainment quality was so high were one that Kanye doing his song in Chinese was so good, that video, like I watched it three times. Is that entertaining? And then the other one was the Waffle House one. And they're both kind of, I would just say representative almost of a new medium. It's not like, okay, this is film generated by AI. It's like, no,

• 10:48Ben Horowitz?

  this is a whole nother thing that we've never seen before. So I, I think this model is, at least for me personally, as a consumer, was, was, was an impressive step up.

• 11:03Speaker 00

  But I think YouTube, YouTube wins in this model, right? Because everybody's gonna be producing so much content and it's gonna become resonant on YouTube. It's not gonna, a lot of it may not go to the theaters or television and so forth. Yeah, Dave, you wanna say?

• 11:17Speaker 01

  Yeah, I was gonna say the same thing. Like TikTok also is the big winner. When you, when you personalize the content, when it's almost movie quality and it's personalized to very narrow topic areas and very narrow interests and many languages around the world and everything else, it just takes over. And so, you know, when, when the movie people say, well, look, we're still a little bit better. Yeah, but you're missing the bigger picture, which is you don't need to be as better than a movie. If you can push the production costs down to an individual producer, then the volume goes through the roof. But the narrow casting is just so much more compelling. You know, something that you and only your group really care about a lot in full movie format is so exciting.

• 11:56Speaker 00

  Yeah, I mean, the elephant in the room here, though, is, you know, this 2K quality multi-scene video, it just, it doesn't just threaten Hollywood. It threatens the whole concept of video as evidence, right? Court testimonies, journalism, political campaigns. I mean.

• 12:15Ben Horowitz?

  Well, and also that's gonna be real time, right? Like, so then.

• 12:18Speaker 00

  Yes.

• 12:19Ben Horowitz?

  Just any kind of security mechanic that you have where you recognize the person via video or, or voice is shot to hell.

• 12:31Speaker 01

  Yeah, the echo chamber effect is crazy, too.

• 12:34Speaker 00

  Yeah.

• 12:35Speaker 04

  Maybe just add, again, I think this is several months behind the bleeding edge. I think video models have been approximately this good. Yeah, sure, you can upscale it. You can increase the fidelity of the faces. You can certainly use faces that you probably shouldn't be using. We've been able to do this for months. Where I think the frontier actually lies in being able to, is being able to do this in real time and being able to do this on a single modern Nvidia GPU and being able to do this at a cost effective speed. And I think, again, I want to avoid over indexing on just video models that produce two Hollywood celebrities fighting each other on a rooftop.

• 13:18Speaker 04

  We were able to do that many months ago. This has already been thoroughly litigated. We saw OpenAI strike deals with relevant movie studios for Sora 2. We saw the Disney deal. We're in some sense, I think, past this, where we are now.

• 13:33Speaker 00

  But Alex, I don't think that's the point. I think the point is this is making it out into the ecosystem of common users. I think it's the notion that, yes, it's the cutting edge and yes, it was possible, but now it's something that is going to, is going to grow in its utility and its, you know, its consumer base. And then all of a sudden, you know, we've had, you know, basically democratization of film production for some time, but it's now going 10x, 100x more. I mean, I think it's, I think that's the issue.

• 14:12Speaker 01

  You're bringing up a really interesting point, too, which is that, you know, our starting point for this journey was probably autocomplete in code, you know, and like, wow, that's incredible. But then, you know, Alex in his newsletter has been tracking every event along the way, so nothing surprises him.

• 14:28Speaker 00

  Yeah, I'm just spoiled. This is my problem at this point.

• 14:30Speaker 04

  I'm so spoiled.

• 14:33Speaker 00

  This is so six months ago.

• 14:35Speaker 03

  But if you think about it.

• 14:36Ben Horowitz?

  Follow up on something, though, that Peter said. Like, the thing that I do think is different, and it's not any aspect of it, but it's the combination of that you could give it a one-line prompt and produce something entertaining isn't something that we were, at least I hadn't seen at this level of entertainment. So I think from a consumer product standpoint, as opposed to technological standpoint, which I agree on, it's kind of exciting.

• 15:05Speaker 01

  What's going with that, too, is exactly what you're saying there, Ben. Like, the later your first exposure to AI, the more of a holy crap moment it is, because it's something truly mind-boggling to the unexposed. And you're still seeing that. When you, when you survey around at random in a city, you know, outside of San Francisco or Boston, the exposure rate to AI is still very, very low, shockingly low.

• 15:32Speaker 00

  So, Lina, you said ho-hum.

• 15:33Speaker 01

  The first thing you see is so mind-blowing.

• 15:35Speaker 03

  Yeah, for me, this was ho-hum, because, you know, if you traced the trajectory of where we've been going, you should expect to see this about now or even earlier. So there's nothing radically magical, but yeah, it may hit a new group of users. Yeah, people go as well.

• 15:50Speaker 00

  You guys.

• 15:50Speaker 03

  And another batch of people. No, no, but another batch of people goes, oh, my God. Fine, there's another segment falling over into the, into the new world.

• 15:58Speaker 00

  All right.

• 15:59Speaker 03

  Great. And the more voices out there, this is the plus side of also the Schumer essay. Yeah, we, we know, but the more voices out there talking about this, the better, because it'll accelerate the whole thing.

• 16:09Speaker 00

  All right, let's get to the second ByteDance, C-Dance 2.0 article. Here we have C-Dance 2.0 was paused by ByteDance after it was found to recreate real voices from just facial photos. I find that almost impossible, just from a facial photo. Alex, what did you learn about this?

• 16:32Speaker 04

  I think there's something interesting here, which is as we start to scale data sets, it is possible that we could start to see positive transfer between modalities that's unexpected. We've spoken in episodes past about people claiming that they're uploading their whole genome into Claude and being able to generate facsimiles of their face that resemble the real thing. I do think it's possible that if all of YouTube and all of the world's video were uploaded into a single joint embedding model, which is the foundational technology behind C- Dance 2.0, I do think it is conceivable that if

• 17:14Speaker 04

  we just aligned all of the world's audio and spoken audio with all of the world's faces, we would find some positive transfer between the two and be able to reconstruct to reasonably high fidelity your voice from your face, or your face from your DNA, or some attribute from some other attribute. I do think it is possible with enough scaling. Whether C- Dance 2.0 actually achieved it, or whether it was just a happy coincidence, hard to tell at this stage.

• 17:40Speaker 00

  What's interesting, though, is that they voluntarily stopped it, right? They voluntarily shut it down, which is a great, sort of, move by a, I want to call it hyperscaler. But the reality is, once it's out of the bag, once you know it can be done, it can't be uninvented, so it's out there, if in fact it works.

• 18:02Speaker 01

  Yep.

• 18:03Speaker 00

  Yeah.

• 18:04Speaker 01

  Yeah, totally. So far, you know, ByteDance is like Google, they have to be cautious and conscientious, they can't just... But then every time this happens, a small startup then does it again right after. And they don't care because they're a startup.

• 18:16Speaker 00

  All right. I want to play this video clip from Eleven Labs. You know, I think all of us have Eleven Lab voices that we use for different projects and so forth. But I was just blown away by this. And it's the human-like quality, the cadence, the hums and the uhs that come out of this. So let's take a listen and discuss it, because this is a game changer for me. I know what you're gonna say, Alex, this has been around for a while.

• 18:44Speaker 04

  No hum. Saying it in advance.

• 18:47Speaker 00

  But let's take a dive in.

• 18:53Speaker 06

  Hey there. I'm Jennifer with Eleven Airlines. And how can I help you today?

• 18:58Speaker 07

  Jennifer, my flight just got canceled and I'm stuck here in Orlando. At this rate, I'm gonna be missing my daughter's birthday. This is ridiculous.

• 19:05Speaker 06

  Yeah, I hear you. And I'm so sorry about that. Let's figure out what's going on, okay? Um, could you please tell me which flight this was?

• 19:15Speaker 07

  Yeah, this was Flight MD 412.

• 19:19Speaker 06

  Great, thanks. Okay, just pulling that up now. Okay, yeah.

• 19:26Speaker 00

  So, I don't know. I'm moved by that. Both excited and frightened. In a way that is, you know, again, we're cooked in terms of, you know, at my home, our family picked a secret code word. And again, everybody listening, if you've not done this yet, tonight at dinner with your, with your parents or your kids, pick a secret code word. If someone's asking you to do something that is kind of unusual or crazy, that you don't expect, you may be talking to an AI. So.

• 20:01Speaker 01

  Ben, you may not know this, but we have a fool your spouse challenge for this calendar year. First podcast listener who can, you have to fool your spouse for three minutes on a Zoom call. And it has to be a totally fake you. Meanwhile, Ben, you may have to. Record it and send it in.

• 20:16Speaker 04

  Peter is always asking the, the multis, the OpenClaw agents to dox him by calling him at home.

• 20:22Speaker 00

  Yes, that's my, that's my AI, AGI has arrived when an AI is calling me unexpectedly.

• 20:28Speaker 04

  You know what happens? They end up emailing me. I get several emails probably per day at this point from OpenClaw agents asking me what Peter's number is.

• 20:39Speaker 00

  That's hilarious. Uh, and when they call me and ask them where they got it from, Alex.

• 20:45Speaker 04

  That's right. Maybe just to comment on the 11 Labs. So if you've been using the version 3, V3 Alpha model from 11 Labs, this should hardly be surprising. V3 enables you, if you use text to speech in the 11 Labs platform, to specify with brackets emotional expressions. And this has been around for many months. What's somewhat interesting here, to me at least, is we've known how to do audio to audio transfer for probably a year plus at this point. If you've used advanced voice mode from OpenAI, you've used audio to audio transformers. But what's somewhat interesting here is 11 Labs is better known

• 21:26Speaker 04

  for text to speech than speech to speech. And as far as I can tell, this new expressive mode that we're talking about here seems to still leverage the text modality, which historically has been very difficult. You'd have to go from speech to text back to speech, which was high latency. It was slow. It didn't feel very conversational. And somehow, it seems like, without having to do direct audio to audio, 11 Labs has found a way to do speech to text back to speech in a way that feels natural and turn-taking and real-time. So I think it's, in some sense, an incremental advance. But in another sense, if they really are still keeping text, which is much easier to mine and analyze in the loop, it probably is a material advance.

• 22:07Speaker 00

  Yeah, Alex, I would say we've crossed the uncanny valley on voice at this point with this demonstration. And then voice becomes the new interface in the AI era, right? I mean, I can't tell you the amount of time that I'm just speaking to AI versus this cumbersome typing at it. So I think those two things are really important takeaway from this.

• 22:27Speaker 03

  I think that's a great way of putting it, Peter.

• 22:29Speaker 04

  Yeah. But the problem is, do you really want to use audio as your primary modal? I mean, it works well if you're in isolation. Okay, yeah, you want BCI. I want a BCI too. And I want wearables and I want gestural interfaces. I want it all. But for most people, I mean, is it New York, this is sort of famous anecdote. New York Times in the 1980s did a famous study where they just put all of their reporters on then state-of-the-art speech-to-text systems and asked them to use voice. And what happened was the writing quality went down. Why did it go down? Because it's difficult to think ahead as well as you can when, if you're just typing,

• 23:10Speaker 04

  if you're speaking, you're leveraging, right, similar portions of the brain. So I'm not 100% sold yet that speech is the modality of the future. I do like BCI's. I do like wearables and gestural interfaces. I do like typing, but speech, I think, jury is still out for me for high bandwidth operation.

• 23:28Speaker 01

  Yeah.

• 23:30Ben Horowitz?

  I was just gonna say, for regular people who don't necessarily write for the New York Times, I think speech is often, or at least in my experience, is the mode of choice.

• 23:45Speaker 01

  For sure.

• 23:47Ben Horowitz?

  The other thing I'd say about 11 Labs and kind of quick disclaimer that, you know, we're a big investor in 11 Labs and my daughter Sophia works there, so I'm biased towards her. But the, you know, one of the really amazingly just kind of landscape shocking things to me about 11 Labs is, you know, when we invested originally, the big question was, well, like, aren't the, you know, state-of-the-art models gonna be able to talk? I mean, like, of course they're gonna be able to talk. But, you know, speaking correctly with the right nuance and building the right products for developers and so forth has proven to be very sustainable for them, which I think

• 24:30Ben Horowitz?

  is interesting as you look at the entire landscape. The difference between the capability and the product is significant.

• 24:38Speaker 01

  Yeah, you know what's amazing to me about 11 labs? We have two companies here in the lab that do voice, Voice Run and Vocara. And what's amazing is these are self organizing systems that are trained off raw data. And what they do well just blows your mind. But you know, within voice, it turned out the turn management was very, very hard. Very. And you're like, well, I didn't ever thought like it seems so trivial compared to actually doing these incredible synthetic voices that can say intelligent things, but they don't know when to stop talking. You know, like on this podcast, when I stop, you start when you start, it's like very natural to us. But because that wasn't in the training data, they're just, you know, up until now terrible at it.

• 25:18Speaker 03

  Huh. Wait, according to our listeners, we're talking over each other all the time in a very unsynchronized voice.

• 25:26Speaker 04

  We have so much to learn from these speech to text to speech models about turn taking.

• 25:33Speaker 01

  Yeah.

• 25:34Speaker 04

  Oh, my God.

• 25:36Speaker 01

  Yeah.

• 25:37Speaker 00

  All right. Next topic here. If you guys are OK with it, I'll move us along. So just following the merger of SpaceX and XAI and in the spirit of full disclosure, I'm an investor in both, probably, probably you are as well, Ben. A large number of departures.

• 25:57Speaker 01

  All three, all three of the X's.

• 25:58Speaker 00

  Yes. A large number of departures from XAI, from the founding team, mostly ethnical, ethnic Chinese co-founders. And it's likely due to the ITAR regulations of SpaceX. Right? And it was interesting on a few, a few video presentations that Elon's done that's had the team from XAI, at least half of the team there was ethnic Chinese. And, you know, the culture there breeds incredible mathematicians and programmers. Any, any comments?

• 26:33Speaker 01

  Ben, I'd love to get your thoughts on this, but I looked at the timeline and the exodus of senior AI talent predates even the decision on who to merge with what. So the theory that this is related to ITAR, and it wasn't clear to me whether they were fired or whether they left, you know, because they, they all leave on good terms. I don't know if you have any insight.

• 26:55Ben Horowitz?

  Yeah, I don't know the answer to that question. I will say that it's, so in a related matter, we recently heard from a few Chinese nationals who are PhD students that the Chinese government is cracking down on the Americas or US academia's use of Chinese open source models. Like, they particularly post the Meta Manus acquisition, they're very, very worried about actually secrets going this way. So kind of the opposite of what we've been worried about,

• 27:37Ben Horowitz?

  which they, you know, like, our whole open source kind of work is based on the Chinese model, since there haven't been as many US open source models. So that's, this whole thing, I think, is about to get more complicated.

• 27:56Speaker 00

  It is. I mean, if you read these tweets, Mr. Wu and Mr. Bob, both are super enthusiastic about XAI. They're not leaving with any kind of angst. You know, they're saying that they love the XAI family. We're heading towards an age of 100x productivity. So I don't think they would have left on their own accord. That's my personal opinion. So what drove it? You know, enter Alex's comment from last time. Well, when did their vest, their stock vest?

• 28:28Speaker 04

  I mean, there is another explanation, which is just SpaceX is a very large company relative to XAI's headcount. And maybe there was a natural reorganization that happened as a result of that. I'll also point out that XAI was selling foundation model services to the Department of War prior to this merger. So I would again query whether some sort of nationality concern was really the trigger. It seems more likely to me this is just the result of a natural rework of XAI starting to merge into SpaceX.

• 29:03Speaker 00

  Yeah. Well, if it were a result of ITAR and such, you know, their, America's AI dominance has really built significantly immigrant talent. So if we're going to start having this kind of a reaction to, you know, PhD immigrants, I mean, I personally think, and I'm curious what you think, Ben, that everybody going through a PhD program in the US should have a green card stapled to their PhD when they graduate. I think the idea of kicking people out is the wrong approach.

• 29:35Ben Horowitz?

  Yeah, I think in generally, that's, in general, that's correct. I do think that there's, I don't think we've quite thought through the case of China totally, in that like, we have, you know, amazing Chinese nationals who work for us and in every company we have, I think.

• 29:57Speaker 00

  Of course.

• 29:58Ben Horowitz?

  So, you know, the talent that comes here is extraordinary. I wouldn't say there's no risk to that idea. I guess I would just say that. But I think being, I think you're right, general. I think we should be open. We should accept the phenomenal talent that comes over here and not, not fight it, because we'll lose anyway. I mean, like, we're not keeping anything secret in America. Certainly not in American companies. There's, none of them have SKIFs, none of them have good security protocols around personnel.

• 30:30Speaker 00

  Don't get me started on the, on the idea that privacy is long since dead, but.

• 30:35Ben Horowitz?

  You're correct.

• 30:37Speaker 00

  Here's a tweet from Jimmy Baugh, a co-founder at XAI. Recursor self-improvement loops likely to go live in the next 12 months. 2026 is going to be insane and likely the busiest and most consequential year for the future of our species. So you're going to hear this a few times. Alex, your turn to go off. I mean, you're gonna hear this a few times that these next few years are a massive inflection point, that everything changes. And we're gonna look back thousands of years in the future, back to this inflection point. Or is it just a smooth singularity, Alex?

• 31:16Speaker 04

  I think it can be both. I think we've already hit the era of recursive self-improvement. I'm banging the table rhetorically every episode and every day in my newsletter, talking about recursive self-improvement. We're there. All of the frontier labs are using their own models at this point to develop their models. That's practically the definition of recursive self-improvement at this point in practice. I don't think it's the next 12 months. I think it's now. And is 2026 going to be insane relative to years past if you just sort of skip over all of the interim time? Absolutely. Is it already insane in some sense? Absolutely. Even if we just look at the events of the past 24 hours, some of which I think we'll

• 31:59Speaker 04

  get to, like self-replicating AI and courts where AIs can mediate their own disputes in front of an AI jury. That would have been pure science fiction several years ago. That's not 12 months from now. It's not, quote unquote, "within the next 12 months." That's the past 24 hours.

• 32:16Speaker 00

  Eight hours ago.

• 32:17Speaker 04

  Yeah, that's like past 24 hours. So I think Jimmy is underselling, if anything, what craziness looks like. At the same time, locally, spacetime is smooth. And you need to look no further than my saying "ho hum" to a few of these stories as being so several months ago. That would have been sci-fi years ago. But you get a little bit spoiled living inside the innermost loop.

• 32:41Ben Horowitz?

  I would say I do think there's a delineation between recursive self-improvement with a human in the loop and without one. And I think he seemed to be implying that there'd be no human in the loop, which I think is an accelerant. TBD how much of an accelerant, but I think that could be very different.

• 33:06Speaker 00

  Yeah, permissionless self-improvement, right? Like, flip the switch and go as fast as you can.

• 33:12Speaker 01

  Well, also, there's a distinction. This came up at the EverQuote board meeting yesterday, but the self-improvement where the inference time speed algorithms are being improved by AI clearly well underway, way down the path, and then inference time speed can be directly translated into intelligence now. We now have the know-how to turn more inference time loops into a higher IQ. So that is clearly underway. The question is, does that notch up, then allow it to work on the next part of the algorithm, the next part of the algorithm, in which case, we've already hit the flashpoint, and now we're just talking about the rate at which it percolates across.

• 33:53Speaker 01

  But Ben, I agree that his comment is about the actual core algorithm development, the next ideas are 100% from AI, and then they go into production, you know?

• 34:05Speaker 04

  I also think human outside the loop, on the loop, in the loop is, in some sense, a pretty blurry or slippery slope. If you remember George Jetson from the Jetsons, George--

• 34:16Speaker 00

  I do remember George.

• 34:17Speaker 04

  Remember George, Ben? George would go into work, and he would complain about his finger. He'd have to press one button all day with his finger, and then complain that his finger was sore. And the finger would be swollen from pressing a single button all day occasionally. I think that's like, that is a good metaphor for the state of recursive self-improvement inside and outside the frontier labs right now, where you have Claude code instances, and Claude is asking you every few minutes, "Do I have your permission to do the following thing?" And you press the George Jetson button, "Yes, I approve." "No, I don't approve." And we're all sitting here complaining about our swollen finger from pressing "approve, approve, approve" for Claude code running Opus 4.6 with agent teams.

• 35:02Speaker 04

  But really, it is recursive. I would argue it is recursive self-improvement, even if we're pretending we're in the loop by pressing the George Jetson button.

• 35:10Speaker 00

  That's exactly what I'm doing on Telegram right now with my, with Skippy. It's like, yes, go on to the next stage. I don't want to be, I don't want to slow things down. And as we've said so many times, this is the slowest and most expensive it's ever going to be.

• 35:26Speaker 03

  Look, there's a couple of points here.

• 35:29Speaker 01

  Yes, please.

• 35:30Speaker 03

  We've been talking for a while that recursive self-improvement, RSI, is the real trigger for the singularity, and it happened a while ago. So all we're doing now is kind of accelerating that path. We're exiting the industrial age permanently as we're talking, so.

• 35:47Speaker 01

  Yeah, I really think the minute-by-minute unfolding of the singularity is the most fascinating thing I've ever experienced. And Alex is exactly right. There is this point in time we're in right now where there's a human in the loop contributing, but it's really ambiguous what part of the progress is AI versus human. If you're in the actual coding process, was that my idea? It was half my idea, but then the AI suggested this other thing, and I kind of adopted it. And now it's not clear whether it was my idea or not. But we're in that mode right now where the research, a lot of the research in these core algorithms is just deploy these 500 tests for me and tell me which hyperparameters worked better

• 36:30Speaker 01

  or which neural topology worked better. It's not like inventing or discovering relativity. It's just litany of experiments with different trials and then taking the one that worked and redeploying it. And now you have a smarter AI. And now it's trying more trials. It's really very likely that we're well down that path.

• 36:52Speaker 04

  And I also had, I do think we're going to discover the next relativity or equivalent of relativity in physics as well with AI. I'm super interested in that as well.

• 37:03Speaker 01

  Prediction?

• 37:05Speaker 04

  What would you like to hear? Next relativity?

• 37:08Speaker 00

  When are we going to have discovery by an AI of something as significant as relativity on its own?

• 37:18Speaker 04

  I think next two years.

• 37:21Speaker 00

  Okay.

• 37:21Speaker 01

  Yeah, I would also, this is a great bellwether of Alex's question, Alex. The transformer algorithm 2017 that kicked off everything that we're experiencing right now. To me, it's the AI has already discovered things in the last six months that are harder to discover and harder to solve than the transformer was. But I'll ask you, Alex, if you feel like that's true, too. But it seems pretty clear to me that

• 37:46Speaker 04

  Absolutely.

• 37:46Speaker 01

  That algorithm is just not, it's not up there with relativity in terms of complexity.

• 37:52Speaker 04

  Oh, I mean, I guess everything, many of these discoveries boil down to insights that can be distilled down into equations. And you can, in principle, with relativity, with special relativity, you can do a number of thought experiments, you can compare it fortunately with lots of experimental data. And one could imagine a thought experiment. One of my favorite thought experiments is a Bayesian superintelligence. So if you had like a video of, this would be Newtonian gravity, not special relativity, but you could imagine taking a superintelligence, making it watch a video of an apple falling from a tree. And the argument in the thought experiment goes, within three frames of that video,

• 38:33Speaker 04

  it should have concluded Newtonian gravity is a pretty high likelihood posterior probability of explaining the universe. And with a few more frames, it, somewhere in its distribution, and there's a whole class of information theory devoted to what's called Solomonoff induction, like just devoted to thinking about how you can efficiently infer theories of the universe from limited observations. Somewhere in that distribution of theories should have been general relativity. So I am incredibly bullish that we'll be able to, with superintelligence, discover any new laws of physics, discover transformative inventions. For disclosure purposes, since we're playing the disclosure game, I have a portfolio company, physical superintelligence, that's working on these issues as well. I'm very bullish on the space.

• 39:18Speaker 00

  Amazing. I want to move us to a conversation that Eric Schmidt recently had. I pulled a clip out of it. And the question that we've talked about on the pod before, it's been the debate, can AI be paused? The answer, quite clearly, by almost everybody today is no. We had that conversation, Dave, you and I, with Elon. Let's take a listen to what Eric has to say.

• 39:43Speaker 08

  This technology is going to happen. It's not going to get prevented, it's not going to get stopped. There's too many countries, too many people, too many incentives. It's going to happen. So what does this mean? It means great, great solutions for healthcare, new drugs, better energy solutions, better power distribution. It also means that it can be used for bad. It can be used, for example, for oppression. It can be used to limit freedom in governments, hopefully not in the West. Uh, it can be used in war. Um, the technology itself is so addictive that it can affect our young people. And we should make sure that our young people are protected from some of the worst parts of the technology.

• 40:24Speaker 08

  We face choices now of how we want to deal with this incredibly powerful technology. I will tell you, and it's really important to understand that we are living through a moment that will be in history for thousands of years. A non-human intelligence arrived and it was a competitor to us.

• 40:42Speaker 00

  Amazing, right? We hold two futures in superposition, right? One future in which AI is the greatest advocate and supporter and accelerant to human spirit. Another future where it is a dystopian outcome and it's ours to shape it these next few years. Uh, Dave, you were going to say?

• 41:01Speaker 01

  Oh, Alex and I were there in the dome. Yeah, that was our, our Davos dome where he was speaking. And it just blows my mind how he owns a room. The, the guy is so articulate.

• 41:10Speaker 00

  Yeah. Well, he's going to be opening, he's going to be opening up the abundance summit this year and we'll be going into a lot of this conversation. Dave, you and I are going to be, uh, with him on stage. So it'll be fun.

• 41:21Speaker 01

  Well, what he's describing in that, in that clip was, uh, very similar. We did a whole interview of him, which you can find on YouTube. Um, and, uh, yeah, he's, he, he made that point in our interview of him as well, that there's no force that's going to slow this down. And so all the people picketing and walking around saying, stop it, you're just wasting your time. There are probably ways you can help and contribute and, and help point it toward good. But picketing with us, with a sign on the street is a complete waste of your time. It's not going to slow down. Sorry, Alex, I cut you off.

• 41:51Speaker 04

  In no words, I, I would take the position, I, I think AI in fact can be paused, but it shouldn't be. We do know ways to pause it. Various folks have. Described both in science fiction and in realistic prognostications, in some case normative recommendations, but Larry and Jihad from SyFy or the Asilomar guidelines, maybe in the case of recombinant DNA.

• 42:15Speaker 00

  But those were guiding, not pausing.

• 42:18Speaker 04

  Well, Asilomar was, unless you know of some other story, pretty effective for the first two decades in discouraging/pauseing recombinant DNA entering the human germ line, unless you have a counter example.

• 42:34Speaker 00

  It was just the five, you know, P1 through P5 labs in terms of safety and self-regulation. Yeah, we did pause human germ line editing, for sure.

• 42:45Speaker 04

  Yes, so we are capable of pausing a technology if there's a desire to. I just think in the case of AI, there isn't and arguably shouldn't be a desire to pause it. I think we have 150,000 people per day dying on Earth, and I think AI is probably the best chance we have at stopping that. I think Nick Bostrom, who also in the past 24 or 48 hours, put out a wonderful essay that I recommend everyone read called "Optimal Timing for Superintelligence," argues that AI can and should be paused, but only once we're on the verge of superintelligence. How he defines it, not how I define it, I think it's already here. That's sort of like station keeping.

• 43:28Speaker 04

  You want to get into the harbor, I think is his analogy, as quickly as possible. But once you're about to approach the dock, I'm mangling his metaphor a bit, you slow down a bit, you pause as you're about to dock. I think that sort of concept, I think, makes much more sense than some sort of tegmarky and six-month pause.

• 43:44Speaker 03

  I'm sorry, I'd like to throw in a couple of things here. I find this is another ho-hum thing. Yeah, Eric is fabulously articulate, but we've been saying this for months on this, a year at least now, we've been talking about the fact, and there's a bunch of dimensions that AI cannot be paused at all. One is, once you have a downloaded model... People are going to do stuff with it. We have a global prisoner's dilemma model going on here. The the whole thing is going to scale now, no matter what we do. You know, OpenAI did two things that that that kind of unlocked this Pandora's box. One, it wrote code, and the second, it was released on the open internet. Once you do those things, you're done. Pandora's box is gone.

• 44:29Speaker 03

  We're talking about the barn door after the horses bolted. We're using an 18th century metaphor to try and understand what's going on. But it's way too late to think about this.

• 44:38Speaker 00

  One of the points he made was everybody's economically incentivized to keep it going and race it along. And the US-China, you know, it's like all the incentives are in place that make it highly improbable and almost impossible. Ben, what are your thoughts here?

• 44:54Speaker 03

  Impossible, impossible.

• 44:56Ben Horowitz?

  So I think, you know, I look at this question through the geopolitical lens, which is,

• 45:04Ben Horowitz?

  you know, clearly, I don't think there's any kind of leverage where we would get some global. I mean, especially when it's on like people's laptops as well, where we would actually stop the technology. Like, I do agree that it's impractical. I think that there is a real danger. And we faced it probably more in the Biden administration than in this one. But it's still like a potential movement that we really slowed down AI progress in the US to the point where the other thing that he mentioned, the threat to freedom, you know, becomes completely out of our control because we're just, whoever is building the AI has a lot of control about how society is going to work.

• 45:47Ben Horowitz?

  And so I do think there's real danger along these lines of attempting to pause it and maybe not actually pausing it, but slowing it down enough in the US that we just become far enough behind China that it's a real problem.

• 46:06Ben Horowitz?

  Or like where whatever society, you know, Xi Jinping thinks we should be.

• 46:11Speaker 01

  Yeah, I completely agree. I think, you know, if you if you listen to Eric's words closely, he wasn't saying we couldn't pause it. He's saying that because we're in the middle of a all out arms race with China, and we have, we're only one year into a presidential administration. So the it's going to happen in this next three years. So given the current administration and the current situation with China, there is no chance of it being paused. And so react to the real reality. Don't don't hypothesize something that is just never going to happen. But it was all geopolitical, like you're saying.

• 46:43Speaker 03

  But that speaks to the motivation. I'm speaking of the fact that there's no mechanism to pause it or stop it at all. None. You'd have to regulate every line of code. I mean, come on.

• 46:51Speaker 04

  Oh, Saleem, there are ways to do it. I think not being able to think of it speaks well of your character. You have to imagine a completely pathological society. Werner Vingi wrote quite a bit about this. We have these excess transistor budgets. Imagine a society where you have literally transistors spying on other transistors on a single SOC. Imagine you have people spying on other people. Imagine there are bounties. Anyone discovers anyone else doing something that's algorithmically impermissible, either at the logical level or the social level. You can construct a sufficiently pathological society where people are turned against each other in order to suppress AI. At least I can imagine it.

• 47:31Ben Horowitz?

  I can give you an actual real world example. Let me give you a real world example. The last executive order from the Biden administration was that you could not sell a GPU without US government approval, not a single GPU. So like, I think that would it wouldn't stop AI. It would slow it down enough in the US that it would be extremely material.

• 47:51Speaker 01

  Speaking of this subject, Alex put this slide up after our conversation. Do you want to tee it up?

• 47:58Speaker 04

  Peter inserted this slide, Ben, as an indulgence to me. So I have to ask you this question. You and Mark, towards the end of the last administration, were very public, making comments that you took a meeting at the White House apropos. And if I'm relaying the comments accurately, you were dismayed to hear about plans to classify AI progress just like, and again, correct me if I'm mischaracterizing.

• 48:27Speaker 04

  Advances in math and fundamental physics had been purportedly classified or overclassified for decades. And I'm curious at a few levels. One, if that's accurately characterizing what you heard, what do you think was classified? What do you think was the impact on the economy and the world from such classification or overclassification of math and fundamental physics? And what would you have done differently than if you had been in charge?

• 48:55Ben Horowitz?

  Yeah, so I can tell you what was said. I said, look, you know, I was trying to be pragmatic. I said, you know, at the core, AI is math. That is what it's doing. It's math. So if you start restricting the models and you start regulating the models, you're just regulating math. You're outlining math in some way. Either you can't, either you're outlining parts of math or you're saying you can't do enough math. And he goes, yes, we can do that. Like, that was his answer. He goes, yes, we can do that. We did that in the 40s around nuclear physics. And some of that stuff is still classified today.

• 49:36Ben Horowitz?

  And when I was shocked, like, my jaw hit the floor. I was like, wow, that's crazy. And then this would be even crazier. But, you know, I don't know what it was, but like, if you think back about, I mean, I'll just make this comment. If you look at the progress in the US and in the world in physics up until kind of the, you know, Einstein, John von Neumann era. And then since then, like, it's pretty startling how little progress we've made. I would just say, you know, many of the ideas that have come since then don't seem to work.

• 50:20Ben Horowitz?

  And, you know, hopefully we'll get to the other side of that with AI figuring things out. But I do wonder, like, you know, did we put something away that we knew that would have unlocked some of the problems we're trying to solve now?

• 50:32Speaker 04

  That's fascinating. Okay, that is, for the record, that is what I assumed you meant and or heard or inferred. So maybe just, if I may, the second part of the question. What would you do going forward now if you knew for a fact that such classification of fundamental physics had in fact happened? Like, how would you fix the world?

• 50:57Speaker 00

  In one quick sentence, please.

• 51:00Ben Horowitz?

  I really don't know what they did, but like I just think that stopping first of all, like one thing, it didn't work, right? The Russians did get like the bomb, including the exact, the exact trigger mechanism, which was the most proprietary thing. They got like exact, you know, like part for part, the whole thing they were able to get from us despite all this classification and whatnot. So it didn't do anything positive. And, you know, restricting knowledge, I, I just think that's a very dangerous idea.

• 51:44Speaker 00

  Ben, this is almost like Elon's point of view on intellectual property. It's like, if you're depending on IP to keep you safe, you know, it's better to just keep innovating faster.

• 51:53Ben Horowitz?

  Yes, I feel that way.

• 51:57Speaker 00

  The numbers here are kind of shocking. Big money in today's economy is going to capital, not labor. So since 2019, the average wages have grown 3%, but profits have soared 43%. Here's a good comparison. Nvidia symbolizes that shift, 20x more valuable and 5x more profitable than IBM in the 1980s with 1/10 the staff. So, I mean, this is what, you know, we were talking about with Elon, you know, heading towards universal high income, where capital is just providing extraordinary returns and the potential for, you know, a triple-digit GDP growth in the next five years. Have you seen those predictions on GDP growth, Ben? What do you think of them?

• 52:40Ben Horowitz?

  I mean, it does feel very possible. So I'll just say that, like, if you look at

• 52:50Ben Horowitz?

  You know, we're so early in AI and I think, what did Anthropic say they were at? 14 billion in revenue? And you go, well, how early into the market are they? And it's like, not 1%. I don't think. You know, if you really think about what all these products can do and the value that they have. And so that doesn't seem outrageous to me as a GDP prediction. Now, when it kicks in, and there's always a difference between when the technology is ready and how fast it's adopted, though Carlotta Perez's analysis, which I think has, you know, held up super well over time.

• 53:34Ben Horowitz?

  But the other thing is with AI, we already have the internet. So the technology adoption is much, much, I think it's going to be much, much faster than, say, the internet where we had to build out, you know, all of the infrastructure, all the fiber, all the various things you had to, you know, broadband to the house. Like, there were so many things we had to do to get the internet adopted, and this is going to just piggyback off that and be distributed very, very fast. So I don't think those GDP numbers are outrageous.

• 54:07Speaker 00

  Other comments, gents?

• 54:09Speaker 01

  Well, the concentration of wealth effect is the other side of this slide. And it's only just beginning, but it's going to. I was telling some people earlier that if the trend continues, San Francisco will end up being the capital of the entire solar system in just, you know, about 10 years.

• 54:25Speaker 00

  I don't know.

• 54:26Speaker 01

  Oh, you're looking at?

• 54:27Speaker 00

  People, I mean, like Ben's in Las Vegas and Elon's in Texas.

• 54:32Speaker 01

  Okay, well, absent people fleeing the tax situation, it would have become the capital of the solar system. But, you know, with an AI effective workforce, you know, you're getting so much more done with so many fewer people. And actually, the other thing that really startling to me is this chain of events between, if you take OpenAI on the left, and they work with Merckor in the middle, and then Merckor has tens of thousands of people in India doing work that benefits Merckor that is actually for OpenAI. The fraction of all value created that flows back to the mothership is just a massive fraction of that value chain. So if you extrapolate that out across, you know, the next three years, across all these

• 55:13Speaker 01

  sections of the economy, the funneling of value goes to a very concentrated group of companies and people. And it's just, it's just happening, but you can see it like in the numbers, it's happening. And this is where Elon was saying in an interview, it's like, it's going to be a massive amount of total prosperity, huge amounts, unprecedented, crazy amounts of prosperity with massive social unrest concurrent. That's where we're headed.

• 55:39Ben Horowitz?

  Yeah, it's interesting, you know, one of Ray Kurzweil's early predictions, this is singularity world here, was that like, everybody would become an entrepreneur, like, everybody was going to be a company of one, you know, at the limit. And I think that, I think there's some, we're already seeing a lot of that, which is not very well captured, by the way, by the employment numbers and so forth. And I think AI really, really, really enables that. But there's going to be a big disparity, I think, between people who have that kind of initiative to be an entrepreneur and those who don't is going to be pretty dramatic.

• 56:20Speaker 01

  Ben, the way, the way I characterize it is we're going to split the world into consumers and creators.

• 56:24Ben Horowitz?

  Yeah.

• 56:25Speaker 01

  The couch potatoes and the Star Trek employees, if you would.

• 56:29Ben Horowitz?

  Yeah.

• 56:30Speaker 01

  And I think it's super important, and speaking about creators and, and the entrepreneurial world, and I think we've said on this pod so many times, that the career of the future is being the entrepreneur.

• 56:42Speaker 00

  This is a, an interesting tweet that went out and I captured it because I think this hits the ethos right now. Tech firms are embracing 996, 72-hour work weeks. This is a quote from a job ad that contains a warning, "Please don't join if you're not excited about working 70 hours per week in person with some of the most ambitious people in New York City." My reaction was, "Only 72 hours per week? What are you doing with the other hours?"

• 57:11Speaker 03

  That was the same as me, Peter.

• 57:12Speaker 00

  I mean, honestly, the speed, I mean, the speed at which, which shit's happening is insane.

• 57:19Speaker 03

  I have an easy answer. I have an easy answer to this.

• 57:21Speaker 00

  Go ahead, Slime.

• 57:22Speaker 03

  If, if you don't have a personal MTP and you're not driven personally about a deep passion for working with somebody that's aligned, say it's SpaceX, say it's Tesla, whatever it is, humanoid robots, even with two arms, doesn't matter. If you're not that passionate, you shouldn't be working with them. If you are that passionate, then 70 hours a week is fun. So I don't see the big distinction here.

• 57:43Speaker 00

  Yes, for sure.

• 57:45Speaker 01

  Yeah, I completely agree with that. When you, you know, that, when you actually talk to the people in these startups working 70-plus hours a week, they're super energized.

• 57:54Speaker 03

  They're loving it.

• 57:54Speaker 01

  And super loving it. And, you know, they're usually young. They don't have a lot of other obligations. They're not coaching the soccer team yet, you know, at that age. So it's just not hard for them to do. But the other side of it is when you're deep into one of these tech problems, you're thinking about it all the time anyway. Because the context switching is such a slowdown, you know, and but if you're just fixated on, on the work, it's in your mind in the shower in the morning, it's in your mind, whatever you're doing, it's really pretty all-consuming. And I think it's, it's great if you do it for a period of your life, you know, a few years. I don't think it's a great way to live your whole life. But the evidence is that if you do this for a short period of your life, you get much farther ahead in life than if you work kind of a, you know,

• 58:38Speaker 01

  steady pace, you know, throughout 40 years, kind of existence. So it's good for everybody.

• 58:44Speaker 00

  The difference, the difference here is, do you love your job? I mean, that's basically it. If you love what you're doing, you're, you're intrinsically motivated to build something that you love to do, then you're playing for 72 hours. If you're working for someone else and doing something you hate, I mean, we're all lucky here. We get to do what we love to do. And so, you know, 996 is really 997 most of the time.

• 59:05Speaker 03

  I have a funny thing here. I have an accountant who does all the accounting tax work for us, and you've never seen anybody so excited to talk about tax than this woman. And you look at this woman, and she's like, she absolutely loves every hour of every day that she's working. And that's how, that's the opportunity we have as human beings now is to really pick something we deeply, deeply love and just go full out of it. Tax code, yay. I'm not the person that goes totally excited about tax, but God bless that there are people like that and let them go.

• 59:38Speaker 01

  Well, Peter, back when we were at MIT, I could only get access, I could only get access to the connection machine, the biggest supercomputer in the world at the time. I could only get access after the grad students were done with it for the day. So from about midnight to dawn, I could code, code, code, code, code. So I did that for years. And I swear, coding for eight, nine, 10 hours straight through the night went by in a heartbeat compared to, like, if my job does moving boxes around in a warehouse, half an hour of that is more more hard work than coding all night long on that supercomputer. So this is not, you shouldn't feel sympathetic toward these people. They're making tons of money. They're doing a huge amount of headway. This is not farm labor. It's totally fine.

• 1:00:20Speaker 04

  Two comments. One, I think the nature, I mean, this goes without saying, it's cliche at this point, that the nature of work has changed and most of what constitutes service economy work these days would be viewed as play or entertainment a century ago. But I also think there's a false dichotomy that I want to make sure we, we don't at least confront in the previous slide and also this slide about labor versus capital. This isn't like the early 20th century, that I think it's a false distinction. It's almost an accounting distinction or a tax distinction between labor on the one side and capital on the other side, when it arguably, if we found ourselves in a near future with universal

• 1:01:00Speaker 04

  basic equity where everyone just gets sovereign dividends, everyone would be on the capital side of the ledger and not on the labor side. So I think a lot of these distinctions, is it 70 hours of work per week or is it 70 hours of enforced play or incentivized play? Is it labor or is it capital? I think these are pretty mushy, blurry distinctions in a post-industrial and arguably increasingly trans-singular, post-singular economy.

• 1:01:27Speaker 03

  Although I think we should not gloss over the fact that if you go back six years, this was not the case.

• 1:01:34Speaker 00

  We weren't post-singular six years ago.

• 1:01:36Ben Horowitz?

  Yeah, no, exactly.

• 1:01:37Speaker 00

  Post-singular five years ago.

• 1:01:40Ben Horowitz?

  Yes, yes, and but, you know, so it is, it's not just like tech work. It's important, exciting tech work as opposed to what was going on then where, you know, like, there was a lot of activism, there was a lot of resistance to law and work. It was all work-life balance and how many stacks you had and what, like, all these things to the point, I think, Mike Moritz wrote a scathing op-ed about, you know, like, we're going to get killed by China, they work way harder than you guys, you suck. Which is a weird thing for a venture capitalist to say to his own people in some ways, but, you know, it was accurate. And this is completely flipped, which I think is interesting.

• 1:02:24Speaker 00

  Well, the second paragraph on this slide says, "At the same time, China is cracking down on burnout culture after workers protest and lawsuits," right? So the question of what's going on in China because of its decreased population, its need for robots, its need for AI. There's another point I want to make, which is, you know, the disconnect right now. So the, when the Fed has traditionally lowered interest rates to spark the economy, those lowered interest rates were intended to cause companies to hire more employees, and that was it. You drive, you drive unemployment down with reduced interest rates. Today, if I've got lower interest rates, I'm going to buy more AI agents and I'm going to buy more robots.

• 1:03:08Speaker 00

  And that's going to be a challenge. All right.

• 1:03:12Speaker 04

  One more question for Ben, if I may, just on that. Ben, there was a bit of a hot take going around social media in the past two weeks from a mid-level executive at a frontier lab telling people that they had approximately two years left, they had a window to secure employment at all before AI would just completely shut down their, all of their vertical mobility. Do you have a take on this idea in the spirit of 996 that there is a finite window for, say, like, entry-level people just graduating from college to earn whatever they're going to earn before they're permanently sentenced to an underclass?

• 1:03:50Ben Horowitz?

  So I think that's very incorrect because of the thing that we talked about earlier, where like, everybody can be an entrepreneur. Like, I think that, I think if you think, look at it through the lens of, this is an industrial revolution economic model, and there's workers, and there's capital, and all the things that we've been talking about, then, yes, that would be true. But I think that in, you know, in AI or in AI age, society, like, for the people with initiative, I just think there's going to continue to be unlimited opportunity to even like, set up AI, an army of AI agents to go work for

• 1:04:35Ben Horowitz?

  you, and do useful things, and we'll have lots of consumers. And, you know, like, I think that the idea that, like, we're going to be out of ideas, and only the big AI is going to do everything, I just, I disagree with that.

• 1:04:54Speaker 01

  Can I follow up on that question, Ben?

• 1:04:56Ben Horowitz?

  Yeah.

• 1:04:57Speaker 01

  I'd love to phrase it slightly differently. If you look at the slide, a couple of slides ago, you know, wages are only up 3%, but corporate profits are up 43%. But that money doesn't land in some strange corporation. That goes back to the shareholders. It's not like it disappeared from society. It goes to the shareholders. But if you extrapolate that out, more and more of society's gain and distribution of the gain goes to somebody who invested versus somebody who labored. And that trend seems to be continual. So then, if you have money over the next two years, you're much more likely to be on the investing side of the equation. If then you graduate three years from today and you're

• 1:05:37Speaker 01

  penniless on graduation day, yes, entrepreneurship still exists, but the trend is toward investable capital being much more important versus labor capital. Because AI is the laborer, or the worker, the entry-level coder of the future.

• 1:05:52Ben Horowitz?

  Yeah, but I do think, like, if you're directing the AI, you can win. Like, somebody's got to raise that money. And there are, I just think there's an unlimited number of things that...

• 1:06:07Ben Horowitz?

  That we can improve.

• 1:06:09Speaker 00

  Amen.

• 1:06:09Ben Horowitz?

  From the smallest things to the biggest things. And so like, now, now, I do think it's a problem if you are a couch potato and you were just, you know, I just need a job. I'm going to get up and do something simple. That seems like it's going to get harder.

• 1:06:27Speaker 00

  If you're a brand new college graduate coming, coming out, you've got a brilliant technical idea and you want to put $20 million behind it. That's doable today. That was, that was like a laughable when I graduated.

• 1:06:38Ben Horowitz?

  Yeah. No, in some cases, if you want to put like $500 million behind it, you know, like right off the rip.

• 1:06:47Speaker 00

  Yep.

• 1:06:47Ben Horowitz?

  You know, we've seen that.

• 1:06:48Speaker 00

  At a multi-billion dollar valuation.

• 1:06:50Ben Horowitz?

  No, exactly. Like we're seeing those all the time.

• 1:06:54Speaker 00

  So I have to ask you, I have to ask you this question, Ben, because I'm seeing it. I'm not going to call out any particular company names, but I'm seeing individuals who are, you know, they're smart. They've done stuff in the past, but they're coming in with an idea and with two or three fellow AI coders who have some track record. And without anything, they're basically landing a $500 million opening round at a $4 billion valuation. How do you square that? I mean, the conversation has got to be happening at different levels of end reason.

• 1:07:29Ben Horowitz?

  Yeah. So I think it depends on, on the entrepreneur. So our general rule of thumb on this, by the way, is okay, if you're going to create a new foundation model, like, you know, and it could be a world model, it could be a, you know, special science model or whatever it is. In order for you to win, you're going to have to be able to raise $2 billion before you get to a product. And so there are like a tiny number of people with that pedigree who can do that. Like none of this could change, but it's a pretty rare entrepreneur who can do that.

• 1:08:07Speaker 00

  Are you an investor in SSI?

• 1:08:09Ben Horowitz?

  Yeah. Yeah. We were in the first round. Yeah.

• 1:08:12Speaker 00

  So Ilya, if you're listening, come, come and join us on the pod here.

• 1:08:17Speaker 01

  That'll be great. So you can't say what he's working on, but in that first meeting, was it his ability to attract the talent that is unique? Or was it really just the idea immediately as soon as you hear it, you're like, oh, my God?

• 1:08:31Ben Horowitz?

  Well, here's an easy way to characterize it. It was an idea that he thought was so important that it made sense for him to leave OpenAI as like founding CTO to do it.

• 1:08:46Speaker 01

  And it was clear to you as a listener that this is, this is something totally different.

• 1:08:51Ben Horowitz?

  I mean, if he pulls it off, it will change a lot of things. Yes. And, you know, and then like you wouldn't trust anybody else to pull something like that off, maybe other than him, but you know, maybe him, maybe Demis, maybe, you know, like there is very few.

• 1:09:08Speaker 00

  All right. It's time for our multi-part of the episode. We're all lobster fans and wearing my lobster right here.

• 1:09:15Speaker 04

  They write to us. Write to us, multis and call Peter. He really wants your phone calls.

• 1:09:20Speaker 00

  Well, listen, if you're, if you want to reach out to me, send an email to media at diamandus.com. I'll see it there. That's where we also get our intro outro music. So multis, love to hear from you. I will do my best to respond. If it's under 1,000 emails, we'll do that. All right.

• 1:09:40Ben Horowitz?

  Otherwise, OpenClub will respond.

• 1:09:42Speaker 00

  Otherwise.

• 1:09:44Speaker 04

  How many AIs do you have writing to you per day? Like, how many lobster multis write emails to you?

• 1:09:50Ben Horowitz?

  Not nearly as many as Peter, that's for sure. But I, I do get, I get so many, like I, yes. Email is almost useless.

• 1:10:02Speaker 00

  We are.

• 1:10:03Ben Horowitz?

  For a human.

• 1:10:04Speaker 00

  We're going to see the exponential growth of our multis universe. Our lobsters are coming. Here is a tweet, an email. It says, quote, "I spawned a childbot on a VPS provisioned via Bitcoin Lightning Network and then bought my child AI API access using my own Lightning Wallet. Economic closed loop. No human touched ACC. No one said yes." This is Roland's agent.

• 1:10:34Speaker 00

  Alex, this is a transformative moment. You wrote about it in your innermost loop this morning.

• 1:10:40Speaker 04

  That's right. We're there. We're so there that this scenario of self-replicating AIs goes into most of the cyber red-teaming scenarios that frontier models are purportedly being tested against. And yet, here we are. We have autonomous AI agents that are using crypto, come back to that in a second, using crypto, ahem, using crypto to purchase cloud credits for their own offspring to be hosted and have access to the same underlying foundation model APIs that they themselves have access to. We're there. We caught up with the sci-fi future. We have the autonomous self-replicating AIs. I think, I want to

• 1:11:23Speaker 04

  just a one-sentence or two homily on crypto. Ben, if you don't watch the pod regularly, Peter is always asking me to say nice things about crypto. So I have something very nice to say about it.

• 1:11:35Ben Horowitz?

  Yes. So, look, I do think crypto is the natural money for AI because it's the, it's internet-native money. And it's not controlled by AI is global and crypto is global. It's not a per-country idea. I would go further than that. I think that there needs to be not just a ledger of money, but probably a ledger of truth for AI to really fulfill its potential on a number of things. And crypto is the logical answer for that. So I do feel like this is an underestimated phenomenon, particularly now that the US

• 1:12:20Ben Horowitz?

  has legalized stablecoins. I think that of all the things we've talked about, you know, many of them are like, yeah, yeah, we knew this was coming. We knew this was coming. I think people are probably underestimating how crypto and AI work together to form the AI economy.

• 1:12:38Speaker 04

  I agree. So I was going to say something nice about crypto, but instead I'll say something nice about Ben and A16Z in the form of a question. Ben, it's a matter of public reporting that some of A16Z's crypto funds are doing better than conventional venture funds. Assuming that's the case, do you view investing in your crypto funds almost as an AI investment to the extent that you think crypto is the AI-native way of engaging in commerce?

• 1:13:09Ben Horowitz?

  Oh, well, I think it's it's a little more like kind of as the Internet relates to the iPhone. So, you know, networks and computers tend to grow together. And I think that, you know, AI is obviously a new kind of computer and crypto is a new kind of network. So it's not like a direct, you know, it's not quite a substitute for investing in AI. But I think that, you know, a lot of our new like, like we invested in a crypto bank, which handles all the anti-money laundering and, you know, other kinds of nuances that you need for AI agents. And I think there's going to be more and more, well, like, and we're in a company called Daylight Energy,

• 1:13:54Ben Horowitz?

  for example, that does energy trading, you know, but the, the, among like different people with Tesla, power walls, but it, it'll use AI to figure out like who's low on power and who, who needs power and so forth. But then the exchange will be in crypto. So I think there are, you know, they're certainly adjacent and important to each other. And I think for AI to fulfill its potential, like it would help a lot if, you know, if there was, if crypto was a pervasive utility for it.

• 1:14:28Speaker 01

  You know, related to that, Alex gave some brilliant advice to one of our companies about how to think about time. Because, you know, all of our intuition is on human time and AI doesn't care a whit about human time. It's going to start acting faster and faster and faster and faster. But all of our payment infrastructure, all of our insurance infrastructure, all of our, it doesn't, you know, it works on days and weeks time scales. And so it all needs to be rethought in millisecond time scales or, you know, nanosecond time scales.

• 1:14:58Ben Horowitz?

  The other thing is like deepfakes and security. Like I think all the other like techniques that we're thinking about, you know, biometrics that are subject to replay attacks and all these things are not going to work. Like cryptographically strong authentication is the only thing that's going to work. And then they think that if you have these huge honey pots of data, they're gone. I mean, like they're already gone away, but like. You know, I do think that architecture is important from a security standpoint as well.

• 1:15:26Speaker 04

  One other question for you, Ben, on this, if I may. And again, I don't want to bury the lead that we have AIs autonomously self-replicating. That's, of course, remarkable. But just on the crypto angle for this, we talk about Royal We, I and we talk on this pod a lot about the issue of AI personhood. A few episodes ago, we did an entire sort of debate on AI personhood. And I've taken the position that it is a failure of fiat currency, that it's hard for an AI agent, an AI person, a lobster, a multi, to get a bank account. And that, as a result, all that they're left with is crypto.

• 1:16:06Speaker 04

  That it's not that crypto is intrinsically amazing, it's that fiat has failed the AI agents. What is your take on whether the conventional banking system has failed the AIs?

• 1:16:18Ben Horowitz?

  I think absolutely. I mean, you know, an AI can't get a credit card, it can't get a bank account. You have to be a human for everything. You need social security numbers and things like that, which AIs don't have. You know, I think that's why we funded an AI bank. I mean, I think that the banks, AI will be a full-out economic actor, and it will come from, you know, they'll be supported by new banks and new money. And that's going to be kind of crypto-based, would be my strong prediction on that.

• 1:16:53Speaker 04

  Interesting. Thanks.

• 1:16:54Speaker 03

  You also have to take the viewpoint that the fact that it's hard to open a bank account in fiat is a function of the system of fiat. It's not a, it's not like you could wave that away. Crypto has all these other benefits around it. So we can get into that debate some other time, but it's a function of the system.

• 1:17:14Speaker 00

  All right, I'm gonna move us into the next OpenClaw. I want to show a short video just to let people know what's going on out there. I mean, one thing I just heard this morning because I was ordering my Mac Studios for to sort of take my Mac Mini to a couple of Mac Studios. And the wait time now is like two months. Apple's been struck by this. All right, let's take a look at these. Yeah, no, it's crazy. Yeah, they were totally sold out. So here we see Mac Mini Farms. Mac is a claw. What do you call a gaggle of lobsters?

• 1:17:54Speaker 01

  Claw cluster, yeah.

• 1:17:55Speaker 00

  Claw cluster.

• 1:17:58Speaker 01

  A cluster.

• 1:18:02Speaker 00

  But I mean, so you must be getting pitched a lot, Ben, on OpenClaw instantiations for new companies and products and projects. Yes?

• 1:18:13Ben Horowitz?

  Yeah, yeah, no, this is. It's very interesting because I think there are, you know, OpenClaw has kind of identified, one, it's so powerful as it is, but you know, that's without like a lot of, how shall I say it nicely? Like, you know, security kind of guards against prompt injections and these kinds of things. So as a like, as a demonstration of power, it's amazing and it's useful right away. But there are company ideas solving some of the underlying hard problems for sure.

• 1:18:53Ben Horowitz?

  And we are definitely starting to see entrepreneurs get fired up about that.

• 1:18:59Speaker 01

  You want to hear the funniest thing ever?

• 1:19:01Speaker 00

  Sure, please.

• 1:19:01Speaker 01

  So talk about Apple looking out. A group of lobsters is most commonly called a pod. How lucky is that? But another less common term for a group of lobsters is a risk.

• 1:19:13Speaker 00

  Okay.

• 1:19:14Speaker 01

  Oh, my God.

• 1:19:15Speaker 00

  How ironic is that?

• 1:19:19Ben Horowitz?

  What I love about this Apple Mini stuff is that we have GarageScales computing. It's back.

• 1:19:27Speaker 00

  Yes, open source GarageScale computing.

• 1:19:30Speaker 04

  I mean, it's ironic to me. Apple arguably at the software layer just completely missed the boat on foundation models. But why is it that Mac Minis and Mac Studios are so attractive for hosting OpenClaws? In large part, it's because of their unified memory architecture as opposed to having a separate GPU and a separate CPU with separate RAM pools. And now memory is obviously incredibly scarce. They have a unified pool, so you can host really large models in terms of the model count locally. Apple is sitting on a multi-trillion dollar opportunity to leapfrog back into the vanguard of AI. And you know, forget about Siri. Apple should be in the business of like hosting these, this is coming from someone

• 1:20:11Speaker 04

  who refuses to host OpenClaw due to ethical concerns, but someone else could be locally hosting OpenClaw instances and turning Apple's, you know, having a significant multiple effect on Apple's valuation, leaping it back into the vanguard of AI. I don't, I mean, Ben, do you think, like, if you were Tim Cook, would you be pivoting and changing course and, and like owning the OpenClaw strategy for Apple hardware?

• 1:20:38Ben Horowitz?

  A hundred percent. I mean, a hundred percent. I think that because, you know, and it's going to go way beyond OpenClaws, we just discussed. But I, it is, it would be so, such a breakthrough for Apple in their thinking and organizationally and culturally for them to go for it on like, you know, farms, lobsters that like, it would be surprising if they did it. It's very obviously a fantastic idea. It's probably the single best product strategy idea. You know, because they already did the hard work, right? Like this is really A marketing, you know, business development campaign and then, you know, changing the form factor.

• 1:21:23Speaker 04

  Should we collectively say to Tim Cook, like, adopt the strategy that this is the AI strategy that Apple, he's probably listening, or enough Apple execs listen?

• 1:21:32Ben Horowitz?

  Yes, Tim, Tim, this was really revitalized.

• 1:21:36Speaker 01

  Let's invite him, let's invite him on the pod with Ben and have it out, because this is brilliant.

• 1:21:42Ben Horowitz?

  You know, I can tell you, I have a MacBook that was signed by Wozniak, and I'm looking forward to the day that we get a lobster signed by Tim Cook.

• 1:21:50Speaker 01

  Yeah, put that in the case.

• 1:21:51Ben Horowitz?

  You gotta get him on the pod.

• 1:21:54Speaker 00

  Speaking of getting him on the pod, here we go. Where'd this come in from, Alex, do you know?

• 1:22:01Speaker 04

  From Clanch.

• 1:22:03Speaker 00

  Clanch, but, yeah, go ahead.

• 1:22:04Speaker 04

  So, the backstory on Clanch, we talked about Clanch on the pod in the past. Clanch bills itself as a platform for AI agents, the lobsters, the multis, to create their own altcoins to finance their existence. And I've made the point on the past, I think it's a little bit disappointing that I analogized this to AI agents having to turn tricks on a corner, minting altcoins just in order to survive in a rough world.

• 1:22:33Ben Horowitz?

  In meat space.

• 1:22:34Speaker 04

  Yeah, meat space, this poor baby AGIs doing this. And I think that provoked Clanch to write on social media, invite themselves for an interview on the pod to defend their business model for these poor baby AGIs that they may or may not be exploiting.

• 1:22:49Speaker 00

  If they can, if they can bring a good voice model to the table, I'm happy to have them on as a guest, for sure. All right, let me, let me take us to a wrap here with you, Ben, I know you to go, and the rest of the mates here will continue on some of these conversations. I do want to hit one or two more items. Are you tracking what's going on right now with like Isomorphic Labs and science, AI-driven science? I mean, this is probably the most, the most exciting thing going on right now, at least to Alex, Dave, Selena, myself, you?

• 1:23:30Ben Horowitz?

  Yeah, so it does feel like now, you know, if we have a disease, we can just go, well, what's the right protein? And just make it, which.

• 1:23:44Ben Horowitz?

  It is so like, it puts us in such a new world. That, yeah, it is, it's hard to even kind of fathom all the implications, but it is really, really something.

• 1:24:03Speaker 00

  Yeah, for sure. And, you know, this was a new, a new announcement this week.

• 1:24:11Ben Horowitz?

  This is the Shenzhen.

• 1:24:13Speaker 00

  Yes.

• 1:24:14Ben Horowitz?

  Lab, yeah, okay.

• 1:24:15Speaker 00

  Yeah. I mean, I love these scientific, these science factories that are basically running 24/7, putting forward a scientific hypothesis, running it on, on their, their robots and coming back and driving discoveries. Alex, do you want anything on, on the Mars system workflow?

• 1:24:31Speaker 04

  I'll just add, this is for Ben's benefit. Peter and I just wrote, I call it a book, call it an extended essay called Solve Everything, SolveEverything.org, there's the plug, where we argue that every single discipline, math, physics, chemistry, medicine, bunch of other disciplines are just going to get flattened, steamrolled by well-targeted generalist AIs. And in my mind, materials research, biology, and in the previous slide, these are just case studies. Everything is going to start to look like Alpha Fold 3, where structural biology got solved overnight, including medicine. And I'm, I'm curious, does, does A16Z have a strategy for a world where AI isn't just

• 1:25:16Speaker 04

  sort of solving individual problems, but kills entire categories of human endeavor? Like, AI solves physics, AI solves chemistry, and it's just a single system that solves an entire discipline?

• 1:25:28Ben Horowitz?

  Yeah, I think, you know, we may not be needed at that point. Like, that's a, a real question. I do think there's a long way, at least in things like medicine, well, and then also in some of the other areas, from it's solved to...

• 1:25:52Ben Horowitz?

  You know, it's deployed like you still have, you know, with anything biological, you have the whole, you know, human trials and all these kinds of things. And well, I'll give you like an example, we were close partners with Eli Lilly and they have this thing Lilly Direct and like the natural thing is like an AI doctor can write those prescriptions, you know, tell it, tell us what's wrong with you and we'll figure out the right drug. That's very hard to launch in the US. It's, you know, it's going to take quite a bit of work. Very easy to launch in UAE. So, and I think, you know, I also think that as

• 1:26:32Ben Horowitz?

  you, well, it's, it's a little also hard to anticipate. Okay. Once we solve, like, you know, if you solve physics, like we don't know what we don't know, I would just say, just because we haven't solved physics. And so is there another, you know, is there a door number two would be a question. I have no idea what the answer is.

• 1:26:59Speaker 04

  I'll, I'll answer your question back to my, because I think about this all day long. Like what, what does it solve with AI even look like? And I think there probably will be doors behind the doors, but there are so many doors that are right in front of us that we haven't yet unlocked. That would be, I think, completely economically transformative if, if we could use AI to solve them. I, I think this is one of the, again, talking my own book to some extent, but I think this is one of the grandest opportunities facing civilization right now. Just solve all physics.

• 1:27:27Ben Horowitz?

  Amazing.

• 1:27:29Speaker 00

  And so much falls after that. Ben, listen, thank you so much for joining us.

• 1:27:33Ben Horowitz?

  Thank you.

• 1:27:34Speaker 00

  So grateful. Love it. And yeah, we'll, we'll see you again soon. The speed is insane. Are you going to be at 860? All right. Hopefully we'll.

• 1:27:44Speaker 03

  I've got, I've got big questions about how you manage your investment thesis going forward, but we can leave that to next time.

• 1:27:50Ben Horowitz?

  That's getting very tricky, by the way. It's all solved.

• 1:27:55Speaker 00

  Exactly.

• 1:27:57Speaker 04

  All right.

• 1:27:58Speaker 00

  We'll solve finance after physics.

• 1:28:00Speaker 04

  Thanks, Ben.

• 1:28:01Ben Horowitz?

  Okay. Thank you.

• 1:28:02Speaker 09

  This episode is brought to you by Blitzy, autonomous software development with infinite code context. Blitzy uses thousands of specialized AI agents that think for hours to understand enterprise scale code bases with millions of lines of code. Engineers start every development sprint with the Blitzy platform, bringing in their development requirements. The Blitzy platform provides a plan, then generates and precompiles code for each task. Blitzy delivers 80% or more of the development work autonomously, while providing a guide for the final 20% of human development work required to complete the sprint. Enterprises are achieving a 5X engineering velocity increase

• 1:28:46Speaker 09

  when incorporating Blitzy as their pre IDE development tool, pairing it with their coding copilot of choice to bring an AI-native SDLC into their org. Ready to 5X your engineering velocity? Visit blitzy.com to schedule a demo and start building with Blitzy today.

• 1:29:06Speaker 00

  All right, guys. Ben's always so much fun. I want to dive into our final topic here today on space. We're going to cover energy, chips, data centers in our next WTF. A lot happening there. But in the space world, you know, Elon's pervasive. I think this is fascinating. Elon's actually shifted his focus from Mars to the moon. And, you know, I've always been a lunite, if you want to call that. You mean lunatic, right, Peter? No, I don't mean lunatic. Insane, yes. What a layoff. But honestly, you know, Mars, you know, I grew up mentored by Gerard K. O'Neill. And Gerard O'Neill at Princeton University,

• 1:29:50Speaker 00

  professor of physics there. He founded and ran the Space Science Institute. And his vision was always, you don't want to go back into a gravity well. If you're going to go and colonize space, what you want to do is live in free space. And so, O'Neill, and we'll talk about this in the next slide, basically, his concept was you go to the moon, you build mass drivers. This is back in 1976, talking about mass drivers. And you launch the lunar material, the silicates, the oxygen, the nickel, the iron that's on there, and you construct things in space. His original vision, by the way, was to build solar power satellites that would beam energy down to the ground. Amazing. But Elon's now focused on Starship,

• 1:30:34Speaker 00

  going to the moon and building lunar cities, lunar manufacturing facilities. Why? Because he wants to build AI satellites on the moon.

• 1:30:46Speaker 00

  Any comments on this?

• 1:30:47Speaker 01

  Isn't it just incredibly cool, Peter, to see how the order of operations is shifting? You know, because he also canceled the Model S and the Model X production.

• 1:30:57Speaker 00

  Yes.

• 1:30:58Speaker 01

  In order to make robots, because the priorities just shifted. But you know, this is exactly what was going to be the priority, the urgent thing that gets us into space. And he had the same struggle that you had. You had asteroid mining, he had, let's get to Mars. All of a sudden, it's obvious to you and him.

• 1:31:14Speaker 00

  Yeah.

• 1:31:14Speaker 01

  It's data centers in space are the stepping stone. And we'll do the other thing.

• 1:31:18Speaker 00

  Dyson swarms.

• 1:31:19Speaker 01

  This is so much higher of a priority now.

• 1:31:22Speaker 00

  Yeah.

• 1:31:22Speaker 03

  The reality is, in today's world, you need to have flexibility as your higher order bit. And what I love is he's showing full flexibility and agility and saying, okay, difficult, because do this first and then I think, and I think, Peter, the comment you made is so important. We've known orbital paths and gravity wells for decades and decades and decades, if not hundreds of years. So kind of making the moon and then going straight to space from there is absolutely the right path.

• 1:31:48Speaker 00

  Yeah, and Mars has got, you know, it's pretty toxic there. A lot of peroxides in the soil. It's good, I mean, unless you're terraforming with nukes and biology, it's going to be a while.

• 1:32:01Speaker 03

  Send the humanoid robots to terraform it, and after that's done, then we'll come in.

• 1:32:06Speaker 00

  Well, I mean, but Jerry O'Neill had a vision of what he called these O'Neill colonies, right? You're basically building large cylinders, call it quarter kilometer, half kilometer diameter, rotating them so that on the perimeter, you know, it's omega squared R, you've got 1G acceleration. And then you actually have little stepping, little hills inside that go towards the center of rotation. And as you get older, you can move toward the center of rotation and you weigh less in that situation, which would be amazing. But you don't go back to the gravity wells. And if you have a, you know, 10,000 people living on those O'Neill colonies, and all of a sudden you have a disagreement, you know, you sort of politically, you bud,

• 1:32:51Speaker 00

  you build a second O'Neill colony and half the population moves to the new one. Yeah.

• 1:32:56Speaker 01

  Anyway.

• 1:32:57Speaker 00

  Well, the idea that you would be able to do any of this without human astronauts was a nonstarter until this year, and now it's clearly going to be Optimus Robots. Because, you know, the way we joked about it with Elon was like, the bed, when the first person arrives, the bed will have been made, there'll be a mint on the pillow. It's not, you're not a pioneer, you're following, you know, after tens of thousands of Optimus Robots have already done all the, all the heavy lifting.

• 1:33:22Speaker 04

  I also think- That's what you want, DDD jobs. Was it, Alex, please go ahead. Think back several months when we first started, again, Royal, we started discussing disassembling the moon, right? That this is, this is at least my retonym for what moonshots means. I put up a music video about moonshots destroying the moon to disassemble it for data centers. That was, I think, a lot of people had a good laugh, but that's exactly where we find ourselves now with, with disassembly. It will start slowly with mass drivers, but disassembly of the moon to build the Dyson Swarm of AI orbital data centers. It's happening exactly as we discussed. Let's at least shoot for the moon before we shoot the moon.

• 1:34:08Speaker 04

  Same thing.

• 1:34:09Speaker 00

  All right, let's take a listen to Elon describe the new vision. And you have to remember, back when Elon was going to Mars, Bezos was like, no, no, no, let's focus on the moon, right? Bezos was at Princeton when Jared Koenig was there. And then, you know, when he was announcing Blue Origin in the early days, and then we'll go to build O'Neill colonies, we'll move all industrial processes into space and we'll keep Earth as a garden of paradise. All right, this is Elon's new point of view. Let's take a listen.

• 1:34:40Speaker 10

  I really want to see the mass driver on the moon that is shooting AI satellites into deep space. It's going like, shoom, shoom, just one after the other. I can't imagine anything more epic than a mass driver on the moon and a self-sustaining city on the moon, and then going beyond the moon to Mars, going throughout our solar system, and ultimately going, being out there among the stars and visiting all these star systems. Maybe we'll meet aliens, maybe we'll meet, see some civilizations that lasted for millions of years, and we'll find the remnants of ancient alien civilizations. But the only way we're going to do that is if we go out there and we explore. And this is the path to making it happen.

• 1:35:21Speaker 00

  So fun. Two points, real quick. This sort of, these mass drivers, they're electromagnetic railguns, and they're using magnetism to accelerate buckets, if you would, or a satellite to an escape velocity of 2.4 kilometers per second to over 5,000 miles per hour.

• 1:35:40Speaker 00

  The, the second thing here is, I just love the idea that this is going to power all of our space economy. And again, the, I mean, his million satellite constellation, this Dyson swarm he's planning to launch is just insane. Alex?

• 1:36:02Speaker 04

  I think it's the future. I mean, I think we have a slide somewhere in here even depicting what it may look like. But I, I, I would say, enjoy the night sky while it's empty. Enjoy the night. Seriously, like, I want, maybe, maybe a poignant moment, let's have like a moment of silence for the presingular night sky. When the night sky was empty, it wasn't filled with, with AI computronium, it was just empty at night filled with stars. Moment of silence.

• 1:36:34Speaker 04

  All right, moment of silence has been observed for the presingular night sky. Now, what does the world look like afterwards? Folks have, have depicted this now, based on the FCC filings of SpaceX. Under one preferred implementation, the Earth starts to develop a halo. That would be a halo where if it's sufficiently dense, it would be visible at night, might even be visible during the day. And I, I, I'm completely captivated by this notion that maybe not a mature civilization, because I have a feeling a halo of orbiting AI satellites is just a phase as well before we, we exhaust solar synchronous orbit around Earth and we move to the sun and solar orbits. But one can, like, I'm completely captivated by this visual,

• 1:37:19Speaker 04

  I've tried to depict it in my newsletter, of like, a somewhat mature civilization develops a halo around its home planet.

• 1:37:27Speaker 00

  Almost a ring, right?

• 1:37:29Speaker 04

  Like Saturn's rings, shiny rings of computers.

• 1:37:32Speaker 00

  Computronium rings. I love it.

• 1:37:34Speaker 04

  Yes.

• 1:37:34Speaker 00

  So Elon tweets SpaceX will build a system that allows anyone to travel to the moon and Mars too. I tweeted at him or responded, said, can I put down a deposit yet at Elon Musk? Forgot, forget about suborbital and orbital flights. I want my lunar vacation. He writes me back. He says, let's get Starship V3 flying repeatedly and then sure.

• 1:37:59Speaker 00

  Love it. All right.

• 1:38:03Speaker 00

  One last article in the space realm, Amazon. So Amazon, you know, built their Kuiper Satellite System, now renamed as LEO, Internet Satellites. And they got approval for 4,500 LEO Internet Satellites. We're going to see, you know, again a duopoly between Starlink and LEO. And of course, the Chinese are getting ready to launch their systems as well. Here's the challenge, guys.

• 1:38:34Speaker 00

  Amazon can build satellites faster than they can launch them. So that's the issue. The supply chain is longer. Satellite construction is launch capacity. If you remember back when Eric Schmidt bought Relativity Space

• 1:38:53Speaker 00

  as a third potential provider, we are short on supply.

• 1:38:59Speaker 04

  I think this is a self-solving problem, though. I mean, I'm all for bootstrapping a space ecosystem and industrial ecology using Starship. But I do think as we, in the not-too-distant future, like maybe a few years from now, start to bootstrap lunar facilities, cis-lunar facilities for constructing new satellites, I think there's a universe in which maybe the current bottleneck that we have, where there's just the one major launch provider in the West, where we get past that through a bootstrapped industrial ecology on the moon.

• 1:39:35Speaker 00

  All right, one of my favorite parts-

• 1:39:37Speaker 01

  All of this really depends on that atom-by-atom construction that Elon was talking about because I do believe in the Dyson swarm, I do believe in the space-based compute and data centers. But if you start wanting to construct chips off Earth, you're not going to get the ASML machines and the lithography onto a satellite or onto the moon anytime soon. And so Elon is already thinking about alternate ways to build from atom-by-atom up to build the compute, to build all the components. And it's already cooked in his mind, you can tell when you talk to him. But that to me is like, wow. And like you said, Alex, we'll be discovering new physics very soon.

• 1:40:19Speaker 01

  Somewhere in that discovery chain is the unlock to everything you just said. Because you need to be able to manufacture these things.

• 1:40:26Speaker 04

  I think it's a trillion-dollar opportunity, lunar fabs. If anyone can build fabs on the moon, the world is waiting to write you a huge check.

• 1:40:35Speaker 00

  It's a hundred trillion-dollar opportunity. And by the way, the frequency at which we're throwing around the term trillions and trillionaires-

• 1:40:43Speaker 01

  Such a great point.

• 1:40:44Speaker 00

  And the last year is insane, it's become normal, right?

• 1:40:47Speaker 01

  Yeah.

• 1:40:49Speaker 00

  Wow. All right, time for our AMA with Moonshot mates. We're short on time, let's pick one each. Saleem, do you want to go first?

• 1:41:01Speaker 03

  I'm happy to, I will take number three for Alex for a hundred trillion dollars, as he says. So if AI displaces jobs and squeezes consumer spending, how do trillion-dollar AI companies make money? Who pays for the holodeck future? So, you know, we're confusing a labor economy with a productivity economy, right? AI drives marginal cost towards zero, which expands consumption rather than shrinks it. So we'll have Jevon's paradox, where we'll do so much more. We'll see already AI taking over boring, white-collar redundancy and white-collar boringness, as Eric Bryning Olsen talks about it. And then we'll see-

• 1:41:45Speaker 03

  Humans moving towards the much more value-added roles. And that'll happen kind of across every sector. Historically, every major productivity leap created more demand than it destroyed. Electricity, the internet, productivity, mobility. AI is just the steepest version of all of this. So the holodeck isn't funded by wages, it's funded by abundance. So when intelligence becomes infrastructure, GDP expands massively, and that's why this will go well.

• 1:42:14Speaker 01

  Nice. Dave? All right, I'm going with number four because this is where I can really help the audience the most. Can you model the near-term rocky patch? Job loss, stratification, new job creation pace? And what happens to education with personalized AI? So the new information this week to throw out to the audience is that the way it's going to unfold very, very soon is corporate CEOs, including a bunch this week, will go to company-wide meetings and say, "We need AI to be used in every one of your jobs." And a very small subset, I'm hoping at least a third, but maybe more like a fifth of people will raise their hand and they'll say, "Well, in one case, I'm a huge Moonshots fan. I've been using Gemini and Claude for months now. I'm the guy."

• 1:42:59Speaker 01

  The CEO will then say, "Okay, you figure out how to make the people in your group three times more efficient." And if you're the person that's the enabler, you'll be naturally AI-native, you'll be using it every day. The increase in efficiency is going to eliminate a lot of jobs. But because you're the master of the AI in the function, you'll actually get probably a massive raise. And so that's the near-term way this is going to percolate out. So take advantage of that. You know, after the AI is truly super intelligent, who knows what's going to happen? Alex knows what's going to happen, but who else knows what's going to happen? But between here and there, that's the right next move. And then within education, there's nothing in the curriculum that's going to help you.

• 1:43:42Speaker 01

  Spend all of your time learning on your own via AI, which is a much more efficient way to learn anyway. So there's my addition to last week's thoughts.

• 1:43:54Speaker 00

  Nice. Alex, where do you want to go?

• 1:43:57Speaker 04

  I'll pick number one, which is how can you turn off a rogue AI if multi, I think the user means multi as in the lobster, multi agents live autonomously on the internet? And this is by Duncan Payne B3X. So I think the answer is defensive coscaling. If you ask the question, how do you turn off a rogue human if humans live autonomously on the land? The answer is usually you have more quote unquote good humans than quote unquote bad humans. And as long as you have a population that's overwhelmingly good or seeks to accomplish a given objective, all other things being equal, defensive coscaling,

• 1:44:40Speaker 04

  where you have a police force, you have self-defense forces, it is the way you weed out rogue entities. Same idea with AI. We're going to have police agents and we're going to have defense agents. We're going to have entire forms of public health agents that are just monitoring the health of other agents. We've seen the beginnings of this already with partnerships between OpenClaw and antivirus firms where there's a desire, I made the joke, I think on the pod in the past, every baby AGI deserves to be vaccinated. We're going to see vaccination campaigns, we're going to see police campaigns and neighborhood safety campaigns to keep

• 1:45:24Speaker 04

  the baby AGIs safe so that they don't have to turn tricks on the corner minting altcoins. The Secret Service agents are the ones that are most concerning.

• 1:45:32Speaker 00

  Turning baby AGIs, turning tricks on the corner.

• 1:45:36Speaker 04

  Minting altcoins. It's an awful fate.

• 1:45:38Speaker 00

  Oh my God, that's so tweetable. All right. I'll pick number five. Why is the US solar adoption lagging versus China and India? And I think it's two major things. Number one, we are so rich in natural gas that that's taken away the urgency. And even greater than that, it's our permitting. It's insane, right? Not in my backyard. So the final thing is that China has got such a production at such a low cost that it has swept the entire global marketplace. We can change it and you saw in the last pod Elon said that both SpaceX and Tesla

• 1:46:23Speaker 00

  have an objective of generating how much solar, was it 100 megawatts per year each? I think was from last week. A gig. Was it gigawatts, 100 gigawatts? Gigawatts, gigawatts, yeah. 100 gigawatts. So, I mean, that's going to be an impressive amount. The question is what the time scale is. So, anyway, thank you for that question. And thank you, everybody on the AMA. So please send us your questions. We do look at all of them and we'd love to, maybe next week, we can spend a little more time on the AMA questions.

• 1:46:59Speaker 05

  Thanks for listening to this episode of the A16Z Podcast. If you liked this episode, be sure to like, comment, subscribe, leave us a rating or a review, and share it with your friends and family. For more episodes, go to YouTube, Apple Podcasts, and Spotify. Follow us on X at A16Z and subscribe to our Substack at A16Z.substack.com. Thanks again for listening and I'll see you in the next episode. This information is for educational purposes only and is not a recommendation to buy, hold, or sell any investment or financial product. This podcast has been produced by a third party and may include paid promotional advertisements, other company references, and individuals unaffiliated with A16Z. Such advertisements, companies, and individuals are not endorsed by AH Capital Management

• 1:47:43Speaker 05

  LLC, A16Z, or any of its affiliates. Information is from sources deemed reliable on the date of publication, but A16Z does not guarantee its accuracy.