The Experts Weigh in on Intel, Big Family: The Ultimate Status Symbol | Wade Foster, Dan Wang, Nicolas Sharp, Shan Aggarwal, Vikas Enti, Bryan Pellegrino, Logan Kilpatrick, Rylan Hamilton

26:46And so the way she does that now is when she releases an album, she goes on a tour that becomes so big that, again, there is scarcity because there are only so many tickets at SoFi Stadium and you have to be there because it's a moment. One of the best monetized recording artists of all time. Yes, yes. And so she gets the long tail eventually. And eventually her music is essentially free and played in grocery stores. So I'm going to read you something from their Instagram caption announcing the engagement. Yes. It's not just a marriage. Dash. M-dash. No way. It's the perfect love story. No way.

27:24No way. Back on X is calling this out, claiming that Taylor and Travis are one-shotted. No way. What'd you got, Tyler? What was that? No. I was going to say something about her monetization. Kyla Scanlon has this good post. She said Taylor Swift's engagement is actually very relevant to monetary policy because she's the most powerful stimulus tool we have. Oh, really? How so? Well, I mean, her tours are like massive. Like you can see the rise in consumer spending. Yeah. Oh, interesting. Especially like smaller countries. It's like very. Yes. Yes. But but how does her being married like she could just go on tour and be single, right?

28:02Is she going to sell more tickets because she's married now? Does it allow her to tour more? You could also see it in, a lot of people are talking about birth rates, right? So maybe a new boom, baby boom. Catherine Boyle just posted a picture and just said, with the captioned, their pictures just said American dynamism. I do think this will be good for the birth rate. It's also going to be for anybody that's in a longer-term relationship in their late 20s, early 30s. Pressure is now on. These photos are going to be getting sent around. They're setting a new bar, right? Quite the floral display by Travis.

28:40Got to give him props there. How many kids do you think she'll wind up having? She's 35. She's 35. She's got to get going. Oh, yeah, 1989. And hopefully a ton, right? I mean, I think we'd like to see. It is the last luxury status symbol, apparently. I'd like to see at least 10. Yeah. It'd be great. well

29:07if she wants to manage her family's plans she should get on linear linear is a purpose built tool for planning and building products it's a family planning tool you can make it work for it for modern software development, streamline issues, projects and product roadmaps, start building I don't know she could probably use linear she's probably developing software and websites and stuff Anyway, what we should talk about is this post from Robert Sterling, the only form of wealth that's ever mattered, screenshotting the Financial Times, the ultimate status symbol, a big family. So the Financial Times writes, a flex of Uber wealth in 2025 is multiple children, all the while preserving pre-parenthood lifestyle, interests, and physique.

29:51What is the ultimate luxury status symbol, Jordy Hayes? Once upon a time, it may have been flaunting a rare handbag, sports car, or flashy watch. But with the soaring costs of living and shrinking household sizes and depreciation of Ticons, sports cars are now— They're giving Ticons away. They're giving Ticons away. The only thing that you can flex your wealth with is having kids, specifically lots and lots of them, says the Financial Times. For most working parents, quote, and particularly those living in cities, even to have one child comfortably is a major economic calculation that requires considerable financial stability, says Eliza Philby, author of Inheritocracy.

30:32It's time to talk about the bank of mom and dad. while that was also true historically when having an enormous brood was bolstered by a need for labor religious imperatives and preserving famous family legacies in the face of high infant mortality children today aren't a source of positive household cash flow they are a drain on it but i think we gotta turn this around we gotta put these kids to work we gotta put them in the minds in the in the minecraft or get them on if you if you got a if you got once your kid hits three you should set them up with a wop account and i was like hey pay for college yes do it don't don't make mistakes actually make some mistakes but figure it out hey i mean uh in the stevens household you have to pay the bills uh making making tech decks start making start making courses for other three-year-olds how to be a three-year-old yeah don't of course don't expect the the anduril chairs to pay for your college buddy there's only so many of those to get around we're not selling you're still long you got to pay for those with with skateboards the reality is that middle class parents do not support kids to 18 philby says that for families who can afford to it's often closer to 30 with costly child care a lack of family support systems and a global trend of women.

31:52Parents hiring child care for their 25-year-olds? I guess, yeah. In some cases. Conceiving children later, there is a feeling that one child is increasingly the norm for dual-income households, Philby says. Two is a bit of a push, and three or more is not just an anomaly. You need to be super rich. Official birth rate figures in many countries have hit all-time lows. In the UK, the Office for National Statistics says that the fertility rate is 1.44 children per woman, while in the United States that figure is 1.6. In some Asian nations such as Japan and South Korea, the rates are 1.2 and 0.75 respectively.

32:29I was talking to somebody who is very bullish on South Korean tech because of AI and a lot of the companies that they sell. SK Heineck. Yeah, and there's a whole ton. I think they have their own S &P 500 of really, really great companies. They have their own TBPN. They do, they do. But I had no idea that the birth rate was so low. That's crazy. 0.75 is really, really low. So they got to turn that around. But while one and done may be booming as a cornerstone of modern middle class parenting culture, so too has a fetishization of the opposite end of the spectrum. In other words, ogling extremely wealthy figures in the public eye whose picture perfect families never seem to stop growing.

33:12Behold the fixation of frazzled millennial working women and others with the Trad Wife Movement, a social media phenomenon that glorifies homesteading, milkmaids, and taking care of your husband and offspring. Its brightest star, Hannah Nealman, Nealman? I don't know. A ballerina farm has attracted a 10 million strong following on Instagram, thanks in no small part to her eight immaculately disheveled blonde children with her husband, Daniel Nealman, the son of JetBlue founder and airline mogul, David Nealman. Hopefully they sold before JetBlue became a penny stock. Oh, my God. Yeah, get out. Get out.

33:49Get out. In the celebrity orbit and beyond the trials and tribulations of the ever-expanding Kardashian clan, Alec and Hilaria Baldwin sent the internet into a meltdown earlier this year with a TLC reality television series called The Baldwins. I had no idea this was a thing. That's insane. Which documents life between their Manhattan apartment and East Hampton. Try to pull up this picture. East Hampton Mansion with their seven children and eight pets. I had no idea Alec Baldwin had seven children and eight pets. I wonder what pets he has. Huge families. You know that Haley Bieber is Alec Baldwin's daughter?

34:25No. Is that right? Am I right on this, guys? Wait, what? That's true? No way. Okay. Learn something new every day. Pretty sure that's true. Yeah. Might need the truth, don't it? No, no, no. Uncle and niece. Uncle and niece. Okay, so they are related. Okay, okay, okay. Yeah. Got it. Interesting. But here's the picture of Alec Baldwin and his family still just pumping him out. Looks great. Looks great. Take those kids to a wander. Find your happy place. Book a wander with inspiring views. Hotel-grade amenities, dreamy beds, top-tier cleaning, and 24-7 concierge services. It's a vacation home, but better, folks.

35:03Anyway, huge families are increasingly visible among the 1%, especially in some of the world's wealthiest and most competitive enclaves, such as New York city in her widely read, if controversial book, primates of park Avenue Wednesday, Martin member memberably observes that massive families. They were everywhere on the upper East side, a neighborhood with some of the most expensive real estate, schools, nannies and hobbies in America. That was about to dive into that. Four is the new three previously conversation stopping, but now nothing unusual. She writes five is no longer crazy or religious.

35:34It just means you are rich. and six is apparently the new townhouse or Gulfstream. According to a study by Forbes of more than 700 American billionaires, at least 22 have seven or more children, often from multiple marriages and adoption, including perhaps most famously Elon Musk, regularly named the world's richest man. Musk has become a poster boy for billionaire procreation in a vocal proponent of pronatalism, which he sees as vital to humanity's survival and declining birth rates. fascination with the uber elite with how they live their lives and run their households is not new a raft of high-end fashion and consumer lifestyle brands have turned to notions of parenting with almost unlimited resources and incorporated them into their product offerings and marketing efforts artipop offers 800 velvet and cashmere baby carriers that have been heralded as the birkin of mom gear heralded but yes heralded heraldo rivera heraldo uh rebranding wearing a baby as an ultimate luxury accessory.

36:34It is the ultimate luxury accessory. Much like having Graphite in your software development team. 100%. It's the Birkin of code review. 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. If they don't mention Pavel Durov in here, I'm going to stop reading. When I'm walking around with my one-year-old and my my three-year-old people often assume that I'm the uncle and, and are surprised because you're young. Yeah. Oh, just, it's, it's still. Sure. Seems wild. Even though if you go anywhere slightly off the coast, it starts to become commonplace.

37:17Yeah. Super normal. Yeah. It is interesting that if you, if you actually look at, if you look at birth rate trends by income bracket for a long time, the higher, the, the, the, the, like the lower income cohorts would have more kids. Yeah. And so, uh, technically I think from an, from an economics perspective, having children was an inferior good. It was something you did more. It was like, uh, the equivalent of Costco. It's like the opposite of what they're saying here. Yeah. But really what we're identifying is like, is like, there's a, there's a barbell effect where at the high end you get a lot more kids and at the low end you get a lot of kids, but in the middle there aren't that many kids.

37:58Yeah. And this is, I'm going to make a movie reference that you haven't seen, so I won't even try. Hit me. Hit me. There's a 1 % chance. Idiocracy. Nope. We should pull up the intro to Idiocracy because it is hilarious. John Ross in the chat says Nick Cannon has 12 kids. That is crazy. That is a big number. Wow. Very good. I think, I mean, it's also, I'm guessing that's across a number of different relationships. It is. It is. Yeah. In a canon. Okay. So, Mini-Me designer collections and$600 high tops allow an army of children to dress like their parents as well as one another. And white labels such as Patek Philippe and Dolce and Gabbana have long placed glamorized visions of family at the heart of their advertising campaigns.

38:46Of course, you never own a Patek. You merely hold on to it for the next generation. Or Patek, as they say. Newer efforts from Patek Veneta and Burberry. Emonition. Oh, yes. You want to play this? I think this is the intro to Idiocracy. Let's play this.

39:27intelligent. But as time went on, things seemed to be heading in the opposite direction, a dumbing down. How did this happen? Evolution does not necessarily reward intelligence. With no natural predators to thin the herd, it began to simply reward those who reproduced the most and left the intelligent to become an endangered species. Having kids is such an important decision. We're just waiting for the right time. It's not something you want to rush into, obviously. No way. Oh, shit, I'm pregnant again! Shit! I've got too many damn kids! I thought you was on the pill or some shit! Hell no! Must have been a thing with Brittany.

40:11Brittany? Don't give me a question! There's no way we could have a child now. Not with the market the way it is, no. Oh, God, no. That just wouldn't make any sense. Come on over here, bitches! We don't care about you! Yeah, well, there must be something he likes over here. You mean nothing to me, baby? Oh, shit. It wasn't me! It wasn't me! Well, we finally decided to have children, and I'm not pointing fingers, but it's not going well. And this is helping. I'm just saying that before I have in vitro, maybe you should be willing to... It's always me, right? Well, it's not my sperm count. I'm gonna fuck all of y 'all!

40:54Okay, okay, we've got to cut this off. You get the idea. I get the idea. Decision fatigue from the 138 IQ. Simply too smart to have kids. Too smart to have kids. Anyway, let's continue. Yeah, this is an interesting line. So there's been other brands, but Tega Veneta has had a campaign with ASAP Rocky, which has been cool. I thought this campaign was brilliant. um the uh there's a line here healy thinks these campaigns come at a time when the luxury industry is struggling to maintain its cultural relevance reflected in a cooling appetite for products as social signifiers and a fall in global sales focusing on family reflects an attempt to move into the realm of behaviors and values and away from physical things you can't dupe having a family healy says it's uh you know you get a dupe of a of a birkin or something like that Oh, that's what Dupe is referencing.

41:51Yeah, or something from The Row or whatever. Every luxury brand gets knocked off immediately. Healy says it's a very long-term commitment, not to mention one of the most significant and transformational that you can make in your life. But it's not just having kids. It is venerated by brands and vaunted as a status symbol. It's about how you are having kids and to what extent you can maintain your pre-parenthood lifestyle. For many celebrities, this means an army of child care support. Kim Kardashian is said to have 10 nannies on a 24-7 rotation for four children, while the Baldwin family have two on staff.

42:25That feels low for how many kids I saw in that picture, to be honest. Child care positions in the U.S. of this nature can easily command annual salaries of more than$200 ,000, though his more standard wage would be closer to$85 ,000. Still paid support of any kind remains an unobtainable luxury for the majority of families. yeah for for me I noticed this I noticed the jump in you know how I was living at 25 pre-kids realizing as soon as you add kids everything that you do whether it's where you live and how you travel becomes travel instantly becomes you know if you're staying at nice hotels and now you're traveling and you got to get a hotel for your nanny you got to get a hotel for your kid then a hotel room for your kid, then you add a, uh, then you add a second kid.

43:14And so now you need, in some cases, you need three extra room, one for the nanny, one for each kid, one, one for the couple. Um, and so the actual like cost of, of something as simple as a weekend trip can, can four X pretty quickly. Um, and, uh, it certainly inspires you to grind harder in my experience. Uh, this version of parenthood doesn't mirror the experience of most people. hey uh healy says there's a version where you're still constantly meeting friends returning to your pre-baby body showing up the club or late night event not being tired all the time moving out to the suburbs or wearing clothes with milk splattered over them fashion has always traded on allowing us to project something a little bit better than we are but having a large family also sends a signal of confidence in the future we're also the generation that grew up with suburbs being bad and something to avoid.

44:05And as an adult with kids, I realize the suburbs are incredibly underrated. Yeah. Heard it from Dr. Drew yesterday. He lives next to a golf course. It's amazing. A demographic that may feel more secure than most, those with more money. Musk paints his pro-natalist vision in stark terms and has some highly unorthodox personal family values. But among the traditional conservative right, whose social media sway is growing and who champion Americans having a large household. The question of who should be having more babies is an uncomfortable one. A sizable portion of the extremely wealthy pro-natalists say they want us to use breeding to fashion a better human race, Philby says.

44:46But the current fetishization of being a mother of multiple children can reduce women to a kind of birthing machine, fulfilling her social societal duty and demonize women working incredibly hard and contributing economically toward society who are having fewer or no children, whether by choice or necessity. I think Nadia Asperuhov had a good rebuttal to that, which is that tech can certainly embrace and kind of invalidate this false dichotomy by creating a more welcoming ecosystem and not like noveling when a woman shows up. Yeah, I mean, it's interesting to note that a lot of the technology that we take for granted today makes parenting like a little bit easier, but not dramatically easier, right?

45:30I think people tend to judge the family out to dinner with the kid on the iPad. And if you haven't had kids, you don't realize that many kids would not be able to... It could be so much easier to strap the VR headset right on. Yeah. You'd have to deal with them at home. And just bolt them down to the kid seat. Yeah. And then just fully locked in. No. When I go out to... The high chair. Yeah. Yeah, when I go out to the restaurant, I bring the full racing rig with the steering wheel and the pedals and the VR headset. And I'm just like, lock in, brother. Lock in. You've got to beat this time on the Nürburgring.

46:07You're doing the Nürburgring. You're doing the Nürburgring. And I don't want to hear a peep out of you. No, I'm not wasting time with an iPad kit. No, but I mean, we've tried the iPad here and there. We don't do it anymore. but the kid is just, the kid is just, you know, my three-year-old will just run laps around the restaurant for like two hours straight. And it's cool because he likes to socialize, walk up to other tables and like try to grab a French fry or something. Yeah. Oh yeah. That's a little bit. Yeah. No, I love it. I love with him running around the restaurant just being crazy. It's so funny.

46:44Nothing brings me more joy. Cause I'm like, that's what I want to do. I want to get some, stretch my legs right now. I don't want to be sitting here. My little guy was stood up. We were at we were at Little Beach House in Malibu and he's just stood up. We last time we were there was a Sunday. There was a DJ playing. He just put his chair up next to the DJ stand and just stood on it. Yeah. Yeah. The iPad thing is like a is like an endless source of trolling for me. So the four year old's like, oh, for my next birthday, like, I think I want like a tablet. I'm like, oh, like what? Like you want to like because, you know, some kid I'd told them that they're the most incredible thing in the world, of course.

47:22And I'm like, yeah, let's make one out of cardboard. And so we're like six hours into a crafting project where you're like, what do you want on the screen? Let's draw it on with crayons. We're just six hours into this thing. Oh, you want to have a, he was like, I want one for my birthday. I was like, oh, so you want the theme to be like tablet-based technology? Like we'll bake you a cake that's the shape of an iPad. And he's like, no, I want the actual thing. I'm like, iPad cake coming your way, buddy. Don't worry about it. Don't worry, I got you. Paper mache iPad pinata that you will smash and get candy out of that we will make by hand.

48:03But I think you'll have to wait. You'll have to assemble one on your own with a box of scraps and a cave. That said, following trad wives or billionaire spouses with lots of children on social media isn't necessarily a retreat into the home and history, as with many objects of aspirational focus, sometimes these trends are rooted in a type of parasocial cosplay, exploring styles of life and livelihood that are far removed from or will never be our own, nor do we really want them to be. Much of the internet is consumed as if it's a type of speculative fantasy fiction, Healy says. The ballerina farm family, for example, represents a vision of economic security.

48:42There's just so much diet coke. Clearly defined gender roles and a certain degree of purpose. People might engage with it as a type of entertainment or escape, but many will also look at it and think in the real world in 2025, in no way do I actually want that. Why not? Move out to the farm. I don't know. The more the merrier. The more the merrier. Within reason. Within reason. No, I do think that anecdotally from what I've seen in my social circle outside of work. I do think that kids are 100 % a status symbol. I also think it's incredibly unfortunate that it's gotten to that point. Yeah. I wish there was a Polymarket on the birth rate.

49:28I don't know if there is. We are, of course, sponsored by Polymarket and you can see the Polymarket's on the ticker. We will need to spin up something. I want to see if Southgate There's a play market on the Swifts. Oh, really? I guess it would be Taylor Kelsey. Okay. And who is she marrying? Travis Kelsey. What does he do? He is, I think from what I can remember, he's involved in the pharmaceutical industry. Okay. As a spokesperson? I can't quite remember but I think he also played some sport we'll have to look into it well I'm sure he'll be more popular now that he's marrying Taylor Swift people will probably dig into his backstory and learn about him it'll be interesting, we'll be following it for sure anyway, let's go to this post by Young Macro one of my new favorite posters he's been on an absolute tear he has been on a tear and he highlights the dynamic that he's experiencing right now.

50:33He says the hyper-reality version of the bedridden industrial magnate envying the dashing young socialite archetype is that basically every one of your billionaire idols, Elon, Teal, P. Marka, Sama, et cetera, feels intense jealousy every time they witness an Anon Zoomer surf the cutting edge of the algo purely organically. Ooh, Donald Boat. Shout out Donald Boat. I bet you every other day they see a random zeet that makes them ache to have once experienced being the guy to have posted it. Enjoy every second of it, Anon. You're the nubile maiden bathing at the village creek as the laboring brutes admire from behind the bush.

51:11This is actually sort of tied to what we were just talking about. This is great. So if you're Anon Zoomer out there surfing the cutting edge of the algo, just know that you can parlay that into a$50 million Series A if you do it right. Yeah, that's great. Well, we should move on to this story in the Wall Street Journal. a billionaire, a psychic, and a bad investment, the friendship breakup from hell. Of course, you want to stay away from these bad investments. You want to get on public.com investing for those that take it seriously. Do not take it casually. Take it seriously. They got multi-asset investing.

51:47I think we're allowed to do that as advice, right? It's not financial advice. It's not financial advice. We just say take it seriously and don't be a casual. Anyway, a woman convinced her best friend, an heiress to the Thomson Reuters empire. we've of course had the CTO of Thomson Reuters on the show. Didn't get to talk to him about this. I need to have him back. I'm back on. To invest tens of millions of dollars into a crypto meme coin created by a psychic, it went to zero, of course. And this is in the Wall Street Journal magazine. So over a decade, Taylor Thompson and Ashley Richardson were inseparable.

52:19A fight over money destroyed it all. Oh my God, said Taylor Thompson, clapping her eyes on Ashley Richardson for the first time. You have those fabulous, heroin-chic arms. In 2009, both women were lounging in the backyard of a Malibu home of Beau St. Clair, a film producer and mutual friend, Richardson, wearing a muscle tee over her bikini, basked in the sun. Muscle tee? Muscle tee like a stringer, you know? Okay. Kind of like that. Gold gym enthusiast. I literally don't know the name for the non-bodybuilding version. It's like a tank top, a tank top. A tank top. Yeah, tank top. Richardson, wearing a muscle suit over a bikini bass in the sun, while Thompson sat carefully covered in a flowy outfit and hat.

53:03Her then 10-year-old daughter clutched a hot pink Minnie Birkin. Richardson, a lanky six-foot-tall blonde, was a free spirit who went on to build a career designing social media campaigns for companies like Ford Motors and McDonald's. Thompson was an heiress to Canada's wealthiest family, an eccentric with a self-deprecating sense of humor. She went to dinners and parties with wild hair and drapey, distressed clothes by California designer Rick Owens. Let's give it up for Rick. She was a subversive secret billionaire, says one mutual friend. But a few crucial commonalities, assured silliness, a love for animals, and a deep spirituality drew them together.

53:37Thompson liked that Richardson had spent five years in India working with a spiritual leader named Amma, the so-called hugging saint. Richardson bonded with Thompson's daughter, whom she recalls as precocious and a quirky little being. Taylor is holly, go lightly, adventurous bohemian spirit, says Richardson. She has this big, beautiful heart when she lets her guard down. Both women have ADHD, she adds. Because of that, I think we got each other. Richardson soon became part of a small Los Angeles friend group that Thompson called her inner sanctum. Whenever Thompson would land in L.A., says Richardson, her phone would blow up.

54:12The women, often with Richardson's girlfriend, would order Nobu takeout and pair it with tequila at Thompson's beach house. Richardson made Sunday dinners and frittatas at the heiress's Bel Air mansion. When Thompson had staff there, her assistants chopped onions. When she didn't, Thompson was the first to do the dishes, and Richardson taught Thompson's eager then teenage where— Daughter, how to roast potatoes for New Year's Eve. They planned a menu that included chocolates and truffles, also an onion to chop for caviar. Reminded Thompson in a text message at the time, I can see the problem here already.

54:50They were not doing enough data analysis on Julius. Julius could have changed everything in this equation. Totally. What analysis do you want to run? Chat with your data, get expert level insights in seconds. The story would have ended here if they were running their life. Yeah, if they'd just become data analysts powered by Julius. But instead, we're forced to continue here. Over the next decade, the friends traveled to Italy and London, bought the same Via Life vegan feta and stocked up on Rite Aid hand cream. It's very random. Richardson remembers the... This is so many details. Insanely unhinged article.

55:25Remembers the duo peeing in the bushes at a Rolling Stones concert. It was Thompson's first ever. What is this writing? I can't even read this out loud. We got to get to the part where they start slanging checks. Okay. Let's scroll down. Richardson had grown up in an affluent part of Monterey County, California, where she attended elite prep school and was skiing by age two. There we go. That's crazy. But Thompson's wealth was a different ballpark. We would go somewhere for a few days and she'd be buying houses like other people buy mugs. Though there was a lot of it. Money didn't interfere with the relationship.

55:56The reason we'd be friends because there was no financial connection. Richardson could be defensive about the way people used Thompson at the Freeze Art Fair in London. Richardson recalls attendees descending on the heiress. People know who she is and that she will drop millions, says Richardson. It was nauseating. Then there were one-time friends who Thompson later assumed were reusing her for her money. She had a nickname for them, Taker. Don't be a taker. Now, the relationship between Thompson, 66, and Richardson, 47, has exploded into an epic battle that landed the pair in court. But this time, the billionaire is accusing her one-time best friend of doing the taking.

56:33The story is based on interviews, blah, blah, blah, blah, blah. What went so wrong? Okay. By the time Thompson met Richardson, the heiress had spent decades trying to prove herself to her family. For the Thompson's control of the empire passed down the male line, according to people familiar with the family. Taylor, the middle child and only girl, felt less favored than her two brothers, and her efforts to forge her own path led to occasional periods of estrangement. But much of the tension with her brothers revolved around money. The Thompson family in 2024 was the world's 10th richest. Wow. Let's give it up for the Thompson family.

57:03Let's hit the gong for the Thompson family. There we go.

57:1010th richest family, according to Bloomberg, and Taylor wanted equal access. The family holding company Woodbridge owns a majority stake in media and technology company Thompson Reuters. Her older brother David is the chair of both companies. Thompson pursued a career in acting training at the American Repertory Theater in Cambridge. She performed with Shakespearean theater companies in Massachusetts and Los Angeles and had roles in the TV show Matrix. What? Canadian drama. It was a TV show? Canadian drama Forever Night about a vampire working as a detective. Okay. That's very cool. If you're big into TV, maybe you should go.

57:39No, no, I'm not big into TV. I'm a cinephile. Thompson often suspected people were trying to use her for her money, according to people that know her. She insisted that those around her sign non-disclosure agreements from assistance to an ex-boyfriend of her then 20-year-old daughter. And she could get aggressive if she thought those terms were breached. Thompson even hired an infamous Hollywood private investigator, Anthony Pelicano, to spy on her former nanny, who she suspected of sharing personal information with her daughter's father during a custody battle. I want you to do whatever you can to get information on Pamela, said Thompson to Pelicano on a recorded call played during Pelicano's 2008 wiretapping trial, after which he was convicted.

58:20Just get it. I'm not going to ask questions. Crazy. By the fall of 2015, Thompson and Richardson's friend St. Clair were losing their battle with ovarian cancer. Her closest friend surrounded her in her final months. Richardson and her partner at the time had weekend slumber parties at St. Clair's ranch home. St. Clair's producing partner, Pierce Brosnan, then came by. Wow. Thompson decorated for Christmas with Dr. Seuss original green sketches from her personal collection. Okay, we got to get to the crypto stuff. There's some crazy stuff in here. Let's just keep scrolling, keep scrolling, keep scrolling.

58:53Thompson's focusing on this is all false, blah, blah, blah. With the pandemic, Richardson's work was drying up in a fit of anxiety. She Googled a burning question. Had anyone predicted any of this? She stumbled upon Michelle White Dove, a celebrity psychic. She dropped$25 a month for White Dove's newsletter, then largely ignored it until the spring of 2021 when she noticed the psychic had recommended a new crypto token. Persistence, White Dove wrote. This is another dark horse that's going to come up on everybody and be a big dog. Get it and sit on it, which is what I would tell myself. The value of persistence's token called XPRT quadrupled from less than$3 in April to as high as$13 by the next month.

59:36That summer, Richardson brought up White Dev's predictions to her friend. What is the link to that website of the channeler, Thompson texted. She followed up with Robert Sabella, an astrologer and intuitive she regularly consults. Combining astrology and crypto is so good. It's a match made in heaven. Absolutely. What more do you need? In Intuitive, she regularly consulted for advice, both personal and financial. Once, when Thompson tried a Xanax after a panic attack, she asked Sabella for his thought. He replied, I'm getting that Xanax is not good for you right now. Stop taking it and find another, I mean, fair advice.

1:00:13He also weighed in on her investment. No blazing there. Those answers I promised you. You're so right. Yeah. He said, but other coins show promise. Theta has a very high reading on it, a 10. Persistence does as well, even higher. Oh, no. And so they're using the astrology readings as financial advice. So with Richardson's help, Thompson invested more than$40 million in messages after the initial purchase. Richardson cheered Persistence's early success, but at times she pushed back at Thompson's request, including when she suggested buying$60 million of the token. Are you sure you want$60? I will try.

1:00:51Still think you should diversify, Richardson wrote her. That same day, Thompson told Richards, Thompson that she would seek access to what she viewed as her share of the family money. I'm going to write to my brothers directly. David and Peter Thompson wrote, as I know you both are more than aware, our family branch, Thompson and her daughter, has been completely locked out of any voice of control of stewarding our future financial wealth. She acknowledged having always been one for riskier investments than the rest of the family, but said she had been exposed to valuable opportunities like crypto.

1:01:33She accused her brothers of discrimination, the 2020 infrastructure around my family branch, and the inherent restrictive compliance, which was put into place, completely removed any opportunity of me stewarding meaningful investments. To be blunt, it is no longer acceptable. For the next several months, Richardson spent as many 20 hours as many as 20 hours a day hunched over her laptop researching cryptocurrencies and executing trades for thompson richardson says thompson provided instructions over daily hours long phone calls and messages they just become a hedge fund they're just day trading this is crazy richardson they get into physical custody look richardson stored physical wallets like usb like devices holding millions in crypto around her home including at one point in her then partner's underwear drawer Richardson, whom Thompson didn't pay and who has no financial background, says that moving that money was exhilarating and stressful and then debilitatingly scary.

1:02:29So at this point, they're managing$140 million in crypto based off of astrology. I looked up the chart for XPRT, the one they tried to put$40 or$60 million into. At the time that they were investing in it, it was about, I guess,$3 per token. Now it's sitting at less than three cents. Less than three cents. Wow, that is like a complete and total loss. Like, wow, they had$140 million. They probably walked away with like$140 ,000. That's really bad. Today, Richardson is driving an Uber up and down California's central coast, a job she took to make ends meet. I had this charmed, magnificent life. She sounds ejected and wistful.

1:03:13her life wasn't charmed because of Thompson, she now says, but in spite of her, she had a lively community of friends in Los Angeles, a partner she loved and a thriving career. There were dinners and parties, she says, a chosen family that included Thompson and her daughter. I don't look at them and say bad people. I look at them and say this is the catastrophe of having money. Now she lives in her childhood home in Monterey County. A private investigator's unannounced visit to the home of her mother and stepsister this August has led her to worry for her family. She said, I was hired by Taylor Thompson, says Richard's mother of the woman, a former FBI agent who now works for the consulting group Guidepost Solutions.

1:03:51Richardson has applied for food stamps. She can't afford paid medication for her dog, Jasper. She is buried in legal work. Wow. This story has been a disaster. She lost more than$80 million in crypto associated with Richardson's involvement. Thompson did. Asked about the private investigator. A Thompson spokesman said guidepost was making continued efforts to recoup the tens of millions of dollars of Miss Thompson's money under Miss Richardson's control. because, of course, any one of those thumb drives could still have, like, a bunch of Bitcoin on it that you, like, forgot to sell or trade into something else.

1:04:27And so whenever a crypto deal happens, there's always a lot of, like, exhaust nonsense going on. Persistence apparently rewarded Richardson by giving her a secret kickback of$783 ,702 worth of XPRT, the lawsuit alleged. So they were like, you're bringing us a whale. We'll kick you back. Yeah, that's not good. Which is not good. And you need to disclose that, obviously, if you're getting commissions or something like that. You're not acting as a fiduciary. You're also not licensed or trained or have any reason to be doing this. But what a mess. Yeah. The lesson is clear. Would have hoped there was a white pill in here.

1:05:05But I think it's just total disaster. Total disaster. Should have gone long T-bills instead. Wow. A friend of the show just texted me something very funny about the story that I cannot read off without completely doxing him. But this goes deeper than we thought. Okay. But anyways, let's move on to the next story. This was a total black pill. But anyways, if the, I guess, moral of the story, don't day trade crypto. Don't day trade your friend's crypto based on your astrological readings. Yes. And. Yeah. Just stick to betting on horses. Trust the experts, really. Go down to the track. I mean, it seems like stick to real estate.

1:05:58Like, you know, she was known for just like buying houses on a whim. And that seems like kind of ridiculous and silly. You can overpay for something, but it's hard to have a total loss. I bet almost all of those have done very well, even since the real estate market's been booming. Anyway, Pavel Asparujov, friend of the show, says, Travis Kalanick's corporate catering service is insanely good. This is brilliant, by the way. Their cloud kitchen faux Chipotle is better than Chipotle. Delivery is the exact same time every day without fail. Cheaper than competition. TK is a beast, man. I mean, absolutely.

1:06:31We've got to get on this. Legendary move. This would change. Chipotle's fall from grace has been devastating. This could be the thing that gets us from 100K to a million. Devastating for many technology brothers out there that used to be able to rely on Chipotle for a quality meal. Yes. And it has just gone down the drain. The problem is with Cloud Kitchens, not very good name recognition, very much under the radar. they need to get, Travis needs to get on Profound. He needs to get his brand mentioned in ChatGPT. He could reach millions of customers who are now using AI to discover new products and brands.

1:07:12We've got to try this out. There's another post from Polymarket in here. Breaking Bitcoin is now predicted to crash below$100 ,000 this year, 67 % chance before 2026. Currently, at$110. $110. 10. I remember this, this viral video of like Bitcoin goes up and then it crashes down to a hundred dollars and then it crashes down to a thousand dollars and then it crashes down to $10 ,000 and then it crashes down to a hundred thousand dollars. It's like, we're kind of using like silly terms for like what is like a pretty remarkable rise over a very long time. Anyway, I'll leave it to my astrologer to decide whether or not I should buy or sell Bitcoin.

1:07:55Just kidding. not today that uh nick here has an as a very funny post um we uh there was a new uh google image generation uh product released today it's very cheap it's very good uh banana and nick is what a great name uh always taking advantage of the moment to to post a funny picture of sam he loves it um but this is a good meme where did he get this from valley or something yeah he went deep in the archives. Yeah. Some paparazzi uploads a bunch of photos and Nick finds the most ridiculous one. But it is interesting to see the cost of these different models. If you're doing high volume image generation, it probably makes sense to check out Google's offering.

1:08:40And we will have Logan joining too today to talk about release, state-of-the-art image generation editing, incredible character consistency, lightning fast, and available in AI Studio and the Gemini API today. Fantastic. In other news, Alex Wang has unblocked Yaxin. This is shaking up the timeline. Maybe we need a trading card unblocked. Unblocked. It is huge news in the tech world. So Yaxin says, can someone tell Alex Wang to unblock me? The reason I was being annoying was because of the whole MEI thing. because the whole MEI thing was sensitive to me. But now that Zuckerberg crowned him the king of AI, I actually think it's really bad for my career that he doesn't like me.

1:09:29We got to get a Yassine up. We got to get him back on the show. Yeah, so Bone GPT messages Alex Wang and says, please unblock Yassine. He's very sorry. Thank you for your attention to this matter. And Alex Wang says, unblocked. LOL. I love it. And Bone says, you must send a raven to rune for me. I don't know what that means. Good post here from High Yield Harry. Do we have the video? We have to pull up the video. In the meantime, let me tell you about Numeral, sales tax and autopilot. Spend less than five minutes per month on sales tax compliance. Go to numeralhq.com. I got the video here. It is in the timeline.

1:10:09This is, of course, Mike O 'Hearn, the famed bodybuilder and known for memes and video reactions around the internet. And here he is. We need the audio too. It's the audio. Are you following generally accepted accounting principles?

1:10:31Giga Chad.

1:10:36I'm just smiling. I just look good. What are you asking me? I don't care. Just look at the chart. Yeah. Just look. It's fine. It's all good. It's all good. Stop asking me questions. Yeah, this guy is just born for the internet. Absolutely fantastic. Anyway, in other news, Mark Zuckerberg gifted noise-canceling headphones to his Palo Alto neighbors because of the nonstop construction around his 11 homes. So thoughtful. This is awesome. Chris Spacky says, this is so funny. I'm deeply sorry I've been jackhammering the concrete of my old wave pool to build a new wave pool for the last few years at my$120 million home, which is next to your 18 million home, 18 million dollar home, in order to remedy the situation, I've gifted you these$199 headphones.

1:11:18Who knows? I think it's a thoughtful gift. Yeah, it's the least you can do. It's a thought that counts. He's like, It is the thought that counts. Eventually the jackhammering will end. Hey, I mean, are you, is it time to build or not? Yeah. Throw it on the gauntlet. Yeah. Just say, I mean, if you're angry about this, just say it's not time to build. Yeah, just say you're anti-reindustrialization. You're anti-growth. Yeah, you're a de-growther. You're a de-growther. If you're not into growth. Anyway, in other news. This has been. Cadillac is coming to F1. This is the most important story of the day.

1:11:50This is huge. This is huge. Sergio Perez is returning to F1 to drive for the Cadillac team. Join the grid in 2026 along with Valtteri Bottas. And this is particularly important for John, who dailies a Cadillac. I do. Blackwing. I do. Much like Zuckerberg. Been waiting. Does himself. Yes. if you want an American-made, if you just have a few requirements in your daily, you want it to be American-made, you want them to have an F1 team, you want it to have a supercharged V8, there's really only a few cars that fit that, that fit those basic criteria. You don't have a lot of options. And so it eliminates a lot of other options.

1:12:30Pretty much everything. Yeah, so you kind of forced my hand on that one. Totally, totally. Sergio Perez had this very serious announcement. And Botas just posted him in a tuxedo with an American flag. The tuxedo with an American flag is fantastic. Looking very sharp. He has a water bottle or something. It's a green can of beer or something, but it's unbranded. It's fascinating. Very interesting stance on the forward, too. I have to admit, the first time I saw this picture of Sergio Perez, I thought it was the guy from Fire Festival. You thought he was passing his back? No, I thought it was Billy McFarland.

1:13:06Why did you think that? Billy McFarland looks kind of like that. Don't you think? I don't know. Maybe I'm crazy. Anyway, Fin.ai, the number one AI agent for customer service, number one in performance benchmarks, number one in competitive bake-offs, number one ranking on G2. All eyes on 2026. Their driver lineup is locked in. I can't wait to see an American team. And they're going to be working with Tommy Helfiger. Oh, really? And so this team is going to be very, very American. I love to see it. Yeah, this is a good opportunity for, you know, Trey was taking shots at Palantir, his former employer, because Palantir sponsors an F1 team.

1:13:44Anderil sponsors a NASCAR event. This is the chance. If you're a re-industrialized company, if you're an American dynamism company, getting your logo on the Cadillac F1 car feels extremely high ROI. It's probably not going to price like Red Bull, like Ferrari, but it will certainly get a lot of attention, especially online. And then you'll fill up the top of the leads bucket. You'll be able to put everyone's information in Adio. It's customer relationship magic. That's right. Adio is the AI native CRM that builds scales and grows your company to the next level. And Nick, the founder of Adio, will be joining to talk about their new round later today.

1:14:23Excited for that. What else? And we have a post here from Mike Anuzia. He is highlighting an article from Bloomberg, Data centers to propel infra securitizations past$110 billion by 2026. We had been joking about data centers crashing the global economy. Was it last week or the week before? Hopefully it doesn't happen, but the tech lash. Someone was saying that Mark Zuckerberg has the ability to crash the economy with how much he's spending on CapEx, but I don't know. I think that's the Derek Thompson take that the economy is data centers and dialysis. It's kind of an interesting take. But I don't know.

1:15:10The data center build out, like the CapEx numbers are growing, but they're not 10xing yet. I mean, even the most AGI-pilled folks are like, the training runs will increase by 3x the size per year. Like that's a ton, but we're still early. and like all the CapEx numbers are growing, but the cloud bills are growing as well. Like there are some crazy deals out there, but it's certainly not the size of the mortgage market yet. Yeah, and the big question is how quickly will these assets depreciate? There's some conservative estimates. There's some very extremely bearish estimates. Bloomberg is reporting the market for securities backed by various digital infrastructure, including data centers could grow by 46 % by the end of next year to roughly 115 billion.

1:15:54Data centers account for 61 % of the current$79 billion market for digital infrastructure securitizations, according to Bank of America. Fiber infrastructure makes up another 20%, while cell towers are at 18%. The estimates take into account both asset-backed securities and commercial mortgage-backed securities. Companies have been developing large data centers to help support demand for AI, and banks as well as private lenders have been competing to underwrite such deals. Meta platforms picked Pacific Investment Management and Blue Owl Capital to lead a$29 billion deal for its data center expansion in rural Louisiana.

1:16:31JPMorgan Chase and Mitsubishi Financial Group, meanwhile, are leading a loan of more than$22 billion to support Vantage data centers plan to build a massive data center campus. Investors concerns about the development of data centers cloud adoption in the path of investment in AI have been

1:16:51Have been Sorry cut off The four big data center developers in particular Microsoft alphabet Amazon and meta platforms discussed higher cloud business revenues and reiterated plans for capital spending increases risk premiums on all three types of digital infrastructure ABS data center fiber and cell towers have tightened significantly over the past two years while B of A Strategists expect limited spread tightening going forward they see such securitizations as offering attractive relative value compared to other types of ABS so Okay, so There's a post that I found through a Martin Screlli quote tweet Congratulations again to Martin I'm becoming a father but it starts with this cuppies corner post global crossing is reborn and it's a picture of mark zucker i got a fever and the only prescription is more data center build out micro cap david says this is insane i had no idea it's this bad shares this post which leads you to the first shocking revelation the ai data centers to be built in 2025 will suffer 40 billion of annual depreciation while generating somewhere between 15 and 20 billion of revenue the depreciation is literally twice what the revenue is.

1:18:04Global Crossing was started in 1997. And it's like, it's the textbook example of like the dark fiber overbuild. So the main business was a global fiber optic network. They wanted to sell wholesale bandwidth to telecom companies in the 90s. It completely mooned. The market cap was over$47 billion around 2000. Didn't really have that much profit, but it was like key infrastructure for the new economy. The bubble burst and they had already spent a lot of money to lay cable, but the demand for bandwidth is far lower than expected. The company filed for bankruptcy in 2002. And so a lot of people lost a lot of money and that's what people are, that's what people are memeing around.

1:18:46Martin Shkreli has a take here. He says, so this assumes$400 billion spent on AI data centers. So much of this math is wrong. A100s depreciated 50 % after five years, not 10 % after three to five. A B200, which will depreciate over eight to 10 years, costs roughly$62 ,000, 500K for eight. You can produce some amount of human-like intelligence with this machine. Would you pay for around-the-clock, always improving human-level intelligence for$8 ,000 a year? Chips are a much bigger percent of the cost than applied here. So this breakdown says like it's a third depreciation, a third energy, a third GPUs or something like that, which I think is not quite right.

1:19:34So Scrawley's putting it in the truth zone, but obviously it is worthwhile to like reality check the, the, the, the overbuild. You don't want to get over your skis too much, but at the same time, bubbles are important. Bubbles are important. We get a lot of, a lot of capacity to build the next thing. Joe Weisenthal hit the timeline at 3 a.m. this morning. He wasn't sleeping on AIDSleep, but you can be. Go to AIDSleep.com, get a pod five with a five-year warranty, 30-night risk-free trial, free returns, free shipping. What were you doing, Joe? I have to assume it was actually 6.30 Eastern. Okay, he's up early.

1:20:13He said, rather than take equity positions, and he's talking about the Intel deal, rather than take equity positions in private companies, perhaps the government could confiscate 20 % of pre-tax earnings. And then when it comes to control, Congress could pass laws governing corporate behavior. There could be something here. Gregory replied, that is just the beginning. Your scheme could be expanded to confiscate a quarter of any increase in the value of those companies when people sell their shares. Another potential play here. Yeah, this is interesting i think we're getting somewhere um there was an interesting um post here earlier dan primack uh highlighted that uh trump's pentagon is thinking about taking equity stakes in defense contractors according to lutnik so um they're going all over they're doing they're they're they're fiending they're in deals mode deals mode uh lutnik was on a squat box with Sorkin earlier and what?

1:21:25Someone, apparently the team put up a breaking news of TBPN, Taylor Swift engaged and somebody quote tweeted, Kana coach, would you never in my life? Did I think I would find out about Taylor Swift getting engaged via TBPN notification? I mean, she's a beaner and she's a private jet enthusiast, right? Why would we not be highlighting that? This is fantastic. She's an incredible entrepreneur. Yeah. An artist. This is great. Anyway, should we do this five-span stake in the New York Times, this activist investor? It's kind of interesting. Bloomberg has a story. So the financiers that led a high-profile activist campaign at the New York Times three years ago have now taken a stake in the iconic newspaper via a new investment firm, Five Span Partners.

1:22:14New York Times, Mike Isaac, your favorite activist is back. They missed you guys. Yep. So the activist investment firm started last year by Dylan Haggart and Sarah Coyne has built a position in the Times and is pushing it to use artificial intelligence to expand its subscription base, according to a letter. They're just like every VC. Write a check. Hey, you guys should be using AI. I've heard a lot about it. I think there could be something here. AI is a clear tailwind for the New York Times. Our work shows it can more than double the company's long-term revenue and profit potential, enhancing growth by reaching broader audiences, converting more readers into paying subscribers and unlocking new lucrative profit pools.

1:22:55What's the steel man for this? I think that if the New York Times goes AI slot mode, it's very bad. It's a short-term win. Well, so they're saying, they're talking about using it to more, I guess, more efficiently convert non-paying. They're saying the Times can use AI to broaden its international reach, Love that. Low-cost text and audio translations. Yep. That makes sense. Yep. Develop more dynamic paywalls and optimize pricing. Yep. Low-cost video offerings could also boost revenue per user, according to FypeSan. Yeah. And so AI doesn't necessarily mean LLMs generating text of articles, but translation, nailed.

1:23:36Audio, nailed. And already, I don't know why you're laughing. Well, I just... What? I think it's pretty funny. to just, we should do this bit with TB Nation, where we collectively take a stake in something like a Google and tell them that AI, we think AI is very important. We think it could be good for your business. Yes, we sent a letter. We'd like you to take this seriously. Yes, yes, yes. Hey, Satya, we're going to take a position. We've taken a position in Microsoft, our whole team. Yeah. And we want you to take this seriously. The ask, if we're being activist investors in Microsoft, I know the ask.

1:24:15Bring back Clippy. Bring back Clippy. Clippy, is Clippy still around? Can you find him anywhere? I think there's some merch out there, but Clippy is not productized. Clippy needs to be... That is a billion-dollar IP. It's locked in. Imagine a billboard campaign on adquick.com centering Clippy as the new spokesman for Microsoft. Out-of-home advertising made easy and measurable. Say goodbye to the headaches of out-of-home advertising. Only adquick combines technology, out-of-home expertise, and data to enable efficient, seamless ad buying across the world. Clippy should become CEO for a day as like a little stunt.

1:24:48Oh, I like that. Give Satya the day off. Yeah. Let Clippy run the show for a day, just one day. We got to bring back Clippy. It's so good. It's such good lore. And it's so fun. It's cute. It's like inoffensive. It's the perfect mascot. Clippy's not going to get canceled. So we've been joking around about Five Spans' ass, but I think it is, I'm happy to see that they're not saying, you know, lay off, you know, too. AI will definitely not be able to do the fact finding that the New York Times does. Like, journalists are particularly good at that. And it requires, like, building a relationship over time, getting someone to share a piece of information that you can then verify.

1:25:26That's really intractable for AI. Yeah. And then even the instantiation. You know, Teal has that line about, like, AI will hurt math people more than writers, which is kind of counterintuitive. because you'd think that, well, AI is really good at writing, but AI is not actually that great at writing, and you can still clock it when there's the M-dash. And we saw this in the Travis Kelsey Taylor Swift announcement. Going to ChatGPT. It's not just a marriage. It's a love story. It's a love story. That is bullseye. The problem is people do write like that every once in a while, but now you just can't use those.

1:26:04It's not this than that. They hit ChatGPT. You think so? You're convinced. But on the tinfoil hat, Jordan, if you're going to make that strong. We got to keep it closer. We do. We need the hat rack over there. A representative from San Francisco-based 5Span declined to comment. That's the most boring line I could possibly read in this article. Today, we believe AI represents transformation opportunity. There is something interesting here about instantiating low-cost video. So the New York Times will occasionally take a story, if it's really big, and turn it into a video, like a video essay. And they could probably do that for more content where they take the video or they take the facts that have been found by human and the writing that's been found by human and then use an AI voice and then use AI to select different stock imagery or generate images.

1:26:47So five spans saying they could slop it up. Yeah, they would be slopping it up, but it would probably be like value additive to just like get the stuff. You know what they should do? They should do the clearly strategy. They should take a position in the New York Times and be like, you need 500 ,000 clippers. You need to just be clipping everything. You just need to clip it all over. This is the thing. Maybe that's the strategy for Clippy. Clippy will clip your content. Clipping as a service. So you do a campaign for Clippy, and then you have WAP clip it all over TikTok, and then you buy TikTok.

1:27:21Anyway, we went through the family stuff. Let's go to this gentleman over at Waymo, Ethan. I love the Waymo team posting. Everyone in Google feels like they're getting more comfortable on the timeline right now. I don't know if they're just emboldened, but they're having a good time. They were cooking, now they're posting. Exactly. They've been sort of building in silence, but now they're getting out there. So there's a video of a Waymo on the 101 driving from San Francisco to San Mateo, rider only, doesn't touch the steering wheel. Very impressive. Congratulations to the team. And Ethan at Waymo says, multiple sensors together is a fool's errand.

1:28:00What do you do when the sensors disagree? Surely no one has solved that. Anyway, here's one of our freeway rides that don't count in our 1 million plus fully autonomous paid rides per month because freeways are for employees only for now. Interesting that they're not doing freeways because the risk of serious injury is much higher on a freeway because you're going much faster. So Mercedes has done the same thing with their level three autonomous system that only works like stop-and-go traffic I think you're like under 40 miles an hour because with the modern seat belts and airbags Like it's really really hard to get into a serious serious like injury accident if you're just going 35 miles an hour So the way most can kind of drive around it's much lower stakes But they're testing it and they're gonna be bringing them on to the video So congrats to everyone over there on the cogs debate Tane shared a chart of marginal differences AI startups recent gross margins This is where High Yield Harry pulled that video of Mike O 'Hearn for sure Anthropics at 60%, Perplexity is at 60%, Open AI is at 50 % gross margins StackBlitz at 40%, Lovable at 35 % and Replit at 23 % And there's an important note here Perplexity, Replit, Lovable and StackBlitz do not incorporate the cost of running AI models for non-paying users and calculations while open AI does.

1:29:25Anthropics accounting couldn't be learned. Interesting. So that's why. Which I think is fair. They don't need to, I mean, it's not Anthropics' job to tell journalists their approach to accounting. Well, you know what? We have the perfect guest to talk about building a business without burning a ton of cash because we have Wade Foster from Zapier here in studio. Zapier, the French pronunciation. You know, you were in my YC batch in 2012 and I called it, what did I call it? I called it Zapier, like a rapier for a long time. And then people were telling me that's wrong. And so I started calling Zapier and then I learned that was wrong.

1:30:09But it's Zapier, like Snappier, right? Yeah, Zapier. It's funny. Snappier was actually what we called it first. Oh, no way. Yeah, we had to switch to Zapier because there was, you know, copyright issues. Of course, of course. And of course it was one snappier with one P. And so we just doubled down on the mistake and went zappier with one P as well too. Well, it certainly hasn't hurt the business. You've been on an absolute tear. We've enjoyed talking to your co-founder a ton about RKGI and that stuff, but we'd love to get just into the rest of your world and what's going on with the business.

1:30:41And do you have any advice for current generation startups who are maybe playing a little bit fast and loose with the accounting. Did you ever have gross margin problems? Was there ever a moment where you were like, okay, like we have customers that are actually making so many calls to our APIs that we're losing money on them? Or was that just never an issue? We became profitable in 2014. So basically two years in and when we raised the, yeah, when we raised our seed round at OYC, we said, Hey, we want this to be the last money we ever get. And so I like looked at every expense on our PNL and approved everything and was just like vigilant about not spending money if we didn't have to.

1:31:28And, um, yeah, I think in 2010, like that decade, we were looked like as crazy. Totally. And now it's very trendy. Like, in fact, I heard, uh, this term popularized, I don't like six months ago, seed strapping, where it's like, yeah, one round and then, you know, don't raise more. And I was like, Oh, that's what we did. Yeah. Did you believe it at the time? Were there specific entrepreneurs that you were looking to as like mentors? Like where did that idea even come from? Because the meme in Silicon Valley at the time was much more on the Reid Hoffman blitz scaling, raise a ton of money. And every venture capitalist would echo that because they want to put dollars to work.

1:32:06And so there was definitely like a mood in Silicon Valley of like, hey, it's more than okay to raise an A, B, C, D, burn, burn, burn, win, win, win. Like, did you believe that when you said it or did it come from somewhere? Like, talk to me about the thinking at the time. We had two examples. So we founded the company in central Missouri. All three of us worked at Veterans United. This is a company owned by two brothers, 50-50. They never raised a dime. I was employee 500. And when I left 10 months later, there was a thousand employees there. So this company scaling like crazy with mortgages during the financial crisis.

1:32:44Wow. And so we see that and are like, well, clearly you can build a successful company this way. Then one of our favorite products, MailChimp, they were one of like our earliest partners that was most successful, entirely bootstrapped. And then coming from the Midwest, we're just we just don't know anything about like this Silicon Valley world for companies. Yeah. Yeah. And so it kind of makes us skeptical of any time, you know, venture capitalist is. Yeah, I remember one said no important company has ever been built this way. And, you know, when you make grandiose statements like that and you have counterfactuals, you're just like, well, what other advice are they saying that might just be wrong?

1:33:24yeah uh and it doesn't mean that all the advice is wrong but i think we just ran it through a thicker filter than most people who are like well i guess we're going to raise an a and a b and a c but i think it's important it doesn't my read and i guess what i've heard over the years is it's not like you said we will never raise another dollar it was more so we we hope to just make a great business and not be forced to and i think the issue when i hear people talking about seed strapping. It's like seed strapping. I don't know if it's like the right goal. It can be something that you hope to be able to achieve.

1:33:59But if you're raising a$5 million round and you're telling everybody, oh yeah, we don't want to raise more money. It's just like not the right, oftentimes not the right thing to be fixated on. It's like be fixated on like, we want to use this money to find a really sound business model that will allow us to scale efficiently. And we will tap capital markets as needed is like a better goal than we just don't want to raise more money. And because I usually feel like that comes from founders that are like, you may as well just say, I don't want to dilute myself anymore. And that's not necessarily the right thing, because ultimately, if you want to create value, you should just make the share price go up.

1:34:37And like there's some companies that need to dilute a lot in order to create more value. I think the exercise to go through is to look at your growth and really understand where is there a bottleneck in your business? What is the long pole in the tent that is holding you back from achieving your goals? And every time we did that exercise, we came back with an answer that wasn't more money. It was management horsepower or, you know, we need this particular feature to go launch, which we can go build. Like there was just other things that were the thing to go address to help us grow faster. And the place where I see folks making mistakes is they get caught up in the noise and the logic will be like, well, so-and-so raised a lot of money.

1:35:27Or what if something, what if this happens? Or what if that happens? Which are not actually grounded in factual truths of reality. What you do know is this is where my company is. This is what it'll take to get there. And if my bottleneck is money, great, go get money. But if the bottleneck is something else, you as the CEO, like it's incumbent upon you to spend time on the most important thing instead of go chase another thing that actually is, is not the most important thing. And so that's kind of how we always thought about it. And, you know, I think it raising more money never was the most important thing.

1:35:58And so we spent our time differently. Was there ever a moment where you blinked like, you know, halfway in or a few years past being profitable where you were like, oh, like this idea, maybe it's worth, you know, dipping in, raising some money, getting back on the train? Or was it just like, OK, now we've been going for a few years. We're good. well once you get it going yeah it actually gets harder to want to raise more money because you start to look at your growth rate and going you know what like why would i raise at this valuation when i can just wait six months yeah and you know my revenue will be up by 50 or 100 like the valuation just never makes much sense now obviously and this is i think why you see a lot of like really runaway valuations right now because the founder kind of just looks at the math and says, well, it's, there's no, there's no sense.

1:36:50And so a venture capitalist has to come in and be like, well, I'll give you a 40 X or 80 X or a hundred X to get them to go, well, okay, maybe for that, I would do it. Yeah. Yeah. It's very funny that we're seeing in the market right now, like, like tough margin stuff, crazy burns, crazy valuations. And then simultaneously we have another company in the AI space mid-journey, David Holes, that hasn't raised any outside capital and is just growing, growing, growing. And so it's not like it's something unique to the fundamental structure of AI. There's clearly a way to do it without raising an incredible amount of money.

1:37:30And yet not everyone's on that path. Everyone seems to be on the path of like burn, burn, burn. Yeah. Well, and I think it depends. The foundation models obviously have huge capex and expenditures. And so that's a different beast. You know, whereas if you're at the application layer, your cost structure obviously looks a lot different. So, you know, this, this advice, this way of building obviously has to be filtered to your, your context. It doesn't make sense for all companies. Yeah. Was there something about your particular business or where you fit into the landscape where you, it wasn't, it wasn't positioned as a product where on on day you know 200 aws is like we're competing seriously uh or you had like a million competitors immediately like was it less competitive because of the position in the market and kind of like how you how you shape the product i like the product felt like i don't know somewhere in between like a developer tool it was almost like a new market that felt less competitive but I don't know what your experience was.

1:38:38It definitely was an odd product. Like we had folks like comment to us, like how exactly did you come up with this idea? And I think partly what made it so defensible was integrations, middleware is just inherently sloggy. Like there's not a lot of secret hacks to do that. You know, now with AI, it's getting easier, but it's still not easy. You still just have to like, you know, just kind of chunk away at this stuff. Like I would put us, Plaid, others kind of in this category where it's not actually easy for a big company to just decide, oh, we are going to go compete in this. And most of them just don't have the fortitude to kind of keep chipping away at a bunch of like really boring engineering work.

1:39:22Like they are hiring all these engineers that want to work on more sexy things versus chip away at integration code all day, every day. Yeah, now you kind of have such a platform that you're in. I feel like you're just like a natural beneficiary of the AI wave because you're already plugged into so many customers, so many integrations, so many data sources. There's already so many primitives that AI can take advantage of. There's two areas that stand out to me as potentially valuable. I'd love to know which one's more valuable to you. And then if there's a third that I'm not thinking of, let me know.

1:39:57First is just having AI co-pilots help write new integrations, help accelerate your software development, just do more with the team. I doubt you've done like massive AI layoffs or anything like that, but you're probably being more productive. That's maybe one area that is interesting. And then the other is just AI as a product that can be drawn on in the midst of moving data around. So if I have one connection over here on Zapier, and then I want to have an LLM transform that data and then send it to another integration, that feels like a really amazing use case. Have those both been equally beneficial?

1:40:34Has one stood out to you as particularly great? Or is there something else that I'm not thinking of? They're both really valuable for different reasons. So the first, this idea of like a co-pilot or text to workflow is really valuable for helping people build new types of workflows. Sure. Inherently, like these no-code workflow editors have a fair amount of configuration involved with them. You know, we always invested a lot in design and making that experience good. Yep. But it's still a lot to configure. And Copilot just gets closer to how humans think about this stuff. Most humans are not great systems thinkers.

1:41:14They don't break the task down bit by bit by bit. And so as a result, a Copilot can be a really good way to attract folks into the market that maybe struggled in the past. But it doesn't fundamentally change the use cases that folks can do, which is where the second one gets really interesting. and i think the this is where we've had some really interesting learnings because we have launched zapier agents which is a purely agentic experience and we have our classic workflow editor where you can now inject ai steps yep now both have been successful but the latter has been far more successful and i think it's because when you look at the um and the latter sorry to clarify the latter is in this case is workflows.

1:41:58Yes, exactly. And I think the reason why is because agents inherently just don't quite have the reliability yet to get you to solve like sophisticated use cases. You know, you think about it where it's like, well, if you're asking it to do anything complex, call it like 10 tasks in a row, 20 tasks in a row, but the error rate is 10 % every time. Well, by the time it, you know, rips through 10 or 20 tasks, like it's messed up on a handful of these things and it's way off in left field. Whereas within a workflow, you can put it in a harness and say, Hey, mostly I want this thing to be deterministic throughout all these parts of the steps.

1:42:37But right here, I need to, to summarize something, or I need to analyze the sentiment of this, or I need it to generate an email. Uh, and you can put the agent part, uh, the AI part in the spot where it does the thing best. And then you wrap it back in, fold it back into a deterministic workflow. And that has opened up a huge amount of usage. We have like a quarter billion, more than a quarter billion tasks that have run through Zapier with AI like that alone. And I think that's the, honestly, I think it's kind of a sleepy category. When did you guys realize what a good position you were in here?

1:43:14Because you basically spent, before agentic workflows were like a term that people were throwing around every single day, You guys spent a decade basically building out all the really, really hardcore integrations with hundreds or thousands of different services and getting just simple workflows dialed so that you can. Yeah, you can just slot in, you know, reasoning or slot in LLMs at different at different steps. But have you been surprised at how many people are coming in to kind of like compete in this space? because it feels like I've just seen a lot of pitches over the last 12 months that are, I just look at it and be like, okay, this is Zapier, but just pretending that Zapier doesn't exist.

1:43:59But AI first, right? Yeah, yeah, yeah. AI first. It's Zapier, but you can buy the stock in the private markets. Yeah, it's Zapier, but we're doing a Series B. There you go. Zapier, we sell software. We're Zapier, but we sell stock. Yes, exactly. There you go. Yeah, you know, I do think it kind of surprised us that we were as well positioned as we were. You know, we had started to dabble with generative AI a few months before ChatGPT had come out. And I had started to see some customers building workflows on our private developer platforms. We have a developer platform where folks can build their own stuff.

1:44:45And so we'd started to see some of that happen. And as the models advanced, in particular, when GPT-4 came out, we were like very unsure if workflow was durable. Like we started to think like, actually, this like pure agentic experience is probably going to be the future. Now, I still actually believe that it will be the future. I just don't think it's going to get there as fast as we think it will. And, you know, over the course of the next year, we just kept seeing agent like AI use cases on the traditional workflow builder grow and grow and grow and grow. And then we started to realize, huh, like maybe what we need to do with our product roadmap is actually help people surf the wave, like help them go from just old school, classic deterministic workflows and move steadily into purely agentic ones and allow them to kind of mix and match these experiences along the way.

1:45:42Take it back to gross margins for me. How important is offering the full suite of Pareto Frontier models within those deterministic steps? And is there a world where you want to offer even smaller and more focused and cheaper models? We talked to one company that's training transformer-based models, but just for sentiment analysis, just for censoring bad words, just for JSON to CSV or something. Stuff that could almost be done deterministically in many cases, but just like you need a little bit of AI there. And they train them on GPUs or on like gaming graphics cards. And they run like super, super cheap.

1:46:30And I'm wondering if you're in a high, if you built out Zappi or workflows that are high volume, the cost per inference feels relative there. And it feels important. And I feel like you would have a good insight into how some of these bigger models can kind of blow up on you if you're not careful. Yeah. Well, this is also where it's nice. If you can encode your workflow into something that's deterministic, you get reliability. You get cost advantages. You get speed. you get a lot of things that are really, really nice. And then, you know, insert AI only where you need it. Now, the way we've thought about this is we really want to abstract as much of this away from the customer as possible.

1:47:13You know, we sell predominantly to folks in GTM orgs and GNA orgs and folks that are not paying attention to the subtle differences between, you know, which model is best for coding, which model is best for writing, which model is best for this, that, or the other. so we're trying to abstract away a lot of those choices and that extends into how we price for this stuff as well too so we want to make it a little bit more like you just buy it we charge you for tasks doesn't matter if it's an ai task or another task we'll just sort of it all the math will kind of all work out for you in the end um but i imagine if you were selling do you guys have a lot of do you guys have a lot of learnings from like pricing on on pricing based on on outcomes versus traditional SaaS?

1:47:58Because that's, again, so many of the ideas that VCs have gotten excited about with agentic workflows and just agents broadly. I feel like our ideas that you guys have explored for really long time. You've basically always been consumption-based, correct? We've always been consumption-based, yeah. And I think the challenge that we have seen... Sorry to interrupt, but the other way to look at that is you've been pricing based on work done is like the sexier way to describe that at least in 2025. We do a task, like you pay for tasks. And I think the challenge that exists for horizontal products is the way in which folks use the software, they get variable value from it.

1:48:43And high usage doesn't necessarily correlate with high value and low usage doesn't necessarily correlate with low value. And so that's where, you know, if you're building horizontal usage-based software, you run into a trap where you're like, hmm, where along this spectrum are we going to price? Are we going to pick one credit system where everything is going to cost one and we're trying to figure out, well, is it a premium thing or is it a more commoditized thing? Or do we risk making our pricing more complex and start to say, we're going to have a credit system that's tied to different value.

1:49:21But now your customer is having to do a bunch of different calculations in their head, figuring this out. This is where I think vertical agents have an advantage right now where they can say, hey, we are hyper fixated on one use case. And as a result, we can maximize the value that we are going to sell for. We're going to say, hey, you know, if you're in customer support, we're going to charge you if we close the ticket. And you can do very direct math to say, well, today I have a support rep that I got to pay this much money every single year to answer this many tickets. If I can turn that into an agent and they can answer those questions just as well, then I'm willing to pay.

1:49:58You know, this is sort of like the maximum I'm willing to pay for this. So that's where these vertical software do have like pricing advantage. Whereas if you're charging horizontally, you have a much trickier challenge because you have so many variable use cases at the end of the day. Yeah. Are there new markets that you're noticing open up that are uniquely available to you because of AI? I think the thing that has surprised me is the certain like more traditional industries have been fairly quick to jump on this. And I think my hypothesis is partially that their use cases are well suited for this.

1:50:46So if you have, if you're working a lot with like PDFs and like paper documentation, the SAS stack was like kind of tough for you because it had to be structured. And there wasn't tooling that could turn all this like unstructured data into structured data for you. However, LLMs are uniquely good at that. And so as a result, these companies have seemingly jumped on this stuff pretty quick. The second thing that's been interesting is we've watched particularly media companies be relatively fast at this, which also surprised me. In some ways, I felt like they were faster at it than certain tech companies we've spent time with this.

1:51:28and i have to wonder like how much of this is because they saw so much disruption in their business model during the internet craze that they were like we are not going to let that happen to us again whereas technology companies have kind of had it pretty cushy yeah and so like yeah we're we're doing the ai thing it's fun when you're doing the disrupting but getting disrupted yeah not so exciting and so they're like act two we're gonna be on it this time around Yeah. Where do you stand on estimating the impacts of AI on the job market? Because I feel like a decade ago, or around the time when you were getting Zapier off the ground, Zapier, we said it both ways.

1:52:12I'm sure people were saying like, automating business processes will lead to, with software will lead to job loss. Did people say that? I guess, yeah. I mean, people said that with the internet they said you're not going to need sales reps anymore everybody's just going to be able to go and you know you're just going to be able to go to a website um and i just yeah i'm curious like there's two data points that we've been pointing at one is uh the wall street journal had an article today saying that uh new grad unemployment is uh is not tracking as well as it should and there might be a real effect of of ai not driving companies to hire less new grads But then on the flip side, there was this report that came out that said that 95 % of AI tests at Fortune 500 companies did not make it past the test phase.

1:53:05And so AI as an installation into these large companies is not going as well. So there's kind of evidence on both sides, but I'd love to know what you think. Yeah. You know, when I think about this, my take is that capitalism is largely undefeated. Mm-hmm. Uh, and you know, it's, it's, it's easy for us to spot the disruption. We can see the jobs that are disappearing easier than it is for us to see the new opportunities that are being created. Sure. But history will tell us that there is always new jobs created. Yep. And this is what we see play out in our customer base. I talk to small businesses over and over where they're like, I never could have hired somebody to do this work.

1:53:53And now because I'm doing it with automation, I'm actually getting more done. I'm generating more revenues. And now I'm reinvesting that. And when they go to reinvest that, yeah, some of it's being reinvested in AI and more automation. But some of it's being invested in people. So that's kind of where I think this plays out. And the second thing I think will happen is there's this meme that goes around where it's like, well, all these software companies are just going to have crazy high margins now. Like, we just don't need that many people. But again, capitalism is undefeated. If someone looks out and sees like, wow, you're generating so much value.

1:54:32How about I just come in? I'm willing to tolerate. Or if everybody has high margins and they're like, we should just spend three times as much on user acquisition. and just get more scale. Exactly. Exactly. And well, how are you going to get user acquisition? Well, some of that's going to be invested in advertising. Some of that's going to be invested in people. And like, I think at the end of the day, it nets out where, hey, we're all going to still have jobs. But what jobs we do have, that's where it's going to shift. And I think those reports are kind of telling that story where new grads, the requirements are shifting.

1:55:04What's needed inside the workforce is changing. And so there's kind of a shift going on there. And that's always what's really challenging in these moments is time is the skill sets required to be successful. They're not static. Yeah. What about job creation around the implementation? Like I'm sure there's been a cottage industry around Zapier for a long time now. People that just help integrate Zapier in companies. and we were talking probably at the beginning of this year thinking that in the same way there used to be a big opportunity to just help companies run Google ads or help companies run Facebook ads and that there's still opportunities there but it feels like right now taking Zapier into a company and just saying like I'm going to save you guys$250 ,000 a year if you pay me 20 grand to like set this up properly.

1:55:57It just feels like that is extremely repeatable. And there's so many businesses that as for as many that are coming on and using Zapier or other platforms, there's probably a multiple of those that just aren't even fully aware that these tools exist yet or how to implement them. So 100%, we're seeing the rise of the AI automation engineer. We have some of these internally where for our enterprise customers, if you want us to do it for you, we will come do that for you. And to your point, we're seeing this cottage industry starting to blossom. And it's particularly effective when you can combine that skill set with domain expertise.

1:56:34So if you have somebody in HR who has this capability, they're able to work magic on this. Same thing in legal, finance, marketing, sales. If they have this systems thinking, this ability to break down tasks, to use the tools, they're often able to work wonders inside the stack. And it's been pretty exciting because we this this person is kind of existed for all of our This is who we've served for all of our 15 years. But in the last couple years, they're starting to get more Credit inside these organizations they often were like the underappreciated person who just kept all the plumbing working But now like they're like the automation engineer like I'm making magic happen Can you give me your take on MCP and just explain it and how it fits in?

1:57:23And is it a tailwind or a headwind, an alternative, something that you're just benefiting from immediately? How does it all fit into your landscape? Yeah. So MCP, Model Contacts Protocol, it's a protocol much like HTTP or REST or something like this that allows – it focuses on agent-to-agent communication, letting agents talk to each other. I think right now there's obviously a lot of hype around it whether it's a tailwind or a headwind or both we're still figuring it out I think there's a lot that has to be figured out with the use cases and what not I think the most important thing is to actually get the use cases really dialed in and the place where we're seeing it be most effective is in like Claude, we're seeing a lot of folks that set up Claude projects.

1:58:19They give it access to various tools and then they go deploy that tool like an internal tool for their organization. So, for example, if you wanted to create like a sales briefing tool for your sales reps, you might say, hey, I'm going to set up a project. I'm going to give it a list of a template for how I want the brief to be created and I'm going to give it access to HubSpot. I'm going to give it access to Zoom info. I'm going to give it access to like deep research through chat GPT or perplexity or something like that. And now my sales reps can just come in and say, hey, I'm talking to, you know, Sam Altman tomorrow.

1:58:56Like, what do we know about our deal with OpenAI? And it goes and does deep research. It fesses everything you have from your internal CRM and it generates a brief for you. So you come in way more prepared. So those types of internal tools in the past, those were kind of hard to build. And MCP is making it really easy for folks to build those kind of chat-style internal tools really effectively. That's the use case that we're seeing pop up most frequently. But I have to imagine that as time goes on, we're going to see a lot more. It very much feels like we're in the very early innings. And we haven't seen these like new user experience paradigms like launched yet or popularized yet.

1:59:38Yeah. Yeah. It always just felt like to me with just on the basis of inference cost, if you're hitting an MCP server a million times, you would rather just have your agent like write an API integration and just go direct to the database for most of those things. But there's probably specific use cases where it makes sense, especially if you're like dancing around a lot of different. Well, I think you're correct there. Like at the end of the day, there's going to be some use cases where you want an API call and deterministic code because it's faster, cheaper, more reliable. And then there's other use cases.

2:00:11Yeah, we saw this with ChatGPT pretty early. It was like, yeah, it's pretty good at math. It can memorize a lot of math. But like, why not just give it Python and then it can get the math 100 % right? I don't care. I just want the right answer. You don't need to like stunt on me by memorizing every number of pi. You can just go look that up on Google if you want. It's fine. That worked out pretty good in the past. Yeah, exactly. Let's stand on the shoulders of giants as much as possible. Anyway, this was fantastic. Thank you so much for hopping on. We will talk to you soon. Great to chat. Thanks, guys.

2:00:40Have a good rest of your day. Cheers, Wade. In the meantime, let me tell you about Bezel. Your Bezel Concierge is available now to source you any watch on the planet. Seriously, any watch. You can go to getbezel.com. And we have Dan Wang in the Restream waiting room coming in to the TVPN Ultra Dome. He is releasing a new book today, I believe. It's already being reviewed. You've got to ask him how many pages it has. Is that the question you want to ask? And hit the gong. Okay, yes, yes, yes. Definitely. What everybody's been waiting to hear, how many pages is your new book? Yes, yes, yes. He's a research fellow at the Hoover Institute.

2:01:15The book is Breakneck, China's Quest to Engineer the Future. I saw a video of him going viral talking about the difference between China's government officials and America's government officials, talking about how they have an engineering society. We have a lawyer society. It was fascinating. Can't wait to talk to him. So we will bring him in from the rooster room waiting room. Dan, how are you doing? Welcome to the show. Welcome to the show. Really well, guys. Thanks so much. Big day for you. Huge day. Well, huge day. I get to be on TBPN. This is a nice day indeed. Welcome. We've been waiting for this moment.

2:01:47How many pages are in the book? Less than 300. Wow. Less than 300. I think this should be a quick read. Quick read. Quick read. Talk about the background. Was that intentional, by the way? Were you going for some, like, could you have written, you know, 700 pages and took a lot of work? You wanted to condense it down? To condense it down. Yeah, I think it's much better to write a book that people feel was too short and could have been a lot longer than it is to write a book that everybody feels should have been 100 pages shorter. So you want to err on that side. Yeah. Yeah. Yeah. Give me some background on the path to writing the book, the type of research you were doing, what inspired it, and a little bit on your background.

2:02:31I was in China from 2017 to 2023 working as a technology analyst, mostly for hedge funds, thinking about semiconductors, clean tech, China and science, trying to figure out what's going on. So I was living in China at the start of the first Trump administration. I was in Hong Kong, then Beijing, then Shanghai, trying to figure out what China's going to do on Huawei, how it's going to respond on the trade war. And so I ended up living through the first trade war, the trade war that moved into a tech war that really hurt Huawei. I lived through the entirety of zero COVID, and I was just trying to make sense of all of this experience.

2:03:07I was thinking that, you know, you use these words like socialist, capitalist, neoliberal, democratic, whatever that means to describe the U.S. and China now. These are all words from like the 18th century or something. And let's just come up with a new framework. And that's why I came up with the engineering state and the lawyerly society. The lawyerly society is a great framework. It really stuck out with me. Remind me where your family comes from within China. it's somewhere somewhat far away from Beijing is that correct? Yeah it's in the southwestern periphery named Yunnan so it's really far away from Shanghai and Beijing.

2:03:44That's right and you said that the distance from the government was relevant to the political culture there and I was wondering if there was a similar dynamic that you saw in the United States Like, is being in California or Silicon Valley, is there a distance from Washington, D.C. or Mar-a-Lago or Trump Tower in New York that is at all relevant? Or do you think that that distance from the center of government power, unique to the geography of China that your family grew up in? I think I'm an outsider both the US and China. Well, I'm Canadian, so I'm not exactly American. Now I'm in Silicon Valley, but I've never quite felt like this is exactly where I'm from.

2:04:32This is exactly where I need to be. And also in China, I didn't grow up in these hyper-competitive Chinese cities like Shanghai or Beijing, where everybody owns a bunch of property. You liquidate one of these properties and you can send her kid to UCLA and give her a Mercedes as well. And so I feel like my strength looking at both countries is as an outsider, not really buying into everything that DC or Silicon Valley thinks about itself. Yeah. When you try and define the structure of China's government or the United States government, you've obviously coined these terms, the engineering and lawyerly society.

2:05:19I was looking at data to try and quantify the difference between the Chinese economy and the U.S. economy. And the U.S. has, I think, higher taxes and potentially a higher share of the population working for the government. And I'm wondering if there's this world where you can look at the United States as a socialist country that's trying to be capitalist and China as vice versa. And I wonder if you could like toy with that at all, where, where, you know, the U.S. is actually, you know, like the government is huge. And yet it is the country of like, oh, yes, the government is not involved at all.

2:06:00And in fact, it is extremely involved. Yeah. Well, I think the best line I've heard on this is that the Department of Defense is America's answer to central planning and the Communist Party is just so big. My view of China is that it is 1950s Eisenhower America, giant manufacturing sector. They're building highways absolutely everywhere. Very little immigration, very traditional gender roles, pretty low taxes, very minimal transfer payments. Eisenhower had more transfer payments because he embraced the New Deal. But there is a sense in which I think China is the most right wing country in the world for all of the reasons that I've just outlined.

2:06:39And the U.S. is kind of a capitalist country that's a little bit embarrassed about it and trying to get to a bit more socialism. Yeah. I heard you mention that imperialism was somehow like the final form of capitalism or something like that. Can you explain what you meant by imperialism's relationship to capitalism? Well, that wasn't me. That was a much balder thinker, Vladimir Lenin. You were quoting it. Yeah, that's right. I was quoting Lenin. So imperialism, the final stage of capitalism, which is one of his big pamphlets. I recommend it. Okay. But unpack the thesis there, because I don't immediately understand intuitively the relationship between imperialism and capitalism.

2:07:23Like what was his argument? And then what's your take on his argument? Yeah, well, I think maybe we don't need to get too deeply into Marxist thought here. But they have something about class struggle and how the proletariat really needs to rise up because the people at the top have all the resources. And the capitalists, in order to maintain their capitalist expansion, really just have to go overseas and conquer Africa and Asia and Latin America in order to maintain their expansion. That's the idea. Yeah. And then the white pill of America avoiding that, Like what was the bull case for American capitalism not devolving into imperialism?

2:08:05Well, imperialism, I think, was a thing of the past when, you know, you really needed a lot of rubber in order to power your cars. And we don't need this natural rubber anymore. You know, we can synthesize it through petroleum. And so I think the U.S. doesn't need to be imperialist. That doesn't mean that it doesn't have any other sorts of problems. but it hasn't needed to conquer a lot of territory, in part because it's built this wonderful structure of alliances that you don't really need to conquer them because the elites kind of do what you already want to do. And I wonder if the Trump administration is still going to value all of that.

2:08:37Yeah. How much do you think Trump is actually inspired by the Chinese approach? Well, every time he talks about his buddy Xi, he compliments Xi for his great head of hair. and this is like something that they talk about pretty often. I've always felt that Trump never said a bad word about his buddy Xi Jinping up until COVID. Trump always kind of hated the Japanese and the Germans more, which were the enemies of the 1980s. I think there is an aspect in which the U.S. is becoming like China. The U.S. is nationalizing Intel. Intel is now going to be a stay-doned enterprise with American characteristics.

2:09:18There's a lot less data probity, I think, with the statistics not being perhaps quite right. There's a lot of interference inside the economic agencies. I think there is more construction, but there is more construction around gilded ballrooms and detention centers. And I just want Trump to take a leave from China and build a lot more mass transit, build a lot more subways in New York, where I'm chatting with you from, build a lot more housing for people. Let's make Silicon Valley a much more livable, mass transit friendly place. Is there any hope for the federal government building anything? Really?

2:09:57I feel like all of that is the domain of the state and local governments. When I think of the roads and the bridges and the high speed rail in California, like that's not a federal project. Should it be? Well, the interstate highway system is a federal project. And that was something really built by Eisenhower. Manhattan was a federal project. Apollo was a federal project. There is a lot of local building, that's for sure. A lot of maintenance, a lot of highways need to be built locally. But I think there's no reason for the federal government not to be much more muscular and take over California high-speed rail.

2:10:30Rather than nationalize intel, let's nationalize California high-speed rail and do a much better job at it. This feels like something that Trump could get on board with, like for sure. He would love to pop over to Newsom's backyard. He loves building stuff. He's a builder. He's a real estate guy. What about on the energy side? What do you think America can learn from China's approach to increasing energy production? Well, Trump has this policy of all of the above. And I think China is actually the one that is much more all of the above because they're building most of the world's coal plants still today.

2:11:03They're building all of the wind. They're building all of the solar. so they don't have that much natural gas, but they're building much of everything else, including all the nuclear. So the Trump administration is doing more nuclear now, but they're getting rid of wind projects that are almost completed. They're not really building much more transmission. I guess they're favoring coal as well, but China is doing all of the clean technology, all of the new stuff, as well as some of the old stuff too. um one one question i have on like press freedom in china is uh just why isn't china does does the repression of the press in china reflect a a lack of confidence in how good things are over there because you think if if you're if you're confident about what you're doing if you're happy if you're successful um you'd be like there are no skeletons in the closet let the press run free they will by default report positive things.

2:11:58I think that is a sign of lack of confidence. The Communist Party is super thin-skinned about any sort of criticism. But I think the other aspect of that is that there are a bunch of engineers, and you've spoken to engineers. They're not really good at talking. They don't even want to talk to any of us, right? So they like to just sit with their projects and design a new bridge or something. Whatever they don't understand, they tend to censor, And they're also really prickly skinned about any sort of criticism that they have. And so I think that it is a manifestation sometimes of lack of confidence in China's system.

2:12:33They're really nervous that any criticism will somehow topple the Communist Party. And then whatever sort of jokes or any sort of humor they don't understand, they simply censor it. And so I think the engineering state is also kind of an explanation for why China has been relatively weak at producing soft power. It's not like the Koreans. You know, we're all listening to K-pop. We're all watching Squid Game. The Japanese with Nintendo and the Game Boys and Mario and Mario Kart. They've been amazing. Chinese have not produced quite that much. What about the Labubu? Labubu is the first attempt, I suppose.

2:13:06I think this is kind of a new thing. China's been getting richer for nearly five decades now. And this kind of weird doll, is that the best they can do? I don't know. Well, TikTok. TikTok would be the other. TikTok feels like it. But even TikTok, it's not its own cultural power. It's not its own brand. It's like a bucket that people put different brands into. It's like the hoverboard, the anonymous hoverboard that came. That product was developed in Shenzhen. And it never had a brand around it. It was just five different front brands on top of a technology that was engineered. I don't know. It was an interesting phenomenon.

2:13:43Yeah. I mean, mostly on TikTok, all I'm watching are short clips of TBPN, not the Chinese propaganda, right? I'm only watching American propaganda from you guys. I mean, that is really proof that they're not putting their thumb on the scale. If we have any traction over there. If we have any traction. Have you been surprised to see Elon engaging in lawfare with the Apples of the world, the open AIs? It feels like, you know, if anybody represented the engineering state in America, it was Elon, you know, creating these mega projects, massively impactful companies. Even more broadly, it feels like the lawyerly society kind of had organ rejection from having an engineer in the room.

2:14:26That was like, I don't know, if I'm trying to apply your frame of thinking to the Elon Trump, there were other people that were successful in Washington. David Sachs has a law degree. Peter Thiel has a law degree. Keith Reboia has a law degree. Elon goes in. J.D. Vance. Elon goes in. He's in the same crew. He's in the PayPal mafia. And yet he is, you know, ejected pretty quickly. Yeah. I mean, I think this is one of the tragedies of Doge. I feel like we could have had engineers building a really big mega project inside government as we used to do. We had Robert Moses build a lot of projects in New York, sometimes not very well.

2:15:03We had Hyman Rickover, father of the nuclear Navy, build this giant Navy project powered by nuclear reactors with some ships as well. What did Elon do? Well, he wasn't really trying to build anything inside government. He was kind of destructive to a lot of aspects of government. And I really wish he applied his talents more to build 30 big nuclear reactors for the U.S. instead. And that's not what happened. Would those need to be net new projects? Is that a more fertile ground for government projects as opposed to, say, take over Intel, take over the high-speed rail initiative, take over something?

2:15:40because it just feels like there's a lot of stuff that's not working both in the private sector and the public sector. And that seems to be where we focus as a society in America. Just there's so much that's going wrong. The government is kind of the lender of last resort that can step in. When I agree with you, I think I like the idea of another Apollo project, another Manhattan project, another project that's like completely new and no one would underwrite it in the private sector to begin with. Yeah, let's have both. Let's have a functional high-speed rail project out of California. Let's have many more nuclear reactors because we, I think we had only one new nuclear reactor in something like the last 40 years.

2:16:16And that's kind of pathetic. I'm talking to you now from New York. New York is trying to fix up Port Authority bus terminal. It's going to take something like five years and$6 billion to make a new bus station. Every mile of subway that New York builds costs something like$2 billion per mile. This is just the highest cost in the world. and we can't even build something as simple as a public toilet in San Francisco. And so we need to get better at every sort of thing. We need to do the public toilets. We need to build the bus stops. We need to build the nuclear reactors. Let's go to Mars. Yeah, yeah.

2:16:48Is one possible path here the wolf in sheep's clothing, the engineer with a law degree that they acquired online or something, they passed the bar in their free time. Like, can the lawyerly society be hacked in the way that, you know, an engineer might social engineer a job application or reverse engineer an API and hack a website? Yeah, I want Peter Thiel to hack all of that and be the lawyer that actually is actually the real engineer. I think that would be great if something like that could happen. But I also feel like the U.S. used to be an engineering state. You know, it built a lot of, in the 19th century, built canal systems, built the train systems, built a lot of skyscrapers in Manhattan and Chicago.

2:17:39And then in the mid 20th century, we built the highways, we got Apollo, we got Manhattan, we built the Hoover Dam. You know, that heritage is out there. But right now it feels like we're walking amidst the ruins of a fading industrial civilization. And we need to at least renew this and hopefully build more too. Yeah. Trump yesterday said we're going to get along good with China. I hear so many stories about how we're not going to allow their students. We're going to allow, it's very important, 600 ,000 students. It's very important. Where do you think that's coming from? Feels like a little bit of a pivot.

2:18:16And do you think there's that much demand on the student side for opportunities at U.S. universities? Yeah, Trump is now the most pro-China member of the White House. And as I said, I think I've always, every time I saw Trump, and I was living in Beijing in 2017, 2018, when he was visiting China, he visited the Forbidden Palace once. He's always said nice things about China up until COVID. His real enemies are like the Germans and the Japanese. That's been long the case. And that's not always going to be true. But a couple of months ago, when a journalist asked him, does the U.S. want so many Chinese students?

2:18:56Trump said, it's our honor to have them. You know, it's our honor to have them. And I think that's a really ringing endorsement of Chinese students. And I think there's still a lot of Chinese who are really keen to move to the U.S. They're attracted by some aspect of working without censorship, if they're in the arts, if they're working in journalism. They're attracted by some aspect of pluralism. New York is a great place. San Francisco is a great place. They want to come. And I think more American students should also go to China. Take a look at what Shanghai is like. Take a look at what Shenzhen is like.

2:19:27Let's just have more exchange and let's have more mutual curiosity too. I studied abroad in Shanghai in 2016. What did you think? Within a week of arriving, I decided to stop studying Mandarin. And I was excited to go home. I think that was partly because I was born and raised in California, and I like clean air and healthy food and things like that, which are not in large supply over there. But I was still shocked. It was overall a cool experience. It was interesting because I was working at an interning for a company that was at a Chinese startup accelerator, and I got to experience how hostile the sort of local market was to external kind of startup players in the ecosystem.

2:20:21You also got paid to be in a promotional film about teaching English, which you weren't doing, right? Yeah, yeah. It was a little like local government fraud. Yeah, yeah. It was perpetrated. Can we pull up that clip? Yeah, we've got to find it. I think it'll be tough to find. I still don't have it. Yeah, one day Jordy gets paid to go out to some province where the official was clearly cooking the books on how much English was being taught. Yeah, there was a local, some state government had really bad test scores and they wanted to bring English teachers, which anybody, any student in China can become an English teacher or whatever.

2:20:57but anyways it was a very weird experience this was also during the election cycle which was just absolutely wild but I'm not surprised now that I think there's like only 800-ish American students currently studying in China which is just such a wildly inverted from how many Chinese students that we have here switching gears what do you think of I like the take about like the quantitative nature of Chinese policies, zero COVID, the one child policy, it's very quantitative. My kind of take on the Chinese birth rate decline was that if you can effectively put a gun in someone's face and tell them they can only have one child, you can put a gun in someone's face and tell them that they have to have three children.

2:21:51And so is there actually cause for for like nervousness or fear around true population decline. If you are, if you do have the authoritarian playbook in your back pocket and you can just upregulate fertility, essentially. I doubt it. I think it is. You can stop a lot of births by putting a gun to people's heads. It is really difficult to, you know, coerce some measure of copulation. how are you going to get two people in bed and put a gun to their head I think that's not the best conditions it's much harder is population decline like overrated, underrated is a serious threat to China in the long term from an economic perspective or population perspective yeah Peter overrated in the short yeah Peter Zaihan loves to go on and say how how Cook yeah he's been saying that for a long time but yeah what were you saying overrated in the short term um correctly rated in the long term i mean the curve is not that steep right now the population is really gently declining sure you don't need a giant population in order to have a semiconductor industry right the global population of semiconductors is not that large yeah and so this is really going to bite something like 20 years from now yeah but that's still 20 years when you know 20 years is a plenty long time 20 years ago you know we were dealing with some sort of a housing bubble and the U.S.

2:23:19was bumbling around in Iraq and Afghanistan. 20 years could be a long time and it could be that China fully de-industrializes Germany and Japan and maybe even the U.S. over the next 20 years. And that sounds pretty bad. Yeah. What is your frame of thinking around the Chinese semiconductor industry? It feels like they've been investing consistently for decades, never really hit the leading edge, but they've always been there in the conversation. They've never stopped investing. But how do you think about China's place in the competition against TSMC, ASML, the rest of the Western semiconductor supply chain?

2:24:03They've never caught up, but they're closer and closer. And as they get closer and closer, the profits among the leaders are going to shrink because they're going to be out-competed by China. You've already seen this happen in memory chips, which is simpler, but Micron has been kind of struggling for quite a while. Intel struggles have almost everything to do with itself and nothing to do with China. But it could also be that once China gets much better at Foundry, then a lot of orders will shift to SMIC, which might be able to offer a lot of cheaper alternatives. So I don't think that China will catch up to TSMC in the next, let's say, 10 years.

2:24:38It might never catch up. China also might never catch up to ASML. But if it is just consistently following behind and it's good enough, maybe that extra quality level is not worth triple the prices. And that's always been China's success. Right. They make 80 % as good at half the price and that's good enough for most people. And once that shifts, then a lot of the leaders stop having profits too. Are there any moves by the United States in the competition for semiconductors with China that you would like to see America avoid, or you'd like to see America particularly go down a particular path in terms of the semiconductor competition?

2:25:20I think that America should import a lot more Taiwanese engineers, send them to Arizona, send them to California, wherever it is, and then just bring all of their process knowledge and to manufacture in the US instead. The US hasn't done a great job supporting Micron, hasn't done a great job supporting Intel. I'm skeptical that nationalizing these companies is really going to be a good solution. But the US has everything it needs to succeed in chips. Intel hasn't been good, but NVIDIA and Apple could shift a lot more of their orders to, let's say, a better Intel or a better TSMC. TSMC. Import a lot of ASML equipment.

2:26:00and then just become a lot better at that. It has a lot more resources and much more capabilities than China does now. Yeah, the yields in Arizona seem particularly good. I'm not sure where I heard it, but someone was saying that there was potential, the potential of TSMC spinning out the Arizona assets, which I think might be quickly made a meme stock, but certainly would be an exciting development. How do you expect the trade war to resolve between the U.S. and China, just given Trump's more up-to-date messaging around signaling that we actually might be friends after all? Every time it gets close, there's always the chance that it falls apart.

2:26:41And I've seen that the trade deals fell apart several times when I was living through the first trade war in Beijing. And so I want to be optimistic that these two countries come to some sort of agreement so that we don't have the world's two biggest countries at each other's throats. two biggest economies at each other's throats. But I'm not really optimistic. Trump is erratic, and sometimes he wants to make a deal, and sometimes he doesn't. And even if Trump and Xi come to some sort of agreement, I'm not sure if that's really good for the world. What if Trump decides that he's simply going to gift Taiwan to Beijing?

2:27:18I don't think that would be good for the world. And I think there's a lot of European countries and other Asian countries who are really nervous that they do come to some sort of accord. So I don't have a good sense of what is the right move here. How do you think China is thinking about Taiwan? I think America comes to the Taiwan conflict with a very Cold War mindset, a very kinetic warfare mindset, where I've heard more interesting theories around sort of like slowly shifting the population with propaganda and political influence and just creating a more pro-China-Taiwan over time, being very patient might be something that's like more in the direct playbook of China.

2:28:08What's your current thinking on how China would like Taiwan to play out if everything went according to plan over the next like decade or so? China has had this patient strategy for a really long time and then it's try to shift the population with propaganda. And it was on course to work until they blew it up themselves by being really mean towards Hong Kong. So China promised Hong Kong that it's going to have essentially 50 years of freedom. And it did have about 20 years of freedom. And then Xi went back on basically Deng Xiaoping's word, and they crushed it. And so after that, you saw that the Taiwanese population don't have faith that China would actually respect any of their own wishes.

2:28:49And so I think what we're thinking about now is I don't think that a kinetic action, a big conflagration in the Pacific is the most likely. I think the most likely path is just some sort of status quo, which the two, you know, US, Taiwan, China don't really try to do anything big. And if anything, more likely than a kinetic action would be some sort of a blockade to surround Taiwan. And then the US would be in this kind of a tough position to have to be the initiator to break the blockade rather than, you know, being the defender of Taiwan. And so I think that that is kind of the challenging, politically challenging thing that Beijing would like to put the American position in.

2:29:28Yeah. Well, hopefully it's resolved peacefully. We'll continue monitoring the situation. Thank you so much for joining the show. Yeah. Congrats on the launch. Congrats on the launch. Thank you very much. I get our copy. Yeah. We'll talk soon. Have a great day. We'll talk to you soon. it bye the book is breakneck you can go and check it out and send it you for right up the charts yeah send it up to the top of the charts uh our next guest is in the restream waiting room we have nick from addio coming into the studio sharp what do we got fantastic name the news give us the what's happening welcome to the show hey guys how you doing great uh late for you right Yes, it is.

2:30:13Especially when you've got two toddlers asleep in the house, this is well past bedtime. Yeah, yeah. Well, I know how that goes. And yeah, I'm glad you were able to... always good to get them... it's good to have like a meeting set for somewhere around bedtime. It's like extra incentives to have bedtime. Exactly. Yeah. And also dodge the duties of putting them to bed. So it's all good. Thank you. Good, good. Well, big day, big day. Congrats on the raise. I have the gong mallet here ready if you can give us the details. Okay, absolutely. So we are very, very excited to announce our$52 million Series B, led by Google Ventures with...

2:30:57Sorry to interrupt. Fantastic. Breakdown. Give us, I mean, first time on the show, give us kind of the full, kind of abbreviated, but full history of the company and how you guys got to this point because it hasn't, it's a big milestone, but certainly been at it for a little while now. Yeah, absolutely. No, big, long journey. So, you know, fundamentally what we do is we build an AI native CRM. And what that means is a very, very flexible data model. It allows businesses to mold Atio to go to market motion rather than having to do it the other way around, which has been the way software has worked for a while.

2:31:41And then a massive amount of data ingestion, which gives us all of the context to automate action, essentially. CRM is a big product to build, a very big product to build. and there's not really an MVP of a CRM. It's a big ask for a customer to adopt it. And, you know, we've got about 5 ,000 companies that have now adopted Atheo. But to get there is a lot of engineering. And so we did three years of engineering in a, you know, in this sort of startup desert pre-launch. Pre-launch. And, yeah, so we did three years pre-launch of just hard engineering grind and launched that two years ago. Incredible.

2:32:26I guess maybe talk about the inflection points at what stages, where LLM started to really unlock value for customers and where you had those initial sparks. Because I think like when people, for better or worse, every CRM provider now says that they're AI native. So I also want to understand like what does that actually mean in practice? And what kind of advantages do you have as a new player that hasn't been in the industry for a couple decades at this point? Yeah, I think fundamentally we have two really big advantages. So one is the quantity and resolution of the data that we ingest. and building a data model that can do that.

2:33:13So LLMs work great when they have lots of context and you need to feed them that context. And so one of the things that we worked, one of the first things we worked on was this data model and being able to instantly ingest the company's emails, calendar, product data, et cetera, et cetera. And so that was the bit that really took a long time to build, especially when you need to do it at the scale, you have to do it for CRM. The other thing that is a big difference between Ateo and legacy products is that we think that, we have a thesis that code is becoming the new no-code. So the idea of no-code and these kind of very complex UI that you have to learn and very steep learning because you're not learning have the code, but you're learning something pretty complicated.

2:34:10And, you know, especially when you start to hit up against edge cases, et cetera, then the UI has become extremely complicated. And so one of the things that we've sort of engineered into Atio from the offset was the ability to execute code inside of your CRM. And you think, if you think about what has made Salesforce successful is this infinite flexibility through very, very complex engineering processes with APEX and a language not many people know and huge amounts of difficulty. But what we do with Atio is build in this sort of native code execution very early. And LLMs are brilliant at generating code, especially in these kind of safe, somewhat constrained environments.

2:34:53And so that's another thing that we think is going to, especially as this plays out, it's going to be a really, really big deal. This idea that software becomes so much more malleable because code is so much easier to generate. How do you think about the right, like where's your focus on the customer side? Again, CRM has so many different applications within different companies. And in the long run, you can serve everyone. But what's the focus been to date in the kind of 5 ,000 you mentioned? Yeah, so the focus really is on what we would consider builders. So people who like to get stuck into building things, the rise of the GTM, the go-to-market builder, is pretty apparent now.

2:35:39And so about 50 % of our customers are tech companies. And even the 50 % that aren't tend to be people who really want to get stuck into building something. But, you know, as you said, CRM has a huge range of users. and we think deeply about the consumer of the CRM, that's the AE, the SDR, et cetera, all the way through to someone in RevOps or building out really complex go-to-market motions. And so we have to serve all of those customers and we focus a lot on making the consumption of the CRM super seamless as well. And obviously AI has been incredible for that, massive reduction in data entry and stuff like that.

2:36:28Talk to us about gross margins. They're in the news with a lot of companies, sort of like just reselling inference. Obviously, your business is very different than that. It doesn't feel like I'm just paying you for tokens or just output. But are there any examples of customers where you've looked at their usage and you've been like, wow, this person somehow triggered a million different calls and ran up a big bill? And are there any lessons that you've learned from just like building an AI product that is on the back of inference and does have real costs that are somewhat variable? Yeah, yeah, absolutely.

2:37:07And, you know, like all good startups, the sort of approach we took to this initially was it's not our focus right now. In the last sort of six months, 12 to six months, we've actually made incredible gains there. And we're in a very, very good place now. And that has been a combination of models getting more cost effective and us learning to use them better. One of the things that we can do, for example, is we can decide at runtime which model is best suited for a task. So I think this is now becoming a relatively mainstream strategy. But if we're digesting an entire call transcript, we don't need to use the best reasoning model.

2:37:54We can do compaction first on a faster, cheaper model, and then put less tokens into a better reasoning model. And so it's been a big engineering effort. But with a combination of all of these techniques, we have very, very good margins now. The other thing is a lot of the value that the LLM provides is very apparent to the end user. So it's work done. And that's changing the way that we think about pricing as well. And so CRM has historically been just a seed-based product, but I think increasingly we'll see that that will change. And we now have a hybrid pricing model, which is there is a seed-based aspect and we give sort of very generous automation and sort of AI credits to those seeds.

2:38:46But if you want to go really wild and you want to replace a lot of work, then you can buy credit to do that. Yep. That makes sense. Well, congratulations on the round. Thanks so much for hopping on the show. We will talk to you soon. Have a great rest of your day. I'm back on any time. Great to see you. Thank you, guys. Nice to see you. Up next, we have Sean from Coinbase joining. I want to get his reaction to this front page news about stable coins. Wall Street calls for stable coin rethink as friction with crypto industry builds. There's a loophole on industry yields, apparently. We're going to dig into it.

2:39:19He's in the Restream waiting room, and now he's in the TDPN Ultradome. Welcome to the stream. How are you doing? Good to see you. Hey, guys. I'm doing great. Good to be here with you. The chief business officer. It's the coolest title at any company. That's the coolest title. It's the best. Anyway, give us a little of your background and what you're working on. What's the most interesting thing in your world these days? Yeah, first off, congrats on 100k pumped for you guys in under a year. That's huge. You know, we're just trying to keep up with the crypto charts here. Yeah, my background. So like you guys mentioned, I'm chief business officer at Coinbase.

2:39:54I oversee the broader company's strategy, partnerships, M &A investments and a handful of other things. I actually started out my career wanting to be a brain surgeon. So I studied neuroscience at UCLA. Realized late in college that I had a big enlightening moment, which was that I fainted when I first saw that much blood up close. And I realized, hey, this isn't going to be the thing for me. I ended up working in strategy consulting. I know you guys have been talking a lot about Intel and semiconductors today. Those were some of my early clients there. We're doing a lot of due diligence for private equity firms and the like.

2:40:36Were you pounding the table? saying go pure play foundry in 2013 you're like ben thompson just wrote about it we got to implement this yeah exactly i don't know if you had the same clout back then well fortunately but uh yeah after the consulting thing you know i wanted to get more on the uh the right side of the table in terms of allocating capital so i worked in uh in venture at a fund called graycroft uh you're actually based down in the arts district john down the street i've been to the office yeah So, yeah, we were early movers there to the east side of L.A. But, yeah, and then I jumped on Coinbase in 2018.

2:41:11Yeah, so take me through the recent history of your work at Coinbase, what you're focused on, what is moving the needle for Coinbase in kind of this next phase. It's obviously a very skilled company, but it has to be one of the largest founder mode financial institutions now. and I'm starting to stop thinking, I'm thinking about it less as like, yeah, it's the place where you buy Bitcoin and more just like it is a financial institution that does a lot of things and potentially will do kind of everything in the financial world over the next decade. And that's been extremely inquisitive. Yes. So yeah, give us a whirlwind tour.

2:41:48Yeah, totally. I think two primary dimensions. So historically, like you mentioned, I think Coinbase has primarily been known as this place where we make it really easy and secure for people to buy and sell crypto assets, right? And people know us for the first party applications, the Coinbase Blue app, which has been great. But we're broadening that out now to support a wide range of different assets. I think fundamentally it's driven by we think that all assets are going to get tokenized and be on chain in some capacity. It just provides better infrastructure for financial transactions. And so we announced pretty recently that, you know, for the first time, Coinbase is going to be building what we call the everything exchange and supporting things like equities, prediction markets.

2:42:30We opened up through the Coinbase app now that you can trade assets directly on DEXs. So in general, what we see is basically the blurring of these lines between what was off-chain and on-chain. And we want to make a very simple and seamless user experience for people to access all of these assets that are going to be on-chain. So that's... Yeah, go ahead. Yeah, yeah. I mean, in general, super on board. I feel like liquidity just lowers cost of capital, allows businesses to do more impressive things. It's good. Obviously, there are areas where I'm particularly thinking of like stocks on chain.

2:43:13There's situations where like, I don't know that I'd want someone to be trading shares in my company all the time. So how are you thinking about that market evolving? Obviously, there's a huge benefit to getting access to capital without some of the over, like the cumbersome nature of the reporting constraints of going public and all the different issues with floats and IPOs. But how are you thinking about the path to bringing more assets on chain? What's the most logical and then what's further out on the long time horizon? Yeah, I think it starts with the regulatory front. And that's where we see the biggest uphill battle.

2:43:55Right now it's sort of like in vogue to talk about tokenized stocks and the theme of tokenization. Tokenization isn't really anything new. When we co-founded USDC in 2018, it's a tokenized dollar, right? It is a dollar on a blockchain. And at that time, people were sort of like, what is this thing good for? But now they realize like, hey, you have all these benefits that are just come with the asset when it's on chain. So you get programmability, interoperability, global access, all those types of things. So public stocks, this is like the new thing that people are really excited about. We're really excited about it, too.

2:44:31Actually, you know, in 2021, I was helping take the company public through our direct listing. And we actually explored issuing a tokenized version of Coinbase simultaneous with the exchange-listed version of Coin. And that was always the dream where you could basically have a user that holds the asset on-chain. And if they wanted, they could redeem and hold the asset on NASDAQ or go vice versa, depending on their preference. And we just ran into a number of different regulatory blockers in 2021 where we realized that wasn't possible. And so I think the regime change in 2025 makes it much more collaborative.

2:45:11There's a very healthy bidirectional conversation that's ongoing around, you know, what are the rules and requirements for an asset that's traded on chain? Do you need to be, does it need to live in the environment of a broker dealer and a national securities exchange or does it not? So we're really excited about the regulatory environment progressing with the benefits of the technology. Is there a great argument right now that you've heard for how your average American benefits from stablecoins? Because I've literally paid people internationally in stablecoins. It is amazing. And if they don't have access to dollars otherwise, like in a high inflation country, I totally get that.

2:45:55But is there a way to contextualize it for somebody who isn't doing any international business and isn't American? Like what are the benefits besides like, you know, programmability abstractly? Yeah, I would say right now, most of the benefits are really cross-border transaction oriented, but like, you know, the one that I think will resonate with a lot of people who are Americans even is, you know, if you're hiring freelancers or contractors around the world, like how do you get them money? These are people that you might need for like a one-time job or for like just a few weeks. It's a lot easier to do that.

2:46:29And, you know, we have a partnership with Stripe where Stripe is enabling stablecoin payments through USDC on base. And one of Stripe's key customers is Remote.com. And they start to see that this is just a naturally native, better way for small startups and founders in the U.S. to pay freelancers regardless of where they are in the world. Do you think that with the freelancer, are they more just excited about getting dollars or do they actually care a lot about the yield that comes from treasuries and all of that? Because it feels like they're both important, but I have trouble weighting them.

2:47:05What do people care about? Well, one obvious benefit is just getting the money immediately versus having to wait days, potentially 10 days to just get your money, even after the client has hit pay. Yeah. Yeah. I think the thing that has become clear is there is just insatiable demand for US dollars around the world. Like everybody wants US dollars, but it's really difficult to access US dollars too. So John, I think to your first question, the pain point is really, I want dollars or a safe, stable currency. And then Jordi, to your point, it's like, I want it as quickly as possible in a form factor that I know and I trust.

2:47:45And we started to see this kind of organically a couple of years ago where a lot of people in Argentina and a lot of other countries were just like natively finding their way into dollars, stablecoins. And in those countries, we spoke with a lot of customers. And it's normal, right? You look for different stores of value, whether that's buying real estate in the US or gold or Bitcoin. And stable coins are just a natural and more relatable form of store of value for a lot of these people. Yeah. Can you talk about the significance of the Darabit acquisition that you guys just completed? Yeah, we're really excited about it.

2:48:22We just closed it. For context, Darabit is the world's largest crypto options exchange, specifically. They've had about 80 % market share in crypto options for the last eight-ish years or so, so kind of like up and down through the various crypto cycles. And in the crypto markets, derivatives are orders of magnitude larger than spot markets. Coinbase kind of grew up and got its foundation and its roots in helping people seamlessly go from fiat currency, primarily US dollars and euro and GBP into crypto. and the derivatives markets have just sort of like ballooned in size. So over the past year or so, we've made a big push into the derivatives markets.

2:49:07We have exchanges in the US and internationally that support a handful of products. But Deribit really, for us, brings us market leadership in this options category that we think is going to grow tremendously. It's also a differentiated capability that will basically pull together spot options and futures as different trading products together into one exchange platform. And what that does for users is really two things. One is it creates new trading opportunities. A lot of people will buy spot and short features. And you can do that now across more instruments just on one place. and then the second thing is it unlocks more capital efficiency and this is really a key one is whenever you put on give it up for unlocking capital efficiency love to see it doesn't get better than that yeah uh whenever as you guys know it's like how do i get the most bang for my buck quite literally right so if you can trade across all those different assets you hold one dollar or one crypto asset with coinbase then you basically can get margin and cross margin across all these different products with that$1.

2:50:19So it's just a better trading experience overall for people who move in size. Very cool. I love it. Let's hear it for people that move in size. Let's hear it for people that move in size. Play the eagle sound. There we go. Thank you. I love it. Thank you for coming on. Thank you so much for hopping on. You're always welcome. Yeah, anytime. We'll talk to you soon. And congrats on the new title. Well deserved. Thank you. Well deserved. Cheers. Up next. We need to make somebody the chief business officer here. Yeah, it's up for grabs. It's simply too good. Tyler, are you going to take it? You're going to take it?

2:50:51I'll take it. Chief intern officer, the CIO. CIO, but any day now, you could get promoted. Anyway, in the Restream waiting room, we have Brian Pellegrino from Layer Zero. Oh, sorry. No, we have - Okay, sorry. We have a switch. We are - We have - Ryland from Blue Water. Vicas from Reframe, right? There we go. Sorry. We're having technical difficulties over here. Welcome to the show. Good to meet you. Good to be here. I'm glad you guys have Ryland on too. We did. Old colleague of mine. Oh, no way. That's amazing. From what era? We were both at Kiva and Amazon Robotics maybe 10 years ago. That's very cool.

2:51:30Awesome. I had a friend who just got a tour of an Amazon facility. They said like on site, just on this one facility, they had like 10 ,000 robots or something. And they built the whole facility like six months ago. They said there are already two versions out of date from what the next stage of Amazon's build-out is. It was the most impressive tours I've ever done. Anyway, we're not here to talk about Amazon. We're here to talk about you. Pick us off with an introduction on yourself and what you do. Wonderful. Yes, well, it's great to be here. I'm Avikas NT, CEO and co-founder for WeFrame Systems.

2:52:00We're on a mission to build climate-resilient homes for everyone. And we're doing this by applying some cutting-edge robotics and algorithms with microfactories that we can deploy within 100 days to build pretty impressive, good-looking homes in neighborhoods all around the country. Incredible. And the news today? Give it to us. We've raised a$20 million Series A. Super pumped. Not as pumped as us. Not as pumped as us. Congratulations. Incredible. Incredible. When did you start the company, and I guess what was the initial catalyst? Yeah, we started about three years ago with the pitch deck. And the catalyst was as simple as me becoming a dad.

2:52:45My twin daughters are four years old now. Oh, twins. Yeah, so they were the catalyst to kind of take a step back and think about what else I should be doing in my time besides getting boxes to people's doors faster. And I decided I'll get bigger boxes to people's doors that boxes people can live in right now. You're committed to making boxes go faster, basically. That's my superpower. Obviously, Reframe is trying to provide a solution for one of the biggest problems, I think, facing our country. How have you thought about narrowing in and focusing, go to market? What markets are important and what does success look like in the early days?

2:53:31Yeah, maybe I can kind of talk from what the big picture is for us. We want to build a million homes by 2045. That's twice as fast as how long it took DR Horton, the largest home builder in the U.S., to build a million homes. But the places we're going to be building homes will look vastly different. If you think about where homes are getting built today, they're out in farmlands that have been converted. They're far away from existing communities. That's the only place you have land available for large-scale optimized builders. We believe that by applying software and robotics, we actually change that equation.

2:54:00We can actually be just as efficient as them, but in infill neighborhoods. How do we actually build 100 units, not on a single lot, but on a bunch of scattered sites in a neighborhood in Palo Alto or a neighborhood out in Somerville or Cambridge, Massachusetts? And we believe that by leveraging software, we actually get to a wall that we define as mass customization. How do you convert existing building designs to have it be compliant with any site, but make the marginal cost of this change go down to zero because software's in the loop, robotics is in the loop. And whoever can achieve that end state can get the same efficiencies of a production builder on scat-rich sites.

2:54:35And we believe that's our path for actually getting and creating the type of housing that people actually want to live in, in neighborhoods people want to live in, ideally at price points that are also significantly better than what's out there today. So we're currently focused on markets that are building homes for over$400 a square foot. So there are markets on the coast, so New England, so we're currently based in Massachusetts, will be as part of the series. They will also be launching capacity out in Southern California to help with the LA Wildfire Rebuild. So LA is also a target market. And then we have markets out, so Salt Lake City, Utah, and then Denver adjacent.

2:55:10And then Seattle and Washington DC adjacent would be other markets in the future. How does, what is the process to start working with somebody that ultimately wants to buy a home? Are you working with developers that are developing entire neighborhoods or can somebody just go to your website and say, I want one house, please? Yes. So today we're very focused on working with developers. So it's still a B2B sales motion for us. We believe that the best customer is a repeat customer. And this is how we're trying to build a pretty long-term multi-year pipeline. We have sold to individual customers, but the experience for them doesn't resemble that of a custom home builder.

2:55:48Like you're still building off of our product portfolio. We will customize our product to meet the site. we're not going to design a ground-up custom building, and if customers are fine with that. And that's the experiment we're testing out with LA because now we're working with homeowners who've had their previous homes burned down, and we're trying to help them rebuild. And we're able to make good progress with customers who realize they're going to get a better product significantly faster working with us, but they'll have to make some trade-offs on what the product is and what their expectations are.

2:56:17And that's been an education process. And we've actually been surprised that people actually just want a limited set of choices. if you can give them choices on things like cabinet colors, a few different arrangements of windows, et cetera, they're actually pretty happy with the process versus saying, right, you can only get one color as long as, or any color as long as it's black, right? I think that definitely doesn't work out here. Doesn't apply in a house. And the whole aspect of mass production doesn't work for construction either, which is why we've had to embrace customization. Yeah, that makes sense.

2:56:47um what uh what do any of these stick out as factors for your business uh uh self-driving cars potentially making it easier to people live in farther away suburbs or uh just the the adoption of technology in cities you can live in a smaller house because you can put on a vr headset and you don't need a home theater anymore or you can you know go to a local gym instead of needing a home gym or a library because you don't need a home library um or even like the adu thing in california the idea of building more smaller houses on existing plots of land? Are any of those factors that you think about?

2:57:24Yeah, so the ADU factor has been great. I think it's actually spurring a whole bunch of upzoning. You can actually add more units into existing properties. It changes the calculus even for developers trying to develop new properties. They get to add additional density. And we're also seeing this trend where people are actually okay living in smaller spaces, which also results in us being able to build homes that are more space efficient. But the place that we're also keying off of, I think from the practical standpoint of self-driving cars, solving the logistics challenge, well, that is true. We're also noticing a generational shift taking place.

2:57:56If you actually look at buyer sentiment for Zoomers are coming into the market, eventually what we expect Gen Alpha to also envision, there is a loneliness pandemic. They're looking to build more connection and community. And I think their desires are actually vastly different than what was true for the boomers all the way through the millennials and Gen Buyers. So I actually think we'll see a migration of people coming back into cities or pockets of communities that are more denser and more connected than what they were for how development took place for the last 20 to 30 years. And the only way to be successful in that space is how do you actually embrace scattered site development and infill development and not build a 300-unit apartment building?

2:58:33People only live there because that's the only choice. People don't live there because they want to live with 300 other people who are complete strangers. right i think so this is a place where we get to refactor what housing of the future looks like it's not just the form but just the connection of how it's connected to neighborhoods but the only way to unlock that is again you send up new technology you said a million homes by 2045 when do you want the first home to be built Well, we've already built our first home. It's occupied today. So we built our first home about 18. Thank you. Fantastic.

2:59:06I need to get one of those buttons here. Yeah, for your Zoom course. We'll license you the air horn every time you guys complete a new home. But congratulations on the progress. It's tremendous. We'll talk to you soon. Have a good day. Cheers. Take care. And our next guest is in the Restream waiting room. We will bring them into the studio. I believe we have Brian from Layer Zero. Welcome to the stream. Play that soundboard, Jordan. There we go. Look at this background. There we go. How you doing? How you guys doing? I'm good. You got some news for us? Great. What you got? Yeah. I got lots of news.

2:59:46What do we want to go through? Kick us off. What's the biggest news in your world? Biggest news in our world is we just wrapped up the largest Dow acquisition in history, about$120 million. Yeah. Congratulations. OK, break that down for the audit. What does that actually mean in practice? And kind of why did it make sense for Layer 0? And yeah, basically to combine these two businesses. Yeah, so I mean, there's a little bit of context here that originally we built Layer 0. We actually built Stargate as well and spun it out and sort of was running as this independent entity. and we now have like refolded it back in with this acquisition, right?

3:00:30So it's coming back home. There's been kind of mean that the bridge is coming home finally. And so for us, Layer Zero already today, you know, 86 % of all dollars that move across chains happens through Layer Zero. Largest stable coin in the world is built on us. Largest wrapped Bitcoin asset, largest synthetic yield bearing asset, like first state issued dollar with Wyoming. All of this is already there. So we're doing that set of things. You have on the background of Larry Fink saying, listen, every stock, every bond, every asset is going to be tokenized. So like we're already sitting there.

3:01:05Stargate is the consumer touchpoint, right? Stargate is where the end user facilitates, where we've sat in the background. And so for us, this was just purely a verticalization of the entire stack. It's roll up the whole thing, have a single touchpoint, single offering to both developers, merchants and end consumer and have it in one single place. that's awesome so so breakdown um like what did the what did the what did the entity that was stargate prior to this acquisition actually this was uh dow uh dow was thrown around uh i think our audience some people in our audience are more crypto native but but many others aren't and so they may have heard about dow's back in 2021 2022 uh but but are less familiar yep that was sort of fascinating idiosyncrasy within crypto itself, right?

3:01:54So DAO is a decentralized autonomous organization. We built this protocol, we launched it, and then we basically said, hey, governance is going to be left up to all the people who hold the token, right? So there's a bunch of people who hold the token, they have the ability to stake and actively vote on all kinds of things, how things get priced, how the dollars that the protocol owns, the protocol had accrued about $95 million worth of assets itself internally, like all kinds of things within it. And it was just running. And naturally, as you can imagine, something that gets run by, you know, 50 ,000 people, as opposed to a small group of very agile, fast-moving entrepreneurs runs very differently, right?

3:02:34And so what's meant is a lot of things are slower than you would like them to. Innovation is sort of like slow in that side. So, I mean, it was very dominant from a bridge perspective. That's what Stargate does is move assets around. It's done about$80 billion of assets transferred. So the largest bridge in the space, but purely on the innovation side and to keep moving forward, much slower than kind of you would hope. And so the offer from the layer zero side was, listen, we're gonna basically buy out these SDG, this governance tokens, SDG. We're gonna pay a premium on top of what exists. So we're gonna pay an offer above kind of what's there and the Dow gets to vote on whether or not to basically accept this or not.

3:03:14And so we paid again close to about$120 million at current market price. And so what happens is they get rid of their old token, they swap it for this new one, they now have this$120 million, so they're made whole, they're invested in the future direction of the new thing. And now universally that stack, those assets, everything drives in a single direction. So what was Stargate, what it was accruing value internally, revenue streams internally now all drive directly to layer zero. How does an on-chain acquisition take private in crypto history? Largest, largest DAO acquisition ever in history.

3:03:50Really, definitely not common. I think it might become more common, but definitely very unique. Yeah. Are there lawyers in the room? Is this all executed on-chain? Everything on-chain through the DAO. So it's not like we don't have an agreement that we can sign with the Dow. It really was an offer. So an acquisition is sort of how the industry is internalizing it because that makes sense. What it really is, is, hey, the Dow is going to vote to sunset their own token, take this other asset in place of that, and basically migrate the whole system over. And they're getting a premium for doing that.

3:04:26So it looks more like a GitHub PR than a traditional filing. It has to be one of the most efficient acquisitions from the sell side. from a cost and legal standpoint. Checking on the bank. What should we be looking out for in the back half of this year? I mean, four months, I guess, at this point. But in terms of assets coming on chain and tokenization broadly. Yep. Yeah, I mean, like, right now, what we've become most well-known for and where we sit is stable coins. Almost every, you know, the large majority of stable coins are built on top of those. PayPal, USDT, all of this stuff built moves through layers here, right?

3:05:12So stablecoins have been a brother-butter. That has been the focus of really what we're trying to do is the cleanest and most efficient money technology ever, right? That's really our pitch internally is the way that value moves. And what you're going to see in the coming months and then accelerating very quickly from there is hundreds of tokenized equities, stocks moving on chain, all of this, again, moving out of these same rails. So like the layer zero, what makes it so special and what has, you know, we're talking hundreds of billions of dollars of assets moved is you can move today, you know,$5 million,$50 million for, you know, cents or low dollars, right?

3:05:52Like incredibly efficient to move extremely large amounts of capital around all of these ecosystems. So when you talk Stripe and Coinbase and Robinhood and Circle, all are launching their own networks, right? And all of these platforms are racing for tokenization, for issuing equities on chain. Again, all of this entire sort of Larry Fink 2025 mandate of like, this is the way that the world is moving. None of those have the ability to talk to each other, right? None of them, there is no way right now that you're going to settle between Coinbase and Robinhood or that you're going to be able to move these assets anywhere.

3:06:25So that is where we sit. All of this flow that again has been historical stablecoin flows, these hundreds of billions of dollars of stablecoin flows, now you're talking every asset that exists. And so we're very directionally focused on that. Fantastic. Incredible. Seems like you're set up for success. Yeah, we'll come back on again soon. There's going to be a lot more here and we should have a longer conversation. And congratulations on the take private, the acquisition, whatever we want to call it. Very cool. We'll talk to you a lot soon. Have a good day. We'd love to. Thank you guys so much.

3:06:57thanks for coming on our next guest is in the restream waiting room let's bring them into the studio i think we gotta get the gong ready again rylan what's happening great to see you hey guys good to see you guys sorry to keep you waiting uh give us the news welcome back to the show thank you it doesn't seem that long ago that i was on the show but we raised our series a today led by google ventures and davey nikello 50 million dollars 50 boom So, GV's been busy today. Busy today. Busy day. They're not taking August off. Yeah, that's true. Incredible. Yeah, give us the update since the last time you were on.

3:07:39It must have been probably two, three months. I'm also interested. I think I'm an investor in this round, right? You are. Yeah, I didn't know. You guys have been moving very quickly. But, yeah, what's the update? So we announced our sort of seed round, came out of stealth in the spring. That round was just to get the company going, get the team together, and start developing some of the key tech for what we're building. And we're building and designing autonomous ships for the Navy. And now with this Series A, we're actually able, as one of our investors said, we've left the reservation. We actually build a full prototype ship, which we'll launch next year for the Navy.

3:08:17So when I say full prototype ship, I'm talking like half a football field in length. So where there's a lot of unmanned service vessels the Navy is working with today. Some are smaller in form factor, a little bit of a twirl. They're great for choke points. But we're building ships that are meant for the open ocean that can operate all the way across the Pacific. And they can fit basically a full shipping container on board, if not multiple. And then so you could put a weapon system in there. So in this case, you're saying half a football field. That is what the ultimate size will be. But you can't be starting.

3:08:47It's the Goldilocks. So like we want to carry payloads that matter for the Navy. And so you think about the payloads that today's guided missiles destroyers carry. You can actually disaggregate them and take one or two of those payloads, whether the radars or the equipment look to hunt for submarines or even missile launchers and put them on these smaller sized ships. So they're big enough that you can operate across the ocean. You can carry the payloads that matter. We have the range and endurance for the Pacific, but we're not so big that we cost an arm and a leg. And so we're making ships that are attributable.

3:09:17They're reusable. So you can send them out and they come back. But if you lose them, it's OK. And it's relatively inexpensive compared to like a two billion dollar destroyer. Yeah. We talked to Dan Driscoll about the Army modernization project. What's going on with the Navy? What does it look like to sell into the Navy now as opposed to maybe five, 10 years ago? I would even talk about a year ago. So a year ago when we started this company, we had a thesis around building this sort of Goldilocks type ship. sort of a medium, the Navy would call it the medium unmanned surface vessel. But it wasn't on the shipbuilding plan.

3:09:53There wasn't a ton of money around it. But we thought there was a need. And so we started the company with just a small check to start the company. But what's happened is there's actually$2 billion now for this program. And so there's a ton of investment. There's a ton of activity and a ton of attention around the space. And so right now it's us and a couple other companies that are with our own dollars building what we think the Navy needs. And there's just an AOI that went out. We submitted for it. So we're super excited to see all the progress in this space. How does the Jones Act fit into your business model?

3:10:28Does it affect you? I've heard I've talked to some folks who've said that it's making it difficult for them to source certain pieces if they're building in Europe or something. uh obviously europe has a long lineage of shipbuilding but um how does the jones act fit into your strategy or benefit you or or is it uh red tape that's a good question i don't know who gave that one to you but that's a little bit spicy um so in a good way because of the jones act we have mid-tier shipyards either on the west coast the east coast or the gulf coast that actually can build our ship so we don't need to start from scratch we don't need to go find 100 acres of greenfield and try to go build a shipyard, which would take years to go launch.

3:11:09We'll announce later this year a partnership with the shipyard so we can actually build and launch our ship next year. And if it weren't for the Jones Act, I'm not sure if that industrial base would be around today. But I think in some ways, we're not as competitive as other countries in shipbuilding. And so because of that, we've had to go source some of our long-lead items from other countries that have a sort of a deeper base around sort of shipbuilding. But we think we can bring shipbuilding back to the U.S. And I think it's not by necessarily trying to compete directly with countries like China or Japan or South Korea, but it's about out innovating them and sort of taking a lead on the tech side.

3:11:47And so we believe in maritime autonomy, not only for the Navy, but also for the commercial sector. And we think there are a ton of different use cases that are out there. we're already getting comfortable and you've seen in LA like hopping in a Waymo has zero drivers in the front and just taking that and sometimes it's a preferred kind of method versus having sort of a manned vehicle that's out there we can apply that same technology for the maritime industry and we're super excited to do that that's very exciting incredible very exciting thanks so much for taking the time to hop on congrats to the whole team congratulations and uh yeah well at some point i i assume we'll do a podcast at sea oh can't wait for that on one of your one of your ships the next time i'll the next time i'm on i don't want it to be a funding announcement i want to like do a interview from like a shipyard and show you all the cool stuff we're doing you can call in from your phone it's a zoom link so we can anywhere exactly if you want to see welding and steel and all that yeah i want to see sparks flying that'd be great great congrats have a good day.

3:12:45Up next, we have Logan Kilpatrick from Google coming in. I think I have it right this time. We got a little mixed up in my calendar. We're caught up to speed. Logan, great to see you. Congratulations. Give us the news. What's the launch? Is there a number we can associate this so we can ring the gong for no reason? We're looking for any reason to hit the gong. Number of parameters, number of cost per token, cost per image or something. 0.039 cents per image. There we go. Congratulations. There we go. Step back and contextualize it for us. Give us the actual news, the actual update. Yeah, super exciting day.

3:13:25So we launched, I think folks have probably seen online, this whole nano banana thread. We should have gotten y 'all in banana suits or something for this. Yeah, we launched this new model, state of the art image generation editing model. I think what people are losing their mind over is just this character consistency of how well you can do all of these prompt-based edits and transform whatever people have. I've seen all of these great examples online already of what people are doing. So it's awesome to see. And it's a model that's available for free for everyone to go and play with and try Gemini App, AI Studio.

3:13:59You can Vibecode some cool stuff with it. I was working on a CBPN announcement card generator earlier. Oh, that's amazing. We need that. it's like i'm putting your media team out of work no no no i mean the biggest thing there is speed i mean it's just about the ideas matter that's amazing yeah yeah yeah i love that uh what's the story on the on the research side that unlocked this is this uh new algorithm is it just more scale bigger data center big training run is it some sort of unique advantage to google on the data side? Like what, what went into achieving this? Cause I think people are tracking this stuff very closely these days.

3:14:41Yeah. Yeah. That's a great question. I think there's three things quickly that I'll mention. I think one of them is there is this interplay between image understanding capabilities and image generation capabilities. And it's actually, I was just talking to the VO team two weeks ago as well. And it's a similar story on video as well, which is like, if you have state of the art video understanding, it tends to be that that actually really helps you when you want to generate really great videos. And there's a bunch of dimensions of why that's the case. But so I think that's part of it, which is from the ground up, Gemini was built to be multimodal.

3:15:12So it's been like the core focus from a modeling perspective for a long time. So I think we've seen that now in the generation capabilities. I think another one is like this iteration flywheel that we had. So we initially launched, this was actually before OpenAI launched the gbt image model we shipped the first like native llm image generation model um and got a ton of feedback of things that like didn't work and people were uh you know you know wanted it to be better along a bunch of dimensions and i think so this this launch was a combination of like looking at and like spending a ton of time with folks in the ecosystem who've been giving us all the feedback of like what would they actually want this model to do in order to build a bunch of like real production use cases and use it to to support whatever their work is um so i think that's another angle of this.

3:16:01I think the last one is just this like continued model train that we have, which is like the, there's all of this like infrastructure that's being built in order to make it so that Google can ship models and bring them to the world and have them scale up on day one, like we are right now and all this other stuff. So I think the like new capabilities of image generation, I think benefits from like this huge amount of momentum and inertia that's happening inside of Google to like get this stuff out the door. So I think it like it helps. And we see this on like every other model benefits by like how much momentum there is on the core model progress standpoint.

3:16:40I have two follow-up questions. One is, are you thinking about some sort of Pareto frontier? Do we need a benchmark for image generation that then we can comp to put on a beautiful X and Y axis so we can go viral here with cost to quality trade-offs because I haven't really seen that pop up. We're still in the age of just like, there's like the state of the art's great and then everything else I wouldn't want to use. And so people aren't really making those trade-offs, but you can imagine that as people start to implement this into business processes, if I'm just like, oh yeah, I'm just generating like an icon and like a previous generation model can basically one-shot it because I'm not trying to do anything crazy.

3:17:24I want to be very cost conscious versus this super frontier. I'm going to be using this as a poster and VFX and a Hollywood movie. It's got to be the best and whatever the cost is, let's do that. Are you starting to think about a Pareto frontier kind of emerging in image generation? Yeah. I mean, we should collaborate on this because I feel like y 'all will do a great job on it. So that's maybe an action item. But I I think one of the cool takeaways for me on this launch is actually the like cost per image didn't change between 2.0 and 2.5. So if you see, I think we released a graph of like the level of progress from 2.0 to 2.5.

3:18:02It was literally like almost like doubled in quality across certain dimensions. The same zero point. It's like the pricing is annoying. It's like four cents. Well, we're going to switch to like cost per million images at some point because just like cost per million tokens. If we tried to say cost per token, we'd be in like 0.000000. Four cents per image. Four cents per image or$40 for 1 ,000 images. That makes it easy to do. There we go. Now we're talking. Yeah,$40 for 1 ,000. To do, yeah. I think it's just crazy, though. There's so much that you could build that you wouldn't have thought about building before.

3:18:35At that price point,$40 is not for 1 ,000 images, customized for whatever you want. It's actually not that crazy. Totally. Totally. Yeah, what's the reaction been from some of your guys' partners on the API side in terms of this? I'm assuming they got early access and exposure to it, because I see a lot of them are already rolling it out to millions of users. Yeah, I think a great example of this and something that I'm super happy about is OpenRouter, which I'm sure you all have covered some of this stuff. They have a bunch of actually cool leaderboards that show their leaderboards are all usage-based.

3:19:08Yeah, we had Alex on the show, the CEO. Alex is awesome. He's an incredible founder and entrepreneur. He joined the day that he raised like$100 million and didn't mention it at all. He's a super low-key dude, too. It's crazy. Yeah, he's awesome. But so we are the first image model that OpenRouter has actually ever released. And I think there's like a bunch of philosophical reasons why from a platform perspective. But I also think it speaks to like the level of quality and the progress that's been made on these models that OpenRouter wanted to like actually take the first step in and have like a model that they felt is like conversational enough for their customers to just like embed it into a bunch of their existing workflows.

3:19:48So I think the reception has been super positive and you, we need to get like a live type coding segment and show some of the cool stuff that you all can do. There's, there's literally so much. It's, it's incredible to see what people are building already. I want to talk about naming version numbers and all of this. It's good for releases. It's good for showing that there's some sort of binary change that you should pay attention to and actually see the jumping capabilities. At the same time, eventually, this stuff just diffuses into everyday workflows. And Google has a bunch of products that don't really have version numbers.

3:20:24BigQuery just gets better. Google Sheets, I mean, there might be some version numbers under the hood that I'm not aware of. But like for most people, they just open it up and, oh, there's a new feature today, or maybe there's release notes. Are we close to entering that era? Are you going to miss the era of constant name changes? Banana emojis hitting the timeline. There are benefits and there are costs. There are high expectations when there's a new big number, but it also draws a lot of attention. It's a reason to talk because something happened, as opposed to just BigQuery is growing and it's better and it's good and it just keeps getting better and it's less exciting.

3:21:00So how are you thinking about that in terms of messaging all of Google's products here? Yeah, I think this is a super fine line to walk. I hear, like I talk to lots of customers who have this feedback, which is like there's so much exhaustion with the pace of releases. And they're keeping up with something 0.1 version update. And it's just like all this stuff to keep in your head is really annoying. They're just like, please stop making technological progress. We've had enough, please. I don't think that's not what people, this is the challenge is this now. I don't think that's actually what people want.

3:21:31They just don't want to have to like stay on top of it. Um, so I do think there's a point at which like this stops being, uh, the thing that everyone is talking about. And there is this, like, what's the, like, there's a bunch of good sayings that I'm sure you both know better than me, which is around like the arc of technology and how it like transitions from actually being technology to just being like something that is a part of a product, like a car we don't look at as this like science thing anymore it's just this like object that we use to get around um and i think ai will sort of eventually go on that arc where it's just like diffused into everything that we're doing and we don't think about it and it's um it's part of every product experience and therefore you don't look into like all the nuanced details of the small version number um but i do love the my takeaway is you need a good mascot like i love the the nano banana thing really took off.

3:22:20And I think if it wasn't something that was like easy to attach to that had an emoji that like all this other, like there's all this like subtle, weird marketing lessons that you, uh, I feel like I relearn every time something like this happens. That's great. Uh, talk about jailbreaking bug bounties. Is there some benefit to the flywheel of people trying to jailbreak stuff? It actually helps improve the models. Uh, there's some, There's some post out there somebody figured out some jailbreak the chat wants to know. But I've always thought that the nature of social media kind of encourages, like, if you can get the models to do something weird or funky or rude, then it goes viral very naturally.

3:23:00And, of course, there's, like, the negative, like, oh, this is bad. But it's actually pretty helpful. But how are you thinking about that side of the world of getting these models out to the world? Yeah. One of the things that we're always trying to like, again, walk the fine line of is all the like safety parameters. So I think I already saw a bunch of folks making comments around like, what are we over blocking on? What are we under blocking on? There's the global legislation on a bunch of the stuff is extremely complicated. So I have a bunch of empathy for the teams that have to like think about this more deeply than I do.

3:23:31But like, it is it's actually not easy. There's a lot of open questions. It's super hard, especially on, I was thinking about like the trademark infringement side because I was like, well, well, I want to be able to use my own trademarks. I want to be able to go and generate my own, my own deep fakes of myself, but I don't necessarily want everyone else to be able to do that necessarily without something. So like, it's such a complicated, we've run into that trying to generate images of ourselves. Yeah. And it's a, oh no, you can't do this. And it's like, you've kind of arrived, but now you've got to prove to the model.

3:24:01No, I'm John Coogan. Yeah, I am. I really am. This is a very, I think this is like a non-trivial problem that like will end up getting needed to be solved. So I don't like, I don't think there's any good experiences to solve this right now, but you know, to the extreme of this is like the future of you, you orb scan your iris to like IV to the model that like you're really John. And then that's how this sort of you unlock this capability. That's the extreme version, but. Yeah. Yeah. I mean, yeah, there's a bunch of ways to, to, to make sure that the stuff works. But I mean, also just like, like, I feel like Google's been particularly good with releasing like the thinking behind the decision making in both like a blog post and a technical paper.

3:24:46Like Jeff, do we cover Jeff Dean talking about the amount of energy and water used by various queries? That was crazy. And I just, I love that because it was, it was written in a way that was accessible to the person that just wanted to see a viral post on X, the everything app, which is kind of like, you know, it's a, it's a, it's a mud fight on there every day. And then, and then you also had the blog post, which was like, you know, a press person could show up there. And there was also the technical paper. So if you're a scientist, you could go in there and it wasn't phrased as like, you know, this counterweight to like all the pieces that have come out saying it's really bad.

3:25:19It was just like, here are the facts in three different tiers of complexity so you can kind of get the answers at whatever level you want. And I thought that was well handled. Anyway, Jordy, anything else? Absolutely. No, congratulations on the launch. Always great to see you and have fun out there on the timeline. Have fun out there on the timeline. Let's go vibe code some stuff. I can't believe you think the timeline is a mud fight. I'm like, people in your replies must be horrible. I feel like everyone's very kind in my version of facts. No, no, everyone's kind, but you're always one quote tweet away from getting – From war.

3:25:48From war. The timeline's in turmoil a lot. We monitor the situation every day, and we see the timeline in turmoil nonstop. Someone puts out a bad take, there's going to be a pile on. They're always, always. But it's good content. Good for business. It's good for business. Sure is, sure is. Great to see you, Logan. Thank you so much. We'll talk to you soon. And on that note, you've got to hop on. You've got to get on with London. With Pyongyang. One sec. Oh, we have one more thing from Producer Ben. We have a special guest here. We have a special guest. Who is this? okay let's bring them in okay who is this this is a surprise what's this what's this view this is a joke or something this is a prank I'm genuinely not prepared for whatever's going to happen who's this oh wow no way what is this oh thank you we have a cake in the studio.

3:26:48Come here. No way. That's so cute. You've got a bottle of Dom. Wow, that's amazing. Thank you. Thank you. Thank you. Welcome to the show. Amazing. Look at this. Can you say into the microphone, you're watching TBPN? You're watching TBPN.

3:27:11Amazing. Well, thank you, folks. Fun show today. Can you say, see you tomorrow, Technology Brothers? Say, see you tomorrow, Technology Brothers. See you tomorrow, Technology Brothers. There we go. Yay. Awesome, folks. We will see you tomorrow. We will see you tomorrow. Thanks for tuning in. Thanks for tuning in. Goodbye.