Ferrari Goes Electric, The Return of Nonsense Metrics, 𝕏 Timeline Reactions | John Quinn, Martin Casado, Preston Zhou, Scott Stevenson, George Kurdin, Zachary Townsend

27:02It was within our firm. Okay. It wasn't some outside PR. That's great. But, you know, we came up with the copy. Yeah, it's amazing. Talk to me about a little bit of the journey. How are you defining the firm now? Has anything changed over your career? I mean, it's not exactly a nontraditional path. You know, you went to law school, built the firm. But what were the key decisions along the way? Well, I started out at a big firm in New York that's very well known, Crevasse, Wayne & Moore. But, you know, I'm from the West. I grew up in Utah. Went to college here in L.A. Really liked L.A. I like New York.

27:40But I decided I wanted to come back to L.A. Came back here, bounced around a little bit. And we started this firm in January of 1986. And we started with four lawyers, downtown L.A., not too far from here. and along the way we decided to focus just on disputes work, litigation work. So most large firms, we're now 1 ,300 lawyers almost. There we go. In 34 locations around the world. Most firms of that size are so-called full service firms. So they'll have some deal lawyers. Corporate. Yeah, they'll have corporate lawyers, they'll have some tax lawyers, trusts and estates, and somewhere they'll have some litigators.

28:19And we thought, we got, But we came to realize that this model of only doing litigation, arbitration, disputes work, government facings, kind of dealing with legal risks in all its different forms was a very powerful model. And so we decided just to stick with that. We're not going to try to be... And historically, people would diversify because they wanted the stability of a variety of different businesses or they wanted to be able to add kind of contract value, effectively like add hours to existing relationships. Because I'm sure you would have had pressure over time from people that say, I love working with you.

28:54Can you do more for me? I think the thought is you want to be able to service all a client's needs. So a client may want to do some M &A work. They may want to raise some capital. And then they may need tax advice. From time to time, they'll have a litigation. So the thought is to cover all those. And by the way, that's how our industry is organized. We're one of one. I mean, there is no firm. I don't know what the next largest litigation only firm in the world is, but it's probably not more than 50 lawyers, wherever it is. So you started in 86. How long did it take you to become the most feared lawyer in the world?

29:33Well, so what you're alluding to is there's a consulting firm, BTI Consulting. We had nothing to do with this. They called us up one day and told us they were doing this survey. that they survey the general counsel at like 350 large corporations around the world. And they asked them, what law firm do you least want to see on the other side in the dispute? And they were publishing a list. And we've always been on that list of the top four. They call it the fearsome foursome. And a few years ago, we got them to tell us, okay, of the four, which one actually got the most votes? And they told us. And for five years running now, we have been, we say, we characterize it as the most feared law firm in the world.

30:19It's good to be feared. Well, in our business, we don't find clients very often that are looking for shrinking violets to represent them, to stand between them and trouble. They're looking for lawyers that will cause the other side to think twice. You know, there's a settlement here. This can go away for this amount. Or you can see what's behind door number two. Yeah. And so, you know, we think we can get resolutions that other firms can't because of our reputation and our record. What's the secret to 90 % win rate? Is it case selection? Walk me through all the decisions that make sure that you're...

30:56No, we don't just select winners. Yeah. We don't select just winners. I mean, we're in for the hard cases, the tough cases to win. And what's behind that is, frankly, there aren't shortcuts. Doing what we do at the highest level is labor intensive. And it requires people who are not just smart, but truly dedicated. What's the hardest case? What's the long shot that you actually wound up winning? Well, early in my career, I mean, I'd be embarrassed to kind of tell you the facts. It was an employment case. And we were defending a discrimination case where there were two plaintiffs. They were both African American.

31:39They were both over 55 years old, I think. And they were both let go. And there was evidence. I mean, the plaintiff had some evidence that the motivation behind the they had both been terminated from their jobs, that the motivation, you could argue that age entered into it. Yeah. Maybe race entered into it. And it was a very, very tough case. And I'm kind of embarrassed to tell you on most of these 30 years later. I mean, we won the case. Yeah, yeah. What about in tech? You've been a part of some of the biggest tech cases. Is that something that took you a while to get up to speed on? Is there anything different about working with a big tech company from any of the other work that you do?

32:32So we noticed 25 plus years ago what was going on up in the Bay Area and in Silicon Valley. And we made a decision that we wanted to be part of that and start doing some intellectual property work, patent litigation work, tech work. So we're there for the conflicts. We're not raising money. We're not doing the M &A work. It's primarily patent litigation. We now have the largest patent litigation practice in the world. But we sent a lawyer up there to start an office. And frankly, it was a little bit difficult to get traction in the beginning because there was then, I think it's diluted somewhat now, kind of a Silicon Valley culture.

33:12And people wanted to know how much time did you, how much Valley experience did you have? And we were kind of outsiders. We did recruit some people locally, but it took us a while, frankly, to get some traction. What's different about Silicon Valley? there's so much about Silicon Valley that is weird around IP it's like in software there's a world where if somebody developed a unique user experience you would have been able to really enforce that in the scale of time if a big platform creates some incredible feature it's going to get copied by every big company that it's relevant to how did that kind of evolve because I think early on you maybe would have expected that you could have if somebody like a Snapchat creates a stories feature, that could have been, there's another world where that could have been IP, but it didn't play out that way.

34:03There is kind of a culture. There's a feeling in many circles, especially among the biggest companies, the hyperscalers and the like, that you should compete based on your business. Don't compete on IP. I mean, there is a reluctance. Which is completely different than like almost every other industry in the world. Yeah, biotech would never say that. Well, I mean, you don't see the big automobile companies suing each other either. You don't see the big airplane companies. There'll be trade disputes between Airbus and Boeing, but not so much technology disputes. Interesting. And there is kind of a feeling, and Elon Musk has been outspoken about this, that anybody can use our inventions.

34:47He has said things like that in the past. So you don't see the big players going after each other so much. They will. There are examples. But there's kind of an ethos that you compete on the business, you compete on your offering, and you don't compete with IP lawsuits. What lessons did you take away from working with Elon? Is he different than other clients? Is there something that you were coaching or advising at various points in time? Well, we've represented Elon in a number of cases, including cases that have gone to trial here in California. There was that defamation case, the British cave driver, diver, where there was a tweet about whether this was a pedo guy.

35:30And the case about wanting to take Tesla private for$420 a share, funding secured. We tried both those cases. I was not on either one of those trial teams. So we have a team of people that work very, very closely with Elon. So what I know about him and our experience working with him is secondhand. I do know he values his time. He closely guards how he spends his time. I think it's fair to say that based on what I've heard that spending time with lawyers, preparing for depositions, and trial at the like is not the top of his list of things that he likes to do. Not a priority. I can imagine. So he's a busy guy and obviously he's got great instincts himself.

36:16Yeah. What about. I can say just one kind of one story. So in that defamation case, he had tweeted that I show you this British guy as a pedo guy. And he came in and then lived in Los Angeles, sued him for defamation here. We didn't have a lot of time to prepare him for trial. And when the trial started in federal courthouse downtown, the plaintiff called him as their first witness. You can do that. You can call the adverse party, put him on the witness stand, and immediately start cross-examining him. So the jury, the first witness they see is your adversary being cross-examined. And the question to Elon was, when they put him up there, was, this is a defamation case.

37:00You're Elon Musk. so the world really cares about what you say not a bad question what's he gonna say yeah it's hard to deny there's kind of a beat and he says well well i'm i'm not so sure i i i don't know i i've been talking about our need to get away from fossil fuels for for so long now and it just doesn't seem to be happening i'm not sure people aren't listening to me people listen to me at all. It was like genius. That's remarkable. Yeah, that's a good argument. That's very funny. We, I mean, we're tracking like a crazy AI boom, crazy tech boom right now, mostly from the financial markets.

37:42We have founders on, I'm sure we have seven people on today that are raising a lot of money. Is there a boom in tech litigation? Well, maybe even before that, talking about kind of the dot-com era, you saw this craziness. We opened the show talking about how people were just valuing eyeballs, right? And there was, aside from that, there was like, you know, these round trip, you know, effective, you know, commercial agreements that ended up getting litigated in the years following after the crash. What kind of stood out from that era? What was just, you know, I guess, what were the takeaways from that era?

38:20Did you ever think we would repeat that kind of cycle? Because I'm sure that you were in the, your job even after the crash just got busier and busier. Yes. Whenever there's a crash, as in the financial crisis, for example, 2008, which we weren't very involved in, that plays out in the courts, of course. There'll be people who've lost a lot of money. There'll be claims of fraud and misrepresentation. You know, in the dot-com boom, There were down rounds. There were class actions of all kinds, securities class actions. So every deal gets scrutinized in retrospect. In AI, where we are, look, I'm not an economist, and I'm not a prognosticator about where we are in terms of whether this is a bubble or not.

39:09I can tell you in our world, the disputes world, litigation, we're seeing a lot of AI-related disputes. There's no question. And is that around primarily today? I can imagine the trajectory is like right now it's a lot around IP, right? We saw a case against Anthropic around their usage of, I forget. Reddit. Well, there's Reddit getting litigated, but then the books where they had potentially. They just did a huge settlement. It was a big settlement. And so right now it feels like we could be in the era of like litigating around IP and what even is IP in the world of generative AI. But then it feels like the next wave could be around just rehashing and litigating.

39:52It's going to be a lot of lawyers looking over this announcement that went out. It's like, yes, in the fine print, it said that it wasn't an actual deal yet. It was just kind of a general sentiment from the two companies. But yeah, where do you think, do you think we're at all close to having some precedent around AI and IP? Or is this going to take years? I mean, AI is throwing up so many novel legal issues. I mean, IP for sure. Everybody's heard about the cases that have been brought relating to music, literary works, visual works. using copyrighted materials to train large language models. And whether that's copyright infringement or what's called fair use.

40:44So we have dozens of those cases. We finally have a couple of decisions now out of the federal court in San Francisco where the judges in those two cases, one case involving Metta, another case involving Anthropic, decided that it was fair use, that it was not copyright infringement to use copyrighted materials to train large language models, although both decisions were kind of qualified. And there were some unique facts in both cases. But yeah, we're starting to have those decisions on various issues. But there's so many things. All this money is going into large language models and AI. And you think a lot of founders are just running the calculus of I know that I'm going to have to eventually pay out something for what I'm doing, but I'm just going to run with it because I'll be able to eat the cost later.

41:39And if I wait, then my competitors will. I don't know. Losing one of these cases, the copyright cases, there's so much involved. The price tag could be very, very big. So I don't know that people are sitting back saying it's just the cost of doing business. Yeah. But at the same time if your competitor is like willing to take the risk and they can ramp their market up market cap up into the hundreds of billions they're not they know that and and the alternative is you know being safer around ip and then you just lose the race then you're a zero anyways so i think a lot of people are it feels like a lot of people are willing to just roll the dice and say we'll figure it out later well apple and it's fairly in its early years uh had a reputation for not being very careful with its IP.

42:26You know, they launched the iPhone knowing they did not own the name iPhone. Oh, yeah. It was owned by, I think, Oracle. Oracle owned it. Yeah, Cisco. And so as soon as they launched it, there was a lawsuit filed, Cisco versus Apple, and it was immediately settled. I've never heard what it settled for. But, you know, that was Steve Jobs, you know, driving ahead and kind of doing what you're saying. I think they've gotten a lot more careful. But there's just a host of issues like you have all this money going into AI and AI inventions. How are you going to protect it? You cannot patent an algorithm.

43:02You cannot patent a mathematical application. You know, it's the, these come under the rubric of general ideas. Yeah. And again, laws of nature. Yeah. Yeah. And then you, and then in the talent, the talent wars are a whole other thing. I'm curious if, if what, what your kind of view on, you know, over the summer, we saw all the labs kind of poaching, pulling talent from different places and people realizing that in this person's brain is like potentially a hundred million. You think that's, they're not being hired for what is in their brain? They know. Yeah. Of course they are. And that's an example of what I was referring to earlier, where the major players are very reluctant to sue each other.

43:41I'm kind of surprised there hasn't been more litigation over the, this poaching, but I think it's this phenomenon that yeah our view is that they were very much like aqua hires just unauthorized aqua hires where it's like this team knows how to do something really well that they've learned through spending tens of millions or hundreds of millions or sometimes even billions of dollars and we want what they know how to do we want the uh we want the process that's in their in their head basically yeah you know there's no doubt i mean some of the what you're referring to is trade secrets. Trade secret law, which is another area of intellectual property.

44:18And that's an awful lot in the AI world, what we're talking about, mathematical applications, algorithms, and the like. They can't be patented. They can be trade secrets, and they can be protected as trade secrets. And if somebody has access to those and goes to work for a competitor, it's not lawful for them to be using that information when they go to work for the competitor. But then proving that actual use, you know, is another thing altogether. How often would an issue pop up and you have multiple, let's say it's Mag 7 on Mag 7 and you're getting calls and texts from both of them racing to work with you guys?

44:55And then have you ever had to make a call in the moment? You know, in that world, we never encountered that because we always and forever have been adverse to Apple, Meta, Amazon, and Microsoft. Okay. We've always represented Google, NVIDIA, Qualcomm, Salesforce, and some others. In our world, it seems like you come to a fork in the road and you decide we're going to go this way. And you never come back because people know you're the firm that will be adverse to Apple. It's like being the firm in Seattle that will be adverse to Boeing. That's kind of a job in itself. Yeah, that makes sense. What about...

45:34I mean, it does happen in other areas of practice, like in private equity. I mean, we had a situation just within the last couple of days where two global private equity firms who were opposite each other with respect to a particular transaction, both of us called us and we accepted the engagement from the one that called us first. Okay. So that does happen. Is being litigation only an advantage in being able to create this patchwork of clients that you don't have conflicts because you're also already doing their contracts or anything? So you can be a little bit more. You know, I think as a litigation only firm, we have fewer conflicts.

46:12So for us, the world is divided into companies and individuals we represent and those we don't. And sometimes a party, a company or an individual moves from one category to the other. But we have very few sort of lasting conflicts vis-a-vis the business world in principle. Where deal lawyers, there might be some industries that they know, well, we just can't be on the other side of this issue or represent a competitor or the like because it's a little bit different in the deal world. Where is your firm getting leverage from AI today? We're using it a lot. We're using it to, you know, not only to manage documents, the things that are produced in discovery, because we've been able to do that for a long time.

47:04But to be, if you can imagine, we tried a case in Delaware a few months ago, and we had loaded into the database all the deposition testimony, all the documents, and the trial testimony as the trial is going down. And as somebody is on the stand, I mean, you can imagine, you can query this database. What is the best evidence that so-and-so just lied about that? Yeah. Yeah. In real time. In real time. Yeah. You know, you press a button and it'll give you a list of things in rank order. Yeah, yeah. Now, if you had this - Before, back in the day, in the 90s, early 2000s, 2010s, you had to be - Go back to the filing cabinet.

47:40Yeah. I mean, no, you're relying on your memory and what you know and what you know about the case. And still, that's the most important thing, I think. And mostly, it's kind of a backstop. You know, it's a check. things you would have thought of anyway, but it'll also come up with things that you wouldn't have thought of, or maybe you would have thought of if you just had more time. What are your thoughts on resolving co-founder conflicts? There's been a number of really high-profile cases around a group of people came together, started a company, the company grows and grows and grows, some people leave, and then there's...

48:14Winklebloss twins. That was one example? We settled that case. Yeah. And we didn't file the case. We were brought in to settle it. Snap. Yeah. There was a third guy in the garage. Yes, yes, yes. Who got kicked out. Yes. And he brought a client. Are there any common pitfalls that you see on the incorporation documents or the safe note or specific? People start working on these projects, sorry, all the time. Yeah. And they're not thinking about this might become super valuable someday. Yeah, yeah. You know, maybe they're more sensitive now in the tech world. It's easier than ever to create. You can create a C Corp in a few clicks.

48:52You can see. And they don't think to lay out, okay, who owns what? What's my share? What's your share? We're friends, right? So they don't think to document it. And it's not till later that, and suddenly it's super valuable. And they're getting all these views and followers. And they realize, wow, this is super. Now, remember, this is one third, one third, one third, right? No, no, no. That was not my understanding. Yeah, yeah, yeah, yeah. How much in a case like this, is it worth pursuing a settlement, taking it to trial? Like, what are the key decision points where you realize, okay, this is going to trial?

49:26I mean, usually, so there's a stage called summary judgment where a party can go to the court and basically say, look, on the undisputed evidence that's been developed in the course of discovery, I'm entitled to judgment as a matter of law. It's basically you're arguing to the court. No reasonable juror could fine for the other side. They could only fine for me. and there has to be no disputed evidence. The judge has to find, okay, on the law, and yes, the evidence is undisputed, you win. If it gets, and a lot of cases are dismissed at that level. If it gets past that, then there's the prospect this case is going to go to trial.

50:04There's nothing really going to stop it because that's kind of the last court off-ramp. A lot of cases are resolved in mediation. In any significant case these days, the courts will expect you to go through a mediation. That is to say, engage a neutral professional mediator, spend some time with the mediator, and see if you can resolve the case. And that happens, sometimes it happens right at the beginning of the case, or even where there's a dispute and no case has been filed, you get a mediator together. Sometimes it happens on the eve of trial, but a lot of times it happens in the course of a mediation.

50:42So, you know, if If it's gone past summary judgment, if you've gone through a mediation and that hasn't been successful, then that looks like a case that may go to trial. And is that a function of total settlement? How much is at stake relative to the legal bills? Am I more likely to settle a case if it's$1 million at stake, but$500 ,000 in legal bills that I'm looking at versus there's a billion dollars at stake? I would certainly think so. The legal bills are more than what you would get out of winning the case. Sure. Sometimes there's other things at stake, whether ego or whatever, reputation.

51:20But it's usually a calculus. I think most clients, it's pretty much an economic calculus. What are my chances of prevailing at trial? What do I get? What does prevailing mean? What will I get if I prevail? And obviously, what's the cost to get me there, including primarily attorney's fees? Do you think that AI will speed up cases or is there some element to litigation that just because maybe one side realizes the longer this goes on, the better for me, I'm just going to drag it out? Like, will it actually, even though a lawyer could now do two weeks of work in an hour, maybe, if they're properly using some of these AI tools, it might still just take the full, you know, however many two weeks in order to file.

52:11Because the other side is going to be using AI, right? Yeah, look, I mean, cases settle when parties realize the strengths and weaknesses of their case and the strengths and weaknesses of the other party's case. And they assess that. You know, cards in the discovery process, everybody's cards are basically put on the table. And you have a basis then to make a judgment about am I going to win? What does winning mean? What am I likely to get in damages? I think AI will accelerate that process. There'll be more disclosure and more ability to handicap the potential outcome sooner. So I do think there's a potential for resolving cases sooner.

52:52However, I think AI is going to result in much more litigation. More cases. More cases. Why? There are already early stage companies out there who are putting cases together. They have put into databases all records about permits, licenses, advertising, what parties claim is in their product, what they say their product can do. All this information and information about what is actually in the real world, what's the experience there. They'll put that together and they will serve up to lawyers. You can subscribe to these services and they'll come to you to say, I've got these great class action lawsuits for you.

53:30And so they're finding they're finding cases. So I think there will be more cases, but I think there's a potential they'll be resolved sooner. Do you think the legal profession is a mentorship business? Do you like when you bring on, do you see, you know, young lawyers needing to find a mentor to establish a full career? Is that how you think about it? I really think the best way to learn is working with more experienced people. That was certainly true in my case, whether you call them mentors or just senior people in the firm who you work with, go to court with and see how they examine witnesses and the like.

54:08So I do think that's really important in terms of learning what it means to practice law. What advice are you giving to young lawyers today that want to be where you are in 30 years or so? Well, as I said earlier, I tell people, you know, this is practicing law, litigating high value cases at the highest level is there's no shortcuts. So you have to be prepared to sacrifice and work super hard. It's a it's an hour's commitment. There really are no shortcuts. Is it thrilling, though? Do you love it? I do love it. I do love it. Is the case more thrilling or is it who you're working on behalf of?

54:46Is it the partnership between you and the client? No, the partnership with the client can be very, very satisfying. This is a team sport, right? Exactly. It's definitely a team sport. The team within the firm, the team with the client. What I love most about it is we're constantly learning. We have to be able to cross-examine experts, whether it's on biotech, steel manufacture, AI. I mean, you name it. We have to get to the point where there's an expert on the stand, somebody maybe has a PhD in something, and we're going to cross-examine them. We have to learn. And that's what I love about my job.

55:19We're always learning something new. Yeah, an expert can make you sound really silly. What do you mean by that? Like, I don't understand what you're even asking. And then they could just be, you know. You sound like an evasive expert. Yeah, exactly. That's not going to be great for your credibility. There you go. What about how the media characterizes law? Is there a book or movie that stands out to you as kind of telling the story the correct way that resonates with you, that's something that you don't feel like it's a mischaracterization of how you work. You know, there are a number of good books about trial lawyers and perhaps my favorite was written by a trial lawyer who recently passed away, Jerry Spence.

56:04I don't know if that's a name you know. Famous trial lawyer from Wyoming. He did the Karen Silkwood case, that uranium case. She claimed that she was injured by exposure to radiation. He represented Imelda Marcos in a trial in New York City. I mean, but he was always kind of this Wyoming buckskin-clad country lawyer. A character. And he wrote a book called Gunning for Justice that was about his own life. And it's a very, very moving book. That's one that I really like. I will have to pick it up. Well, thank you so much for coming by and doing this interview. Thank you, John. This was fantastic.

56:48I enjoyed it very much. Thank you so much. Thank you very much. I'm back on it again. Yeah, you're welcome anytime. What a legend. That was a lot of fun. We have our next guest in the Restream Waiting Room. let's first tell you about graphite.dev code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. You can get started for free. And our next guest is Martin Cassato from Andreessen Horowitz. We have a bunch of breaking news. We're excited to talk to you. How are you doing? Welcome, Martin. Good to see you. Can we adjust the levels, make sure that we're hearing him okay?

57:26How was your day? Just tell us how you're doing. I'm doing very well. Thank you. It's been actually a pretty chill morning because I had a couple meetings canceled, which is kind of the best gift you can give me. There we go. Paradise. Canceled meetings, hop on TV to PN. What more could you want in the Silicon Valley life? How's your Q4 been? Pretty mellow? Most of my companies are still in Q3 because we shift by a month, but pretty good. Thank you. There you go. Why do you shift by a month? Is that standard now? Is this something that ties to it? It is very standard. I mean, listen, so I did enterprise sales for my company for 10 years, and I don't remember a Christmas or a New Year's where I could be home because we were like deep enterprise sales.

58:09So 50 % of the year came in in the last month. And so everybody realized that if you're pushing to the end of the quarter, it's probably better to do it when people are still working. And so most will use the end of January. And so that's the majority of the companies I work with by a month. How often are you investing in a company and they haven't made the shift yet? Are you going that early? Honestly, if they don't, I recommend them to make the shift. Yeah, I imagine that's a first. You like the holidays at the end of the year? You like spending time with your family? I would say 2008 to 2012 of my life, maybe 2016, was me on the phone in a beach somewhere trying to close a deal.

58:48And it paid off, and here you are, leading the infrastructure practice. I want to get to robotics in China, but first, I'd love just an update on how you're thinking about infrastructure at Andreessen. You have$1.25 billion under management for that fund. And infrastructure, I mean, we're seeing deals for energy. Sam Altman's doing stuff with SK Hynix and HBM. There's the NeoClouds. There's the hyperscalers. There's so much in infrastructure. And then there's also just like dev tools, which I think could maybe be in there. So how do you define it? What's interesting? Where are the key themes these days?

59:22Yeah, so we do primarily software infrastructure and things that directly relate to software infrastructure. So everything from, like, let's say, chips to AI models to your point dev tools to the traditional compute network and storage. A very interesting thing about this AI wave is, like, it's very hard to distinguish between, like, a deep infrastructure model company and, like, say, an application company. And so we're having kind of a renaissance in the field where everything has to be redone. Like, you have to redo networking. You have to redo compute. And you have to, like, build new dev tools.

59:50And so, you know, we're just trying to keep up with all of this change that is coming from AI. How are you thinking about competition from the hyperscalers when you're looking at a startup? We've talked to a bunch of founders who they're going up against AWS and Andy Jassy's not asleep at the wheel. But at the same time, there's this meme of like, you know, a VC would always tell you, what if Google does this? And then we have, you know, Cloudflare and like, you know, GitLab. And there's like a million companies that we could name off that are, you know, decacorns, public companies that successfully went up against the, you know, the public clouds and won.

1:00:25Or at least saw, you know, massive businesses built. So is there, is it all about the founder? Is it about the particular market or the technology or lock-in or being program record or being, you know, some sort of system of record? Like, how are you thinking about it? So listen, I've been an infrastructure investor for 10 years. And listen, in that entire time, the public cloud has been incredibly dominant, right? And so I would say after reInvent, which is the AWS show, I have to move into therapy mode. I'm actually on the other side of this. I'm never the one that actually worries about the public clouds.

1:00:58I'm always telling the founders. I have this line. My line is, can you name one company that AWS has put out of business? One, just one. That's a good point. They literally have copied everything. They've hosted everything. And so the reality is that if the market is large enough and you have an entire company focused on it, you're going to win. And that's just the reality. And AI in particular is a new behavior, which, you know, incumbent companies are terrible at new behavior. So this is definitely an advantage to startups. It's like a new buyer. It's a new use case. It's a new way of thinking about things.

1:01:33So they don't have the incumbent institutional momentum. And so I think that we're free of a lot of the trappings of the cloud, and I think the future is very bright for startups. So I don't spend a lot of time thinking about the incumbents. That said, I have to say one thing is, listen, Anthropic and OpenAI are startups, really, in the classic sense, and they are incredibly dominant. And so I would say I'll put them in a different class. There is a class of these new types of incumbents. Think like Stripe, Figma, where the founders are still there. They're still hungry. they're executing very well.

1:02:08I think those are much more dangerous than, you know, Google or Microsoft. Yeah. I was talking to a founder who is going up against an AWS point solution that was rolled into stuff. And I was like, you are going up against Andy Jassy, but you're really going up against the particular product manager who's on that particular product. And that product manager might not be taking calls on a beach in on December 25th, you know, like they might be taking a vacation. And so if you're going to work them, it's possible you could beat them. Totally. And, and like, this is a company that, you know, is famous for bragging about two pizza box teams, right?

1:02:43Okay. Yeah. So fine. You've got, you know, the 12 people on this product is dying on the vine. Like if you can't beat that, you kind of deserve to lose. Yeah. Yeah. Yeah. That's a great point. Well, I'd love to go into robotics in China. You recently published a Wall Street Journal op-ed, which I love. I read the journal every day. Before we dive in there, one question around when you're evaluating how you think about revenue ramps specifically. You know, there's always this question if a company is ramping revenue just at an ungodly rate. There's a question of like how durable that revenue is.

1:03:23Is it going to be possible for another company to come into the category and sort of pull off the same type of growth? What are you evaluating for these software infrastructure companies on whether they have actually done the hard work to figure out, develop something that's unique and proprietary? Or they're just capitalized. Or they're just like a, we did talk about it. It's like a barnacle on the side of a whale like opening eye. A reseller. Yeah. We're into the barnacle economy. Yeah. We love the barnacle economy where you just naturally grow at the growth rate of anthropic. And you're like, it's a miracle.

1:03:59Yeah, yeah. Yeah. I mean, listen, so there's always this concern of easy come, easy go, which is kind of like what you're asking about. Like, you know, you got this revenue so quickly. How hard could it be? But, you know, listen, we're three years into this and we've got a lot of data points that this stuff is actually pretty durable. Right. I mean, mid-journey bootstrapped to actual scale and it still is very much a leader in image. You know, OpenAI roared into dominance in text and it's still the leader. Anthropik roared into this. I mean, Cursor roared into this. And I think that these revenue ramps are much more of an indication of the size and the growth of the market, which, I mean, these things have just been so large and growing so fast.

1:04:39And of course, like the leader that becomes the brand monopoly will follow brand effects and then people will know the name and then they'll go ahead and adopt it. So I think that, you know, like there is no indication just because the revenue ramp was quick that it's not durable. That said, I don't know of any endemic defensibility in AI. I think companies have to figure out traditional modes. And there's a lot of traditional modes we know about, right? Integration modes, two-sided marketplace, brand modes. Like whatever the mode is, they've got to get to that. So when we evaluate the company, just because it grew quickly, we're not like, oh, easy come, easy go, right?

1:05:15I mean, that's kind of like intellectually lazy. On the other hand, we do look for like what actually is the mode, right? If you're basically just reselling TPM tokens per minute from Anthropic, that's probably not as durable as if you've got a workflow and a bunch of users that use it. And they save configuration in this and you've got good retention numbers, et cetera. So we do kind of full-fledged investment like we always do. But I don't think it's correct to just say just because the growth is high, it must not be durable. How quickly can a brand moat emerge in infrastructure? Yeah, I mean, it's a good question.

1:05:53The last time I remember brand modes like this was the internet. I think when markets grow incredibly quickly, education is always kind of out of date, and so you default to brand. Remember, Netscape was largely a consumer phenomenon, and very quickly everybody was talking about Netscape very early on. And we're seeing a very similar thing. So I think it's been probably 25 years since we've seen brand effects dominate adoption, especially for highly technical projects. I think the last time was probably the internet and before that, like maybe the PC, like Windows 95, again, was very much kind of a brand name.

1:06:24And so I think it's more, again, an indication that the market's very large, it's going very fast. There's so much confusion in the market so that a few leaders end up becoming brand monopolies. And then the adoption is literally because people can't keep up. So they just kind of adopt the one that they know about. Yeah. Speaking of brand, there's actually a comment in the chat. A16Z blows my mind. How can they be a venture fund and a brand at the same time? And I feel like Andreessen really is focused on brand in many ways, both the Andreessen brand, but also partnering with brands on the consumer side or, you know, internet consumer side.

1:06:58But do you have thoughts or advice for infrastructure founders on brand? Is there an equivalent of like, make sure you have your dot com or make sure that you have a specific brand package? Like, do you spend any time thinking on that? Or is that at all relevant? Or is it just something that, you know, kind of check the box, make sure it's converting customers and think about it in more quantitative terms? Yeah, no, so quite frankly, this is a pretty new phenomenon. Like, I am part of the nerdiest part of Injury School. Don't really think about brands. It's like, no, that's not the conversation we have.

1:07:27Normally, it's either talking about, like, the mechanics of, you know, inbound bottoms-up growth or the mechanics of enterprise sales or something like that. But that said, we've all had to, all of us, the founders and the investors had to become students of this because like you'll have model companies that will be consumer companies. And those are very much brand led. You'll have, I mean, you know, I mentioned Cursor before, like, you know, Cursor is a dev tools company, but it's got so many users like brand effects are at play. Brand is really important. And X now, because this stuff is proliferating so quickly, has become a primary distribution channel.

1:08:02X has its own way of propagating information that's very, very different than going to traditional press or doing a press release or writing content marketing or whatever. We're in a very different world. And so I would say historically, I personally, while the firm has, I personally have not been very focused on brand, like the opposite. But now I've had to learn a lot more about it. It used to just be make the docs dark mode. Do you think some of the – Yeah, yeah, yeah. But now we are emerging. Sell the thing and be technically correct. Exactly. That was kind of my, yeah. Exactly. Do you think public SaaS is generally oversold?

1:08:38Are people too bearish? Or is the conversation we keep coming back to, Brett Taylor was on the show talking about how it's easier to change your tech stack than your business model. And he was referencing the dynamic between seat-based. But that's kind of not as relevant in public company infrastructure companies that might be disrupted by – Systems of record. Because the public – I imagine that most of the infrastructure competitors, the incumbents, are already on consumption-based models. And so you're not seeing a dramatic shift in the business model of the companies you back. It's more of a technology shift, or am I mischaracterizing that?

1:09:20Okay. So I definitely think we're seeing a pricing shift from seed-based to usage-based. And listen, we saw that. When I was doing my company, we saw the shift from perpetual to recurring, right? To SaaS recurring, right? And so we're seeing that type of shift, and that's being accelerated by AI. It was started by the cloud, but it's being accelerated by AI because the tokens are so expensive. I don't think that alone is depositioning anybody, right? I think that the SaaS companies are very good at adopting that and using that. I really think a lot of the story of AI is actually kind of new capabilities and new use cases.

1:09:53And that's kind of like the untold story. I mean, these things now speak English, right? They develop an emotional connection. They can create something from nothing. Like, that's cool. That's net new behavior. That's net new budget. And you can see this in this dramatic growth. Now, of course, budget has to shift to cover that, right? And so I think we are seeing a slowing down of traditional SaaS spend, as you would expect when you have something super cool and new and whatever. But I don't think that this spells the demise of the traditional sector. I mean, it's actually not even good at things like systems of record yet.

1:10:27Now, over time, that will change, and maybe it'll change the consumption layer. And once you have the consumption layer, you can change the back end. But I really think the story we should all be talking about is, like, what are these net new use cases? Like, what is it enabling? Where does that budget come from? As opposed to, like, who is this necessarily going to kill? Because the zero-sum thinking, I think, is the biggest mistake we all make in this AI wave. Totally. All right, well, let's switch gears to robotics. Yes. There is a, what's the major robotics company out of China? Unitree. There's a Unitree for sale on Walmart.com for like$20 ,000.

1:11:02I'm hovering over the buy button. How un-American am I if I pull the trigger? Take me through the landscape. I think we're maybe going to try and hack it and do something weird. But yeah, take me through the current understanding. In general, it feels like, maybe I'm wildly wrong, but it feels like tension between China and America is calming down almost. I don't know. Maybe that's just the news cycle and people have shifted to other geopolitical topics. But, I mean, Polymarket has the chance that China invades Taiwan by end of 2026 down at 14 % from 25%. And we're doing more trade deals. There's crazy news today.

1:11:43But walk me through how you're understanding the geopolitical landscape. So listen, I'm not an expert on geopolitics. I will say today China put expert controls on rare earth metals that are useful in things like lithography. So that's clearly, I would say, not a detente. I mean, that clearly is kind of more of an aggressive posture. Probably this is because Trump is meeting with Xi Jinping. Probably this is just kind of for negotiation. I don't know, right? But I would, you know, again, I am not an expert on these things from my perspective. which is through the lens of technology. Like, there is not a detente.

1:12:16I think that things tend to be escalating. Now, that said, you know, the technical collaboration between China and the U.S. is the strongest it's ever been. I mean, you know, half the founders I work with are Chinese. Half the teams I work with, half the papers I read, half the models I use, half the models the companies I work with use. I mean, like, I think from a technical academic standpoint, it's the strongest I've ever seen in the industry. And so in that way, there's very much a positive. But politically, there's clearly some jockeying. And I think we're trying to all understand what that means.

1:12:50And I'll just tell you my very quick view is like, I don't think anybody wants a generational struggle with China, right? I mean, like, the, you know, the hope is, is that both of us collaborate together and take over the solar system, right? These are two just phenomenally great countries with a tremendous amount of talent. But there is some level of Pollyanna in that belief. And so then the question is, to what extent do we want to depend on China? It's not about being aggressive. It's not about being proactive. It's like, do we want to depend on them for critical infrastructure? And I would say my position is the answer is no, we don't want to depend on them.

1:13:26In which case, it's important to understand what they're capable of, what we're capable of, and what we can do to develop local capabilities. Do you read anything into the fact that Xiaomi was able to be a phone company that launched a car and Apple was maybe rumored to be working on a car and never got it out? Is that just a market structure thing? I'm interested in understanding, like, where is China behind? Where are they ahead? It feels like Tesla was way ahead in electric cars and China's catching up there. but now Tesla's playing catch up in humanoid robots. And I'm interested in the landscape of like what each country does best.

1:14:09Yeah. Okay. So, I mean, listen, this is going to be very polarizing topics. I just want to say up front that I'll send some opinions. These are not rooted in fact, there is no agenda here. Just, you know, as far as your first question on, on the structure of companies like Xiaomi and the car and Apple, you know, Asian companies tend to do more. Think about the Korean shayballs. For a long time, there were five companies that did everything. Everything from the cone to the car. I don't think that that is a China versus U.S. I think that's something quite a bit different. That said, we've seen China be very, very effective at a number of areas in recent memory.

1:14:495G, China basically won, which is critical infrastructure communications. For solar, they've been dramatically good. Then I wrote this post on robotics. They're very, very good. And so it's clearly a country with a ton of capability. On the other hand, like, listen, like, the United States continues to, like, do two things that I think are pretty distinct from China. One, they're very good at complex software systems, right? Like, you know, not building a model, but building kind of like the opening eyes and the anthropics and, you know, and everything around. They're very, very good at complex software systems.

1:15:19And two, this is going to sound funny, but the U.S. is actually really good at building stuff people like. That's good. No, but I think there's a specific reason for that, which is we're more like the rest of the world than China is. Like China, like, caters to a local market, and the local market is just different behaviors. It's different buying behaviors, different political behaviors, different economic behaviors. And so, like, the stuff that they produce is just weirder. And so, in some way, I would think, like, you know, there's two areas where they're just not as strong. It's very complex software systems, which we historically have just not seen come out, especially for B2B, but also like software that the rest of us can like use and adopt outside of like these viral consumery things.

1:16:00La boo boo. I was about to say, yeah. La boo boo, right? Like this consumer thing, but I'm not a consumer guy, man. There's these cars that have like a LED front dashboards and karaoke machines inside and they float and they're all very cool, but I don't know if the American consumer actually wants that necessarily. I think you could reduce it to like, you know, listen, people are people and whatever, but like enterprise buyers are very different in China than they are in the United States. And like, can you name one large Chinese enterprise software company? No. Software. Yeah. I mean, it's impossible.

1:16:34I mean, we, we, we, like, we have the hardware. Does ByteDance get into enterprise in China? Does Alibaba Quen, like, count? Or like, does DeepSeek count? I know DeepSeek's getting inferenced a lot. I mean, I'm like, well, listen, Databricks, Snowflake, Salesforce, Microsoft. Yeah, yeah, yeah. We have more than possible. You know, the U.S. enterprise companies, and then in China, you're like, well, you're kind of trying to look, well, it was a consumer company, but maybe they do enterprise. Like, there's kind of these weird telco companies that people will mention. Sure, sure. It just, you know, listen, I built a team.

1:17:00You know, I ran a team in Beijing. I ran a team in Shanghai when I was working at VMware after they acquired my company, and we sold software in China. It's just different. It's totally different. Like, the consumption pattern is different. How you sell it is different. What they expect from the software is different. And the market is large enough and localized enough that any local company is going to focus on that. They focus on that. They're just like through Darwinistic forces of acceptance are not going to build something the rest of the world really understands. And so I do think this is an area that the United States has a tremendous advantage.

1:17:31Do you think, based on your experience selling enterprise software in China, do you think in our lifetimes we will go back to American companies trying to sell enterprise software into China? Or is that ship sale? because I briefly lived in China. I studied abroad there. I worked out of the China Accelerator office in Shanghai. And there was a lot of international companies that were trying to sell into China. This was the 2016 era. And there was a sense that the Chinese market sort of wanted to learn about what was being sold, but they didn't actually want to adopt it. And it was more like, how do we learn about this so that we can just build this ourselves and utilize our networks locally to just kind of out-compete here.

1:18:22Yeah, so listen, I think behavior follows business. I think that there are some systems that are pretty straightforward to reverse engineer, like an actual hardware system where you can pull out the embedded software and you can look at the chip, like let's say network routers, like classically, Huawei copied Cisco routers early on. I think for very complex software, it's just, you know, It's like all the operational know-how is just a very tough thing to copy. You can look at the code all you want, but what you want to know is what it does when it's actually running, and that's just not something.

1:18:52The state space is too large. No, I think behavior follows business. I think there's a lot of business between the two countries. I think they're both commercial countries. They're full of commercial entities. I do think we're going to go back to having a market between the two. But I do think the landscape is shifting. Like, I do think, like, listen, like, right now, you know, we buy, you know, the majority of, you know, our kind of hardware parts for robotics from China. And I would, you know, assume, you know, we sold a lot of software in China that a lot of the business software, like, they're adopting from us.

1:19:25And, you know, that, you know, what that looks like may shift over time, but I don't think it's going to go to zero. Do you think humanoid robots are going to start as an enterprise product or consumer product? and how confident are you that it will start where you think it will start? I mean, listen, I'm not an expert. Can I just talk about the piece that I wrote very quickly just to put this in context? Because I want to show the area where I am expert versus not expert. You're a VC. You're an expert in everything. Give yourself some credit. I know politics. I know humanoid. So listen, the area I'm an expert is infrastructure software.

1:20:02That's what my PhD is in. That's what my company is in. And so in order to know what is impacting enterprise software and software infrastructure, we do these market studies. We'll study what's going on with energy. We'll study what's going on with microchips. We'll study what's going on with robotics. And we did this study on robotics. And it was in service to our software investing, which we do. Even though a lot of the groups in Andreessen Horowitz do hardware investing. I've sat on the board of hardware companies. We do it. But the purpose of this study was software investing. and we got the results of this and like oh damn like china's kicking ass we're like wow they like got the supply chain locked up they're doing incredibly sophisticated stuff i think something like 50 percent of all deployments for the last three years were done in china for the globally and so the purpose of the piece that we wrote was just to say listen you know we do these studies normal we do tons of studies like this but like this was so dramatic like maybe we should kind of make it a little bit more apparent to everybody how China is doing.

1:21:05And so I am not an expert on any given robotics market. If it's an enterprise industrial thing, I'll have more of a sense to it where from a humanoid standpoint, I think there's huge expert on that. But I will say that when it comes to China and the United States, I mean, they're a very, very real contender when it comes to the full supply chain all the way up to the hardware. It's the software where I think things start to differentiate.

1:21:33How do you characterize the opportunity for America to catch up? Should we be more bullish on the hyperscalers, the mature companies, the Teslas of the world stepping up and taking it seriously or completely new startups from scratch building solutions and then kind of integrating, creating their own Keiretsu, if you will? So here, let's talk about the areas where I think that the US has an advantage. I mean, like I mentioned, complex software, the United States does a great job of. When it comes to building products that people want, I think the United States does a great job of. And so I think that right now, when it comes to Chinese hardware, I think we have to really look at ourselves and say, do we want to start putting import restrictions and tariffs on those not to penalize China, but to kind of incur local capability building for these things.

1:22:31And I think that would be an important thing for us to consider as a nation doing. And there's another very interesting area, which is, you know, for a lot of AI, you want big GPUs, right? And so you can do a lot of robotics with smaller microcontrollers, but you kind of want the big GPUs for the heavy processing, like certainly for AV. And it's pretty well known that China doesn't do anything below 9 nanometer very well. You know, like 7 and 5, they can do it, but the yields are pretty low. And of course, they'll be subsidized by China, but they're well behind. So I think at some level, like we should continue our export restrictions on SME, like, you know, ASML lithography machines.

1:23:05Like, I think we should do that just because it does slow them down. I think we should, you know, do import controls, you know, and tariffs so that like we can do local capability building. But again, I like I have no interest in a generational struggle with China. I just think that we need to understand where we're weak, and then we need to put in the controls necessary so that the United States can catch up. And these are the areas I would say. So I would say it's not the big chips. I think we're pretty good about that. It's not the software. I think we're pretty good about that. But I think actually a lot of the hardware supply chain, I do think that we should put our finger on the scale.

1:23:38Yeah, to me, the interesting thing, seeing Unitree robots selling on Walmart.com when everyone in the tech community realized, probably starting about a couple years ago, that letting DJI flood the American market with cheap drones. Just killed GoPro. Yeah, killed our internal, you know, they were willing to sell products at a loss in order to, you know, subsidize these products in order to make it impossible for American companies to compete. and now we have you know millions of drones here in america that are flying around that we don't own and they're going to continue that and allowing them to do the same thing with humanoids feels like an even bigger national security risk to me and would again just set us back in an entirely new it's crickets it's crickets i haven't heard i mean the implication yeah yeah imagine it right it's like it's like health care it's like economics it's like industrial capacity it's I mean, it's like basically as broad basis software if you take the generalized form of robotics.

1:24:41And so it's incredibly critical that we get on top of this. Yeah, so we have a geopolitical rival, and we're going to allow them to put a human-sized robot that can presumably do human things. Wait, I saw a movie about this once. Yeah, yeah, they could do laundry. If they could do laundry, they can do a lot of other things that maybe we don't want them. So anyways, I think this is something that I hope it gets more attention before it's too light. because they're already selling again on walmart.com. I will say one last thing, too, is we kind of sometimes hope that, oh, listen, this is an issue of scale, and we want to work with our partners and our allies, and so what about you?

1:25:16Can they help out here? But if you actually look at some of their dominant robotics companies, like Kukos, I think it's a 100-year-old company. That's now owned by China as well. And so, I mean, even on the global footprint of industrial robots, China is incredibly dominant. And so I think now is the time for us to map it out and figure out what we do to enable local capability building. And again, the point is not to penalize anybody or to have a lifelong struggle. The point is to not depend on them. And what's happening in India on the robotics side? Obviously, Apple is trying to shift a lot of iPhone production to India.

1:25:50Is that an area that you're... I don't have any expertise on that now. Yeah. What about open source LLMs? Have you changed your thinking at all? I know that you were pro open source generally. We had the DeepSeek moment. Now OpenAI has GPT-OSS. MetaLama is sort of declining in favor, but there might be a new generation of models for Meta. They're certainly staffing up for it. And so do you think that that market has resolved? Is it calcified in any ways? Is it heating up? Is it something that people should be paying attention to? Yeah, so I generally think AI is following open source as we've historically seen it, which is the closed source models get 80 % of the market and open source mops up the remaining 20%.

1:26:43And that's happening now. I think open source is incredibly important to the broader dynamics of markets, like kind of anti-monopoly stuff. It's good for startups. It's good for pricing. It's good for academia. It's good for research. So I maintain that position. I think the Chinese models are phenomenal. I think they're leaders. I think we should keep having the academic partnership. I feel very strong about that. I think any sort of import or export restrictions on software is stupid. Like, it just doesn't really work to begin with. And so my view is software kind of, you know, let it all play out.

1:27:13The U.S. is dramatically ahead, I would say, end-to-end when it comes to capabilities and capacities for OSS. The Chinese models are fantastic. You know, we use them. It kind of will keep us honest. I think that's great. Where I think that we should be more serious is import and export restrictions on hardware. I mean, we've done them in the past. that we've done them incredibly effectively. I mean, you know, especially when it comes to critical infrastructure. And I don't say this from an economic or even from a job standpoint. I literally say it from a national security standpoint. Like, you know, do we want to run Chinese hardware infrastructure for our internet backbone?

1:27:43Probably not. And if you know what has happened in the last few years with the telcos in China, you can see why this is the case. And so I do think of having like an aggressive posture when it comes to like what we run nationally and what our capabilities are. So are you not worried about the national security risks of running an open source model from a near peer competitor? There's this idea of the Manchurian candidate hiding inside of the weights of an LLM. It all seems fine when you benchmark it, but once it realizes that it's inside of the NSA or something, it phones home. Red Star Linux from North Korea does that.

1:28:24It phones home. and there's rumors that TikTok and DJI might be doing or might be capable of doing. Having our physical infrastructure, whether it's in manufacturing, like robotics in a manufacturing context or even a humanoid in somebody's home, having that not be something that is a... No, I get that the hardware... Oh, hardware, I don't agree, but you're asking a specific question about software. I'm asking about software. If hardware is like a 10 out of 10 risk, where is running an open source model from a near peer competitor that might want to sneak something in. I don't know if technically they can, but is it a one, is it two, a five?

1:29:01How would you think about it? Super legit question. So first off, I'd say I'm much more sympathetic to import restrictions and export restrictions. I think it's just a net good for the rest of the world to use U.S. technology and to have that footprint. And so when you talk about export restrictions, I think that when it comes to software, we should never have any. When it comes to import restrictions, I think that you should differentiate between what we use for, like, the government and critical infrastructure versus, like, what you have access to academia. I think, for example, if we did import restrictions of, like, the latest greatest models, we would be handicapping our researches in a very significant way.

1:29:35And I don't think that that would justify whatever gains we might have given the fact that, like, half of what we get is from China. Yeah. There's also like a free speech, like math is should be legal argument here where, yeah, like the government and critical systems and maybe our Fortune 500 companies should be very careful with what they run if they're running a Chinese open source model. But if you're just as a libertarian, you know, you could think like, yeah, if you want to go and run this on your local hardware in the cabin in the woods, like that's pretty American in my opinion. Yeah, yeah, yeah.

1:30:05I'm very, very much about never do kind of export restrictions and import restrictions when it comes to software. Sure, sure. Listen, I used to work for the intelligence community back in 2001 to 2003, and this is kind of the height of concerns around Huawei. And so maybe I'm just a little bit more sympathetic to arguments of not having these components in our critical infrastructure. That was the height of the concern around Huawei, and it took us how many years to actually act on it? That makes me concerned about it. It was one of the many peaks of culture. That makes me concerned about it. By 2045, we'll get around to banning Unitree in the United States.

1:30:41After they get a robot in every home in America. We'll finally be like, you have to pay us 10 % tax on this. That'll be the resolution. As long as we're making money from it, maybe we'll be fine with the result. Anyway, thank you so much for coming on the show. This is fantastic. It was a good time. Good to see you, Mark. We'll talk to you soon. Have a great rest of your day. and let me tell you about Fall, generative media platform for developers. The world's best generative image, video, and audio models all in one place. Develop and fine-tune models with serverless GPUs and on-demand infrastructure.

1:31:17Backed by A16Z. Oh, really? Used by Adobe, Canva, and the other players. Core, copped by Perplexity. Did you see this post by Jacob Rintamaki? He said, Truth Nuke, What are you scared of? Oracle being 500 % leopard and not a thousand percent because Larry isn't AGI pilled enough. And tell us something tangible you want. Porsche 911 that comes with a signed copy of the new Dorcasch-Stripe Press book. We should be able to make this. Those are fantastic answers. I believe that's a screenshot of the Teal Fellowship application, maybe something like that. That's hilarious. Yeah, we, Jacob, just post more.

1:31:54Post your way to being able to cover a 911. You can get a 997 for like 40 grand. Yeah, what's the, yeah, yeah. What would the monthly payment be on that? My first 911 was like$32 ,000. Bought it on Bring It Trailer. Yeah, yeah, yeah. You finance that. You pay for it with the ex-post. I think he's - Harder to finance a 997. Yes. But yeah, just post your heart out for 12 months and you'll be in business. Oh, we of course have a sub stack now. You can go to tbpn.com, drop your email. And you can get a summary of the show, some posts that we like, some important links, the run of show, and sort of my take, my current thinking on whatever the current topic of the day is.

1:32:40I'm also posting those on my personal X account. And yesterday we got a quote post from Zephyr who kind of answered the question that I was asking, which was, you know, we have to trust the Sam Altman plan. What is the Sam Altman plan? Are they building their own chip, their own data center, their own everything? like how mature will this be? Zephyr says, is OpenAI building its own chip? Yes, they are in collaboration with AVGO. Fab by TSMC will be available in the second half of 2026. That's forever in singularity terms, in my opinion. The second half of 2026 feels like it's a decade away, but obviously it takes time to actually design and fab these chips.

1:33:21But I want to know more. I want to know more about the plan. What were you doing on August 7th, 2009? August 7th, 2009 I was in college I was, where was I? Well instead of having a couple beers you could have bought Broadcom at the IPO It was$1.64 a share It's now sitting at$345 There we go That would have been good Go back in time, John, and figure it out Indeed, and I could have gone back in time and signed up for Turbo Puffer and been searching every bite but you can do it today with the power of the internet. Serverless vector and full-text search built from first principles and object storage.

1:34:00Fast, 10x cheaper, and extremely scalable. A Wordle game show is in the works at NBC. Jimmy Fallon is set to produce a Get That Size Gong ready. Harvester Stallone says, well, you know what they say, strike when the iron's cold and has been stored in the cup board for several years. Yeah, Wordle. Was that a COVID phenomenon? I believe it was, 2020, everyone was doing Wordles. I still hit them up every once in a while. I like a good word game. I like a puzzle. The New York Times has been on a tear. I like that. It is not true, but I like the meme that people will. The New York Times makes a lot of money from their games.

1:34:41And they have a separate subscription for games. And everyone kind of says, oh, the New York Times is journalism on life support by games. And it's kind of evolved to the point where they make 99 % of their revenue from games. It's like, that's not true. They got to get into gambling. They should get into gambling. I mean, they have an insane user base. Yes. And they're a$9 billion company. Yep. So basically, the New York Times value the same as Polymarket. Yeah. And what could the New York Times do with? They could do mail order gambling. I would love that. Mail order prediction markets. I would love a daily gambling.

1:35:23a daily the daily gamble the daily imagine the new york times betting podcast yes i have another idea for them since they're into games they're into wordle they should do a first person shooter there you go don't you think it'd be good you know we have call of duty we have battlefield yeah team deathmatch you go to newyorktimes.com it just preloads a vibe code you get to play as the legend of Mike Isaac. Yes, exactly. Love it. You get Mike Isaac. Love it. Russell Paulson quoted our post. We were talking about the Anthropic ad campaign. He added some insight. He says, this is all about values-based marketing.

1:36:01It's not about the hat or what Claude does technically. It's about how it makes us feel and how it elevates our inherent humanity. It's a breath of fresh air to feel like an AI tool makes us better rather than replaceable. Yes. Very well said. I like that a lot. And it does feel like a pivot. Do you think this feels a little bit written like Claude? It's not about the hat. You can't. When I write now, I actively think, okay, if I'm going to try and put something in a it's this, not that, I'll just completely rephrase the sentence to just be a wildly different structure because it just has a tinge to it these days.

1:36:39Dean Ball says, OpenAI plans to detect underage users and give them a model with more safeguards. This may clash with Colorado's SB205, which prohibits algorithmic discrimination based on age. What is Colorado doing? One of literally millions of pro-social AI uses plausibly rendered illegal by that stupid law.

1:37:03Very interesting that a single state is going to be able to potentially create policy, nationwide policy around AI. I think David Sachs has been kind of ringing the alarm bells about this. And there's been a number of folks we've talked to on the show who have said that like the state by state stuff is going to make it really. It sort of turns us into Europe where you have like one cookie policy that then cascades all over the place. At the same time, I feel like is it not possible to just like geofence a feature? Still, I mean, you have 50 states if they all have different laws. Colorado only has six million people.

1:37:43we have 40. I'm not worried about the technical burden. Should be able to one-shot that with Codex. Come on. We're in the future. But I am worried about some weird situation where one of those Coloradoans is on the border and the cell phone tower that they connect to says that they're in a neighboring state and then that creates a liability and there's a class action lawsuit or they're using a VPN and the VPN doesn't count as actually leaving Colorado. And so there's a lawsuit there. And if it just becomes a legal precedent that requires you to treat the entire United States populace like the folks in Colorado, that could be very, very annoying and kind of just dampen everyone's experience.

1:38:27Because when I think about, you know, we have kids. When I think about if they're using AI tools, like I like the idea of more safeguards. I, you know, I like that if If you open up the Disney app on Apple TV, you can choose. Don't show anything R-rated or PG-13. And just, you know, content filtering, adult parental controls. Like, these are good tools that should be in the hands of parents. And they should be surfaced by the tech companies. Also, I don't really want, if, you know, my son's very into dinosaurs. And if he's someday talking with ChatGPT about dinosaurs, I don't really want, I don't feel like it's necessarily fully appropriate for him to feel like he's becoming friends with the dinosaur machine.

1:39:13Yes, I would also. I think we're probably years and years and years away from him using LLMs. I would also just be worried about annoying syntax leaking into his vocabulary. He's like, Dad, this isn't a Tyrannosaurus Rex. It's not bedtime. It's the king of the jungle. He's like, come on, dude. It's not bedtime. It's playtime. It's playtime. which is basically the constant debate. Anyway, did you, do we mention profound, get your metal, get your brand mentioned in chat GPT, reach millions of consumers who are using AI to discover new products and brands. He's got to mention them. P.E. Cooper. Hit the timeline, built his own TBPN.

1:39:54Meme template. Meme template, and he's putting it to use. What is this picture? Is this just how he sees himself as a royal lord? I love the, I love. Very base. I love the enthusiasm to just hardline that you're an 18th century or 15th century king or something. I like it. It's very royal. Bill Ackman. You can go check out. He has a meme maker. You can make your own. People have been adopting the TVPN trading card aesthetic. Thank you to Tyler Cosgrove for, I believe, inventing that whole strategy or something like that. I think you made the first one of the AI researcher that got 5 million views on X and then was shared all over the Internet and wound up getting us on French TV.

1:40:39Did you wind up talking to the French journalist, by the way? I talked to them, but I don't think they wanted me as like an actual interviewee. Okay. They didn't film you? I think they want people that work. They're not real fans of the show, clearly. I don't understand. A true fan of said. You're a heavyweight in the industry. We tried to pitch them this idea that Tyler was beyond the Metis list, the zero-eth AI researcher, since, of course, lists are indexed at zero. Bill Ackman says, in one week, Jared Kushner pulled off the biggest LBO of all time, that is EA, and then negotiated one of the biggest peace deals of all time.

1:41:15A pretty good week. That's wild. Really just running the art of the deal playbook. We have been so behind on our EA coverage. Would love to have Jared on the show. In the meantime, you can listen to his interview on Invest Like the Best if you want to understand more of his investing philosophy. The EA Take Private is fascinating. It's a massive deal. EA has so many interesting properties from Battlefield to – they have Madden, right, and FIFA. They have all the sports. And it's just a very interesting business. We've talked for a long time about what it takes to pivot a business model in the public markets, how difficult that is.

1:41:55Will they be pivoting the business model to what? If so, will they be doing that? Is that the point? Or is there some other restructuring that will happen while the company's private? That's fascinating to me. Ben Thompson had an article in Stratechery Today, which of course you should go subscribe to, all about basically the failure of Xbox Game Pass because Microsoft paid something like$70 billion for Activision. They got Call of Duty. And the idea was they would put people on a Netflix-like subscription and you'd be able to play any game you want. And according to some rumors and talking to some Microsoft employees and some leaked data, it seems like the LTV went down.

1:42:38So it seems like it was actually better to get people to shell out$60,$70 a couple times a year for the big games that they want as opposed to putting them on a$10 a month plan. and then they're playing all these games for much cheaper. And so it's been, you know, the conclusion is that it still isn't working. We'll have to keep checking in on it. But it'll be interesting to see, does EA go private and then find land with Sony or something? Like land with another, you know, gaming platform? Does it create its own store or, you know, pivot into, you know, buying other companies and getting into mobile gaming?

1:43:17Is there some sort of other monetization microtransactions? Like, what are they going to ramp up? They've tried a lot of stuff, but it'll be interesting to see where it goes. So we'd love to talk to the folks over at Silver Lake and Jared Kushner's firm. Will Manitis says, it's genuinely nuts that if you get good enough at New York City real estate, that the only other place where those skills are useful is negotiating world peace. I didn't realize that these two were connected, but of course it's about Jared Kushner. Yeah, and Trump, of course. Yes. Oh, yeah, Trump. That's very funny. Should we cover the China news today?

1:43:53So Dmitry Alperovitch says, huge escalation by China. MOFcom announces new export controls taking effect after December 1st of rare earths to anyone anywhere in the world producing chips or equipment to make chips below 14 nanometer and 256 layer memory due to military applications. And so Dean Ball had a big take on this that we should probably read through to get up to speed. I'm not an expert on this, but I hope to be after talking to Dean. We're hoping to have him on the show soon. Dean says, this is a very big deal. China has asserted sweeping control over the entire global semiconductor supply chain, putting export license requirements on all rare earths used to manufacture advanced chips.

1:44:42So we have the design firms with NVIDIA. We have TSMC kind of on our side. We have ASML. There's a whole bunch of American IP that goes into the Western semiconductor supply chain, but China's upstream of all of that with their rare earth supply chain. There was a comment in the chat earlier about what does China control all of the rare earths. I think the common cited metrics are that they control a huge portion of the rare earth supply today, but they don't actually have all of the rare earths. More processing than supply. Rare earths aren't as rare. We just call them earths. Yeah. And around here, we call them earths.

1:45:25Because they are, in fact, all over the place and you hear about these rare earth deposits being found in Ukraine and in various parts of the United States. We have them. We just haven't actually built the mine and processing and that takes years. And so in the short term, putting an export restriction on the actual processed rare earths is very geopolitically significant. So Dean Ball continues. He says, if enforced aggressively, this policy could mean lights out for the US AI boom and likely lead to a recession slash economic crisis in the U.S. in the short term. Remember, this wouldn't be the first time they tried to, I mean, the DeepSeek moment felt at the time very calculated.

1:46:07Let's ramp this app up the charts that I didn't still didn't know a single person that was using DeepSeek. It felt entirely that the chart run felt entirely manufactured and that the immediate effect on markets was insane. counterpoint maybe they were just trying to take a little bit of air out of the bubble to create a bull catalyst i mean everybody knows if markets go down that's a bullish catalyst yes it also inspires us to grow grind harder yep uh tyler do you know ball dean ball oh i know of yeah oh you do okay cool oh he knows okay good uh do you know linear that's why we hired you it's purpose built tool for planning and building products meet the system for modern software development, streamline issues, projects, and product roadmaps.

1:46:55Didn't start building, but leaning. Some thoughts and recommendations from Dean Ball. Many things are unclear, including how intensively they will enforce this. This is a costly decision for them, too, and not without risk. These controls are being imposed on the whole world, not just the United States. My current assumption is that China is doing this to gain leverage over the U.S. in advance of the president's upcoming meeting with Xi Jinping at APEC. it'd be very funny if they just put the controls in the United States and then you had a bunch of startups that were like, I'm going to do diversion and I'm going to go set up a rare earth buying and selling operation in Dubai or Malaysia or somewhere.

1:47:39That'd be very interesting. Number three, I do not think they are doing this to get the US to relax its export controls on NVIDIA Blackwell chips. In fact, my guess is that they want us to think that's their goal. We buckle by relaxing Blackwell controls, and then the PRC shrugs and says, no thanks. We're very confident in our domestic chip industry. That is a message I'd want to be delivering to APEC if I were advising the PRC. Four, further relaxation of export controls is therefore unlikely to be a useful policy lever here. Instead, it is time for the U.S. to get serious about export controls on semiconductor manufacturing equipment.

1:48:16Five, because we have been mostly unserious about this issue, China has stockpiled most durable equipment needed to make semiconductors. So most controls will not have an immediate effect. Six, he really ran through this. This is great. Six, however, there are some consumable materials needed for semiconductor manufacturing that China cannot make domestically. Even better, those consumables are often degradable, meaning one can only stockpile so much. These would be on the top of my list if I worked at BIS. I would retaliate strongly, but in a targeted manner. The ultimate route to de-escalation here is tariffs and broader trade war policy moves, not AI compute.

1:48:54This only underscores the urgency for the US plus allies and partners to accelerate domestic efforts to mine and refine rare earths. Let's hear it for that. The Trump administration has been especially strong here, and Congress allocated serious money to this goal in O3B. But all options must be on the table, including the Defense Production Act. Policy will be important here, but remember that these rare earths are essential to the functioning of the global economy. It has heretofore been hard to solve with markets alone because China already supplies these commodities in abundance and at low cost, but they are fundamentally just commodities.

1:49:29We can make them. And 10, to the extent China enforces this policy, they will have made the price of these essential goods infinitely high. Remember, most of all, that supply is elastic, not on arbitrarily short time cycles, but faster than most people think. So if the price is infinitely high, there will be a huge incentive to go start mining rare earth elements in the United States. Interesting. MP Materials is actually down 0.6 % in the last five days, which is, of course, the company that the Trump admin took roughly a 15 % stake in, in rare earth mining. Well, Peter Haig says, sir, a second whispering to the president meme has hit the timeline.

1:50:14Is this a real photo? This feels like something that someone would have AI generated, but it looks real to me. It's from Reuters. And sir, a second ad read has hit the timeline with numeralhq.com. Sales tax and autopilot spend less than five minutes on sales tax compliance. I screenshotted that one twice. It was so good. Andy Jassy says, Joseph Carlson, the current CEO of Amazon, has been with the company for 28 years. Wow, Andy Jassy, what a run. In 2003, he played a pivotal role in creating and scaling AWS, which is the most valuable part of Amazon. A significant portion of the modern internet is powered by AWS.

1:50:52Amazon today is worth$2.4 trillion and is one of the most critical companies to the United States and the world. Yet, the CEO makes less money than many YouTubers, comedians, podcasters, professional athletes, and actors that don't manage or guide anything nearly as important or critical to the U.S. or have millions of employees. We need to stop pretending that the CEOs of Fortune 10 companies who have helped build trillions of dollars worth of value for investors should have their salary compared with the delivery drivers working their way through college. This trope is old and should be retired.

1:51:26Yeah, Josh Hawley, senator for Missouri, is firing shots at Jassy, saying, how is it pro-worker or pro-American that he makes at least$40 million a year while the average Amazon worker makes less than$38K a year? Well, he needs to go on a campaign. Bobby in the chat says Andy Jassy doesn't even have 200 ,000 Twitter followers. That's rough. He's also never streaming on Twitch. Get on Twitch, Andy Jassy. We'd love to see you there. I know. They own it. Amazon owns Twitch. And so that's where he should give his press releases. This is wild. Chip Wilson, Lululemon founder, takes out a full-page ad in the Wall Street Journal to roast the brand that he started.

1:52:13When was this in the journal? Lululemon in a nosedive. He says, when I founded Lululemon in 1998, it was built on a relentless focus on innovation, product culture, and customer experience. We invented a quote-unquote technical product business model, created a new apparel category, pioneered community marketing, and developed system processes so simple a refrigerator could manage them. By its 25th anniversary, Lululemon should have been a$100 billion market cap company. Instead, its trajectory has declined. Why? Lululemon directors have systematically dismantled the business model and lost employees who held the institutional knowledge that made the company great.

1:52:50like a plane crash, decline rarely happens because of a single failure. It's a series of mistakes. He's adding some different sections here. One, the founder leaves. When a founder leaves, boards tend to fill committees with operators or finance-focused directors who are beholden to quarterly commitments. A company bereft of a visionary loses its singular voice for product and long-term strategy, a strategy that builds a moat around success. An operations-slash-finance - driven board lacks the moxie to understand the market pulse, gap, gap, evasion loss of creative to the merchants. Every few years, a design team will stumble.

1:53:27When that happens, merchants and MBAs swoop in claiming to be the adults who can deliver consistent quarterly projections for wall street to meet those projections. Merchants follow algorithms and double down on what sold the year before. This kills risk taking and prevents new innovative product from emerging. As a result, innovation dies. The brand becomes predictable. Customers drift, and the best creative people leave. The domino effect. As top talent leaves, short-term results often look strong. The board harvests brand value by rapidly growing store count, while margins expand through economies of scale.

1:53:55Wall Street is ecstatic, and the egos of directors and management soar. Meanwhile, owner-operated competitors seize the opportunity, producing better product. Lululemon's institutional knowledge walks out the door, often to competitors, and is replaced by Nike executives, shots fired, who import failed business. model. Desperate for growth, Lululemon squandered$1 billion on Mirror and wiped out$10 billion in market cap with a wildly inappropriate Disney collaboration. To squeeze margins further, directors cheapened store design and shifted to non-technical fabrics, eroding the brand's premium positioning.

1:54:30Cultural erosion. New executives from old school companies chased short-term margin growth and resume building. Directors unwilling to confront their own role in the decline in an attempt to replace the CEO, only to realize for the third time that they've failed to develop an internal successor. With the brand faltering, no capable candidate wants a CEO job. The nominating committee struggles to attract directors because no great leader wants to join a sinking ship loss of cool. The board insists on operator finance CEOs who can speak Wall Street, rejecting the idea of a product-driven CEO. These types of finance-focused CEOs don't know how to attract or motivate creative talent, or even worse, they think they understand how great product when they don't.

1:55:08Without an eye for outstanding design, the company dies a slow death. Lululemon forgot its muse. The woman who inspires culture not just follows it. By drifting toward the mainstream and trying to appease everyone, Lululemon lost 50 % of its market cap earned from brand power. It lost its edge and with it, the ability to hire the best people. The result, on paper, Lululemon still looks good, but it's losing its soul. It's still a$20 billion company. That is getting absolutely smoked by aloe in their core segment, which is the muse that he described above. And founders back at Outdoor Forces, too.

1:55:44Yeah. So he says, the deeper issue is not just management. It's a disengaged nominating and governance committee that has failed to safeguard the company's long-term vision. The path forward. I believe Lululemon can recover, but survival requires courage. Put product and brand back at the center. Rebuild the knowledge and systems that deliver product in nine months, not two years. Bring entrepreneurial ownership back onto the board. Three, empower creative leadership over merchants. Four, stop chasing Wall Street at the expense of customers. Five, recommit to the muse, the woman who inspires the brand.

1:56:13The world doesn't need another bland, quarterly-driven apparel company. It needs bold vision. It needs Lululemon to fly again. Lululemon can keep growing, but growth alone is not a healthy measure of success. The true measure must be innovation and brand reputation. When those are strong, growth comes naturally. When they're not, growth halts. The path is clear, but only with a revitalized board of directors with diverse capabilities. Sincerely, Chip Wilson, founder of Lululemon. Chad Wilson. When did he actually... Step off the board? No, when did this actual article come out? It feels like...

1:56:48I mean, it's not in today's paper. I checked every page, so... But I think it... But it must be recently. I think it came out... I'm on the journals. Yeah. Yeah, I think it came out yesterday. day. What's interesting is like, I'm surprised he had to pay for this. Like this would just be a good op-ed to throw in there, but it does hit. It's kind of, it hits harder that he's paying out of his own pocket. And the layout also just feels very, I don't know, iconic. It looks a lot different. So we should get Chip on the show. I'd love to talk to him about that. Yeah. I mean, they're just being competed from every possible angle.

1:57:23There's new brands popping up all the time that are more catered to the early kind of core lemon customer. You have Aloe, you have Vori, you have Skims from Kim Kardashian, right? And all these brands are just better serving that initial cohort. So, yeah. Well, let's get him on the show. Let's pitch him Thin.ai, the number one AI agent for customer service. More SaaS, Chip. Number one in performance benchmarks, number one in competitive bake-offs, number one ranking on G2. Did you see Weiji Capital taking a shot at Samuel Gibson? We'll hold our judgments until we actually meet him. So Weiji Capital says incredible things are happening in the nuclear grift space.

1:58:15Meet Samuel Gibson, Zoomer, who graduated from college in 2023. Let's give it up for Zoomers. Yes. And is now co-founder and CEO of Hadron Energy a nuclear vaporware company going public via SPAC at a$1.2 billion valuation. They have, according to LinkedIn, between two and 10 employees. Okay. But 11 ,000 followers. Everybody starts somewhere. I don't know. I mean, you got to get the money to build a nuclear reactor at some point. And they raised a$900 ,000 pre-seed, according to Crunchbase. You went from 900K pre-seed to$1.2 billion SPAC? I mean, built different. Gen Z really on a tear. He had a 3.4 GPA at University of Nebraska-Lincoln.

1:59:05Not bad at all. I don't know. Get the bag. Get the cash. Build the thing, hopefully. Got to build. Yeah. You can't purely cash out and then write it down. You got to turn it. You can SPAC it. But you got to deliver. On their website. You've got to deliver the actual energy. On the website, they have a bunch of renders. Yes. And one of the renders just says Data Center 1 on it. Okay. So they've got data centers. They've got data centers. Yeah, but they're saying the world's smallest, lightest, most versatile micro-modular reactors. Yeah. And the design is awesome. The renders are awesome. Yeah, a lot of work.

1:59:42I mean, I feel like... And they have a Tesla Cybertruck render, pulling a render of their micro reactor. So it's looking, it almost looks kind of like Fortnite. Huge question on who's actually underwriting the SPAC, who's running this thing. How did the capital go into the pipe? Like what actually happened here? You know, there's, SPACs are not inherently bad. you can raise money in a variety of ways. And if you go and actually deliver and generate a lot of cash flow, at some point you'll be worth way more than what you SPAC'd at. So good luck to him, but it does seem like a tall order to be building in one of the most heavily regulated.

2:00:29Gig Capital Global is putting up some of the cash here. And they also, you might remember, they are investors in Big Bear AI. I don't know Big Barry. I didn't think you would. Okay. But anyways, we'll see here. I think we should withhold judgment. Yes. The renders look cool. Plug it in. Plug it in. Plug it in. Just plug it in. Plug in the nuclear reactor. Generate some electricity. Profitably. It's time to plug. Tyler, you should talk to him and see how real it is. Yeah. I mean, I've been thinking about getting to the SPAC game. Oh, yeah? What are you going to SPAC? uh i think i'll probably figure it out later on okay i've kind of started the motions already but just empty shell for now yeah self-spac cabaluraza so could go any direction yeah i'm surprised somebody hasn't tried to spack themselves with some type of income share agreement yet during the nft boom someone did wasn't alex token am i getting that wrong some yeah but a lot of that he definitely i think started a trend of people isn't this the ben pasternak thesis as well everyone will have a coin attached to them or their project.

2:01:45Nathan Summers in the X chat. Are you guys hiring for a narrator intern? I have great cadence, he's a drummer, and tonal variety for an exciting viewer experience. I also would hate to see a vocal cord strain cut down on your ability to run the live show DMs. Imagine being able to say - Tyler, reach out to Nathan. We've got to talk to this guy. Okay, yeah, he's just sitting here and we're saying, read me this Wall Street Journal article. Yeah, a narrator - Narrate, and we will just do the reaction. action, but we want you in person, live, to read us this story from the Wall Street Journal on, can Ferrari persuade the super rich to buy an EV sports car that won't rev?

2:02:21We're going to read this article, but first I'm going to tell you about Adio, Customer Relationship Magic. Adio is the AI native CRM that builds, scales, and grows your company to the next level. So the brand Ferrari is planning to unveil its first all-electric model next year, and it's a bold move. It's a bad time, probably something that was greenlit years ago, but we will dive into it in this Wall Street Journal article. And some backstory here. Ferrari just updated guidance. They had a capital markets day. Wall Street was not excited about it. The stock is down 18 % in the last five days. Okay.

2:03:00Let's get into their next vehicle, which I think is something no one wants. We will see. Extravagantly powerful and noisy engines helped make Ferrari the ultimate sports car brand. Now the company wants to persuade the super rich to buy a model with no engine at all. The Italian carmaker this week started lifting the hood on its first electric vehicle, a years-long project that has cost the brand hundreds of millions of dollars and promises to set a benchmark for how battery-powered sports cars should look sound and drive. At a glitzy launch event at its headquarters in Marinello, Italy, Ferrari showed off the technology that will power the EV, including a new electric axle, motor, and battery pack set to be made in-house at its factory.

2:03:47Wow, going vertically integrated all the way down to the battery on day one. That's bold. Which is wild because Ferraris have always been known to have electrical issues, including the Ferrari that I owned, that I think I replaced the battery, I forgot exactly how many times, but I had to replace the battery multiple times in a single quarter, so. Skill issue. It said the vehicle called the Ferrari Electrica. What do we think of that name? is not quite there. The Dolce Trilindy hits, you know, the Electrica, maybe. It feels like a name that somebody that was like trying to come off as an Italian brand would use.

2:04:30Well, it'll be able to go from 0 to 62 miles an hour in 2.5 seconds, which is fast, but not as fast as a Model S Plaid that you can get on Bring a Trailer for 50K now. So it has a top speed of 193 miles an hour. The range on a single charge will be at least 329 miles. That's not bad. The company won't reveal what the model looks like until next spring, with deliveries slated to start later in 26. Apparently, the Model S Plaid does 1.99 seconds, although I wouldn't want to try that out, to be honest. You wouldn't? Or you would? Maybe on like a runway. Yeah. But on a normal road, I wouldn't want to, I wouldn't.

2:05:15You wouldn't want to go that fast? I just. Maybe when you're an adult, when you're a man. Yeah. Then you can do it. When I finish growing. Yes. The debut comes as the global auto industry adjusts to a slower transition to EVs than expected when Ferrari announced the project. So when Ferrari started working on this, everyone was saying everything's going to be EV. EV is going to dominate. It's not just going to be some like subcategory of the market. It's going to be everything. So you've got to get in on it. And, of course, other sports car makers such as Porsche, Lamborghini, and McLaren have delayed plans to build electric models citing weak demand.

2:05:49While Ferrari is committed to launching its first EV on schedule, it has scaled back its wider bet on the technology. At Investor Day Thursday, which you mentioned, the company said EVs would account for 20 % of its vehicle lineup in 2030, half the share previously indicated. We know that this path is not easy, Ferrari chief executive Benedetto Vigna said in an interview. But we also know that we as a leader need to show the entire world that we can harness this technology. At the same time, the company plans to release many new vehicles with traditional engines. According to its latest product roadmap, the Ferrari Electrica is one of 20 models set to launch through 2030, more than previously planned.

2:06:29Despite the new models, Ferrari issued a cautious sales forecast, 5 % annual revenue growth through 2030, and the stock fell 12%. Since being spun out of Fiat a decade ago, Ferrari has become Europe's most valuable carmaker by persuading uber-wealthy gearheads to splurge hundreds of thousands of dollars on meticulously crafted sports cars. EVs pose a particular challenge for luxury sports car brands, which say roar and rumble are central to their identity's appeal. Ferrari, perhaps more than any other automaker. Imagine Nathan in the chat narrating this right now. I know, it could be so much better.

2:07:09The brand is known for its noise. The Dolce Trilindri is naturally aspirated, meaning it doesn't rely on a turbocharger to push air into the cylinders. An old school approach credited with a more natural open roar. Ferrari said its EV wouldn't mimic engine sounds as some competitors have, like the Kia EV6 GT. I forget what it's called. Yeah, that was an engineer from the M series, right? That's Hyundai, actually. So the engineer who created the M division at BMW went over to Hyundai and created the N division because N comes after M and tuned up a bunch of the Hyundais to be performance cars. And one of them is the Hyundai EV6 or something.

2:07:57I'm getting the exact model name wrong. But it has not only engine noise that comes through the speakers, but a simulated dual-clutch transmission that you can operate via paddle shifters that are entirely digital. And so you can essentially be rev-limited and be redlining the car, even though it doesn't have a real redline. And a lot of the super old school car lovers tried it. Doug DeMuro and a bunch of other folks, Matt Farah from Smoking Tire, all tried it. And they were like, this is amazing. Like, I forgot that I was driving an electric vehicle after just a few minutes of the car. And so the prediction at the time was that every electric vehicle car maker would adopt that same simulation technology because it was popular.

2:08:45And it's a pretty easy feature to add. seems like Ferrari won't though. They're going natural EV. Instead, it will pick up the sound of what it calls the electric engine and amplify it into the cabin to give the driver feedback when required. So you'll hear the electric engine whirring. The company compared the difference between its conventional sports cars and the EV to the difference between an acoustic and electric guitar. Ferrari also faces the risk of EVs losing value faster than traditional models as the battery degrades. The promise of timelessness is a selling point for Ferrari buyers.

2:09:17The company leans into the economics of the luxury industry, carefully limiting supply to ensure that its top models retain their value over time, like Rolex watches or Hermes handbags. And if you want a Rolex, head over to getbezel.com. Your bezel concierge is available to source you any watch on the planet, seriously, any watch. So Elcan also recruited Johnny Ive to work on the brand's first electric model. Oh, interesting. I didn't realize this. We talked about this with Love From having a Ferrari partnership. This is the car that was designed by Johnny Ive. That's fun. To make its electric cars, Ferrari built a state-of-the-art factory, which became known as the E-Building.

2:09:58After heavy investment, the question now is whether Ferrari's wealthy clients will want to drive the new model. Imagine spending your whole life being obsessed with the Ferrari brand learning to work on these engines and then getting the tap on the shoulder and saying, you're actually going over to the E building. Yeah. And I think there's also this factor that it feels like the really highly engaged Ferrari buyer maybe doesn't care about straight line speed that much. It's never been a muscle car. The more exciting number that folks track is like the Nürburgring time and the gas cars are still performing extremely well.

2:10:40So just the feeling, the factors that, for sure. Yeah. Just the feeling of driving the car, what the, the ownership experience, right. The, the, the main critique of Ferrari from some of the more, I would say casual clientele is like the depreciation recently. The cars are just losing value far more quickly than they historically did. part of that is already because uh the hybrids right the the sf 90s the the s um the 296 is also a hybrid right that's appreciating i thought that one was kind of holding its value a little bit more a little bit better but i mean you can buy them for less than sticker with like three you know very less than sticker for like 300 miles on them um i would have expected uh ferrari to try to go more towards something like a nissan cross cabriolet i would love that take the pearl song Give it a drop top.

2:11:34Yeah, iconic silhouette. A rag top per sangue might be the move for them. This is interesting, though. Canadian collector Luc Poirier, who owns 38 Ferraris, said he was open to the idea of buying the EV, having found the company's hybrids engaging and emotional to drive. The SF90 is a wild driving experience. experience. I was in Dubai and rented an SF90 and took it out. It's pretty funny. They give you tickets in the UAE based on you'll get, they have cameras. They'll give you tickets without you even having any sense of it. So you can actually go out on a drive and get multiple tickets if you're not safe.

2:12:20I was, so the rental service was, they keep like a deposit, like long after you've returned the car and insured that they're like okay the car is good but you might get you might have gotten like five tickets that's funny and so we need to hold this um but uh luke luke says my only hesitation is reliability since it will involve so much new technology ferrari started dabbling in hybrid technology on the racetrack in 2009 and hybrids now account for almost half its output uh vigness said ferrari would apply lessons that is learned from hybrids to its ev we will not sell an electric car without considering the expiration date of the battery Many EVs and hybrids haven't performed well in the secondhand market, a problem typical of fast-moving technologies.

2:13:01SF90 and SF90 Spyder, for example, have depreciated faster than comparable models with conventional powertrains, according to Hagerty. Again, there was a lot of people that bought SF90s at the top for way over sticker. It also came out during the zero interest rate sort of bubble. So the auto loan rates were really low, and a lot of people were flush from various crypto schemes. and all sorts of froth in the market. Oh, yeah, everywhere. It was easy to buy it over sticker. Do you know David Lee, the Ferrari collector? He has over 40 Ferraris, I think, are all in red. He has all the supercars, the F40, F50, the LaFerrari, the Enzo.

2:13:40He says to be a top client of Ferrari, you buy every car that they introduce, and he says he's planning to buy the EV. But I wonder how many of those collectors will stick around if they just took a bath on an SF90, they're not doing well in their 296s. Yeah, but this, I mean, so city analyst Harold Hendrowski says, it is somewhat surprising that Ferrari is going ahead. Overall demand for those cars is minimal. We were talking earlier off the air, just saying they had such an amazing opportunity to let the rest of the automotive industry go heavy. You know, Porsche went heavy into EVs, is now pulling back.

2:14:19Ferrari could have said we never wanted to go into EVs we like you know naturally aspirated v12s we're going to continue to focus here if they were the one that never did the EV would have been crazy it would have been a very special thing to the brand and now they're they're still yeah they're still like in the same conversation as Porsche and Lamborghini McLaren but again if you're a client and you want access to the F80 you can better you better believe that you are Pick up a couple of these. Go into your dealer, get pre-order a few of the Ferrari Electricas. At the same time, like I feel like if you're David Lee, you own 40, like you definitely want to own the first EV just because it's going to be interesting to see what Ferrari does with an EV.

2:15:02Like it will be a different experience. It will be, there will be a Ferrari twist or take on EV. Yeah. Even though that's been the major. I'm still so interested to see what happens with the Pura Sangue. Yeah. From a depreciation standpoint, they're holding up well now. It's a naturally aspirated V12. I think it looks, you know, people debate the looks all the time. Yeah, but you can always put a Montserrat body kit on it and get the Puginator, and then you're good. But, yeah, we'll see how these hold up. Maybe lower it, make it off-road edition, Safari edition. I don't know. The Uris has held up incredibly well because it still looks exactly the same as it did in 2019.

2:15:40Yeah, it hasn't been refreshed. People will happily buy, happily, you know, continue to purchase a older Urus and use it to sell courses or whatever they do. But I don't know about the, it's hard for me to imagine people paying half a million dollars for a Ferrari SUV or even$400 ,000 in a few years, right? so well if you want a Ferrari for your bedroom get an 8th sleep pod 5 5 year warranty 38 risk retrial free returns free shipping I'm back on a roll I got a 93 8 hours 23 minutes let's read this post from SMB attorney who says wait what'd you get I got a 93 oh I got an 89 2 90s back to back SMB attorney says people keep telling me the American dream is dead but every day I meet entrepreneurs who know more about SEC football than accounting drive trucks and think getting dressed up means a good pair of jeans who've made millions running businesses that clear land trim trees and install equipment the american dream is alive and well it's just hidden behind hard work and a little bit of sweat we have to get a plumber or an electrician on the show that's cashing in on the data center boom there has to be some guys that are on their way to at least getting a PC-12, you know, turboprop.

2:17:07I wouldn't be surprised. But I think that this post is more just about, like, you don't need to be a janitor at a data center to experience a boom if you're just running a well-run operation and have scaled it. There's a reason, you know, the meme is the private equity guys will come and buy these companies. But the reason they do is because they're good companies. And so maybe there's still an opportunity there. And there's also opportunity in the public markets on public.com. Investing, for those who take it seriously. They got multi-asset investing, industry-leading yields. They're trusted by millions.

2:17:38They now have direct indexing. Yes. I am playing around with. Well, what do you got? Wilmanitis on the timeline. Back. With a longer post. Take us a minute to read through this, but it should be fun. Get started. I'll be right back. He says, you can leave the kingdom, but you're still paying tax. Incumbents realize their industry is too regulated and too consumer unfriendly for growth, so they build a parallel, less regulated, often cash pay alternative next door. The parallel system is easier but causes massive consumer harm. I've seen this myself. The prime example of this over the last five years is the rise of telemedicine pill mills in healthcare.

2:18:19Think of the healthcare industry as defined by two factors. One, extremely limited supply. Medical doctors are scarce and costly. And two, a principal agent problem with payments. Your insurance pays and you don't. This has allowed the industry to behave in incredibly weird ways. Costs can constantly grow. Patient experience can constantly degrade. And clinics will still fill up with wait lists because patients trust it, need it, aren't paying for it, and have no choice. As a result of this, we saw a world where patients replaced their primary care doctor with a fleet of telemedicine pill mills for lifestyle drugs.

2:18:57Further, lax regulations during the 2021 era allowed for these telemedicine services to offer things the traditional system could never offer. Fund lifestyle drugs with limited medical supervision. Sure, it was a better experience, but that patient went from having 75 % of their care subsidized by insurance to having 0 % of their care and drugs subsidized. the parallel system passed costs onto the consumer. Furthermore, the consumer is likely harmed in the form of lower coordination and regulation, opening the door to all kinds of harm, bad drugs, conflicting drugs, et cetera. How many people do you know that are now on four, five, six, or seven drugs of dubious medical need?

2:19:39I actually don't know that many people that are on that, but I certainly hear stories about it. Flip a coin. Either way, the percent of your income spent on healthcare is going up. That seems true. We see this playbook across industries. When the kingdom was built, the merchants paid taxes, coordination, regulation. They hated paying taxes and boxed themselves in via regulatory capture, so they pushed it onto the consumer. Unable to conduct business, both the merchant and the consumer leave the kingdom. Things are easier outside of the kingdom walls, but the merchant gets richer, faster, and the consumer gets harmed.

2:20:12You see this in options markets today. You see this in defense procurement, which I don't know what he's talking about. I've never heard of defense procurement happening outside of a regulated environment. I do see it in healthcare. You see it in crypto. That seems like a great example. You see this everywhere. I would love to know what he's thinking on defense procurement specifically. We build palaces of regulation to prevent consumer harm and force consumers into unregulated legal gray area systems in order to conduct business. It's easy to see why. This is a durable economic effect we see across economies.

2:20:46You can simply squeeze more growth out of an unregulated market. Look at rising Asia. And the U.S. has allowed for a continuously developing set of emerging economies by its glacial pace of gray market regulation. We should view this as an incredible indictment of a regulatory state. The solution is to overregulation of essential industries isn't to allow the progressive construction and delayed regulation of legal gray areas. The solution to overregulation is to remove regulation on existing industries, parallel construction of regulated system harms consumers. And I have a great example of this that I've lived for years, but I can take you through it.

2:21:23me. So Juul started 10 years ago, got immensely popular, and was so regulated by the FDA that at one point they were banned, and then they finally got approved. And all of this was justifiable in the micro. But during the time that they were banned and going through all this, guess what happened? There was an entire massive gray market of illegal vapes that weren't even trying to be approved by the FDA. They weren't participating in the regulated structure. They were not in the kingdom. They were just out in the gray market. And they became so big. You saw that video too of a child testing the vape?

2:22:02Yeah, it's crazy. They were coming off the line of child in China's hitting them all to make sure they're working. And then before they ship them out to America. Yep. So crazy. So the market moved to gray market. And I don't know what the stats are most recently, but at one point, illegal vapes, mostly imported from China, were the second largest category of nicotine consumption after cigarettes. So it was much, much bigger than Juul and the companies that were trying to be regulated. There's obviously regulation that needs to happen in this category. But if it takes 10 years, yes, there is a gray market that's going to develop.

2:22:36And so I firmly agree with Will. I've seen it in the business that I know best, and he's obviously seen it in the medical space. I have no idea what he's talking about in the defense space, but do you have any thoughts on Will's latest long post, the yuga, as he puts it? The yuga. I don't know. I think it's interesting that we haven't seen more about the pill mills and all the telemedicine shops from the 2021 era. And we were hearing rumors. Working through the courts. Yeah, we were hearing rumors of like ATF getting involved or the government getting involved. Some of these companies going bust.

2:23:11A lot of them, they kind of had legal problems, and then they just kind of melted away, and they just became quiet. And I haven't heard from these companies. Maybe they still exist. Maybe they just kind of quietly stopped doing the bad thing. Yeah, I'm sure they're all doing it again, or they're being reborn to do various peptides and semaglutide and all those. Well, we heard about a fantastic company, biotechpeptides.com. Do not go there. I don't know if that's the website. We saw someone. Saw somebody, friend of the show, shopping for Chinese peptides offline. The website was just called Biotech Peptides.

2:23:52Doesn't seem like the most legit company. It seems extremely legit. I love biotech. It's the most important technology. It's like reliablemedicine.com. No, it's called medicinemedicine.com. Yeah, basically. we haven't talked about this first brands bankruptcy I don't know anything about this I'll read through a piece in the journal from yesterday auto parts supplier first brands crashed into bankruptcy last month now banks are sifting through their exposure to the company and its chain of customers and suppliers Jeffries on Wednesday said funds run by an asset management unit Point Bonita Capital are owed $715 million from companies that bought First Brands parts.

2:24:39UBS in bankruptcy court filings said funds it operates, including through its O 'Connor hedge fund unit, which the bank agreed to sell earlier this year, have around$500 million of exposure. The exposure is also mainly to customers that owe First Brands money. A person familiar with the matter said, the twin disclosures from the banks illustrate the tangible financing that flowed through First Brands, which collapsed in late September, owing more than$10 billion in a messy and sudden unraveling. Newly appointed directors have said they're probing irregularities in its financing arrangements. Jeffries said it doesn't have any information yet on what the probe has found, and it would also act to enforce investors' rights.

2:25:16UBS also said in a statement that it was working to protect investors' interests. Of course they are. Neither firm indicated any significant direct exposure. Analysts at Morgan Stanley calculated Jeffries' possible total exposure as an investor in the funds at$44 million. so the timeline is getting into a junk bond. Analyst on X says, Jeffries is in the crosshair as the main investment bank behind First Brands with a history of bringing risky capital markets deals. New facts are coming to light. The unraveling of First Brands comes just one month after Jeffries was pitching investors on a$6 billion debt refinancing deal.

2:25:55Investor concerns caused the refinancing efforts to fail as Jefferies struggled to get diligence information from First Brands. Chapter 11 bankruptcy filing followed weeks later. New information about Jefferies' level of involvement. First Brands set up a so-called side letter with Jefferies for an additional trade financing facility, which enabled First Brands to raise funds at an interest rate above the limits permitted by its loan documents. The workaround, First Brands paid Jefferies a supplemental incentive fee instead. This new disclosure caught investors by surprise, increasing financing costs and breaching rules on existing loans, which Jeffries helped arrange and syndicate.

2:26:36On their website, they have a list of brands. I don't know that many of them, but I do see the Michelin logo. They don't actually own Michelin. I was thinking that they might also own the Michelin guide, but they do license Michelin in tires, I suppose, or wipers, actually, Michelin wipers. They sell. They don't sell the tires. Interesting. I was not familiar with this company before we pulled it up. Anyways, still sort of unraveling. There's a bunch of different groups. There's other reporting that says$2.3 billion has simply vanished from First Brand's accounts. So, again, people. there's a there's a an email here from someone at auric that says wheel team thank you for your help yesterday and in respect of the orders and our clients comments we would like to set up a call with you today to ask two questions which our clients needs for informational purposes purposes first do we know whether fbg actually received 1.9 billion no matter what happened to it?

2:27:43We got almost$2 billion just missing. Second, would you tell us how much is in the aggregated accounts in respect of the factored receivables as of today? And they answer, one, we don't know, two, zero dollars in their accounts. So big multi-billion dollar mystery here. And And yeah, we'll see how this turns out. Well, Clark in the chat is talking about the example I gave about vapes. He highlights that something similar happened in THC with Delta 8 and THCA. I completely agree. It was the same thing. For a decade or longer, people were saying, oh, there's going to be a federal bill on where does America stand on THC.

2:28:36And it just kept being kicked to the states and legal and medicine in some areas and then recreational and others. And in that time, basically, you go to one of these trade shows and every gas station owner is there buying like basically whatever they want. And it's just like a complete failure to just have clarity around what the market structure should be. So lots of work to be done on the regulatory side, unfortunately. Well, without further ado, I think we have our first guest in the mainstream waiting room. Let's bring them in. Let's bring in Preston. Here we go. Welcome to the show. Preston from Relace.

2:29:15Hey, guys. It's great to be on. Great to have you. Give us a quick introduction on yourself and the company, and then we'll get into the news. Yeah, of course. My name is Preston. I've been working on Relace with my co-founder, Eitan Borgnia, and we've recently raised$23 million. There we go. So when John gets back, we'll hit the gong. But how did you get into building Relace? When did you start it? What were you doing before? Yeah, that's a great question. Before this, we were both PhD students. So I was doing my PhD in physics. I'm a co-founder, a time student's PhD in machine learning. Same college?

2:29:55No, I was at Harvard and he was at UChicago. But we went to Caltech as undergraduates together. Very cool. Very cool. So straight from PhD program into Relace, did you drop out or did you guys actually get your degrees? Yeah, we dropped out. There you go. There you go. Chachi Padilla just come out and we were like, this is the time to do it. Amazing. What, obviously, hyper-competitive category, tons of activity. What's been your guys' initial edge and focus? Yeah, so we've been really focused on approaching the developer experience, but from the experience of the agent itself. And so we don't consider ourselves competitors with cursor or cognition or anything like that.

2:30:46We're building the infrastructure that someone would want to use when they're building AI agents. and so what we've focused on to start are all these auxiliary models that make these agents more reliable. As an example, if I'm a human, I love my autocomplete. I love my tab autocomplete. If I'm an agent, I have a different set of models that are really useful to me. So this is like what we've started out with. This is like our fast supply model, which applies diffs and our embedding and re-rank models, which is able to find the relevant code in your code base in less than a second. And this is used by 40 companies, including like Lovable and Figma, et cetera.

2:31:26How do you think OpenAI's agent builder product is going to play out? It seemed like a cool, like new capability, but I'm not sure. I haven't had the pull like the other products that they've released in the last couple of weeks to just like jump in, try it, play around. What's been your read on agent builder from OpenAI? Yeah, I've heard good things. I mean, it looks very beautiful. It's definitely a different category than what we've been building in. But I think the launch went really well. What does success look like over the next 12 months? Yeah, for us, we really want to now, with this raise, optimize the infrastructure along with the models itself.

2:32:16I think, as an example, like Cursor has done such a great job of optimizing the product with the models. And so all the models that you have within your IDE, the apply model and the tab autocomplete, make this product work very seamlessly. We want to do the same thing for the infrastructure. And so we recently also released Relayist Repos along with our Series A announcement. It means that you can now have retrieval embedded directly into your source control. And we're really trying to reimagine, hey, what does source control look like when you have access to models to make it better and when you're exclusively focused on building it for the agent?

2:32:56What's the biggest company that you've pitched or talked to in customer development? I'd love to know the mood and the receptivity. We've seen some news about the Fortune 500 trying a bunch of AI stuff and then kind of pulling back from it. Have you talked to any big companies? And can you share any context on how bigger companies are thinking about adopting AI? Yeah, and have you even been focused there? Are you focused on scale-ups like Lovable and sort of founder mode companies like Figma? Yeah. I mean, we're talking to all of the big Vibe coding companies. Lovable, Thigma are some of the examples.

2:33:36I think the way we think about this is that these LLMs are going to continue to increase in capability. We're seeing kind of what dominoes fall in the first, say, year, year and a half of models that are reasonably good at coding. And so already if I need a marketing website or a simple presentation, I can actually like vibe code that on like Replit or Lovable. I think over the next year, two years, three years, as we continue to see the capabilities increase, more and more things that are now traditionally software will be able to express in code. And this is like what we're very interested in.

2:34:24And this is why we're saying that we're building the rails for software on demand. It's because things we've seen people build like PowerPoints and decks and TypeScript and React instead of using PowerPoint anymore. We've seen people use like consumer TikTok filters that are entirely expressed in code. This is one of our customers' gizmo. And we think that a lot of what we might not consider software today will end up being software. and the thing that runs and creates it is going to be a coding agent. Well, congratulations on the round. Hit that gong.

2:35:00Awesome. Congratulations on the raise. I'm sure we'll have you back on again soon for the B and the way things are going currently, and congrats on all the initial traction. Have a great rest of your day. We'll talk to you soon. Cheers. We have a review from Trip Carolyn that we've got to read through. He says, and of course, as a reminder, if you leave us five stars on Apple Podcasts or Spotify, put an ad for your business in that review. We'll read it on the show. Gulf Streams, wrist hitters, impeccable ad reads, the daily coverage of technology news. What more could one ask for? Between John and Jordy, they have the height and the hair to revolutionize technology and private capital commentary, putting the TB in capital journalism.

2:35:45journalism. I have been an avid fan from the inception and will be for life. Cannot recommend more to anyone interested in business technology or loud opulence. Wow, he really is a student. I'll take over now. He says, TBPN is 100 % electric, just like Flux Marine's 100 % electric outboards. Great transition. That's a true fan. Escape the noise, fumes, and maintenance of traditional gas engines with a Flux outboard. Now offering our high-performance 115 horsepower hour outboards as well as complete boat packages with industry's leading whole partners we got only outboard suitable for recording podcasts at full throttle visit fluxmarine.com and tell them the technology brothers sent you thank you we love you trip uh trip truly has been uh a day one show fan of uh of of uh technology brothers since the very beginning so yeah thank you for riding with this and excited to get the update on Flux.

2:36:41You guys have been cooking over the last nine months or so. And without further ado, our next guest is in the Restream Waiting Room. We'll bring Scott in to the TVP on Altradome. How are you doing, Scott? Good to see you. Fantastic. Thanks for having me. Give us an introduction on yourself and the company, and then we'll get into the news. Sure. I'm Scott Stevenson, co-founder and CEO here at Spellbook. We launched one of the first gen AI products for lawyers back in the summer of 2022, a little before Chachabut were an AI contract review tool. So we allow 4 ,000 law firm and in-house counsel teams to draft and review contracts with AI.

2:37:25My background is in engineering. My co-founder Daniel is the lawyer of the team. And yeah, we're based in Canada. Give us the news today. We've raised$50 million, led by Keith, your boy. Oh, yeah, Keith. He's a lawyer, right? Yeah, he was a lawyer. He was a former lawyer. He doesn't love to admit it. Yes. What's it like building a point solution these days? I feel like this month that there's just been so many different announcements for different legal AI tools. Some are legacy platforms that are launching AI features and have a bunch of different workflows across the law firm. We had Crosby on yesterday, who's taking a different approach, obviously building the entire firm.

2:38:16And then we talked to other, we had someone on the show last week who was building, you know, AI legal, but just in ambulance chaser agents, basically. Personal injury, which is a very small niche. so when i hear cursor for contracts that makes sense but it's like a different slice in the market how do you settle there why are you confident about that slice yeah sure so i guess there's a couple questions there like one yeah we like being a point solution so we think of our product often as like a toaster like there's a lot of broad ai solutions for enterprise right now for the beginning we're like we're going to be a toaster for contract review so really easy to use really easy to adopt you know we don't even do a sales call most of the time it's just like welcome a board.

2:38:58Let's get using Spellbook in five minutes. I think there's a bunch of advantages and I think there's a lot of opportunity to build these kind of, I would say, point solution products. And there's a lot of very broad AI solutions out there. Why we chose contracts specifically, yeah, I mean, the story of how I started the company is I had a small business before this and then one day I had a legal bill that took half the cash out of our bank account to do some very minor things. I didn't have much cash. I was a new grad. But I thought that it was hugely painful to just do simple transactional work as a small business.

2:39:37And I thought we could do it much more efficiently. But then the reason we settled on contracts is because the contract market size and the contract pain goes way beyond just lawyers. Like$30 trillion is estimated to flow through contracts in the US alone every year. It's happening very inefficiently. There's a lot of risks. There's a lot of litigation from bad contracting, which Keith used to clean up as a litigator. And, you know, when you look at the actual overall size of on-track pain in not just for lawyers, but in procurement and HR and sales, that market size ends up being really, really big.

2:40:16Whereas if you just look at the market size of, you know, X number of lawyers times Y dollars per month, it's actually a much smaller market than when you look at, like, the overall pain of contracting. And, you know, I see like the millions of transactions that happen around the globe every day and contracts often being the bottleneck. And, you know, what is it worth if we could speed up all those transactions, if we could remove that bottleneck? I think so much of commerce over the last 20 years has sped up. Like we have credit card networks, we have, you know, we have Stripe, we have digital payroll.

2:40:49But contracts have largely remained the same except for Microsoft Word. and so they've become a bottleneck. So our mission is to make contracts move at the speed of commerce so we can all transact and work together much more efficiently. How often is a lawyer or a firm signing up for Spellbook when they're bringing basically like, hey, we have this contract that we're working through right now. We're trying to create or provide feedback on we want to use Spellbook this afternoon. What's the time to value between signing up and actually being able to leverage the product. Yeah, I mean, it's close to zero.

2:41:29And that was our design philosophy from day one, is that a lawyer should be able to adopt this almost instantaneously. So, yeah, I mean, they can adopt it in almost five minutes without having to do a big rollout or a ton of training. And we've optimized almost the whole company with that in mind. And do I'm not nowadays engineers at every software company have the ability to just sign up for tools themselves? Are law firms at that point yet where they have some amount of discretionary software budget that they can allocate as they see fit? Or are they still or is it still maybe like, you know, Silicon Valley engineers in the 2000s?

2:42:13It depends on the segment. So 70 % of lawyers work for small and mid-sized law firms. So when you envision lawyers, you usually think of big law, but actually 70 % of lawyers are working at these much smaller firms where they actually have a lot of autonomy. They have access to a credit card. They might be one of the founding partners of the firm. So in that segment, yeah, it's very much you can do self-signup. And then we also serve in-house counsel teams. On in-house counsel teams, we have customers like Nestle and eBay. you know the decision making power is still very concentrated um i think like with with whiz one of the reasons why they cited it was such a great motion is because you know the the buying the decision making power for was like with the the cio or chief security officer and you know they're sort of the technical buyer they're also the user they're also the check signer um similar thing happens uh in these in-house legal teams where there's often a gc who's the user they're the check signer, they're like the domain expert, and they're the champion all in one.

2:43:15So you still get these, I think, really fast buying cycles in these large enterprise in-house legal teams as well. What does growth look like? We were growing very quickly. Yeah. So we launched... Let's give it up for very quick. Quickly. Put it in your Excel model. Yeah, we launched in 2022. We have 4 ,000 customers today. We're tripling this year. We're over tripling this year for sure. You know, hopefully even more. Let's go. Yeah, it's been pretty wild. Yeah, I think, yeah, over the last two and a half years, I think we did something like 7 ,000 % growth, something like that. So yeah, it's been fast.

2:44:01Up into the right. That's amazing. Up into the right. Well, congratulations. Well, I'm sure we'll have you back on again soon for the sea. have fun out there thanks for having the show we'll talk to you soon on scott bye uh if you're looking for cursor for billboards head over to adquick.com advertising made easy and measurable say goodbye to the headaches of advertising only adquick combines technology out of home expertise and data to enable efficient seamless ad buying across the globe we are joined by george from monk next the monk himself welcome to the show george what's happening hey guys how are you Thanks for having me.

2:44:34Great to have you. Kick us off with an introduction. You launched the company. How are you positioning the value proposition, the business? How are you explaining the business these days? Yep. We're at Monk.com. We help businesses save time and make money by solving invoices and collections, which now that I say this in the show kind of sounds like ramp. Yep. What is the typical collection process? Is this like just automated emails? is that an important part of getting paid as a business? Yeah, it's a huge part. It's a huge part. I mean, most of it is email. And up until now, it's all been like these very boring, done emails that get sent and then get ignored.

2:45:16And then like statistically, a big piece of delays is just an error in these emails. It's like a tiny thing that we accept, but it's like a$3 trillion problem, right? Yep. What were the legacy solutions in the category? why start this business now? What was kind of the key unlock? When I saw the launch, it made a lot of sense, but at the same time, you look at a business like this. I think every business owner can go to Monk.com and think, okay, I'm having these problems. This feels like an elegant solution for it, but why was this not a solved problem yet? Yep, it's a good question. And it's a red ocean.

2:45:58There's a ton of players before OLM, and now every person with a Wi-Fi connection is going to rip a competitor. Just diving into your real shark. It takes a real shark to dive into a red ocean. I like the honesty. I appreciate this. This is much better. It's true. We're the only person that thought of this. Capitalism is inevitable. A few things. A, the selling into the office of the CFO changed twice since the Zerb. So A, there was a Zerb, and it was super frothy. And then there was a lot more financial rigor. And then now with PostLM, there's just a lot more appetite to experiment. So I think the GTM motion got easier.

2:46:40And then the owners on the CFO or the business leader, even the founder, the sole founder, get a lot harder. They have to do a lot more with less. And then B, when Sim Altman and Satya compete and they pour billions in CapEx and OpEx and R &D, we at the app layer just get to rip that into advantages. And so like we are applying LLMs and three parts of this workflow and it just works. And like every, every week that they improve, we improve before. Obviously not possible. How'd you get monk.com? That's a great domain. What's the story? Uh, one of the poker game. You wanted it. You're a pro poker player, right?

2:47:17No, I'm kidding. I'm kidding. I bought it. I bought it. Okay. But you made money in poker and then so you paid for it. So you kind of wanted it a poker game. I like that. Kind of wanted it a poker game, kind of shield. BTV gave me some money. No, we used to be called Atlas because I think Greek philosophy is cool. And I thought a science function holds up the world. But then there's like 50 Atlases. Oh, sure, sure. Yeah, Stripe Atlas. There's a bunch of other. What makes poker players so elite at the sport of business? Yeah. I know a buddy of mine still plays like pro-am, pro-poker. It's because business is basically degenerate gambling.

2:47:56Professionalized, just kidding. I mean, don't you think you can sort of like cross-supply from, I mean, you run a business and now you run a very popular issue. You just cross-supply, like a lot of the same principles govern all these domains, right? Yeah, totally. Totally. Is this founder bluffing or will they come on the show? Will they come answer hard questions in the TBP Ultradome? Will they have a poker face on? Expecting values thinking, you know, risk, discipline, et cetera, et cetera, et cetera. Totally, totally. Awesome. Take us through the round real quick. What happened? You mean like the money?

2:48:31Yeah. The money. I want to ring the gong. How much did you raise? From who? The greenbacks. Four led by Sheila Nihar on BTV. Yes. Thank you. Thank you. Thank you. We love Sheila. Is this from his new fund? He came on the show just like a week ago. He's already deploying, ripping checks. Or maybe you were the last one in fund, too. I don't know. I think I'm the last one in fund, too. Last one in fund, too. Okay. Okay. Well, GTM fund and V rail also helped, but yeah, Sheila and Nehar led. Very cool. And tell us about the last company. Super interested. The last company. Streamlabs? Oh, Streamlabs.

2:49:07Yeah. It's a, it's a big business. It's a creator economy business. It's like, we were a big developer for Twitch and YouTube. So like if you're a content creator and you live stream on Twitch, YouTube, or like back then when Facebook was Facebook, we're using our tools. Yeah. So live streaming, patronage, moderation, et cetera, et cetera, et cetera. And we sold because we're a developer. And Twitch was just ripping our roadmap, which is fine. Oh, interesting. Yeah. Yeah. Who'd you sell to? Logitech. Okay. Logitech. There you go. Awesome. Well, yeah, congratulations on the raise. And yeah, it seems like there's quite a lot of demand.

2:49:46So good luck keeping up with it. Awesome. Thanks for the support. We'll talk to you soon. Our next guest is in the roostrym waiting room. But first, let me tell you about Wander, Find Your Happy Place. Book a wander with inspiring views, hotel-graded menus, dreamy beds, top-tier cleaning, and 24-7 concierge service. It's a vacation home, but better. We have Zach from Meanwhile coming in to the TVP and Ultrodrome. Zach, how are you doing? There he is. Welcome. Great. Happy to be here. Great to have you. Please kick us off with an introduction on yourself and the company. We'd love to learn something.

2:50:16Yeah, my name's Zach. I'm a long-time fintech entrepreneur. I've built, meanwhile, meanwhile is the world's first Bitcoin life insurance company, but our intention is to build the world's largest life insurance company. How does Bitcoin life insurance fit together? How's the normal life insurance market work? How is this different? Walk me through just the like one step deeper in the thesis or the product. Yeah. So the normal life insurance market, you know, this is about 3 % of global GDP. You've probably never talked about it once on the show, but it is gigantic. It is not just life insurance.

2:50:53It's about retirement savings. It's about protecting your family when you die. But what also happens to your family when you live longer than you're supposed to? That's deferred annuities. It's insurance bonds. It's all sorts of saving products. It's a huge, huge market that is completely untouched by technology. And the reason it's untouched by technology is because you need to be big for technology to matter. So that is why you have a bunch of companies who have built marketing engines on top of the global life insurance ecosystem. What we've done is we have built a full stack, vertically integrated life insurance company in Bermuda.

2:51:35It just happens to entirely be denominated in Bitcoin. And what that means practically is if you want to save for the long term, you can now do it in Bitcoin. And there are tax advantages and there's estate and tax planning reasons to do that. But fundamentally, what we see is that if I bought a life insurance policy in 2018 when my son was born, I've seen the purchasing power of that dollar policy go down 25, 30 percent. whereas if I had done that in Bitcoin, I've seen the purchasing power go up hundreds of percent. And we want to bring that not just to everyone in America, but if you're a middle-class person in Argentina, you don't save for your retirement because you are going to live longer than the Argentinian peso.

2:52:18So that person should save in Bitcoin. Okay, so you want a million-dollar life insurance plan to run your life 10 Bitcoins by paying in Bitcoin? Because most people think about their life insurance policies is, okay, it's going to be$100 a month. I have a job. I can pay that. If it's in Bitcoin, it goes up, and then all of a sudden I can't make the payments. How does all that flow? Yeah, so the current product that we do direct-to-consumer, but this round is actually about all this other stuff we do, is called Whole Life, so it lasts your whole life. It's very well named. In our case, you might pay, I'm going to use big round numbers, you might pay one Bitcoin a year for 10 years.

2:53:02So you've paid us 10 Bitcoin, and then whenever you die, you get 15 Bitcoin. But yes, you're absolutely right. In a way, you're short Bitcoin. So I suggest do not buy a policy bigger than your existing amount of Bitcoin. Interesting. You mentioned you just raised$82 million, so we should acknowledge that. $82 million. From Bain, Apollo, Northwestern Mutual, that's big, Han Ventures, and Pantera, a couple others. What was the catalyst for this round? What are you working on outside of the direct-to-consumer product that you just mentioned? Yeah, so actually we raised a Series A in February that was$40 million from Framework and Folger and Wences.

2:53:49And what happened after that round is that a ton of the largest life insurance and retirement companies in the world approached us and asked us if we could work with them to bring these products to their markets, if we could help them do indexed annuities, indexed life insurance that had exposure to Bitcoin. And that is how this$82 million round came about, is that Han and Baincap Crypto in particular, they had sort of been hovering around. They reached out to us in the summer. They said, you're one of the most important companies and going to be one of the most important companies in crypto. Can we invest?

2:54:28And I said, for a large enough check, you can do it. I love the honesty. What is it about, give us the Bermuda lore. I've only been there on vacation, but it's been - It sounds un-American to me, but I'll let you defend it. Well, it's long been a sort of financial hub and has some deep history and ties between Wall Street and Bermuda. You're not winning me over. It's not in America. Yeah, so the fundamental history is that Bermuda used to be a tourist destination when you had to get on a boat to go to your vacation spot. And once there was an airplane, you're like, I'm going to go to the Bahamas or I'm going to go to Jamaica.

2:55:09Bermuda is actually way out in the middle of nowhere in the ocean and it's just not as hot as you'd like, right? And it's the Bermuda Triangle out there. Bermuda Triangle. Not a great history on boats. So basically, they had some really enterprising people in the 1950s and 60s who decided to make it an offshore financial hub. And now it is like the insurance capital of the world. So I'm surprised you went there on vacation because there is about as many actuaries, lawyers, and accountants on Bermuda as there are Bermudians. You walk around and it's like Munich Re, Chubb, Swiss Re. It's like huge financial institutions.

2:55:49and I guess the next huge financial institution there is going to be Meanwhile. Congratulations. That is wild. Is the Bermuda Triangle fake news or are you taking a wild route when you go there? Are you avoiding that? I'm gonna call it fake news but they're actually Bermuda has this really crazy history they did a lot of things they like ran a ruthless assault monopoly over the Caribbean. They were like gun runners in the civil war. They had ran a penal colony. They like, why not find it? They have found ways to, um, be innovative for hundreds of years. So if, uh, so if me and you are flying in a Cessna to Bermuda, you're not telling me to fly around the triangle, you say fly right through the, through the triangle.

2:56:36That's right. I've only been there on commercial jets, so I'm not prepared to support the Cessna plan. Sure. But you're Fearless, generally, around the triangle. Generally fearless. That's great to hear. That's what it takes to take on a$3 trillion industry. I like it. I like it. You shouldn't be afraid of the triangle. You shouldn't be afraid of the financial markets. How are you innovating on the direct, like the actual experience? I got life insurance through my neighbor who happened to be an agent. It was kind of this weird, you know, it's not a, it didn't feel like the most seamless digitally native process.

2:57:11So there's also, I think, a lot that you can do on the actual... Yeah, fundamentally, look, we want to do with 1 ,000 people what Allianz does with 100 ,000 people. We didn't mention it, but Sam Altman actually led our seed round. And our pitch to Sam was this is a once-in-a-century opportunity that we can use digital money in the form of Bitcoin and stablecoins to reach billions of people with these savings products, and then we can use AI and automation to serve them profitably. So we have built everything from the ground up. We are a fully vertically integrated life insurer. So yeah, the experience of filling out the application is quite seamless.

2:57:51People can get underwritten and we can close policies in a single day. We close our books. This sounds so boring, but we close our books in two and a half hours at the corner end. Most insurance companies, we're talking 45 days. But we do all of that. We have a chief risk officer, we have an enterprise risk manager, we have all this boring stuff. But because we built it feature by feature, step by step, system by system, we are building a durable advantage in the life insurance space. fantastic well congratulations on the progress hope to see you out in Bermuda one day have a great rest of your day happy to host a show there sometime if you'd like gentlemen I would love that thanks for joining have a great rest of your day speaking of life insurance Dylan Patel over at Semi Analysis says life insurance in Taiwan is delaying NVIDIA's expansion in Taiwan remember Sam Altman was talking to Ben Thompson just earlier this week about wanting TSMC to ramp up.

2:58:55He obviously wants more chip production. And so Semi-Analysis tells the story. They say, at Computex this May, NVIDIA announced plans to establish its Constellation headquarters in Taiwan. However, the project now faces uncertainty. The proposed site for the Taiwan headquarters was the T17 and T18 plots in the Baituxilin Technology Park. NVIDIA had signed a memorandum of understanding, an MOU, with this life insurance company, a Taiwanese company with total assets exceeding$100 billion USD, but the MOU expired on September 30th and is no longer valid. The key obstacles are as follows. Shin Kong Life completed the registration of building rights for the T-17 and T-18 plots in February of 2022, securing a 50-year leasehold.

2:59:49Shin Kong expressed willingness to directly transfer the rights of T17 and T18 to NVIDIA. However, the Taipei city government opposed this, arguing that Xinh Kong must first construct the buildings and obtain an occupancy permit before transferring the rights. The government is concerned that if Xinh Kong gains additional benefits from NVIDIA through a direct transfer, that may be seen as favoritism. So furthermore, NVIDIA requested that the road between the two plots be incorporated into the site. They want the road as well, not just the buildings. And that would require changes to the urban planning regulations and it would increase the value of the land rights and again the concern of favoritism so shinkong proposed constructing the headquarters for nvidia and transferring it afterwards but nvidia said no nvidia wants full control over the design and construction of its headquarters you got to get a nice corner office for the big man who knows what shinkong is going to do if you let them build a closet for leather jackets yes a massive walking closet that makes entirely the same jackets lots of leather jackets and the jackets come flying out of every every possible closet the taipei city government suggested terminating the existing contract and signing a new one directly with nvidia shinkong opposed this citing its 50-year lease hold and expected rental income the city's proposal did not account for shinkong's anticipated future revenue regarding rumors that nvidia may abandon its taiwan hq plans and move to india uh semi analysis has the following view.

3:01:14Taiwan's supply chain remains more concentrated and efficient than any of the other countries. And with the MOU expired, NVIDIA is free to seek alternative locations or quit the site entirely. Therefore, we believe NVIDIA will not give up on establishing its headquarters in Taiwan. That's where they want to be. The city government is reportedly leaping toward persuading NVIDIA to give up sites T17 and T18 and instead accept a different site, T12 along with surrounding private land. Discussions with NVIDIA are expected soon. They believe the likelihood of NVIDIA accepting the T12 proposal is low as the site is similar and surrounding private land involves many owners, which would drive up costs.

3:01:52So NVIDIA is duking it out with life insurance provider in Taiwan. In other news, REI announced it's shuttering its Soho flagship. It's also closing flagships in Paramus, New Jersey, and Boston. These shores will shut down in late 2026. Also, apparently, OpenAI is opening an office in the same building as the Soho flagship. I think Michael Miraflor was saying that on X, something about that being very symbolic. I have a lot of happy memories of being and visiting REIs as a kid. It was a wonderful, wonderful place. So hopefully this isn't a sign that it's so over, and hopefully someday soon we can be so back.

3:02:42I did want to close out. There's the figure three robot. Okay, let's take a look. We have a video of that. Thought we should pull up the video. I'd also like to pull up the video of Mark Zuckerberg talking about the barbecue, you because isn't isn't it the anniversary of that video at some point let's watch those two videos and then we'll get out of here uh but first while you pull it up yeah uh apparently so mark cuban went in sora yes and said uh free free for all with his likeness i went free for as well he did something that was even smarter than you which you prompted it to always depict you as a bodybuilder uh cuban uh added a tag somehow that says brought to you by cost plus drugs i was going to do that with the ramp logo on my forehead he did yeah he did it and it works yeah it works very funny dylan posted a couple different screenshots very clever okay let's watch this video thank you for pulling this up uh we were just talking about humanoid robots with martin casado from a16z nobody does trying to do it nobody does videos like figure gotta give him credit fantastic launch videos where this is the third so this is version 3 but they're still in alpha and they haven't they haven't shipped these or they they installed some into a demo plant with bmw right but so what's notable here is they're not that i i scanned the video i don't think there's any they're not just depicting it in a factory setting at all okay it seems like this is a pivot away from going more consumer commercial which was one of my questions for uh martin what do you think tyler right at the At the end of the video, it's in a factory.

3:04:23Oh, it's in a factory at the end? And the other thing here that's notable, they followed 1X Robots. That's right. Who was doing the fabric covering. Fabric looks way better, I think, than the hard metal. It just looks, yeah, it looks cool. I can see why. Oh, there's a house. Okay. I can see why the retail army just loves figure. It's not a retail army. We're seeing the emergence of a new army. The private market's retail army. Yeah, we need a new word for it. It's not retail necessarily. if it's private, but people have been pumping. It's a special army. It's a special purpose vehicle army. People have been asking questions about how much pumping is going on on the timeline and whether it's too much, but I mean, it's a great video.

3:05:05It's a$39 billion company, John. Is it still? Wow. Still, yes. That's remarkable. Hey, look, that dog. That was an opportunity to make that a robotic dog. It is interesting that they're not, They're only focused, you know, straight shot to humanoid robot. No Boston Dynamics project. No four-legged friend first. A lot of the other robotics companies have kind of taken a more incremental progress. Do you think they trained on Biden's? It does seem like the walk of an elderly man. But it's a start. It's not. But it's not. I mean, the Boston Dynamics demos are always way more agile. And same with the Unitary demos.

3:05:52So the question here for the in-home use case is going to come down to, one, how effective is it? If we both have kids, kids tend to destroy the living room. You have to go to bed. You've got to clean it up and stuff. A humanoid should be able to clean it up, you'd think. The question is how effective and what's the cost? Are people going to buy? can they come in anywhere near Unitree at 20K for a comparable robot? Yeah. Seems unlikely. The big debate here is just how much does this stuff generalize? Because, I mean, we've seen in all sorts of other AI applications, like it's simple to RL on one certain problem and get really good at it, but then does that learning generalize?

3:06:37And so you can have a model that's fantastic at IMO level math, but can't complete Arc AGI, which is so simple that a 10-year-old can do it. It's doing dishes. And so, you know, if it's trained to be really good at this one dish with this one faucet system, can it generalize to your faucet system? Or when will that generalization happen? It certainly happened in LLMs, so you would hope that it happens in humanoid robotics very quickly. It's an interesting time. Brett shared that none of this is tele-operated. Okay. Yeah, none of this is tele-operated. Which I still think even a tele-operated humanoid would be very cool.

3:07:19Yeah. I mean, you're still paying by the hour more than just power. There is something that's very important economically about getting away from tele-operation over the long term. But I don't see a problem with tele-operation as a data flywheel, as a way to actually get these into people's homes. You're tele-operating them. you're collecting more feedback and data and you can train that and roll that into reinforcement learning. I do, yeah, the question is like, did he comment on whether or not this is scripted? Because it could be recorded motion, right? Because that would not be teleoperated.

3:07:55It would be running on the robot locally, but it could have been, you know, teleoperated at one point and then memorized and then played back effectively? There's a big question about how long has this robot spent in this particular environment? Yeah, for figure to develop legitimacy within our bubble, people are going to need to be able to do some type of live demo setting where they can be in the house with it and they can task it with doing things that aren't necessarily pre-planned, right? Yeah, there's something about like with the other AI companies in the LLM world, like the models go out, the APIs are available, the benchmarks are independent.

3:08:41Because if you can just walk into a room with a figure robot, dump a bunch of laundry on the bed and it just folds it and puts it away, that is going to be mind-blowing, but I don't know how... I mean, to be fair, Noam Brown, I think at OpenAI was saying like, okay, this is a nice example of the robot folding laundry at a table. What happens if you raise the height of the table six inches? And Brett Odcock raised the height of the table six inches and posted another video and said it still works. It still works. So he is duking it out on the timeline. There it is, more commercial setting, delivery robot.

3:09:18They had it at the opening of a hotel. I wonder what makes it so much slower than the Boston Dynamics. example. Like it feels in some ways more advanced than what Boston Dynamics does. Certainly the demos here are more generalizable. Like I haven't seen a video of a Boston Dynamics robot doing any of these like useful tasks. It's always just like dancing and doing parkour, which doesn't seem very commercially viable. But the Boston Dynamics, like the actual motion just feels so much smoother and so much more fluid. And same thing with the Unitary robots. The Unitary robots, you kick them over and it does a 360 to jump back up.

3:10:02That doesn't feel very economically useful like folding laundry or delivering packages. But it's certainly impressive. MKBHD responded and said, I have so many questions. I wonder, how long do you think until MKBHD can do a review? He did a huge review of the Boston Dynamics dog. He got one. I think he might have even paid for it himself. and like been able to really play around with it and really put it through its paces. And I love that. I would love for him to do a proper review on a figure robot. That would be very cool. There was something else that was top of mind on the MKBHD thing. Oh yeah, did you see that?

3:10:42Somebody posted their product and MKBHD responded with a meme that said like interesting and it was like a funny just gif or image. and the company took the text out of the meme and put it in quotes on their website and attributed it to MKBHD. It was like, he said, it looks interesting. And it was like a meme that he sent. And then MKBHD was like, this is ridiculous. Like, I didn't give you a quote. Like, you can't just use me on your website. And the founder of the company was like, oh, well, we also emailed with your team and they said, like, it looks interesting. Like, we'd love to, you know, like, hear more about it.

3:11:22and then they took that as an example. It's like, that's not quite the same as like... We'd love to hear more about it. Yeah, it's like they were just being nice. The chat right now, AV says it's slow for dramatic effect. Yeah, it's actually moving at five times the speed, but they filmed it with a slow motion camera. They filmed it with a slow motion camera to make it look more sci-fi and cinematic for sure. I love that. But anyway, it's a long road. No more information on when it'll be available for sale. But I'm sure somebody at the right point will pit the, I mean, the first thing somebody's going to do is pit the Unitree G1 against a figure and just make them compete in a variety of contests.

3:12:06Yeah. Looking forward to that. I mean, Christopher Mims has a whole deep dive on America's manufacturing resurgence and the robots that are doing that. More focused on co-bots, which we've talked to some companies around, less on the humanoid side. but it's a fascinating deep dive. If you want to go check that out, maybe we'll cover it on the show at some point. That would be great. Is there anything else you want to react to? That's it. Okay. That's it. We can react to this iconic video. Let's head out. Let's head out. On this day. Friend of the show, Mark Zuckerberg, TV PM. Hey, everyone. We are live from my backyard where I am smoking.

3:12:39Imagine Andy Jassy doing this on Twitch. It would be so much better than another conference call for earnings. It's smoking. Is this the edit where they just show the smoking meats? Yeah. Someone asked me, do I smoke meat? Smoking meat. This is the least charitable. You can edit. Yeah, I saw it. You can edit any video. You can edit us to look like idiots too. If you took our 15 hours a week of content and dialed it down to all the times we say something silly. He's smoking. He does like coke. And he went so much farther. He has his own cattle ranch. He vertically integrated his meat smoking operation.

3:13:18So cheers to Mark Zuckerberg. Thank you for tuning in today. We hope you have a fantastic afternoon and evening wherever you are in the world. And we will see you tomorrow. Here in my garage. Sam Altman, live on TBPN, noon Pacific. Tune in. We'll see you then. See you tomorrow. Goodbye.