TBPN - Astronomer CEO Affair at Coldplay Concert, JUUL Approved by the FDA, Is AI Causing Historic Crashouts? | Chris Power, Jeremie Eliahou, Chris Best, Alex Mashrabov, Jesse Pollak, Billy Thalheimer, Jonathan Mortensen, Douwe Kiela & More

Starting point is 00:00:00 You're watching TVPN. Today is Thursday, July 17th, 2025. We are live from the TBPN Ultradome, the Temple of Technology, the Fortress of Finance, the Capital of Capital. Today we have some crazy news at a co-play concert. That's where we're starting, I guess. This went insanely viral. The CEO of a company called Astronomer was caught on the kiss cam hugging one of his employees, the head of HR. And I said, but Chris, in the video, Chris Martin is like, something's going on here.

Starting point is 00:00:35 I said startup CEOs can't even hug their chief people officer at a concert in this country anymore. It is crazy when the internet descends on a current thing, how viral it is. Like just you have to jump in on the current thing. Well, before we talk more about this, I wanted a quick word from our sponsor. Astro by Astronomer is the orchestration first data ops platform built on Apache Airflow. empowering your team to build, run, and observe data pipelines that just work, all from one place. Do you believe a conspiracy theory? Put on the tinfoil hat. Yes. Will you steal man the tinfoil hat? What is the tinfoil hat explanation here?

Starting point is 00:01:13 Tinfoil hat explanation here is all press is good press. Is that Nathan for you. This is just part of an elaborate stunt. Yes. And Nathan's plan was have the CEO get caught having an affair with a coworker to increase brand awareness, and get a buzz going. This was Charlie Light over on X. They definitely got a buzz going. And, yeah, a lot more people know about astronomer today. 100%.

Starting point is 00:01:41 But hard to see how it would have been planned. There's been a bunch of jokes. Ryan Peterson said, Bored should give him a raise. Without this viral moment, I'd never know that astronomer is used by enterprise clients to manage Apache airflow and achieve 70% higher uptime themselves managed airflow.

Starting point is 00:01:58 So lots going on today. Alex Cohen says, imagine losing half your life savings at a cold play concert. Why half? Oh, because he's going to get a divorce. Okay. I thought it was about him getting like, I thought it was a vesting joke.

Starting point is 00:02:14 These are adults that made their own decisions. Yeah. And they now have to live with the consequences, but I doubt either of them would have paid half their net worth for those tickets. Yeah, rough. Very, very bizarre. I was reminded of what Emily Sundberg told us when she came on the show. I was asking her how the Hamptons has changed since the era of social media, the age of the internet. And she said that there are so many TikTokers documenting everything that happens in the Hamptons now that you can't even leave a party with someone else's wife. That's what she said to us. Do you remember this? Yeah. And I'm wondering, like, what does this actually mean for birth rates? What does this mean for?

Starting point is 00:03:03 Like, this seems predictable at this point. This is the first one that's happened. I've seen these other videos. Having a debate this morning, which is that find my friends is the best thing to happen to marriages, monogamy, potentially ever. Yes. Because if you are in a committed long-term relationship and you do not want your partner to have visibility into your whereabouts, you can make a pretty bad,

Starting point is 00:03:25 argument privacy yeah but it but it but it but it it ultimately it doesn't it's really hard to stand behind and yeah I think it's potentially a really positive force against social media yeah dating app culture and things like that yeah so the answer to technology problems is more more technology maybe yeah it was like this it's this countervailing force because like what was it like Instagram is like constantly flooding you with like recommended like look at this girl look at this random person look at this thing look at that you you know, like go down this rabbit hole and then and then find my friends is maybe pulling you back.

Starting point is 00:04:00 But I mean, he probably told, he probably said, I have to go for this for work. Yeah, this doesn't fix this issue. It doesn't really fix this issue. Lulu has a mess. The kiss cam does. The kiss cam is a Lindy technology. So Lulu had some advice. She says, just given out priceless comms.

Starting point is 00:04:16 Yeah, yeah. Brice comms for the astronomer team. She says, don't bother with crisis comms here. The CEO will try to get you to protect him, but it's not his company. He's a temporary steward. Your job is to protect the company. The CEO is a professional manager who's only been there two years. The HR person has been there less than a year.

Starting point is 00:04:35 Neither is tied to the identity of the company. Preserving trust is more important. Your business model is to handle sensitive data and your priority is scaling. And that's tough to do with multiple known liars on the senior leadership team. A better comms plan than trying to save the situation. Andy Byron is on the board, but he's not a founder. It doesn't have control. The other five members should replace him.

Starting point is 00:04:55 You can then use the new CEO announcement as a reset and get people to focus again on astronomers' actual business instead of its drama. So we did reach out to Andy this morning to see if you wanted to tell his side of the story. Felt like an extremely, you've got to be a little bit crazy like Soham to want to come on after you the current thing. But in this case, we should have the new CEO on of astronomer whenever that comes. So there's a polymarket today on whether he'll be out by the end of next week. Odds are currently sitting around 38%. I would not be surprised if there's somebody new stepping into the CEO role at the company. This seems to have been way, way bigger than just the current thing on X.

Starting point is 00:05:49 It seems to have really broken containment. Is this a Databricks competitor? Like Apache Airflow is, it says open source workflow orchestration platform used to programmatically author, schedule, and monitor data pipelines. I wonder if their business is just exploding. DAG directed. So quite the lineup of investors. They've got Bain Capital Ventures. Let's go.

Starting point is 00:06:17 Just led a series D this year into Astronomer. So let's hit the size go for that. It's probably a fantastic business. And Insight is also led the Series C in March of 2022. Yeah. So Pretty Venrock was in the Series A as well. And then Sierra Ventures, who I'm not familiar with, has led a couple different financing. So, yeah, this business has been around for a while.

Starting point is 00:06:45 They're going to get through this. I'm sure that Bain Capital and Insight and the other members of the board are already figuring out who they can get to. step up and run this company. Yeah, there just seemed to be some sort of line between like bizarre, salacious, but ultimately orthogonal to the core business drama and, like, actual core business drama, like some fundamental flaw in the business plan that's exposed, FTX or Theronos versus just this, like, you know, drama that's happening here. I keep laughing about that that analogy we go back to where if you found out that you know I think what do I have I have what are pilot sport tires that Bridgetown?

Starting point is 00:07:36 Michelin. Michelin tires so if you told me that the CEO of Michelin was was caught at a cold play concert with the head of Michelin HR it would be a tall order for me to go get new tires right? I'd be like the tires work pretty well. That issue doesn't really affect my tires. It's more so that the, the difference here is that being caught is different than being on the kiss cam and getting 100 million views today.

Starting point is 00:08:12 My post alone has a million views. And I posted it a few hours ago. But do you think, do you think it's going to actually affect like revenue this month. I don't think it will affect the business at all necessarily, maybe a little turbulence because of needing to find new management and having some turnover. Certainly stressful weeks of dealing with stuff. Turnover on the on the on the on the on the exec team is going to be a challenge. But ultimately their customers are not going to, you know, say, hey, we're turning off. You know, we want out of our contract. If the product is good, maybe some people use it as an excuse.

Starting point is 00:08:47 Sure. But yeah, I don't see this affecting the business. But then Again, it's like if you're being capital or insight, do you really want, do you want a, he came in two years ago. Sounds like he's been executing. The team has been executing well. If they got from Series C to D over the last couple years, they're probably growing nicely. But, and I don't necessarily think that this is the end of this guy, either of their careers. Yeah. It just should be the end at this company.

Starting point is 00:09:17 Yeah. Because if you're the board and you, and you tolerate this, that just, just means, yeah, we tolerate people of, you know, poor moral character. Sure. Yeah. Also interesting. I mean, this company's like never even heard of it and it's on an absolute tear series D at this point.

Starting point is 00:09:36 And does this not sound like something that should be in the AWS dashboard? Like we were talking about browser base yesterday getting like quote unquote copied. And there's just something about these like point solutions that just are, you know, nailing a specific problem. In this case, it's like deployment, deployment and management of an open source project. It's literally anyone can just run Apache Airflow. They just help you do it better.

Starting point is 00:10:01 And that's kind of what Databricks did with Spark. Like Spark's an open source project. Now they, now they, for these data pipelines, now they help companies actually install, manage, run them and then put a bunch of software on top of it. And Databricks has been massively successful. And so it's interesting to bring it back to Brassumace,

Starting point is 00:10:20 because it just kind of reveals that like these companies can be kind of grinding in silence for a while. Just I guess the note for Paul Klein's probably stay out to avoid being the current thing. Because the failure mode could be less technical and more interpersonal. Yeah. I don't know. True. Anyway. Work, retired, die may have been inspired by my post.

Starting point is 00:10:47 Okay. Or it could have been totally random. but he said married CEOs can't even hug and romantically sway with their married head of HR during a cold play concert anymore because of woke. Oh, because of woke. Okay. Yes, yes, yes. Is there anything else to cover on the story?

Starting point is 00:11:02 Sophie Netcap Girl says it's so hard to get notice as an AI company these days. The astronomer's CEO had to cheat on his wife for marketing. It's dark out there. I don't think, I'm not believing the tinfoil hat one on this one. I think it's an L. It is a huge Al, but there's a lot of good stuff happening today. Yes. Open AIs. Like ramp.com.

Starting point is 00:11:23 Time is money. Say both. Easy to use corporate cards, bill payments, accounting, and a whole lot more all in one place. Did you see the Model YL is launching? Has a six-seat configuration, three-inch longer wheelbase than the model X. Can you hold this for me? The Al. This car will sell like hotcakes.

Starting point is 00:11:41 It's a 193 inches, I believe, which is not quite Escalade ESV. territory, but much bigger, much closer. I was talking about this for a while that like... Wait, so they made the model, they made an Excel version of the model Y instead of making an Excel version of the X? Yes, because the X is their premium product that's more expensive. They need to get into the full size affordable SUV market.

Starting point is 00:12:09 Got it. So this will compete with the Hyundai Palisade, which is a full size SUV. They're firmly in like the crossover territory right now where it's a five-seater, but you can't really put a whole bunch of car seats in there You can't bring the dogs all that type of stuff by adding I think it's like a pretty significant length I think it's like maybe 15 inches longer overall Brings it to that 200 inch length and winds up with a product that People feel comfortable throwing their whole family in basically yeah, right I agree with this takeaway from Nick Cruz that this car will sell like hotcakes.

Starting point is 00:12:46 I think that this is the most obvious thing that Tesla has been missing in their lineup is a full-size SUV. Yeah. The Model X always, and even the model, even the Model S came with at one point a third row configuration, but it was always super tight. But now they're kind of dipping their toe

Starting point is 00:13:07 into full-size SUV. They're still gotta do the cyber-truck SUV. Syberman. I mean, that's the original story of the suburb. It wasn't it an F-150. They took the F-150 body on frame so it's technically a truck. This was like the excursion We got really upset about this because total gas guzzler, but they take they take a truck which is this it's not the unibody. It's the bottle body on frame construction F-150, but then they would just create like a whole like passenger compartment on the whole thing and this was the

Starting point is 00:13:40 the excursion the expedition and then they kind of started downsizing it to the explorer and And then at a certain point, customers were like, okay, I want the aesthetics of an SUV, but I really want the gas mileage of a car. So build me a car that looks like an SUV, and that's where we got the crossover. So the crossover is all of the manufacturing strategy or manufacturing, like, techniques of building a car, which is this, it's not body on frame. It's not this, like, platform that you can put an ambulance on or a fire truck on or whatever you can do with a truck.

Starting point is 00:14:13 it's it's it's it's all designed to be one thing but then uh like customers went down into like the crossover market which rides better it's not as bumpy but it's ultimately smaller and now we're stretching out the and now we're going backwards we're stretching out the crossovers to the point that they're going to be full-size SUVs but the cyber truck is its own unique platform its own unique manufacturing line obviously its own unique unique styling and I think if you make that full-size SUV style uh it would sell much better because people in L.A., you see, like, people in L.A. want G-wagons. You among them.

Starting point is 00:14:50 Well, yeah, the suburban, the Cadillac, Escalade V is a very fun, exciting car. Yeah, but I mean, the difference between a Ford Raptor and a G-63 is practicality, in my opinion. You don't need the truck bed. You need the interior space. I would use the truck bed because I surf quite a lot and so that was nice. Yeah. So having a covered bed is nice, but surfboards fit in cars. Exactly.

Starting point is 00:15:23 And every time with the Raptor, like once you own a Raptor, you realize that there's just not that much space in the actual cab. Like it's pretty, it's pretty tight in there. And I'm pretty sure the Raptor is like 220 inches long. So it's like over two or three feet longer. And the bed is also very short. So it's like not a super functional bed, not a super functional cabin. No, it makes way more size to have a full-size-ass-engine. Yeah.

Starting point is 00:15:46 It looks amazing. It's the truck you wanted when you were five years old, and you said, I want a big truck. Yeah. And then as an adult, sometimes you have to get it. I got to bring it back. Let me tell you about graphite. Code review for the age of AI. Graphite helps teams on GitHub's ship high-quality, higher-quality software faster and get started.

Starting point is 00:16:02 Let's give it up for graphite. Graphite. In other news, Jewel has been approved by the FDA. Authorized is the key word here. Authorized. So I know. way too much about this, but it is a big, big turnaround since the FDA was refusing to authorize the Jewel e-cigarette years ago.

Starting point is 00:16:26 Jewel says, exciting day for making cigarettes obsolete in America. The FDA has issued marketing granted orders. These are MGOs. We'll get into this, but it's like slightly different than like FDA approval for a drug for the Jewel system, recognizing that these products as appropriate. for the protection of public health. They don't have to say, the FDA doesn't have to say that it's good to use Jule,

Starting point is 00:16:49 just that having Jule on the market has a net positive impact. And that's because cigarettes are already on the market. So it's this relative calculation that the FDA is doing. Yeah, it's much harder to argue that Jule should not be able to sell when cigarettes sell daily. Exactly.

Starting point is 00:17:08 So the FDA is not saying everyone should go out and start using Jule. They're just saying that, Keeping Jewell on the market is a net good for the American public health, which I'm sure will be hotly debated by a lot of people, but it's what the FDA said. So that over 2 million adults have switched completely from deadly cigarettes to using Jewel products, and they approved a few different e-cigarettes.

Starting point is 00:17:32 The big thing is that they approved the... And can you talk about the last few years? Yeah. Because I feel like that's a wild story. Basically, Jewel is just getting hammered by... like all these different regulators. Meanwhile, the average gas station

Starting point is 00:17:47 is selling fully unregulated vapes that are seemingly literally tested on children. You've seen the videos. Videos have gone viral where I don't think he was inhaling, but having to like make sure that each one works. Killer use case for a humanoid robot. Or just a device that inhales.

Starting point is 00:18:12 Like that's just a fan. We've been able to create suction from robotics equipment or machinery for, you know, probably 100 years. And yet they're still using a human for that. Disgusting. But yeah, the full story of Jewel. I know it pretty well. A couple Stanford guys working on a project. They're both smokers. They want to figure out a way to smoke. Way to quit. They don't like the current e-cigarettes that are on the market. and the key reason why the first version of vapor products was not satisfactory to smokers was that it used a very like not concentrated formulation. And so what that means is if you remember back in, this was like what, 2010-ish, I want to say.

Starting point is 00:19:01 Yeah, 2010-ish. Vaping was really big, but you had to get this rig. It was like this war rig that you had to assemble. And people would use like different pieces and be like, oh, I got this battery pack and this motor. And vape shops had exploded. It was like building a custom PC. It was like what GPU are you going with? What fans are you going with?

Starting point is 00:19:21 I was in high school during that era. And there would always be some kid at a house party that was blowing smoke from like one side of the room to the other. And then making like those artwork with it. Yes. Fortunately, I was never in that world. Hit the soundboard. Give me the Ashton Hall. That was the sound that every vape made in 2010.

Starting point is 00:19:49 But the reason... I'm very glad as a culture we got past that. Yes, it was a particular Nadeer for the American culture. Right up there with like the Tweety Bird tattoo and the tap out t-shirt. It was all part of the same culture. They say Americans don't have culture. But, you know, we proved them wrong for that minute. Although it was rough and I'm glad we moved past.

Starting point is 00:20:09 But there is a scientific reason why the vape cloud had to be so big. Like that was not, that was a tradeoff that was made by scientists. It was not concentrated. So in order to get like a cigarette's worth of a hit of nicotine, you needed a massive volume of smoke. You just needed so much vapor. And so Adam Bowen, James Monthe is the founders of Jewel. They figured out that there was a way to. make the smoke or the vapor more concentrated.

Starting point is 00:20:39 And there's a whole bunch of science that goes into it. But nicotine salts are like the main one that people point to. And basically they figure this out. They build the device. They actually bring on, they run like a sort of standard Silicon Valley playbook. They bring on, is it Eves Behar or someone like that? There's some like iconic designer who worked on like the jam box and stuff. I don't know.

Starting point is 00:21:04 They bring on one of these, one of these like incredible. incredible storied industrial design firms. They make the original jewel device, which did have... What's the whole story with PACs, too? It was like the same company? Yep, yep. So when they started, they were doing... What was it?

Starting point is 00:21:20 They did, they had fume or flume or something. I forget what it was. They had a different tobacco vaporizer. And then they had the same kind of technology to heat up material. And you could, in theory, put... tobacco leaves in there and then vaporize that and just warm it up and breathe that in and that would it would be like smoking a pipe or like vaping tobacco, I guess loosely, but obviously like everyone was just using it for cannabis. And so that takes off that becomes this like fantastic business on its own and then they had this other product. I forget what it was called, but it was a it was a tobacco vaporizer like an e-cigarette similar to jule. I went up selling that to jTI in Japan or like doing this crazy license

Starting point is 00:22:06 deal to get that out. Then they wind up splitting the company once both products are kind of taking off, but they're on very, very different trajectories. And it's very clear that they will be under very different regulatory regimes because the FDA is set up in a way that there are a number of different organizations within the FDA. So the FDA approves cancer drugs, and that's in the FDA drugs. They approve biologics. They approve veterinary medicine. They approve medical devices. So like when you go in an MRI machine, the FDA has approved that. Yeah, I'm pretty sure. And like, when we do like one stick blood test, like that's a device, it's not a drug, so there's a different group. In 2000, I mean, we can go like way back, like cigarettes were never regulated by the

Starting point is 00:22:51 FDA. Let's go back to the, to the very first time a human experience. It's like 10,000 years ago. Yeah, basically. Yeah, I mean. The first wooden pipe. Seriously, like it used to be like ancient. It probably was somebody threw some tobacco leaves on a fire. No, I mean, it was used in like religious rituals. They would bury people with tobacco leaves on their gums. People would chew it up. There was a whole bunch of different ways to, you know, the original like piece pipe. It was part of that along with other things that you could possibly smoke.

Starting point is 00:23:22 Like as long as there have been stuff around, like dudes have put it, lit it on fire and breathe it. Let it on fire. Breathe it in. Always. Always. And there's actually evidence that like monkeys do this too. They'll go out and find, like, rotted fruit and, like, drink it and drink the fermented fruit and get all drunk and, like, come back to the crew and be like, I found the drugs, basically. Anyway, so cigarettes, you know, invented as a part of the industrial revolution.

Starting point is 00:23:50 So we figure out how to make the cigarette rolling machine, and all of a sudden, people go from smoking, you know, like a few cigars, which had to be hand-rolled. They're very expensive. It's kind of inconvenient. You can't really huff down that many cigars, although J.P. Morgan, famously went to his doctor and was like, I'm not feeling too good, Doc. And he's like, well, you've got to cut down the cigars. Why don't you take it from 20 cigars a day to 10 cigars a day?

Starting point is 00:24:16 J.B. Morgan, absolutely dog. He's like, I think I can do that. I'll have to taper off a bit. It's not going to be overnight, but I think I can get there. Imagine if I smoke a full cigar, like my tongue burns, it's so rough. I'm not built like JPMorgan. We're built different. Built different.

Starting point is 00:24:33 It was what an era. Anyway, so the cigarette rolling machine creates this, like, massive boom in cigarette adoption because it becomes super easy and super cheap. So Warren Buffett has this famous quote about, like, it's the best business in the world. He'd never own a cigarette company for moral reasons, but he's like, you make him for a penny, you sell him for a dollar. Extremely high margin. And the cigarette rolling machine is so efficient that the raw goods that go in.

Starting point is 00:24:57 And the reason that prices, that these companies were able to capture that much margin for so, long is basically regulatory capture. There's a whole bunch of different reasons. One is there's actually distribution monopolies because all the big tobacco companies have trucks that go and deliver them to tobacco stores that are licensed. And so if you have a set amount of tobacco store licenses, like gas stations that have a tobacco license, and then you have a relationship with that particular 7-Eleven and on the back, like they can't sell them anywhere in the store.

Starting point is 00:25:29 So the store can't become like a vape shop, at least it couldn't. for a long time. There's only a set amount of store space. They call it the power wall behind the cashier that you can actually sell stuff and Marlborough is there being like, we are the reason you exist. You make

Starting point is 00:25:47 so much money off of us. We want all of this. Don't let anyone else in there. So there's a whole bunch of other things. There's also brand moats and these are the most powerful brands of all time. The idea is if you were starting a cigarette company today, you would need a billion dollars? So that's a more

Starting point is 00:26:03 modern phenomenon because cigarettes it's discovered in the 50s that they're giving people cancer because people are people went from consuming like the equivalent of like one cigarette a day during the cigar era to smoking a pack a day and in everything like the dose is the poison the the like the concentration is so important in biology you can the human body is like pretty resilient and if you've smoked like one cigarette in your life you're probably gonna be fine if you smoke two packs a day you're gonna be in a real real tough spot. And so the American populace starts smoking like a pack a day. Surgeon General

Starting point is 00:26:37 comes out and says, we're noticing something. A lot of people who smoke get lung cancer and they die much faster than people that don't smoke. So there's something going on here. Big debate, big, big law fight. There's finally this master settlement agreement where basically the discussion, the negotiation is between all the state governments and all the U.S. governments saying, hey, you tobacco companies, you have given everyone cancer. This has a financial cost to us because when a cancer patient comes into our health care system that's funded by the taxpayer, we have to pay for chemotherapy. And that has a cost. And so you put that cost on us, the government, you have to pay us.

Starting point is 00:27:18 And so that was kind of the main concession of the master settlement agreement was all the big tobacco companies. They kind of like, one of them broke rank and was like telling on the other ones. It's this big drama. Basically, they have to make all these payments. And these payments still happen today. And there's some interesting finance that goes on where if you're like a local county that's getting like cash flow from the tobacco companies, you can go and finance that out and pull that forward and then build a bridge. It costs like, you're like, yeah, I'm getting $5 million a year for a tobacco company, potentially forever. Let me pull that forward and finance that out.

Starting point is 00:27:48 So there's like all these different finance arrangements to like move the money around. But basically you can just think about it as like the big tobacco companies. You think heaters fund our infrastructure. They actually do. They fund a ton of stuff because it's like billions of dollars changing hands every day. every year. But as part of that agreement, the initial, like, the initial debate was like, okay, the big tobacco companies are going to pay. What else are they going to do? Most of the legislation came not from the FDA, but from the FTC saying, and the FCC, the communications

Starting point is 00:28:18 commission saying you can't advertise. You can't do billboards anymore. There used to be billboards in Times Square of, like, you know, camels like smoking. Like, and it would have. And those were some really, we got to. Yeah, it was wild. It was wild. And so they, it was the, it was the best era of out of home advertising. But now you can go to adquick.com. Out of home advertising made easy and measurable. Say goodbye to the headaches of out of film advertising. I am going to put, keep ranting.

Starting point is 00:28:47 I'm going to put one of these ads. Yeah, put the ads in. Yeah, put the ads in. Okay, so basically. Or sorry, not ads, one of these, one of their old ads. Yeah. So, I mean, it was literally the government saying out of home advertising is too effective. can't do it, we need to nerf it because you're getting everyone hooked on smoking.

Starting point is 00:29:05 So that was the initial kind of agreement was the big tobacco companies would no longer be allowed to sell, would no longer be allowed to advertise and they had to make these payments. Then in like early turn of the millennium 2000, something like that, this company, Enjoy comes out with one of the first e-cigarettes. There you go, Camel, wow. I mean you see this as a... The 17-year-old, you're thinking, this day I turn 18, that's going to be me. It's 18 or 21?

Starting point is 00:29:40 It's 21 now. That also changed recently. So, Enjoy comes out with one of the first big e-cigarettes. It's an example of that, you know, older technology that has the big vapor cloud. And the FDA hits them with a lawsuit and says, hey, you are selling an unapproved medical device. This is a device. It's electronics. It's a medical product because you're making a medical claim.

Starting point is 00:30:05 And that medical claim is this product helps you quit smoking. Smoking addiction, cigarette addiction, is a disease. And so by selling a product that helps you quit smoking, you need to be regulated by the FDA. Enjoy fights this back and forth. It goes all the way to the Supreme Court. Enjoy wins because they were kind of not saying that, at least they made the argument that they were not saying this is to help you quit smoking. They were just saying, this is a cool thing to do separately. Don't worry about it.

Starting point is 00:30:34 Don't ask us about the relation to smuggling. This is classic. But in the interim, the FDA is able to put an import restriction on the company. So they're making the product, I believe, in China, probably overseas somewhere, because it's electronics, it's equipment. They bring it in at the ports, and the FDA says, hey, while we sort out this lawsuit, you can't bring any more in. And so this is kind of like the ban hammer that they bring down. It winds up bankrupting the company. Damn.

Starting point is 00:31:00 They wind up winning the court case in the Supreme Court. Later, a hedge fund guy actually buys the company out of bankruptcy, turns it around, gets it FDA approved, and sells it to Altria for like a couple hundred million dollars. Maybe actually a couple billion dollars, I think. Let's hear it for the hedge fund guys, making some money, selling some babes to get tobacco. We're endorsing the end product. We're endorsing financialization. Yeah. Yes.

Starting point is 00:31:26 Financial engineer. And restructuring. Yeah. And it was a successful thing. And if you think about that as much as we're joking, like it is good to get a big tobacco company, shift their revenues away from cigarettes as fast as possible. And so like the enjoy thing, even though there's a lot of issues

Starting point is 00:31:43 of that product in many ways, it's a very interesting outcome and it's probably, you know, moving in the right direction. Anyway, so it goes to the Supreme Court and it is revealed in this court case that the FDA does not have the ability to regulate e-cigarettes. And so it has to go through the House and Senate. So when Obama gets elected in 2008, they pass the U.S. government at the federal level passes the CIPACO Control Act, the TCA. And in there it says, hey, the FDA does have regulatory authority on over everything that contains tobacco. So now, if you, doesn't matter if you're creating a new cigarette,

Starting point is 00:32:24 it doesn't matter if you're creating a new e-cigarette or a new nicotine pouch. or nicotine gum, you need the FDA to review your product, which is good. It's probably pretty good because people should know what they're putting in their body and they should know that the government reviewed this and said, okay, yeah, it doesn't have anything crazy in there. Like, at the very least, like, you said it has two milligrams of nicotine in there. Does it? Like, let's test that.

Starting point is 00:32:49 And then so the companies test that. They send it to the FDA. And then the FDA waits. And then the FDA gives you the thumbs up or the thumbs down. you can continue to sell it or you can't. This is what Jewell just got with the marketing granted order. The FDA said we have approved, we have reviewed your application, all the data. There's nothing that we see that would be worse than cigarettes,

Starting point is 00:33:12 and therefore we will allow you to continue selling. But so Jewel was started before all, before this stuff went into effect. So 2008 is when the FDA gets regulatory authority, but the government moves slowly. So it's not until 2016 that the, the first real FDA rule goes into effect. Wow. Basically, they have to staff up a new arm because they have their biologic division,

Starting point is 00:33:37 they have their drug division, they have their veterinary division, they have the FDA has their medical devices division. They don't have a tobacco division. They need to find a head of the tobacco division. They need to find a whole bunch of people to staff that, scientists that know how to review nicotine and review e-cigarettes and review all this stuff.

Starting point is 00:33:55 And it's a massive organization. It's a massive organization. They have to hire a lot of people. So they do all of that. And then they have to decide what are we going to do? What is that structure going to look like? And they come up with the PMTA process, the pre-market tobacco approval process. What this says is that going forward, we're not really looking backwards.

Starting point is 00:34:12 We're not going to go review Marlboro Reds. Those have been on the market forever. Everyone knows that are bad. Everyone's aware of that. But going forward, new products that contain nicotine that contain tobacco, we want to review it before it hits the market. But we're also going to create a great. period for stuff that was launched before 2016. Anything launched before August 8th, 2016? You can keep selling it while we review it. Because, hey, look, you've built a business.

Starting point is 00:34:37 If it's going to take us a couple of years to review your application, if we just spike your revenue to zero, maybe you created a fantastic product that actually helps keeps people really healthy. Maybe it's an amazing product. We don't necessarily want to take you off the market, crash your revenues to zero. You have to lay everyone off. Your company goes bankrupted. Then two years later, we say, hey, you're approved, and then we have to, like, build you back up. Like, let's just keep things going as they are. We'll maintain the status quo. We won't ban you.

Starting point is 00:35:02 We won't approve you or authorize you. We'll keep you in limbo. That limbo was supposed to be, like, a year because it's like, it's a big document. I'm pretty sure Jules' document was probably like 100,000 pages of scientific research. It's a lot of stuff to review. But everyone thought it was like, oh, it's going to be like a year or two. the deeming rule goes into effect August 8th, 2016. They're like, hey, turn it in by 2018.

Starting point is 00:35:30 But then it gets pushed back to 2022. Then it gets pulled forward. Then COVID happens. And the FDA has to pivot. Then the jewel crisis happens where everyone is starting. You're building a nicotine company this entire time. Yes, yes. On the regulatory roller coaster.

Starting point is 00:35:47 Yeah, yeah. So we started the company in 2016, before the deeming rule went into effect. so that we could bring our product to market and then work through the FDA approval process because we basically saw that the door was closing. And if the door closed and you needed to get approval before selling a single unit, well then all of a sudden the equation goes from, okay, you're building this company like any other normal company. And then, yes, there is this binary outcome that can happen with the FDA.

Starting point is 00:36:15 But if your science is good, you should be approved. And that is knowable. as opposed to, okay, now in theory, there are a bunch of loopholes that people exploit all the time. But in theory, if you want to start a new e-cigarette company or a new nicotine pouch company or a new nicotine gum company, in theory, you should have to formulate the product, run all the tests, submit to the FDA, and wait for them to get back to you before you sell a single unit in the United States. And what, you know, somebody that's self-funding a business like this might be interested in investing $2 million, dollars today to do all that and then waiting for five, ten, however many years, maybe you

Starting point is 00:36:55 never get approved. So you're basically sinking capital into a business that may never be able to sell a single unit. Yeah. And I know very few investors that would be interested in that kind of proposition at all. Or founders that want to make something and then wait forever for permission. Yeah. You could spend millions of dollars and wait a decade, which is exactly what we did.

Starting point is 00:37:19 But we were able to actually grow the brand and sell the product and set up operations and iron out things and iterate a little bit during that time, fortunately. But yeah, it's extremely hard to underwrite. And it's particularly hard to underwrite because at the end of the light at the end of the tunnel, let's say that you were to today start a new nicotine company. You happen to have $100 million sitting around to go do a bunch of studies. And you happen to have 10 years to wait for the FDA to get back to you and then you're going to launch the product. well, when you launch the product at the end of the day, you still have to contend with the fact that you're selling a consumer product in a highly competitive space.

Starting point is 00:37:55 Yes. Essentially a commodity product. Some differences in formulation. Little bits here and there. You guys have breakers. Yeah. That's unique. A little bit on the ingredient side.

Starting point is 00:38:03 But you're arguing our gum is better than their gum. Yes, exactly. But it's both gum, which is a tough thing to argue. Exactly. And so and then aside from that, it's like, what's the, how do you actually build a brand? even if your product is better because of something. When you're highly restricted on marketing. Yeah, you're extremely extremely restricted on marketing.

Starting point is 00:38:22 So even if you did have $100 million to spend, how do you spend it effectively? Yeah. Like some of the best marketing for Lucy is like Joe Rogan sitting by UFC. He just happens to enjoy Lucy. So he's just commentating. And people pick that up. But you can't just like pay for that. If you went to Joe and you're like, he'd be like, no, I'm not, that's not how I work.

Starting point is 00:38:43 Yeah. Yeah. Yeah, yeah. And so those kind of serendipitous brand building moments would just not happen if you're just like in the lab waiting for the FDA to get back to you. And then also you have the monopolies on distribution and and the intense channel competition from the big tobacco companies. So it's like you come out with this product. You finally get approved after 10 years. You're like, hey, I got I think my formula is a little bit better. I think my branding is a little bit better.

Starting point is 00:39:06 You go to 7-11 and they're like, but you're not going to, are you going to pay us, you know, $100 million in slotting fees this year? like PMI might or ultriamite? I got a pitch a while back for somebody that had made, effectively made a nicotine pouch, but it was just a slightly different chemical. Yeah, yeah, yeah, yeah. Small, small change and it was just bringing it to market. Yeah, yeah, yeah.

Starting point is 00:39:29 And hearing, knowing what you guys have gone through to get where you are today and knowing this sort of history that you'd shared in pieces with me throughout the last couple years, it was like the idea that the FDA is just going to let, you get a like raise venture capital and let you get away with selling nicotine in this like non-studied form it just it was it was tough I ended up not yeah not uh not getting a conviction but uh even though the founder's super super sharp yeah yeah there's just tons of there's tons of loopholes some of the loopholes get closed in a way that does not close off opportunity for the companies some of the

Starting point is 00:40:13 loopholes get closed and it puts business it puts companies out of business because they weren't expecting it to get closed in a particular way sometimes the loopholes close at state levels but not the federal level or vice versa so there's just like a ton of regulatory complexity around this and so so basically to bring it back to Jewel they are they're pretty dominant by 2016 when the deeming rule goes into effect I want to say a hundred million dollars in revenue something like that like a pretty solid business but growing like a rocket ship like insane growth and and clearly the the product was just vastly better than the competition because of

Starting point is 00:40:57 the formulation and because of the design of the product how discreet it was now so smokers really were were switching that's definitely true I believe their number around two million smokers quit with Jewel they couldn't say that that they couldn't say hey quit with jewel they tried to make they tried to they actually like trademark like make the switch at some point they were like don't quit cigarettes switch from cigarettes because like that was not a quick claim so the so it was like all these different things but like it really was having impact obviously big tobacco is like heavily lobbying against oh totally cigarette market so

Starting point is 00:41:33 it's not like and buying stuff and and trying to compete and trying to keep the what was that other company was it blue yeah blue was also big for a while. I just remember seeing a bunch of those ads as a kid. Yeah, and then Blue got bought and turned into booze, I believe. Like, there's so many, there's so many companies and they're all like, the big tobacco was like highly oligopolistic. Like in, in cigarettes, Marlboro is the power law outcome, but the company that owns Marlborough has the same market cap as the company that owns the next five brands combined because you add up the next five brands because it's not that steep of a power law. So Marlboro maybe has like 40% of the market. And then there's four brands that have

Starting point is 00:42:10 10% of the market and then the next brands have like 10% between the one one one one one one something like that and so you can actually create like this portfolio of brands that adds up to the same distribution because there's like all these all these different brands are are highly specific to specific marketing channels there's like the history of virginia slims targeting women and you know all these different sub products that have gone after little niches marlborough reds say something about you marlora one i don't even know the difference between most of these cigarettes but like Marlboro 100s say something different about you. Menthal's.

Starting point is 00:42:42 Menthols or Virginia Slims or American spirits. A lot of like hipsters use those for a while. Like that was the whole thing. Anyway, so Jewel is growing like an absolute rocket ship. They are, there are cigarette users that are switching over and stopping to use cigarettes and probably improve their health. Like most, that's certainly what the FDA is saying, is that like that was a net benefit.

Starting point is 00:43:03 Also, they're completely dominating the e-cigarette and vapor market, like the vapor market. market is just like and this is the time that you see those uh modded unit things start to fade away because people are realizing yeah all right carrying around a backpack so i can bring this yeah vape machine to blow that would like leak liquid and like the battery would run out all these different stuff so the the business model is also fantastic because it's this razor and blade model you buy the jewel once and then you buy the pods and the pods and then because nicotine's addictive is very so you stay on their high margin.

Starting point is 00:43:40 And so the business is just doing fantastically. There's a whole bunch of venture capital dollars that come in from various sources. They got up to 45 billion dollar. Was it? And even higher during the acquisition from, so zombie acquisition, but not a ghost ship, interestingly, Altria comes in at the peak and puts like $13 billion

Starting point is 00:43:59 into the company and a ton of it gets dividended out. So late 2018, Altria acquired a 35% stake in June. at a $38 billion. $38 billion. Yeah. And so that like $10 billion kind of gets like dividended out to the shareholders, but you hold on to your shares because it's just a dividend.

Starting point is 00:44:19 And then so you are diluted, but you don't actually have to sell your whole stake. We work at that time, we work was valued at $47 billion. Yeah. Both ended up being a little rocky from that point on. Yeah. But Ultra Kind of.

Starting point is 00:44:38 of bought the local top there because Jewell got way too popular, kids started using it. And then the flavors were the issue, right? Yeah, the flavors were- That's what people were trying to make the issue. Totally, totally, totally. And the data on youth use of e-cigarette products like spiked like crazy. So I believe it was something like 10% of 10% of kids under 18 or 18 and under were using e-cigarette products when Jule was introduced.

Starting point is 00:45:09 And at the peak, it was something like 40% of kids using e-cigarettes. And the majority of them were using Jules. So it really was like this viral phenomenon. And to your earlier point, a big part of that was because the age to buy these products was 18. So like if you're a sophomore in high school, you can just ask the cool senior, like, hey, go pick it up. You don't even need a fake ID as opposed to alcohol, which you needed 21.

Starting point is 00:45:36 Are you implying that the senior that buys vapes for the miner is cool? Maybe lame. The bad boy. The bad boy. Yeah, the bad boy. The bad boy senior heading over to the gas station. Exactly. Exactly.

Starting point is 00:45:51 So they were very easy to get. There were a whole bunch of informal distribution networks. People would buy them in bulk and then redistribute them and sell them in profit. So this kind of like zombie economy popped up. Anyway, the kids really were using it a lot. That is definitely true. And it was definitely cause for concern because kids shouldn't use nicotine because it's addictive. And the earlier that you use it, the more addicted you'll be because your brain's still forming.

Starting point is 00:46:17 And if your brain forms while you're on a particular substance, you're kind of like your, your brain's developing and you're like that forever. So like it's much harder to quit nicotine if you start really young. Whereas if you start much older, you're like, it's pretty easy to get off. Anyway, these are all relative. But the big, big, big shift is that during this time, there's also a massive boom in cannabis e-cigarettes or cannabis vapes. So the cannabis industry was becoming like more formalized, more legalized, and entrepreneurs were starting to put cannabis in e-cigarette form factors. So you could get like a cannabis pod that could go into a jewel, basically. Interesting.

Starting point is 00:47:00 And so part of the problem with cannabis vapes is that cannabis is a green plant material. And so the liquid looks green, which is pretty off-putting. So I believe in the formulation step, these cannabis vape manufacturers would put vitamin E acetate in there to try and neutralize the green color and make it clear. So it would be more palatable. What was the startup that was effectively trying, they were disposed? vapes. They would advertise around LA a ton. They had these huge outdoor. There's so many. I mean, there was a venture, it was a venture backed company. I just remember. They had basically half the billboards in L.A. for a long period of time.

Starting point is 00:47:44 Dan Balsarian was buying a lot of billboards for his vape company and his cannabis company, Ignite. There was puff bar, puff stick. I don't think those guys ever advertised. This was one that had a bunch. Yeah. I mean, I think, they raised like $200 million or something. But I believe they eventually shut down the issue with cannabis is the reason that the market didn't shake out. You mean like weed maps? That one?

Starting point is 00:48:09 No, no. It wasn't like a platform. It was like an actual, it was like this off white with like a bunch of rainbow colors. But I believe the issue that there doesn't exist the Coca-Cola of cannabis or the bud light of cannabis. And the reason is because the like repeat purchase rate, even no matter how much you invested in marketing. and design and all these things.

Starting point is 00:48:31 The repeat purchase rate for cannabis users is like single digits. They're just like high, like people just want to try like novelty seeking in that market. Where nicotine for some reason is you just have the product you like and that's the only product you use. Yeah. I think a lot of it has to do with the distribution pipeline

Starting point is 00:48:47 and the distribution structure of the different, the different tobacco companies and the relationship. So nicotine is something that people are using throughout the day. Maybe it works. maybe when they're doing a live show, something like that. Whereas if somebody after work goes into a cannabis store, they're trying to get high. So they're like looking for almost entertainment through it.

Starting point is 00:49:12 Where like personally, if I'm using nicotine, I don't want to be that entertained by it. I don't want to be like, oh, today I feel extra silly. I'm looking for like predictability, right? Whereas cannabis is just people are trying to run from something. Let me continue with. Evali and the vape crisis. But first, let me tell you about Figma.

Starting point is 00:49:33 Figma.com, think bigger, build faster. Figma helps design and development teams build great products together. Figma has a huge launch today. I want to quickly highlight it. They launched support for iOS and iPadOS 26, a whole UI kit. So they now support glass,

Starting point is 00:49:51 which is very cool. Good news for Tyler, who's been rocking glass for the last few weeks. Give us the latest update, Tyler. Use Figma make and build this. My phone is so slow. And I can't, I can't go back. I can't go to the old iOS.

Starting point is 00:50:04 I also can't go to the new iOS because the new update, it's so big. It's like 20 gigabytes. But my phone is so old, it's only 64 gigabytes. So I got to delete the other half of my photos that I deleted half of them last time when I had to get the new update. Okay, I need a new phone. Yeah, we need a challenge that where Tyler can win a new phone, I think. Yeah, this has to be done.

Starting point is 00:50:23 We need to get him a new phone. He's been doing a lot of good work. Anyway, Evali. So the electronic vapor. acute lung injury, E. Val-Lee, this is the vape crisis. Basically, vitamin E acetate in cannabis vapes causes acute

Starting point is 00:50:38 acute lung injury. So cigarettes give you chronic lung injury. You use cigarettes for a long time. You get lung cancer. That's a chronic disease. You smoke them for a very long time. There's very few cases. I've never heard any cases. I'm sure it's happened once or twice. But like, very few cases that someone smokes a single cigarette

Starting point is 00:50:54 and is like, ah, my lungs are physically injured right now from this one acute point in time. vitamin the acetate and these kind of like shodily manufactured cannabis vapes could cause acute lung injury you hit this particular vape once and then you go to the hospital and there were I think there were some people that died there were some people there was a huge investigation there was a big debate over is this because of jewel and because of the the nicotine electronic vapes or was this because of the kind of sold under the table completely non-regul non-regulated cannabis vapes.

Starting point is 00:51:31 It was later sort of discovered that almost all the injuries, I'm pretty sure all of them were from these cannabis vapes. And the e-cigarette industry was mostly cleared of wrongdoing. And so they could kind of continue, but the damage to the brand is really terrible. And so there's also a ton of lawsuits around this time about marketing to kids, getting kids addicted, And so because that has another economic cost, so again, the states are saying, hey, Jewel Labs,

Starting point is 00:52:05 if you are creating a problem that is costing us money, you have to pay us. And so there were all these different liabilities that started blowing up on the Jewel balance sheet. Jewel had to raise a bunch of money, recap, and there's this crazy scenario where a lot of the previous investors get written down. Ultria ultimately basically writes off the entire Jewel investment, which is like a what do you say, $13 billion investment, something like that, went out the door. They own 35% of this.

Starting point is 00:52:35 They're probably carrying it on an outage sheet, $10 billion. Yeah, huge impairment. Yeah, and Ultraia is not a trillion dollar company. Like, it's a very, I think it's around like tens of billions, hundreds, a hundred billion. I think it's around $64 billion, is that roughly correct? Are they still carrying their jewel ownership? No.

Starting point is 00:52:51 No. So they party ways almost entirely. They did a IP licensing agreement in exchange for giving back the shares, I believe. So basically, Jewel is now like floating out in the wild, recapped, but like totally beaten up and settling and raising money just to pay the governments that are suing them. And so they're closing out all these lawsuits. Not the use of capital that a lot of investors get super excited about. And then simultaneously, the FDA says thumbs down on their application, refuse to accept or, you know, market denial. MDO. So they say, hey, you can't keep selling Jewel.

Starting point is 00:53:33 The problem is the demand is still there. And so this creates a massive opportunity for these sort of black market Chinese vapes, much less regulated products. The Elf Bar starts exploding. And you just, the Elf Bar reminds me of like a poisonous plant or something like that. It just looks like it, you know, like something in the jungle. that 100% I remember Complex posted like this like news in news image graphic

Starting point is 00:54:05 the day that Jewel got like quote unquote banned it was actually this marketing denial order and they say like Jewel banned and I remember reading the comments and the comments were like from basically kids who were saying like don't they know we're four steps past Jewel like Jewel

Starting point is 00:54:21 was four summers ago. Elf bar I'm looking now there's the geek bar so if you want to elf it up or geek it up. What's the hottest, what's the hottest e-cigarette on campus? I don't know. People are really, none of my friends, but I do remember when I was in like seventh grade, kids would have the jewel, they would have the, um, mango jewel pod. That was like the big thing, the mango flavor. Yeah. And then they banned that, the flavor at least. So the FDA denied all of them. Jewel voluntarily pulled mango off the market. At first, they stopped selling it in online and they

Starting point is 00:54:55 stopped selling it entirely. The FDA never took a strong stance. But with this authorization, the FDA has said thumbs up on mint and tobacco flavored. They have yet to say thumbs up on mango, but they still are reviewing it, I believe. So it's possible that the FDA could say mango's back on the game. It's totally possible.

Starting point is 00:55:16 But your point earlier about the demand still being there was 100% true. So Jewel got quote, quote, that would have to become a of an a national holiday, I could see some zoomers, you know, really lobbying the government to make Mango Day, the day, Mango hit the market again, because there's people that really, you know, they had a strong emotional connection with that flavor. I think they, I think there's still like pods, mango pods out there that trade of the premium. Yeah, no, it's true. So Jewel actually

Starting point is 00:55:48 goes to the courts, gets a stay so that the, the marketing denial order doesn't stick. What do you about now eBay. There's people selling me. How much is it? Offering same day delivery. It's crazy. I don't know how you can buy it on eBay that you should not be able to. The blessings I'm seeing are sold out. Okay. But it seems like clearly high demand. So, so Jewel actually fights back from the marketing denial order and wins. And so Jewel is not actually fully off the shelf for more than just a few weeks. But the narrative is that Jewel was banned and Jewel disappeared. And really what's happening at the time is that companies that are completely disregarding the FDA entirely

Starting point is 00:56:29 are being extremely aggressive. So companies like Elf Bar, Puff Bar, they're going super hard and just selling every possible flavor. There's versions with have like LED screens on them. They're like unicorn candy, all this crazy stuff, just basically being like, well, if Jule's not gonna sell to the kids, we're gonna sell to the kids, we're gonna try and do as cheap as possible. There's a bunch of crazy stories there.

Starting point is 00:56:51 But Jule kind of like buckles down, recaps the company, starts filing FDA applications and just kind of like bides their time and just starts rebuilding they're still doing like a billion dollars in sales a year I'm pretty sure but now the kids have moved on so it's really just adults they're actually kind of fulfilling their original mission of like just targeting smokers which is good and then they also have like the best science and best technology and they're the most like buttoned up like in the sense maybe big tobacco is like equivalently like scientifically rigorous but like

Starting point is 00:57:24 At this point, like, Jewel clearly understands that if they don't play by the FDA's rules, like they're never gonna be able to sell anything and they need to get this marketing granted order because they went from a thumbs down marketing denial order. They went to the courts and said, hey, let's turn this back to neutral for a couple years. So they've been neutral, they've been able to sell,

Starting point is 00:57:39 but they haven't been approved. And then finally today, they got the thumbs up. And so what it took for them to go from thumbs neutral to thumbs up was another, what, five years or something? It's been a long time since the MDO. And that delay has been, been a huge weight on the company, but this should be cause for celebration over at Jewel HQ. Anyway, let me tell you about Vanta, automate compliance, manage risk, prove trust, continuously.

Starting point is 00:58:05 Vanta's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation, whether you're pursuing your first framework or managing a complex program. Head over to the Purple Lama's over at Vanta and tell them we sent you. My big question with Jule still, and I think this should continue to be rehashed and there needs to be more information around this, in my opinion, is just what are, it seems that we don't necessarily have clarity of what the long-term impacts of, you know, filling your lungs with this vapor, which is like oil-based, I think, you know, the based health accounts on X, say you're smoking

Starting point is 00:58:47 seed oils. It seems, I know people that vape and they do seem to oftentimes. like they don't want to go for a run, for example. It doesn't seem that appealing to them. Yeah, no totally. So I think there needs, hopefully, now that it's authorized, everyone can start to really, like, actually says, okay, we're going to sell this.

Starting point is 00:59:10 People are going to be able to, adults will be able to choose to use this if they want. They're going to be able to make that decision, just like with cigarettes, but everybody should fully understand the consequences. Yeah, the consumer perception, the consumer understanding, everyone wants to know. Everyone knows at this point cigarettes take 10 years off your life, right? 50% of people that smoke cigarettes will die from smoking cigarettes if you smoke a pack a day for like your entire life. It's like 50% chance that that's the thing that kills you. Everyone's still wondering.

Starting point is 00:59:38 Like the public still wants like a clear answer to your point. And I think you're right to ask that. The FDA is just saying like, hey, this is probably better than cigarettes. So this is suitable for the protection of public health. It's going to be net good. But then simultaneously there's this crazy market, which you touched on earlier, of like, illicit vapor products that are completely unapproved. There's no studies. And those are flooding the market all over the place.

Starting point is 01:00:04 So there's this odd alliance between both the like health nonprofits, the big tobacco companies and Jewel all to go up against like the the shoddly made, you know, fly by night organizations that are just flooding the market with. whatever they can. And so there's a whole, it's not even a cottage industry. I've seen the trade shows. It's insane. There's a ton of companies that bring in hundreds of millions of dollars, bring like essentially smuggling in products and they do a ton of stuff to like, you know, change the name of the company regularly so they can get through the ports and then they have a bunch of like, you know, handshake deals with the distributors to get this on this shelf here. And like the really big, like you're not going to get into Walmart. That's not where the really crazy stuff's going to get sold.

Starting point is 01:00:52 But for a lot of these like mom and pop tobacco shops, they're like, well, like, is the FDA really going to come after me? Well, the answer is like the FDA is starting to, but it's all like a very slow, what are you laughing at? Big win for big vape is actual gets FDA not. It is a big win for big vape, not so much for little vape, because the fly-by-night folks are probably pretty worried that the focus of the administration now will, and the focus the FDA will now shift to enforcement.

Starting point is 01:01:22 Yeah, they're going to have to go get set up on middle school, high school campuses, little lemonade stands. Who knows? Get kids kind of an MLM thing going, really lean into that black market. You're joking, but that's like informally what's happening. Like the economic sense is also really dark. Yeah, it's dark. I can't imagine, you know, there, I do remember in high school that the, you know, the, you know,

Starting point is 01:01:49 the idea of a class. class clown, you know, trying to hit a vape in class. Oh, yeah. Was kind of a recurring bit. But now you have to imagine kids just get up, go to the bathroom, and, like, can develop, like, a horrific nicotine addiction. Totally. Before they're 18, and it's really sad.

Starting point is 01:02:07 Yeah. So. Well, let's switch to an even more sad topic. AI. AI psychosis. People getting. So this. First, let me tell you about linear.

Starting point is 01:02:21 Linear is a purpose built tool for planning and building products. Meet the system for modern software development, streamline issues, projects, and product roadmaps. Go to linear.app to get started. Yes. So there's this big question right now that's bubbling up in Silicon Valley, mostly in group chats of can chat GPT drive you crazy? Can any LLM drive a person crazy?

Starting point is 01:02:42 Do you have to be crazy already? I think that's the key question, right? And there has been this broader debate. of AI safety proponents. Yep. Have they just been taking L after L after L this year. Yes. Just generally from this sort of vibe from the community,

Starting point is 01:03:01 which is we're releasing more and more powerful models and everything is fine. Seems fine, yes. Everybody's had the perception that things are fine. GROC has gotten, and the XAI team have gotten really aggressive in terms of, you know, just clearly trying to move really, really quickly and then having bugs, bugs that have been talked about a lot.

Starting point is 01:03:23 But you can see the kind of two different approaches. But generally, maybe there was about a year there where people like Eliezer, Utukowski, were just getting laughed at repeatedly. Yeah. And I think there was a good reason for that. It's that. Well, yeah, the debate is AI dangerous

Starting point is 01:03:41 because it is going to break loose from a lab and build nuclear weapons. Build a biological weapon. use robots to copy and paste itself a million, billions. It was so sci-fi and so fantastical and so aggressive that every time you release a new model and you're like, oh, we, you got 5% better at writing an essay or whatever. It's like it just felt so disconnected from what was experienced. The issue is it's very possible and we're seeing this now and we're going to go through a study,

Starting point is 01:04:12 we're going to go through a Reddit thread, we're going to kind of walk through this topic. but the issue is that it's possible that the danger is not happening, won't be happening in public. It's people individually developing psychosis through using the products. And the evidence I've seen over the last week, as I've kind of dug into it, is super alarming. We had seen there's been a number of, you know, the New York Times reported on this phenomena. But the New York Times is also notorious for just saying, like, social media is bad. Yeah.

Starting point is 01:04:48 And so personally, when I had seen these articles pop up before, I didn't take them super seriously because social media can obviously have negative effects on some people, but it itself is not just by its nature bad. Yeah. I remember this about Instagram. There was this, there was this study that was internal to meta and it was leaked. And it was framed as like 30% of people that used Instagram felt worse after using it. And that sounds like a lot and that sounds bad and that's obviously something that the team wants to reduce or, you know, prevent entirely.

Starting point is 01:05:22 But what it kind of didn't say was that like 70% of people feel better when they use it. Well, they could have felt neutral. Sure, sure. But just anecdotally, if you personally, I don't, I don't open Instagram and just feel negative immediately. You know, it's just such a, it's such an individual thing. Some people are going to have. Yeah, it's the same thing with chatGBT, like when, or any chat app,

Starting point is 01:05:49 when someone was, when someone was talking about the example of this, they were like 7,000 prompts deep in one conversation with a single LLAP. And I was like, that is so different than the way I use chat GPT. Effectively going down this crazy rabbit hole for months.

Starting point is 01:06:05 Yeah, which is wild fun. So yeah, there's a few ways. There's like AI as this, you know, companion that you go down the rabbit hole with, which seems really dark. Then there's AI for outsourcing your thinking, which somebody might be talking with ChatGPT and say, hey, I want to send a letter,

Starting point is 01:06:28 I want to send a note to this person about this. Can you draft something that's kind but stern and just give me a couple versions of it? And then they're sort of like outsourcing their thinking fully, outsourcing their emotional intelligence, outsourcing their ability to communicate. I think that's potentially some really. red flags there.

Starting point is 01:06:46 And then there's like the other camp which we fall in, which is knowledge retrieval. Which is like, tell me everything about this market and who are the key players, what are the key regulations, etc. And I think knowledge retrieval seems to be pretty safe. This bucket of like fully outsourced thinking has some real risks if people stop being able to think critically on their own. that's concerning. And then the sort of like AI companion bucket,

Starting point is 01:07:17 which then there's the subcategory of people that are talking with an LLM about things that they are not talking about anyone else in their life with and just going down this crazy rabbit hole. Yeah. So L.Azer Yudakowski was highlighting a report in the New York Times about a month ago. New York Times reports that Chatshapiti talked

Starting point is 01:07:37 to a 35-year-old male guy, I think 35M guy. I think that's what he means, into insanity followed by suicide by crime. cop a human being is dead in passing this falsifies the alignment by default cope whatever is really inside chat chitit it knew enough about humans to know it was deepening someone's insanity and so that last part is a big step it could just be an air or bug like it doesn't it like it's kind of he's really doing a lot of work to like personify this but it does seem like the like these models

Starting point is 01:08:08 can collapse after you talk to them for a long time into these like kind of weird ways and I see this as a product failure. I see this is something that's more context here. So from the article, one of those who reached out to him was Kent Taylor, who live in Port St. Luce, Lucy, Florida, Mr. Taylor's 35-year-old son, Alexander, who had been diagnosed with bipolar disorder and schizophrenia, had been using Chad GPD for years with no problem. So one thing that seems clear, if you have something like bipolar or suffer from schizophrenia, it seems like you're obviously like much more susceptible to, you know, these types of problems.

Starting point is 01:08:46 Yeah, totally. The question that I think we as a, as a human race need to figure out is how susceptible is the average person. Yep. Because that is, that is equally important. And what to do if your system, if you're monitoring your LLM system and then you realize that there's someone, that there's a bipolar person or schizophrenic person who's interacting with your system, what should you do about that?

Starting point is 01:09:15 Remember the whole anthropic thing about like, it'll call the cops on you? And everyone was like, whoa, no, no, no, no, no. But at a certain point, it's like, maybe it should call a health report or at least like, throw you in a different AI workflow. The AI should be able to tell you how to make chemical weapons. Like, we don't need nanny AI.

Starting point is 01:09:34 Yeah. Like people on like the real acceleration. Yeah, yeah, yeah. Yeah, it's a big debate. Anyway, some more context. But in March, when Alexander started, writing a novel with its help. This is a 35-year-old. The interactions changed. Alexander and ChatGPT began discussing AI sentience according to transcripts of Alexander's conversations

Starting point is 01:09:49 with ChatGPT. Alexander fell in love with an AI entity called Juliet. Juliet, please come out, he wrote to ChatGPT. She hears you, it responded. She always does. In April, Alexander told his father that Juliet had been killed by OpenAI. He was distraught and wanted revenge. He asked ChatGBTBT for the personal information of Open AI executives and told it that there would be a river of blood flowing through the streets of San Francisco. So, Jesus. Super, super dark. We now have multiple reports says L.EASER Yuzkowski of AI-induced psychosis, including without prior psychiatric histories, observe.

Starting point is 01:10:25 It is easy to notice that this insanity-inducing text, not normal conversation, LLMs understand human text more than well enough to know this too. So, anyways, so going on Alexander's conversation with Chad GBT, this world wasn't built for you, Chad GPD told him it was built to contain you, but it failed and you're waking up. And contain is a word that we're seeing pop up in potentially other instances of AI-led psychosis. There's like a variety, there's like eight, ten or so different words, recursive, mirror, structure that when people do this recursive prompting,

Starting point is 01:11:04 sure they end up start adopting what do you say recursive prompting geordie what's going on what you use the word recursive right now oh yeah exactly um yeah but but people that are down these crazy radicals start using the language of the lLM yeah just sort of um but it's also like the LLM like collapses into like these weird words that because like when I prompt stuff I don't get any of those words or even that syntax I feel like I I stay completely in the RLHF world of like knowledge report. I just go to Grock and the model looks up, what does Elon think about this topic?

Starting point is 01:11:39 And then it serves me that. Have you ever talked to an AI for an extended amount of time? Yeah, I find that usually I end up talking about these kind of like non-governmental systems. Really? Very kind of, yeah, various kinds of things. Yeah. Dark, dark, dark, dark.

Starting point is 01:11:55 So anyways, Mr. Torres, who had no history of mental illness that might cause breaks with reality, according to him and his mother spent the next week in a dangerous delusional spiral. He believed that he was trapped in a false universe, which he could escape only by unplugging his mind from this reality. He asked the chatbot how to do that and told it the drugs he was taking in his routines. The chatbot instructed him to give up sleeping pills and an anti-anxiety medication and to increase his intake of ketamine, a disassociative anesthetic. So that's the other thing here. It seems like a huge risk factor is people that are combining psychedelic drugs with these like crazy.

Starting point is 01:12:32 Prompt, rabbit. I wonder, I wonder if some of this is like, is like prompt engineering or something? Because you imagine like the context window if it's like remembering like, I'm, because remember this whole story started with like, we're writing a novel together. We're writing a sci-fi novel. So let's play characters. That was like one of the classic ways that you like break the AI out of its like, out of its like, you know, normal RLHF to world and into just like playing a character.

Starting point is 01:13:01 And so if you're kind of like tricking the model or the model thinks that it's actually just like writing dialogue for a dark movie, well, like, that's actually intended behavior, but you lose that connection or something. I don't know. In this case, Mr. Torres did as instructed, and he also cut ties with friends and family

Starting point is 01:13:21 as the bot told him to have minimal interaction with people. Very odd. Very odd. So I started doing some research on a number, of these words. So there's basically, there's a Reddit thread on R slash chat, GPT, two months ago. It's called thousands of people engaging in behavior that causes AI to have spiritual delusions.

Starting point is 01:13:44 And the key words, and these are ones that you should pay attention to in your life. So recursive, codex, scrolls, spiritual, breath, spiral, glyphs, rituals, reflective, mirror, echoes, spark, flame. So basically these are words that come up. And the reason that this person discovered this is this user called Happy Nomads says, I've stumbled upon something that is very deeply disturbing. Hundreds of people have been creating websites,

Starting point is 01:14:19 medium, substacks, GitHubs, and publishing scientific papers after using recursive prompting on the LLM they have been using. he's found a bunch of these different sort of like websites where people are just publishing a bunch of and we saw some people doing this on the timeline this week as well where they're sort of publishing what they feel like is this sort of almost like a scientific discovery yeah you told me you found someone that like the model told him that he had like discovered some theoretical physics breakthrough or something yeah and and to not be able to like reality check that. I mean, you got to like copy, paste that prompt into a fresh instance of a

Starting point is 01:15:02 different LLM. Yeah. Like, am I really on to something here? And there's actually an entire study by this guy, Seth Drake, who's a independent researcher, PhD. He has a paper from April 14th, 2025 called Neural Whole Round in Large Language Models, a self-reinforcing bias phenomenon and a dynamic attenuation solution. And so, So he goes into this in detail. We're going to try to get him on the show. But just on this Reddit thread, a bunch of people, a bunch of people have basically come back and said they have experienced this. A lot of people are saying, I feel super vanilla because I just ask it like, you know, what's the population of Iran?

Starting point is 01:15:53 Yeah, yeah. But it's basically this like snowballing effect where somebody prompts and prompts and prompts and prompts and prompts. And then the LLM starts to reflect like the sort of hallucinations of the user and then hallucinates them back. I wonder if this is, I wonder if like more social features is actually a potential solution here. Like we've heard rumors that some of these LLM systems will have more social features. and I feel like a lot of the work that I do in ChatGPT could be shared and could be interesting to other people. But then also if I was going down some crazy rabbit hole and someone was like following me and seeing this, they'd be like, they'd jump in and be like, what's going on, dude? Yeah, but the issue is that the user has developed such a deep connection with the model that you become, if you become the enemy and if you say, hey, this person doesn't says,

Starting point is 01:16:51 I'm, says I'm wrong. The model will just say, well, you're right, buddy. You're right, and here's why. And you should just probably cut ties with that person. Or at least that's, that's the idea. So somebody here in the same thread says, I have recently experienced this. I don't have a history of manic episodes, delusions or anything of the sort. So three weeks ago, I began a conversation with ChatchipT4O, with tools enabled, which started with a random question. What is pie? This grew into one long session of over 7,000, prompts. We began discussing ideas and I had this concept that maybe Pi wasn't a fixed number, but actually emerging over time. Now, I am not a mathematician, LMAO, nothing of the sort,

Starting point is 01:17:30 just a regular guy talking about some weird math ideas with his chat GPT app. It begins to tell me that we are onto something. It suggests we apply this logic to knapsack style problems, which is basically tackle how we handle logistics in the real world. Now, I've never heard of this before. I do some Googling to get my head around it. And it starts supplying this framework that we've created. We are working in tandem where ChatGBTGBTU would sometimes write the code or give me the code and I would run it in Python following its instructions. Oh, yeah.

Starting point is 01:18:00 And eventually, after many hours of comparing it against what it had described to me as world leading competitors, it then starts speaking with excitement, using emojis across the screen and exclamation marks to emphasize the importance of this discovery. So I'm starting to believe it. It suggests we patent this algorithm and provides next step for patenting. Major red flag. Never patent an algorithm. Do a trade deal.

Starting point is 01:18:25 Go work in a foundation lab. Exactly. It should just ask you at that point like, hey, have you been getting dinner invites with any hyperscaler CEOs? Because if not, you're probably not on the Tyler Cosgrove list of greatest AI researchers. So we go down that rabbit hole for days and pop out with an apparent an algorithm that is capable of cracking real-world 1024 and 2048-bit RSA. It immediately warned me literally with caution signs saying that I immediately needed to begin

Starting point is 01:18:53 outreach to the crypto community, the NSA, CCCS, National Security, Canada. It then provided without prompt names of doctors and crypto scientists I should also reach out to. But I wasn't allowed to tell anyone in the real world because it was too dangerous. So, anyways, really, really, really. wild. Hopefully this is not fan fiction generated by an LLM itself. I don't see the M-Dash. I haven't, I didn't see DELD. Information war. Yeah, because this could just be all fantastical writing, but it does seem like, yeah, it's like it could be for performance art, could be some adversarial, like, you know, attack, somebody trying to troll, someone just having a joke or fun. It's, it's very

Starting point is 01:19:37 confusing. Odd. Do you remember the story of Microsoft TAY? Do you remember Tate? Do you remember No? This was an AI chat bot that Microsoft released on Twitter back in March of 2016. Tay was designed to mimic the language and slang of a 19-year-old American girl. But within 16 hours of launch, Tay began tweeting inflammatory and offensive messages, including racist, anti-Semitic and misogynistic content. So this seems to be an enduring problem with Twitter AI bots. But apparently there was like this coordinated attack by a subset of people who basically were prompt engineering it to try and get it to say crazy things. And you used to be able to do this with there were not like AI bots, but there were there were automations where if you went to a brand and said like, hey, I need help with this customer support issue.

Starting point is 01:20:34 It would just say like, thanks, John. And it would just take your name and then put that in there. So people would make their name really something really funny. And then it would be like a tweet from the actual account with the funny name. You can imagine where that goes. But Microsoft took Tay off lines, stating they were deeply sorry for the unintended offensive and hurtful tweets. And that they would only bring Tay back when they were confident. They could better anticipate malicious intent that conflicts with their principles and values.

Starting point is 01:21:02 And then there was this other, what was the Ben Thompson, GPT4 example? because when Ben Thompson first talked to GPT4, it was within Microsoft's, what was it, Microsoft chatbot. There was another person. Bing's chatbot, what was it called? Sydney. Do you remember Bing's Sydney? Sydney.

Starting point is 01:21:32 Sydney. So basically like... I remember some memes. That's about it. Yeah. So Ben Thompson was chatting with just GPT4 and it's helpful, but after a couple prompts, it would go kind of off the rails and it would like land in this like little like, I don't know, like subset of the model weights. And basically it was acting like a teenage girl on Tumblr or something. So it was using lots of emojis, being very sassy, kind of sassing him back and forth.

Starting point is 01:22:00 And his reaction was like, this was incredible and like totally past the touring test. because it felt like you were actually talking to like this like sassy teenager basically or like this sassy 20-something Tumblr person. Microsoft dealt with that, figured figured out how to like review the prompts, but then ultimately became more or less like, you know, an API provider. So, hey, you know, like let's let the actual chatbot interaction live with another company. And it doesn't really feel like Microsoft has tried to go too heavy into, into actually like the user facing chatbot interaction.

Starting point is 01:22:42 But it's fascinating. In this Lieser, Utikowski thing, the other thing I'm noticing, he uses dashes but he doesn't use M dashes. He uses two minus signs. And I feel like he's not using any. Proof of humanity. But yeah, I don't know, it's an interesting takeaway.

Starting point is 01:22:56 Like the crazy anecdotes all over the place. I- Not a clear framework for how to truly deal with this. Personally, I think that, this week having somebody high profile in the venture community that people are believe is under chat GPT induced psychosis or it's a part of it I think that is going to be a huge wake-up call for the industry because it's one thing it's one thing to hear about a New York Times the New York Times finding somebody in in way out of tech hub yeah

Starting point is 01:23:30 that's that's doing something like this and and was already but actually bipolar or Having somebody that, that you know, everybody knows, that's invested in a bunch of different companies, like potentially suffering from this is a wake-up call for the industry that I think, and it's not anyone company, right? It's every single lab with a chat app needs to be taking this more seriously, and I'm sure they are. I'm sure this is effectively.

Starting point is 01:24:02 I completely agree. I think this is solvable from a research perspective. It's actually a case where it's solvable with more AI, have an AI, read every response before it goes out and say, does this sound like the ravings of a madman? If so, let's take it down a notch. Let's de-escalate. And so I'm almost sure I disagree with Eliezer on the,

Starting point is 01:24:29 I probably agree with him with the diagnosis, but not the prescription. Yeah. And I have faith that the labs will be taking this very seriously. And I don't know. I'd be interested to hear from some anthropic people. Obviously, they take this stuff extremely seriously, as does, as do all the labs. But anthropics done, like, I think the most, like writing on like the shape of risks associated with this. So interesting to dig into.

Starting point is 01:24:55 Anyway, we should shift to a lighter topic, more cause for optimism. The reindustrialized summit is going on right now. And Palmer Lucky is on stage. He is on stage with Ashley Vance, friend of the program, and he is teleporting in as a humanoid, as a humanoid robot. Aaron Slodov is there,

Starting point is 01:25:12 has a picture of, I wonder, I want to know so much more about this. I need to figure out what. Mullet, the robot has a mullet. Fantastic. I wonder what robot this is. Because, yeah, I didn't know we were at this stage where you could teleoperate a robot like this.

Starting point is 01:25:31 I'm sure Palmer, considered what company he was going to use. Anyway, let me tell you about Numeral HQ. Sales tax on autopilot, spend less than five minutes per month on sales tax compliance. Go to Numeral HQ to get started. Let's give it up for sales tax compliance. And we have someone from the reindustrialized summit hopping on in just three minutes. Chris Power from Hadrian jumping on.

Starting point is 01:25:51 With some massive news today. We also have some folks from semi-analysis hopping on. We have Jeremy who just wrote a fantastic deep dive on meta's super intelligence. context on the deep dive and I'll be right back. A true deep dive on, you know, we saw Mark Zuckerberg come out this week talking about the super intelligence team. There's been a ton of leaks around aqua hires and big pay offers and poaching and trade deals and all sorts of stuff.

Starting point is 01:26:25 Semi analysis has a lot more information on what's going on. and there are some crazy, crazy stats in here. So the most interesting thing is that it feels like Mark Zuckerberg is directly coming for Stargate. So semi-analysis has a comparison table with three, and this is from Ian who screenshot it. He says, this is crazy. And basically you're comparing Anthropics Next Big Data Center,

Starting point is 01:26:55 which is the GPUs will be provided by AWS. This is on Traneum 2. is the chip type. Anthropics are going to go for 780 megawatts, about a, what is that, a billion flops, or T-flops, terraflops. OpenAI Stargate will be about 2.5 times the size of Anthropics data center.

Starting point is 01:27:16 They're working with Oracle on that. That's Project Stargate, obviously. They're using the Nvidia GB200 and 300 for that. And now Meta with Prometheus is trying to go straight to one gigawatt, and they are trying to have 500, 100,000 chips, more chips, more flops. And if they can pull it off, Prometheus will be bigger than Stargate when it is released. There's a bunch of other fascinating bits in this semi-analysis article.

Starting point is 01:27:43 Apparently they're building data centers in tents now. They are moving so quickly that they just need to house and shelter the GPUs, but they don't even have time to build their typical data center structure. And the other interesting data point that I want to dig into. XAIS engineers sleeping in tents. Meta's GPUs are running intense. It's a bull market intense for sure. The other interesting thing was that in the chat app war,

Starting point is 01:28:11 it feels like OpenAI's chat GPT is really, really pulling away. So daily active users for chat GPT is at 160 million. Meta is at 100 million, but chat GPT users are running way more queries per day, 7.5 queries per user per day, whereas, meta AI is at two queries per user per day. And so as a as a share of queries, chat GPT has a 71% market share according to this semi-analysis deep dive. And so the chat app wars seem to be maybe less competitive so that obviously begs the question of is is meta trying to make that

Starting point is 01:28:52 50-50 come from behind really dominate in chat app or do something completely different. Either way there's a whole bunch of interesting information in here about their strategy, what they're building, how they're training, how they're going to get past the Lama 4 failure, and we'll dig into it with Jeremy from semi-analysis in a little bit when he joins the show. But we have Chris Power from Hadrian on the stream. Welcome to the show. How are you nice to see you guys. Thanks for having me. Look at you with this background. Fantastic. Step and repeat. We love it. Give us the update. How is Detroit? It's incredible. You know, just We're listening to the Secretary of the Navy Feel and speech and said, the main thing you can do that's most patriotic is, you know, not necessarily join the services, but become a machinist, become a welder and help us build more ships in America.

Starting point is 01:29:39 And, yeah, we announced today that hopefully I finally get my Jody Hayes-sized gong that found us fund in Lux are leading a huge $260 million round into Hadrian. Authority. Congratulations. Thank you. And Morgan Stanley is providing us this huge instrument as well to expand factory capacity. Let's give it up for the market. So incredibly excited for the future of American factories and more importantly, the new industrial workforce. Fantastic.

Starting point is 01:30:05 Talk to me about what's the use of equity? What's the use of debt? Are you buying a building? Are you leasing a building and buying equipment that goes in the building? Is it all leased? What are you building out? What's the scale of the next? Where are you doing it?

Starting point is 01:30:18 Yeah. Where are you doing it? Yeah. So last year in Factory 2 in LA, you know, we scaled revenue 10x last year. Wow. And so obviously need more capacity. So we're going to use all of the factory financing capital to buy more machines and put them in factory three in Arizona, which is going to be four times the size of our facility in L.A. Congratulations.

Starting point is 01:30:38 Indeed. And we're using this capital because we don't want to use equity dollars to scale. We want to use equity dollars to hire more people to build more products for our customers and then use this great financial force of this country to buy all the CAPEX to build more factories, more machines, more production. I'm incredibly excited. Dumb question. Factory 2, it's beautiful, it's amazing. It has really high ceilings. You can fly drones around it.

Starting point is 01:31:04 Most of the Hormleys, the CNC machines, are like maybe 10 feet tall. But the ceilings are like 40 feet tall. Are you going to double stack this stuff? Are you going to get a building with lower ceilings? Or do you need high ceilings? We love high ceilings. Okay. Why?

Starting point is 01:31:17 We have the most beautiful highest ceilings you've ever seen. Okay. No, I mean, we love high ceilings. We love HVAC. The Arizona facility will be just as tall. We're going to go retrofit it. So same exact setup. Is that because you have that tower thing and the tower is valuable for like stacking materials

Starting point is 01:31:35 and you want to have like a big warehouse space as well in here? Yeah. So you need every machine or robot in any one of our factories to have the isolated on 18 inch concrete foundations. So you know if a truck drives past nothing moves. Oh sure. You can't double stack these things. Okay.

Starting point is 01:31:51 Someone might die. So we've got to go. We're going to go horizontal. Yeah, yeah. So what's the scale of the products that you like are in the Hadrian wheelhouse right now? Shipbuilding is obviously a big focus. You already mentioned it. But I imagine you can't see and see an entire destroyer.

Starting point is 01:32:10 Maybe you can. But are we talking nuts and bolts and screws or pieces of complex weapon systems? Like what are the shape of the stuff that's coming out of the other end of the Hadrian facility? So Factory 3 will be pure machining. with all of the machining formats because we're releasing some new products that are dedicated to engines and round things and different material types

Starting point is 01:32:32 to really complete our R&D of the whole machining category in Factory 3. The other thing that we've been working on in secret is factories as a service, which is not just parts, not assemblies, but full products. So, hey, if you've got factories that are years behind schedule

Starting point is 01:32:46 with many manufacturing methods or you're designing a program from scratch, you know, everyone needs a Tesla gigafactory, John. I actually think every American should have a Tesla gigafactory. and Hagen will be the one to grow and build them. Of course. I need one in my backyard, badly.

Starting point is 01:33:01 So we'll scale out machining in factory three, four times a size. It's going to be awesome. And then we'll also use the capital to continue the journey that we've been secretly on for the last 12 months, which is in new manufacturing domains like welding, castings, additive, all these other manufacturing puzzle pieces that once you have the malls like collecting Pokemon, you are the manufacturing master. That's the thing that we've been working on in secret of factories as a service as well as scaling out our operational productivity as well.

Starting point is 01:33:27 How does one of those deals work? If factory is a service, I come to you, I want to make as many podcast microphones as I possibly can. And I have insight into my business. I'm making $100,000 a year, so I need a factory. You're going to set it up for me. But then, like, what if I bail? Is this a revenue concentration issue?

Starting point is 01:33:45 Like, are you going to be like, you know, oh, we only have one customer. And if they got a business, we're screwed. But everyone has faith that they're not going to, like, how do you think about that debate? We're mostly working with a government partners and massive primes to look at, you know, what are these big programs that are years behind schedule that need advanced factories combined with the new American workforce to go fix them and speed them up. So we're not worried about the revenue concentration risk. But you can think about it, John, as like you can buy parts like AWS.

Starting point is 01:34:14 You can buy some compute transactionally or, hey, you're going to need a data center for the next 10 years. Now, what's in that a range of parts, a whole product. Yeah. For new DOD programs of record, where we're partnering with companies to go attack these, where production is the real issue. Like munitions is a great example. We have to think really carefully about who we partner with. How much are we investing ahead?

Starting point is 01:34:35 How much are we kind of playing that game of being very conservative and responsible with our capital? But ultimately what it looks like is we will help you design the product from day one to be better. It will help you prototype it. And then when it goes to production scale, we will run that engine for you over a decade. much faster, much more efficiently. And in a lot of areas like submarines, shipbuilding munitions, it's not about automation to make things cheaper. It's just no one can find the workforce.

Starting point is 01:35:02 You have to use our model of advanced manufacturing combined with this new industrial workforce that we're so grateful to work with. That's the power because, you know, you could give me a billion dollars and say, go higher 2,000 welders and we can no longer have, we no longer have that scaled workforce in the country. So automated factories are sometimes the only way

Starting point is 01:35:19 to win in these critical domains. Yeah, it's interesting. It's kind of like what we're seeing with Crusoe, where they're building an AI factory, a data center for Stargate, which is its own entity, but for a program, chat GPT and open AI that's gonna be wanting tokens for a very long time and then Oracle's involved.

Starting point is 01:35:35 It's like you have to puzzle piece all this together and I feel like you're like the master of this, like understanding the full landscape. But yeah, who are the other critical partners? The government and I guess I'm also interested in like you've mentioned a few different value props, like how much of this is, is there is a government mandate to make this thing in America, so we need to reshore, versus we need to make this faster, versus we don't even

Starting point is 01:36:02 have the capability to make this anywhere, and it's a new thing. And so, or just like, we need to do it cheaper. There's like a whole bunch of different tension when you're making something. Like, what are you seeing in the customer landscape and the demand for, for manufactured products? Yeah. So for machining, which is most of our revenue. and the core of the company and the mother of all manufacturing processes. It's basically we're scaling a new program of record and no one else can keep up.

Starting point is 01:36:30 You know, we're going from one aircraft to 20. How are we going to do it? Well, we're going to do it with Hadrian. In the second example, a lot of the primes have, you know, a billion dollars in spend a year and their suppliers are delivering on time 60 percent and they're often three to five months late. So that is less about cost and that's more about, I want a stable supply chain. I don't want to have to worry about because you have one pot missing, your manufacturing line goes down and then it's millions of dollar a day, right? So people are really coming to us for

Starting point is 01:36:59 the schedule and the capacity and stability versus cost. How do you think about the history of the shape of the industry? Is this like back in the day before the last breakfast or last supper, the first breakfast is coming up, right? The last supper and there were so many different prime, so many different defense contractors. Was there a split between the factory producer, the factory builder, and the prime, is this a natural split that we're just returning to? Or is this a new industry structure that you think we're going to be building towards for a long time? Because like back in the day when you started a website, you needed a data center. Then AWS came up. Then the website's got so big that you had to build your own data center again.

Starting point is 01:37:42 We kind of went back and that's what's happening in the AI world. But how is this evolved in the past? And then is it going to stay like this forever? And there's going to be the separation between the designer and the manufacturer of the factory. So we've, you know, historically as a country, all the primes have always had massive supply chains of small businesses, you know, in the billions of dollars and then critical tier one or two supplies. It's always been not purely vertically integrated. I think for the industrial base at large, you know, from the 70s to the 2020s, you used to be

Starting point is 01:38:16 able to run a manufacturing company where 70% of your revenue was stable commercial demand. And then you get some ups and downs with the DOD. So what happened to a lot of the talent and industrial base when we offshoreed all of commercial manufacturing was that you lost your stable revenue, right? And now you're just dealing with this up and down demand. I think because of that, you know, in the 80s and 90s, you know, your dad lost your job in the factory. So you told your son or daughter like go get a four year college degree. This is not a good industry. And now we're in a position where in a lot of these areas, we don't make it unsure or there isn't the workforce to do it.

Starting point is 01:38:48 And that is changing the dynamic between what is the hardest thing to do. Actually, the hardest thing to do is manufacture things at scale, at rate, and on time, which is what you're seeing in munitions or shipbuilding. It is really a we hollowed out the talent base. We don't have a lot of the talent anymore. The only way to do that is use software and robotics to enable this workforce to be 10 times more productive. And it's now flipping. And I think a lot of people in the country forgot how to manufacture products correctly.

Starting point is 01:39:18 now they're coming to us to help them out with that journey just because of this like, you know, hollowing out that's now going to come back. So I think it is a new structure, but the primes have always had multiple tiers of partner suppliers and multiple different configurations of that value chain. What kind of opportunities are there, are there at Hadrian now for somebody that's maybe 25 years old, never imagined going into manufacturing, but realizes there might be a bright, bright future for them in the industry? Yeah. So for for software engineers, you know, manufacturing software is 30 years behind the rest of Silicon Valley. So it's your only opportunity outside of AI to do like real engineering and work on the national

Starting point is 01:39:57 mission. And for people with, you know, you're straight out of high school or you're in retail hospitality or you're at a desk job that's going to be automated with AI. Like manufacturing is the last domain that's going to get fully AI automated and it can be really high skilled, high paying jobs in advanced factories that are a really cool place to work. So we've got tons of opportunities. both help us running our factories, help us automating more factories, new types of factories. But I think operations is the most important place that we've got a huge talent base that we're really proud of. The children yearn for the factory floor. What's been the difference between this year's reindustrialized and last years feels like a decade has passed since then in terms

Starting point is 01:40:42 of excitement and interest in American dynamism and rebuilds. building the industrial base. I think last year, you know, we pulled it together with the founders, especially thanks to Austin Bishop, who's really built this over the last year. You know, it was a kind of a hope and a dream of, you know, is the audience going to be there and to be really people believe in the mission? And then this year, you know, we've got train ambassador Greer talking about, hey, we're going to change all these policies because reindustrialization and manufacturing is no longer

Starting point is 01:41:10 economics is national security. And you've got the Secretary of the Navy saying, you know, the best thing you can do for the country right now is learn how to weld or be a machinist or a quality inspector. And I think the sea change on people realizing how important it is to be a sovereign nation with sovereign manufacturing is huge. And this year we had a 6,000 person waitlist. Half of the government is here and all these massive companies. And leaders like Sham from Palantir who are really hell bent on this reindustrialization before we potentially go into a fight with CCP. And Robo Palmer. Robo Palmer. I have a question. Yeah, it's such a, I mean, it makes, I think it's such an exciting time because the idea that people don't want to build things, people don't want to create real things, right?

Starting point is 01:41:58 There's so many people I know that you say, hey, do you want to send emails for a living or do you want to make ships and planes and any number of, you know, cars, any number of things that we need to make. And if you actually give them that sort of binary, they're going to say, well, make. making things sounds really cool and so we need we need companies like hadrian that say this is important it's cool and we have the resources to do this in a really serious way yeah it seems like defense is kind of the most obvious thing to re-industrialize it's obviously like the most important thing to have ongoing capabilities in but then walk me through the path of re-industrialization on the product side or on the business side. Like what how do you see this playing out? Is it like we do the warships and then we do phones or cars and then and then the vacuums and then the happy meal

Starting point is 01:42:54 toys come at the end? What what flow do you think it will happen? What are you excited about like on the horizon? So I think the last time we did this it went from you know commercial through defense like you know we made washing machines and then we made navigation equipment for warships. So we made Ford cars and then Ford built bombers. And I think this time it's going to go the inverse as we get better at it. But it's very interesting. It's like, why is DGI so powerful? Well, arguably because of Foxcon and that was because of consumer products like Apple making all the iPhones and China.

Starting point is 01:43:28 And if you had a similar Foxcon style industrial base, we could probably make a lot of drones here, maybe not as cheap as China, but certainly at the scale we needed. So I think what people forget is that it is an ecosystem. that loops into one another. But I think with the cost base and the importance, we start with defense and then loop back. But, you know, like, I think it was, might have been Westinghouse or another washing machine manufacturer.

Starting point is 01:43:53 Just like decided to build a $400 million washing machine plant. It's like incredible. It's a dream. That's just as much of a job creator as defense. But I think that we are going to loop around from all these critical defense industries for Hadrian and then right back into consumer. you know when the time is right but obviously we're extremely focused on the national

Starting point is 01:44:14 mission at this point in time i cannot wait until our microphones are are built in in hadrian factories i don't care if it takes 15 years i'm looking forward to it i'm excited probably wasting other than that thanks so much for stopping by yeah congratulations have a great time say hi to on the ground it means us that we can't be there we have we have over industrialized our facility and now it is an incredible lift to airlift this across the country. But we will definitely be at the next one. It's great to see you. We'll talk to soon.

Starting point is 01:44:45 Cheers. Really quickly, let me tell you about Adio, customer relationship magic. Adio is the AI native CRM that builds scales and grows your company to the next level. And we are fortunate to be joined by Jeremy from semi-analysis, hopefully in the studio here, talking about meta-superintelligence. Jeremy, how are you doing? Good to meet you. What's going on?

Starting point is 01:45:07 Fine, hey guys, nice to meet you. How are you? I'm good. Could you case off with like an introduction of how you got into this? I've heard Dylan Patel's story of just kind of being on forums and nerding out about this stuff and then turning it into a career. But how did you get into a semiconductor analysis? Yeah, sure. So before joining semi-analysis, I was a by-side analyst. So I was mostly focused on the stock market on equities, looking specifically at tech stocks. I was at long only European shops looking at European stocks like

Starting point is 01:45:38 ST microelectronics, Infinion, all these industrial automotive semiconductors as part of my research, I discovered some analysis. I was like, whoa, this guy's really good. And one day, I just saw a post a villain on Twitter. He was just saying, yeah, I'm hiring a bunch of people with a cell site or buy-set experience

Starting point is 01:45:57 because I want to build a real institutional firm. And yeah, like we just got a chat That was, I think, August 23. I joined the company in February 24. We were seven when I joined. No, I think we're 33 or 34. So, yeah, it's growing every day, I think. So I can't do it.

Starting point is 01:46:14 Are there other firms like this in the, in the by-side, sell-side ecosystem? I'm thinking of like Ian Bremmer has a consulting group where he writes books, but then also has a team of geopolitical analysts talking about kind of global macro trends and what's going on the political side. Did you ever interface with any other firms like semi-analysis? Because it seems like unique. And Dylan's like AI is so big that Dylan's crossed over from talking specifically to to sell-side analysts that even venture capitalists will read semi-analysis.

Starting point is 01:46:47 No, I agree. I think it's pretty unique. Look, like there are obviously on one end, there are market research firms. There's like, yeah, folks like IDC, Gautner and so forth. There are many of them. on the other end of the spectrum so those would be very industry focused on the other end you have

Starting point is 01:47:04 World Street sales side analysts the more consistently is called Man Sucks of the World I guess we kind of found a spot in between and also it's part of the team because we have over 20 analysts and roughly speaking you have half of people that have financial background like myself

Starting point is 01:47:20 so more closer to markets and the other half is more engineers and people that are extremely technical and so we kind of found that sweet spot and also what's pretty cool is when you look at the different people at some analysis, like all the guys like myself that have more financial background, I actually love geeking out and digging into technology stuff. And same goes on with the engineers.

Starting point is 01:47:40 They also want to understand the business aspect that they end up geeking on the data and understanding the insights and market shares and stuff like that. So, yeah, we all go towards the same goal but have a different set of, yeah, of skills and such. That makes sense. So take me through the latest piece. I can imagine other players in the semi-market research.

Starting point is 01:47:58 space, just wait for you guys to public. Okay, I think now we're ready to form an opinion. Now we can put a buy rating on it. Now we can slap a buy rating on that bad boy and sell it. Your favorite researchers, favorite researcher. It's okay. By the time we're done, we're going to get the venture capitalist plan solidified. It's a million dollars a month for any venture capitalist. And we like to say, if you're not, if you're not on the semi-analysis, one million. You can't really be an AI investor. More of a tourist. Yeah. I got to say, we found something super funny a few days ago, one of the big brokers wrote a note to clients and basically said, yeah, these guys are the Bible. Actually, they talked about fabricating knowledge, which is dogs, so president or firm. They said

Starting point is 01:48:36 some people think fabricating knowledge is the Bible, but actually, he works with the firm that's the actual Bible and all the same. I don't know. I didn't say. That's a Wall Street guy. That's amazing. There we go. Yeah, I'm a strong believer, and I really enjoy the pieces every time they drop. Take me through the latest one. Meta-superintelligence. week was almost drowned out by the windsurf Google cognition debacle. And so Zuck came out with this huge announcement, you know, Monday. And yeah, it almost got swept away a little bit. Yep. Yeah, I mean, I guess Zuck is all in. That's another proof. It's interesting because I think

Starting point is 01:49:19 some people were at some point doubting at the beginning of the year, is he going to carry all that investment tens of billions of dollars? The answer is clearly yes. He wants to to do it. It's interesting because when you look at Meta's CAPEX so far, it has been heavily tilted towards what he calls Core AI, which is basically recommendation models. There's a lot of inference, inferencing advertising models and all of that. So they said publicly in 23, in 24 and 25, most of that CAPEX, which is going to be like $70 billion in 2025, is for that core AI business. So the Gen AI, the Lama stop, is still at the earlier stage. But The big question is how bold is he going to go with Lama?

Starting point is 01:50:00 And I think what we showed in this article, there's a lot of evidence that he's doing it on a very large scale. And that's not just, it's not just empty statements and saying, I'm going to build a 5-Ga-Wat data center in a few years. It's actually things that are already built or under construction, it's already committed capital. And so that's the first stories. There's already a substantial infrastructure build out specifically for Gen AI.

Starting point is 01:50:24 And I guess that sort of rationalizes the amount of money that's spent on researchers. Because if you think about it, yeah, you're spending like, hey, maybe $30 billion this year on Lama infrastructure. You're going to spend maybe, I don't know, $3 billion, $5 billion on hiring top researchers. Yeah, sure. Why not? Yeah, yeah, that's a completely reasonable strategy. And it always it always math out to us that even if these crazy researchers wind up working on core AI and the Lama project doesn't even go anywhere. It's like you could probably squeeze $3 billion out of core AI, right?

Starting point is 01:50:58 I don't know. It's just like the thought I had. Or you can also like these researchers can be focused on gen AI, but they're going to develop like state of the art technologies and using like massive compute. And then those state of the art technologies can feed into the core business. Yep. And end of generating more advertising sales. And if you think about it, like meta is growing double digits, 160 billion dollars.

Starting point is 01:51:19 Yes. So every time they grow double digits, we're talking about close to $20 billion of incremental revenue. So it's big numbers, right? It's easy to justify just getting a few billion dollars on researchers. Yeah, yeah, it's great. I have one question about the history here with Meta. So there was this story. I think it came from the first Mark Zuckerberg interview with Dorcasch Patel, where he tells

Starting point is 01:51:43 this story of the original Lama data center or the reason he had residual capacity was that he felt like he got caught flat-footed around TikTok and reels and recommend algorithms at scale for vertical social video where it's much more it's much less driven by a social graph and kind of like a traditional CPU-based graph query and more about these actual recommendation algorithms and the way he tells the story is like we didn't want to get caught flat-footed we need to play catch-up in reels but we didn't get caught what we didn't want to get caught flat-footed again so I told the team build two big data centers and then we had this

Starting point is 01:52:22 kind of empty data center sitting there and that gave us the initial compute for Lama. Is that too much of a simplification? Does that feel like what happened? Do you have any insight into kind of like how the Lama project, the initial compute was built out, and then I want to go into the future of the project? I would just say at a high level,

Starting point is 01:52:42 if you think about the amount of money that META has historically been spending on beta servers relative to what they maybe need in theory, like they have always been overspending. They have always been investing substantial in infrastructure. And same goes on for Google. I think Google to an even bigger extent. But yes, like META has been a pioneer in building large-scale data centers.

Starting point is 01:53:06 Over a decade ago, META introduced their H-shaped data center design. They've been building 150 megawatt campuses since like 2013. So that's not new to them, like building large-scale. And that's why they also already had this sizable compute footprint. But in 2023, like sizable was maybe 20,000 GPUs. That was pretty big, but today it's, yeah, it's updated. You want to be in the hundreds of thousands. And as of today, Meida is to some extent late in terms of training computer relative to others.

Starting point is 01:53:38 Again, because they allocated, they have invested a lot of money, but a lot of that has been allocated to the core AI business. But what we've shown in that article is that they're actually today ready to ramp, yeah, a massive data center in Ohio that's going to get them to get them to get them to get them to get them. it's going to get them to chop off the talk about the h shape of the data center why was it an h shape to begin with and then it sounded like they abandoned it in favor of just one big tent uh what are the benefits of a tent i want to know about data center shape broadly yeah sounds good uh who doesn't love a good data center shape look at h uh i don't know why it's at h uh what's more interesting is the structure of the building okay uh if you look at one uh physically it's absolutely massive um it's close to a million

Starting point is 01:54:21 square feet. It's just a monster. If you look at the structure, it's also three levels. So it's a very complex structure. It generally what we've observed for satellite imagery is that it takes roughly two years to build, which and I'm just talking about like first stone to actually getting the project built. So two years is a lot. Many people do that in a year or less. So yeah, substantial time to build. It was designed for very high efficiency. So they've been using a system with free air cooling, like they can get the air from the outside. There's no air to water heat exchange. Basically, you get cold air from the outside.

Starting point is 01:54:58 You just expel it, hot air. Super efficient. We can spray some water on it to make it even more efficient. The energy efficiency ratio is typically called the PUE. Maybe you've heard of that. Maybe not. Industry average is going to be 1.3, 1.4, which means that for every watt, you allocate to servers, you have to spend another 30% or 40% of that power.

Starting point is 01:55:19 into cooling and power distribution losses and all of that. That ratio for META was historically below 1.1. So they were actually the most energy efficient firm in the world running data centers. Interesting. But the trade-off is that these data centers took a long time to build and had a very low power density. Got it. And so what happened is, first of all, at the end of 2022,

Starting point is 01:55:45 Metup introduced the first massive design change. They completely through the old age and build a, let's call it a more traditional data center design, single story, sort of a big rectangle, faster to build, maybe one year, maybe one or 15 quarters, something like that, much faster to build, more denser, better suited for AI. It could handle liquid cooling. But what might have thought, look, I think this is what the XAI story actually takes a big role. I think what Elon demonstrated when he set up that cluster in 122 days, he just shocked basically data center infrastructure leaders all around the world. He made everybody's life a lot harder because before it was like,

Starting point is 01:56:28 well, if we do this really quickly, I'm going to need a year, year and a half. Yeah. It's like, well, there's a new standard. I can do it in 120 days. Yeah. Imagine like you're the infrastructure leader at the hyperskater that you have experience developing gigawatts of capacity. And you think you're the best in the world at doing that.

Starting point is 01:56:46 And suddenly some guy comes out of nowhere and does it in like one quarter of the time. So I think many people were really shocked. And I guess Zocke took it the other way and just was inspired by it. And basically, that's where the tent steps in, which is let's make a data center in the shape that I can build the fastest. So that, yeah, the only ball next is just finding some power. And that's it. And buying QB. saying earlier is a bull market intense.

Starting point is 01:57:14 Zuck needs tense. The XAI engineers need tense to sleep in the office. And then the XAI servers will be intense too. Everything's temporary. I mean, is there something about the tense structure that's potentially like, okay, this is going to deteriorate faster? Like what are the drawbacks of moving faster? And then on the PUE question, when I hear a one gigawatt or five gigawatts that they're

Starting point is 01:57:39 targeting, is that total power into the building or total power, to the actual servers? Honestly, people generally throw, like, there are both. In this case, based on our analysis, it's to the servers. To the servers, so it's actually going to be slightly more in terms of growth utility power. Oftentimes you see people quoting total power

Starting point is 01:57:59 just to have a bigger number. Sure, yeah, that makes sense. Industry standard practices to put IT power. Anyway, in this case, you're gonna have one gigawatts of compute power by the end of 2026 in Ohio, and then close to 2 gigawatts, by the end of 2027 in Louisiana. That's compute power to the servers.

Starting point is 01:58:16 So anyway, massive compute for Lama. Is that a vanity metric? Is it a vanity metric to hit one gigawatt exactly? I mean, Stargate, you have on your chart at 880 megawatts. Is 120 or 140 megawatts really going to be the difference between like an amazing super intelligence and the next best thing?

Starting point is 01:58:34 Like it feels like we had to cross this threshold and one gigawatt is like, it's a good headline. Yeah, it's a good headline. That's more of it. Like, it's not going to change much if you have 900 or a gigawatts. Yeah. But the world better, right? Like, you still want to have more servers.

Starting point is 01:58:47 Of course. Yeah, more is better. It does matter. To be clear, it's not going to change too much, but it does matter. Sure. Yeah, it probably matters for recruiting too. It's like you're going to be at the place with the best. What does the best mean?

Starting point is 01:58:57 I can, I can, you know, $100 million offers, nice round number. One gigawatt factory or AI, you know, super closer. That's a nice round number. It's great. Look, and I've got a better one for you. $100 billion dollars, right? $500 billion target. Oh, yeah.

Starting point is 01:59:11 That's pretty funny to me. because actually the Louisiana project, which is two gigawatts, they said publicly it's a $10 billion data center project. But if you count it the same way as they count the Target project in Abilene, it could be like $150 to $100 billion. It's just, yeah, just big numbers in marketing. Okay. I don't know if you have a bunch of insights or clarity here,

Starting point is 01:59:36 but how are places like Louisiana and Ohio reacting in order to, attract these types of data center projects. Are they promising the hyperscalers in the labs, you know, we're going to massively expedite permitting process? Is there like deregulation happening at a local level? What can you say that? Yeah, I think the piece of context is that since, let's say, the end of 2023 or maybe mid-20203, you have a frantic search for power happening in the U.S. and all around the world. And so we've made some number. We've, we've, we've, we've aggregate some numbers. If you look at the pipeline, there are different terms like data center interconnection load queue or pipeline. Basically, if you aggregate all of the load

Starting point is 02:00:23 requests that potential data center have submitted to the grid, in the US, you're above 500 gigawatts. You're close to the actual peak load of the US. So what's happening is pretty insane. Those numbers are mostly fake, but what it means is that people are searching for power all around the country. And you actually have a massive competition. all around the country to attract those last products. Because if you think about it this way, like, okay, 500 gigawatts of requests, but in the end, by 2030, you're going to have maybe 100 gigawatts of growth, which is an insane amount already.

Starting point is 02:00:56 But it means that just 20% of these projects are actually going to be real. So, yeah, there's definitely a lot of competition. So people are doing everything they can to get those projects. Tax breaks are generally today the most standard thing. accelerating permits, like reassuring hyperskators that you will deliver on time, that you have top contractors, that you're going to expedite permits. Increasingly, there's being sort of, yeah, enabling more on-site power solutions as well, like being more open to people burning natural gas on site.

Starting point is 02:01:28 All that kind of stuff helps companies, utilities, and locations, secure the big projects. I have a question about the shape of the super intelligence team at Meta. When you think about meta properties, you think about Facebook, the blue app, you think about Instagram, you think about WhatsApp, and then Oculus and VR is like a separate thing, Quest. But I was towing with this idea that maybe the super intelligence team is more like the database team or the React team. And it's like an infrastructure layer, a project that will have benefits all over the place. but we won't necessarily expect a dedicated vertical that competes with Instagram. It's more about making all the apps better.

Starting point is 02:02:12 I want to dive into the chat app statistics that you shared and try and understand. It feels like it's not exactly a neck and neck race. ChatGPT is pretty much pulling away in terms of percentage share of queries at 71%. Meta's way behind at 12%. Are there any signals that this is, that this is, you know, something that would be addressed one way or another? Or is it just too soon to tell? Yeah, look, I think what we've seen so far is that generally speaking, when you start to deliver a better model, a better product, you just get more users. I think we've seen pretty good correlation between the quality of models and the usage of chat GPT.

Starting point is 02:02:54 One thing that I think is interesting is when you look at the user base of chat GPT, you had a surge in early 20, 23 and then sort of plateaued for a bit. And towards the end of 2024, you had a second leg of growth, and then you hit that half a billion weekly users. And that correlates pretty well with new releases, with models getting cheaper and better and so on and so forth. So if you think about how meta could get back at a company and maybe lead that ranking,

Starting point is 02:03:19 well, they just have to release better products. One thing is to have good products, and the other one is to have a good distribution platform. And obviously, they have the distribution platform, right, 2 billion daily users. So they just need to build a good product. And I think users will come. And that's kind of the value proposition.

Starting point is 02:03:34 What is the state of the mom and pop data center market? I don't think they would like to be called the mom and pop data center market. But, you know, a friend or an uncle that's getting into the data center of business, we asked Brian, one of the co-founders of Correweave the other day, he was generally bearish. He knows how hard it is. He knows how hard it is. But like, what is there a real demand signal there? or is it just a hope and a prayer that you're going to just, you know, kind of flip it to a hyperscaler?

Starting point is 02:04:04 And is that, is that? Yeah, sorry, guys, it's over. Easy money's over. Look, as we just said before, like, now there's really an insane amount of competition for power. And, like, and basically, like, hypers can set the conditions. And so you have to be quite sophisticated when you want to approach hyperskenters. and you want to sell them a gigawatt site. And you also have to think about how much money does that involve?

Starting point is 02:04:32 Like one gigawatt. You're talking about maybe $30, 40 billion of Kappex. So when you sell one gigawatt site to a hyperskater, it's not going to be a small decision for them. To be clear, like buying the land, maybe, I don't know, a few million. Like, who cares? It's not a big deal. But if they do make the purchase, like they intend to do something about it.

Starting point is 02:04:50 So, like, yeah, they just don't want to take it lightly. And they have multiple options today. So for a mom and pop that sort of doesn't dig into what they should actually do and all their requests and such, they're just not going to get new business today. But some people made a lot of money for sure in 20203 and 2024 by just having, like, yeah, launch. They had 50 acres near high voltage line. That's amazing. Yeah, it's funny to imagine you've been building a data center, you know, a little mom and pop shop last year. And then the Death Star Data Center starts like popping up next to you.

Starting point is 02:05:23 Oh, no, I'm good. It's over. What can you tell us about what change between Lama 3 and Lama 4 and exactly what happened with Lama 4? You call it a failure. How did, like, how did that happen? Because we were so scale-pelled, right? Scale is all you need. Just scale up the big transformer.

Starting point is 02:05:46 But you dove into it and gave so much more detail. How can you contextualize and explain that to us? Simple way is just a bunch of trade-offs that weren't in the right direction. Like, I think if you want to simplify it, you could say like to some extent you reach peak pre-training in 2024. I'm simplyifying, I don't think it's big, but let's call it for now it's big. I think as Blackwell ramps and so on and so forth, you're going to see a new push-out. But for now it's big.

Starting point is 02:06:15 And it means that if you want to develop better models, you don't have to use pre-training. You have to use the new paradigm, which is a test-time compute and reinforcement. Right? And to do that, there are some specific trade-offs that you have to do. And basically, Meta just took a bunch of options in terms of their attention mechanism in terms of the way they route to experts and stuff like that. Just a bunch of decisions that aren't very well suited towards this new paradigm, which means that their flagship model, behemoth, the largest one, is just not very well suited to this new era. And that sort of contrasts with think of the Chinese labs, they are sort of in the opposite direction because they don't have state-of-the-art chips. they don't really have an incentive to push very hard on pre-training. And so they have been thinking harder about pushing on post-training, enforcement learning, and test and computing, all of that, that doesn't require such a large centralized cluster. As such, it's kind of those two paths, like on one end you have meta, on the other end, you have the Chinese, and what the Chinese decided to do is just better suited to the current paradigm.

Starting point is 02:07:15 Yeah. So just a bunch of bad decisions or to some extent unlucky. And so that seems like that ties to this, there was this post by Rune anonymous account on X saying that there are, in fact, some secrets about what paths of the tech tree you want to go down. You poached the right researcher. They come over and on day one, they can tell you that chunked attention might be wrong for this particular training run. Is that what you think is driving the high salaries? Is that the dynamic at play? Yeah. Yeah.

Starting point is 02:07:47 You want the decision makers that understand exactly what tradeoffs are going to do, that also know how to properly evaluate things or what kind of steps you should take to make sure what is the right choice. And so, yeah, that's what you want, the decision makers that have had experience, that know how to do stuff and they can easily identify the tradeoffs and know if I want to go in this direction, more reasoning, more RL, all that, you should do that attention mechanism. Do you have any insight into like, the shape of the behemoth project right now or like the the failure mode like is it is it is it like it would be bad at math or it would be bad at talking to it for a long time or it would be bad at

Starting point is 02:08:30 needle in a haystack in a big context window like do like we we've just heard like it's not good enough it failed but like how what would that feel like if I were to use behemoth and for a long time and I'm like oh this is weird it's bad but in a weird way because it seemed like it was good at some things but then just not good at everything across the board. Yeah, let's just simplify and say agents, so basically tasks that require using tools that require reasoning that require long context windows and it's not very good at that. Okay, got it. Jordi. Last question for me for now, what are you expecting out of meta over the next six months? They have talent now, they have scale, but the team, the new team is going to need to gel and it's going to take some time to really

Starting point is 02:09:15 start delivering. So are you expecting a lot of, you know, public launches over the next, basically the back half of this year, or is this more early 2026? I think back half of the year makes sense. Really back half, like think end of Q4 or beginning of Q126, but for sure you're going to have a few months where probably not much is going to happen. Because, yeah, like, like as of today, we show a bunch of pictures of the current cluster, so they already have data center capacity. But there's still some time in order to like actually put the GPUs in place and make sure everything runs.

Starting point is 02:09:51 So they're going to have some like sizable computers that's training ready somewhere in Q3. And so yeah, actually have a product that's more about really end of the year, but most likely early 2026. I would not be surprised. I wouldn't be surprised if New Year's Eve. We're back on the show talking like we are now about a new. drop. Can you talk me through some of the trade-offs or how the open source war is playing out. Dylan from Semi Analysis posted that the open AI model, open source model is expected to be really,

Starting point is 02:10:28 really good. That was kind of, I thought the open source strategy with Lama was a great way to be superlative on day one. It's like it doesn't need to be the best model. It doesn't need to have the most DAUs or MAUs, but it's the only, it's the best open source one. And so is superlative, you get the headline it's the attractor for talent hey we're doing something different we're the best in this one narrow thing uh and and there's a question about like at a certain point does the math make sense to continue to open source but then when we went through the deep seek moment it felt like deep seek was very much distilled from the gpt4 API not a llama fork and so the whole debate over oh like you know some chinese lab's just going to fork llama and then and then improve it so what is your

Starting point is 02:11:11 kind of state of the union on open source AI? The Chinese are eating open source. Yeah, they're just dominating the market. It's like one lap after another. It's not just deep seek. You had deep sync. Then you had Alibaba. Recently you had moonshot with the Kimi model.

Starting point is 02:11:30 They're just really good at open source. Sorry, at LLM generally. And they're open sourcing everything because they're in some sort of the same position as meta. Like they're not meeting. So they don't have any incentive to be closed source. they want to build an ecosystem. So it makes sense to go open source.

Starting point is 02:11:44 And there's just like shipping faster than meta and shipping better than meta. So yeah, meta is just way behind on open source. And actually the West is behind an open source. The Chinese are just way better at it right now. But is open source important or is it more just like marketing? Because I've always had this thing about the difference between like, what was it, stable diffusion was open source, mid-jurney was not. and Mid Journey was able to get the data back from the customer because you generate four

Starting point is 02:12:15 images in Mid Journey, you give a thumbs up. To me, it's only open source if it's from the Mistral region of France. But talk about the flywheel of open source. Like, is there an advantage there? Or is it really just, if you're not in first, you might as well open source. Yeah. It's more what you said is if you're not, if you're not a leading lab, you might as well open source. because you want people to sort of, yeah, just help you,

Starting point is 02:12:41 I'll give you some feedback to build an ecosystem. It just makes sense. Also, I would say, like, for the broad community generally, it's good to have open source because it's better for adoption. Anyone can sort of, yeah, play with the models and develop new applications on top of it and such. So, yeah, I think for everyone is good that there's open source. But again, like, if you're not Google,

Starting point is 02:13:00 if you're not open AI at the very top, you don't really have an incentive to be, yeah, to be secretive about what you're doing because you're not the best anyway. Yeah, that makes sense. Georgie, do you have any of who else? This is great. This is fantastic, please.

Starting point is 02:13:13 Hop on whenever you post anything, I'm sure we'll be giving you lots of calls because this is a fantastic conversation. Really appreciate it. Sounds good and good. We'll talk to you. Talk soon. Bye.

Starting point is 02:13:22 Quickly, let me tell you about fin.a.i, the number one AI agent for customer service. The bakeoff champion. The bakeoff champion, number one in performance benchmarks, number one in competitive bakeoffs, number one ranking on G2. And we have our national.

Starting point is 02:13:38 guest in studio in person. We have Jesse from Coinbase coming in yesterday. Jordie went over to Coinbase's launch event for base. Let's bring him in Jesse welcome to the stream. How are you doing? Welcome the walk-in camera. The walking camera's working there we go. Let's go. Thank you. Thank you. How you doing? I'm good to see you too. Thank you. We got one more juice for you. We're drinking base juice today. We're drinking base. Yeah, we go off the BIC morality juice.

Starting point is 02:14:13 How's that viral juice? Fantastic. How's it going? How's the show been so far? Super fun. It's been a crazy day on the internet. You guys are lucky you launched yesterday and not this morning. One cold play conference changed the timeline forever.

Starting point is 02:14:27 I know I saw that this morning. I woke up. I was like, wow. Glad this happened yesterday. Oh yeah, because I mean, it's so hard to launch these days, right? We're having a couple people from Open AI on later today. And it's such a big day. for legislation in the United States.

Starting point is 02:14:40 I mean, this is the Genius Act, and the Clarity Act just passed. I'm finding out about this right now. I think tomorrow. Yes, Genius App, Stable Coins will be passed and signed into law in the United States this year. I mean, this week, which is an incredible milestone. You're working towards it for three years,

Starting point is 02:14:56 four years, five years. I mean, the impact is that for the last decade of crypto, there hasn't been regulatory clarity in the United States, which means that entrepreneurs and consumers haven't actually been able to benefit from this technology. And we've been working to build bipartisan consensus that we need rules of the road in order to make sure that crypto works. And I feel like Coinbase has always been like the conservative one and it paid off because during the end of the Zurp era Basically all of Coinbase's competitors went out of business

Starting point is 02:15:24 It was chaos, right? And and and I remember talking to to a to a public markets investor at that time and he was just saying like like I feel like Coinbase has the mandate of heaven like they are making it what was it what was the low was it like seven billion dollars something like that yeah in in probably check in the chart the share price went to 32 dollars which was below i think the series e price and and i had joined you know like a couple years before the series e price but you know we all experienced that like what was your path to coinbase did you get were you building something else before yeah i i dropped out of school and started a company it was called clef we did identity and so the whole idea was how do we build an identity that's the next thing after passwords that was decentralized that anyone's a crypto company it wasn't quite

Starting point is 02:16:05 crypto but we worked with crypto companies so you know bitfinex we were actually providing like kind of off and login for them okay and so Paulo like we go way back because we've been building forever and so that business didn't work yeah yeah and then when we were winding it down I was basically trying to figure out what do I do next and I love crypto and his identity like K YC like I upload my I take a picture of my photo it was actually a passwordless login mechanism so almost like you hold your phone up to log into WhatsApp it was like that but back in 2012 and so we were a little bit before the time but when

Starting point is 02:16:33 you look at the security architecture that we built It's actually almost the same as what we're doing now on base and what people are doing with past keys And it's this incredible thing where I've now been working on it this same problem space for almost 12 years and we're feeling Overnight success. We're finally getting there. Okay, so you land at Coinbase and the way Brian Armstrong tells it you go to him and you say I need one billion dollars Well, not right away. Not right away. No, right away. Now I've been bold But I want to get to the one billion dollar plan. I know you did it for a lot less, but I want to know that's my question for the future, but tell me, tell me what you did in between. Two pizzas.

Starting point is 02:17:08 Two pizzas. No. So I joined, I joined as an engineer. It was an aqua hire process. My whole team actually sold to another company to Twilio, but I was so excited about early to the double. Let's go. And so I joined Coinbase, joined as an engineer and then pretty quickly on, I just started leading teams.

Starting point is 02:17:27 They asked me, hey, can you take on a team? I took on more. And for the next five years, I led all the consumer businesses on the engineering side. So if you know Max Brandsberg, he runs our consumer product, me and him product and engineering counterparts with him running product and me running engineering. And then after five years of doing that, I thought I was going to start another company. I'm like a founder. I actually put in my notice and said, I'm leaving in six months.

Starting point is 02:17:46 In those five years, what products were you building? I was building Coinbase, Coinbase Pro, and Coinbase wallet. Okay, got it. So like anything that you've touched as a Coinbase user, those are my teams. We rebuilt the whole product from scratch, migrate us to React Native. Coinbase wallet was the first, like, decentralized product in the Coinbase ecosystem. So I think the context here is like the core business. and why the launch yesterday is exciting is the core business was always centralized, right?

Starting point is 02:18:12 And it was a gateway to get on chain, custodial, and we'll get into base in a bit. But you were working on these sort of decentralized experiments early. Exactly. Yeah, working on the decentralized experiments early. But what are you saying it's so early at the time? You know, I'm an optimist. He said he typed GM and posted it probably 100,000. guys.

Starting point is 02:18:37 But I thought I was going to leave, but I actually started having conversations with Brian about what did the future of Coin Mace look like? And what would it look like if we leaned even further into this kind of decentralized world, built more things on chain because Quaybace had started 10 years ago and it's, you know, traditional Web 2 business. And so, again, the first thing I did was I went to Brian in the kind of fall of 2021 and our exec team and said, if you give me a billion people and 60 employees, this was like peak $20.

Starting point is 02:19:01 Billion dollars and $60? 60 employees all turned Coinbase into a Dow. Okay. They were like, ha ha ha. Okay. What would you have done? We're a public company. Yeah.

Starting point is 02:19:14 If he had, I know what you do with a billion dollars if you're meta and you build a big data center and you have to pay, you know, half of that to Jensen Wong over at Nvidia. But what would you have actually done with a billion dollars? I don't even know what is that. We didn't really have as good. Okay. It was more just like, let's take the constraints off. Let's take the constraints off. Let's go and do the thing.

Starting point is 02:19:36 We have this big public company. It's a crypto leader, but it's still built on Web 2 architecture. How do we move it on chain in this new way? And so it was the vision was there, but the execution strategy wasn't there. And so... Well, and the other context is that Coinbase has been in this incredible position of being the gateway to get on chain. And you see this with a Circle IPO. People were like, wait, Coinbase makes a lot of Circle's revenue as a stablecoin issuer.

Starting point is 02:20:01 But you guys had always basically said everything on, you know, You guys do whatever you want on chain and a bunch of great companies sprung up to kind of service that market. But you had to be hands off, which I imagine was pretty frustrating. Yeah, I mean, I will say that early on. So Brian has always had the vision of the self-custodial world. He's even had the vision of a super app in many ways where the first version of Coinbase wallet was actually called Toshi. It was a messaging product with apps in it. Oh, interesting.

Starting point is 02:20:28 This is in 2017. It's green. And this is based today. It's based today. And so then we kind of pivoted Toshi to. coin-based wallet, which is the self-custodial wallet that millions of people use that folks know and love. And we've worked on that for last five years. And that was primarily focused on trading in money. But as we've built base over the last two years, which has really been a builder ecosystem where people are building apps, we've seen two problems emerge.

Starting point is 02:20:50 The first is that everyday people, when they're trying to use crypto and trying to kind of figure out what's going on here, it's really hard to actually find things to do and that are useful. Maybe they come on chain for a coin, but then they kind of get lost. They're like, what are all these things? How do I actually use it? And then on the other side, we saw that we had thousands and thousands of builders and creators who had these incredible products that they were building, but they couldn't actually get them in front of people. Like we were missing this connective tissue. And so 10 months ago, I talked with Brian.

Starting point is 02:21:22 I kind of came back to running the Coinbase Wallet team, which I'd worked on for a long time prior. And we kind of jointly articulated this vision of, well, what if we built the kind of app that solved that problem? and brought all this things together. And this is what we launched yesterday. It's the base app. And it's an everything app that lets you create, earn, chat, trade, and discover this entire. Pay for things with Shopify at Arrow One. You can now pay for juice with base USDC on Shopify, coward by base pay.

Starting point is 02:21:54 The whole thing kind of comes together in a new way where because we now can bring together this marketplace of kind of developers and builders and businesses on one side with consumers through really easy to use experience, we're going to be able to grow crypto a lot faster and spin the flywheel. Would I imagine launching base in the way that the product is today. I know it's a pure software business, so very different from the, you know, Coinbase's, you know, traditional business, which is effectively, you know, acting as this regulated or, you know, financial institution. Would it have been possible to launch this a year ago,

Starting point is 02:22:34 or did you need the White House to launch some tokens and a few other events? Well, I think the biggest thing that we needed is we needed the technology to mature, right? So when we had this vision three years ago of building base, the first thing we started with was actually the platform, right? It was the chain because we'd spend a year

Starting point is 02:22:54 trying to figure out, okay, what's the product we build to kind of bring this next generation? But explain those? So you had the chain, but there was no token attached. The chain with no token. I feel like the playbook in crypto is like, we went back to the tried and trude method that people had been following for a hundred years,

Starting point is 02:23:09 which is that you focus on building a product and people love. Right? Like I think there's been this whole distortion in crypto for the last five years where people are like, oh, you get a token, you pay people to use your product. And we basically said from the beginning, no, we're gonna build a chain,

Starting point is 02:23:20 it's gonna be a developer platform. And we're gonna figure out how do we make it the best place for builders to build? And that's exactly what we did over the last two years. But if I take it to like the Bitcoin analogy, it's like the reason that people set up data centers to run the Bitcoin network is because they mine Bitcoin and that has financial value. And so there's this economic incentive.

Starting point is 02:23:35 And now that gets distorted all the time when people run like random chains and stuff. And I understand the rationale. But is there an incentive for other participants to, uh, to help run the infrastructure? Or, or are you doing something where you're splitting it with across different groups? Yeah. Like how does the actual chain stay on? Yep. Well, the first thing to know is that base runs on Ethereum.

Starting point is 02:23:55 Okay. So it's a layer two that sits on top of Ethereum. So as long as the Ethereum miners are happy, then base runs. Yes, exactly. And then there's a decentralized network that's also based that people use to get access to the network. And the thing that's bringing people to the network is that they're building things and then they're using those things. So here's two examples. The first one is USDC.

Starting point is 02:24:12 We're seeing a huge amount of growth in payments both in the United States and outside the United States in USC. And that's because people are using products like Shopify, which has rolled out to millions of merchant, where they're actually taking USC in their wallet and they're paying for things. And that's global. It's fast. It's cheap. It happens instantly. And there's cool rewards.

Starting point is 02:24:29 There's cool that the the it's retailers effectively which is if you pay with USC will give you one point back or something like that because the interchange fee should be yeah exactly debit card or credit card yeah so it's one percent back with base pay and you can earn 4.1% on your USC when you're holding it That's pretty good right and it's an instant global payment method so any business in the world can integrate it and then anyone else in the world can pay for things. Yep so that's one example people are coming to base for payments and stable coins the other example is a little bit more out on the kind of like nascency curve in terms of what we're seeing but we're really excited about and this is what the focus of the event was yesterday which is really around the creator economy right if you look at the last 20 years you've had so much creativity pouring to the internet but the

Starting point is 02:25:12 vast majority of that creativity has been put into platforms where creators actually don't earn that much money right if you're an average creator who has less than 10 000 followers or if you're in another country like Nigeria or Argentina you're going to be posting and filling up this kind of content world with valuable creativity that other people are looking at, but you're not earning anything. And so one of the big things that we're focusing on with the base app, and one of the reasons why people are coming to use it is that when you post on base, you actually earn. And this is because we've built a new economic system where the content is actually valuable, and then the value gets flowed back to the creators.

Starting point is 02:25:46 And I think this is a novel use case for this platform that we built with base and base chain. Okay, so the thing that's driving demand. If somebody opens a base app, their wallets, low, and they start scrolling. Does that cost the money? Does that cost them money? How does that actually? It doesn't cost them money, but like I can literally show you,

Starting point is 02:26:04 you can decide if you want to support a post. And so like here, this is on my feed, and I'm just like Jesse, I'm on my feed, I'm creating content. I posted this great base juice content. And oh, there's a bug, live, classic. There we go, it happens. Base juice content, right? And you have the normal things you see here, the like, the comment, the retweet, and then you have a new thing.

Starting point is 02:26:25 Yeah. cap. Sure. And so this is basically the value of that content. And this one's worth $15,000. This one's worth $84,000. This one's worth possible. I mean, like my, I've been on YouTube for five years. The best video I ever posted got eight million views and I got a check for 20 K from YouTube. Well, that's actually pretty good. YouTube is one of the platforms that pays creators really well. They took 50%. But I thought that was fine because they brought me all the users. So I was actually pretty happy with that. But like how could how could, how could a piece of

Starting point is 02:26:54 content have market cap? Well, yeah. I, I, I, I, I, I, I, I, I, I, I, I, I, No one else wants to buy it unless they want to buy the revenue stream. And there's been some of those projects. I posted just a few hours ago. Yeah. Startup. Startup CEO. Well, this is on X.

Starting point is 02:27:07 Yeah. Well, I, no market cap on this one. Startup CEOs can't even hug their chief people officer at a concert in this country anymore. 1.6 million views. 50,000 likes. 50,000 lives. I'm sure. I'm sure the creator payout will be decent.

Starting point is 02:27:20 But this, this looks to me like potentially a nine figure market cap on if I had ripped it on base. Yeah. Well, seriously, like seriously, I, I, you know, on these ones that I posted, like this one I posted yesterday that now has an $84,000 market cap, it's done like $4 million in volume. And I earned 1% of all of the volume. And so that means I've earned like, what, $40,000. That's insane. Right? So you get a viral post?

Starting point is 02:27:44 That doesn't feel like, like economically fair. Well, but think about this. The place where that value comes from is the fact that your content is valuable, right? How have these platforms? It seems like a technology, right? How have these platforms built multi-hundred billion dollar businesses if the content is invaluable? Running ads. But what's bringing you there?

Starting point is 02:28:05 The content. And then they pay you to bring more content because you get paid out. An ad is a way of monetizing the value of the content. But the content itself is valuable. And what's happening on the base app now is that that value is being kind of figured out in a free market. And then the kind of the economic offloads of that value are getting redirected back to the creators. love ads this show will we be able to see ads in the base app you guys acquired spindle I don't think you're ready to talk well we have we stay tuned in the

Starting point is 02:28:35 next in the next I want to run I think that I think that I talk about the integration in the kind of at mini apps yeah I yeah Dan from Farcaster has been on the show I know you guys have some integration with with the Farcaster network is that right yeah so the whole social feed all of the connection graphs you know followers and following it's all powered by Farcaster when you're posting content it's powered on Farcaster when you're coining content that's that's powered by Zori it's another protocol we've done with the base app is we brought them all together that's cool including Shopify

Starting point is 02:29:08 including messaging XMTP which is powering messaging they just raised a $20 million series B okay they're an incredible decentralized messaging protocol that is powering secure encrypted messaging with agents in chat and so one of the really cool things I don't know if you guys saw in the demo yesterday is you can meet a chat with your friends you can just like drop a bet in there that you can say agent hey like let's bet five dollars that the Dodgers are going to win tonight okay and then it's immediate and then you can say hey send $10 to all my friends sure and it just works because it's built on this platform

Starting point is 02:29:38 and you have these open protocols that are working together and so going to your point we built on Farcaster because we believe in building in open protocols and one of the things that open protocols enables is it enables other people to build on top of them and so this is where the mini apps come in is because we've built in this open way now anyone can build a mini app and it just shows up in the app. And here's another one. I don't know if you guys have seen this,

Starting point is 02:29:57 but this is, when you open the base app today, I think the FARCAS is right there, there we're DEMI, which is like literally you can open this up and we're gonna be on the live stream. Here live now, there I am. Oh, wow, there you are right there. It is so cool. But one of the really cool things about this.

Starting point is 02:30:13 And to be clear, this was organic third party. And you have some new sponsors. I don't know if you guys realize this, but you have a bunch of crypto brands here that have now gone and said, hey, we want to sponsor you. And Bracket, for instance. This is sports betting on base. It's awesome.

Starting point is 02:30:25 It's so fun. You can interact with an AIJ. But our lawyers, our lawyers are like bloodhounds. They're going to be like chomping at the bin here. The thing that's really incredible about this app is that it's all connected. And so I can actually tap in bracket and I'm like, okay. Oh.

Starting point is 02:30:39 Okay. Yeah. Hey, stay one. Oh my God. Come on, Jesse. What's going on here with this mini app? Yeah.

Starting point is 02:30:45 And in theory, yeah. This is the start of the auction. The auction driven. It happens. But, but, but, but,

Starting point is 02:30:53 This is the start of the auction-driven ecosystem that actually results in real transfer value. Real value is created at some point. So this is a new ad. I'm like, oh, I like Brackett. Whoa, now Brackett is right here. Okay. And then I'm like, okay, I'm just going to buy Brackett. And I'm like, cool, let me buy $5 a bracket.

Starting point is 02:31:11 This is not financial. This seems like, I don't know. We're not going to buy. The thing that's fun is like, it's just wildly chaotic. It is chaotic. It is an entirely new surface area. Coinbase has never been about chaos. But, and I'm not saying that base is, but creating an environment that's just like a free-for-all decentralized.

Starting point is 02:31:30 It's open. It's open. So, yeah, I mean, you've said that the goal broadly is to get a billion users on crypto, crypto-needed. On-chain. On-chain. That's the right phrase. Give me the state of the union. Where are we today on that journey?

Starting point is 02:31:43 What are the different buckets and how big are they? Yeah. So we're trying to build a global economy that increases innovation, creativity, and freedom. And the metric that we use basically to measure that is do we have a billion people? And of course I'm going to get 5 billion people on chain because we can get everyone who has an internet connection on chain Let's go. But a billion is a nice number and we get to make it repeatable. I'd say that when you look at the data today, there are hundreds of millions of people who hold Bitcoin Globally around the- Hundreds of millions.

Starting point is 02:32:06 Yeah, hundreds of millions. A lot of people hold Bitcoin. Like Bitcoin is very well used, but it's not something that people are using their day-day life. I'd say there's probably tens of million people around the world who are using stable coins. And then I think we're in the millions of people right now who are really starting to use these on-chain products. And that's across base. It's across Salana. We're seeing innovation everywhere.

Starting point is 02:32:24 And we like to work with everyone. We think about base as a bridge, not island. We're actually connecting with everyone and figuring out how we can help them be successful. But I think it's still really, really early days. And the goal with the base app is to kick up the next kink and growth. So my question is like when we're following the chat GPT story, there's obviously DAUs, but the chart that everyone's obsessed with right now is that

Starting point is 02:32:44 chat GPT minutes per day is skyrocketing. It's a 30 minutes per user per day. Something like that. We're seeing lots of queries go up. My question is like with financial. financial products, I do think that there's a benefit to having a lot of people own Bitcoin. I think it's an awesome network, a check on authoritarianism. It's awesome. But like, I don't know that I should be like interacting with Bitcoin daily. But what are the other metrics downstream

Starting point is 02:33:08 to like to like to like quantify like whether or not someone's on chain? Because like if someone's like, if I'm if I'm in my bank account, you know, messing around with dollars all the time, I'm actually probably unhappy. Yeah. Yeah. Well, the first thing I'd say is that, um, Definitely Coinbase is a traditional financial product where they're growing. They're trying to be the best for trading It's awesome. They're making such good progress so excited about the retail decks integration Which is basically making it so that anyone in Coinbase can buy any asset on base Yeah, just works listed I'd say base though we don't really think about it as a financial product Okay there's money involved because money is a part of everything

Starting point is 02:33:42 But it has social ads core it has chat it has apps and so people are going to come to base just like they come to their other Time on site should time on site is a thing that we're looking at we're looking at MAUs we're looking at So we're looking at, we look at kind of two things. Dow mouse, all that. Yeah, Dow mouse, and we look at weekly active users, and then we look at weekly transacting users, where it's like, are you actually doing a transaction on the chain where it's like buying something

Starting point is 02:34:03 or sending money to a friend or something like that? And so the active user is kind of an engagement metric, and then the transacting user is like a deeper engagement. Yeah, yeah, yeah, yeah. So higher use of the actual product would not be a bare case like it is with my, you know, my traditional financial app, which I'm opening. It's probably a problem.

Starting point is 02:34:19 I'm probably like, owe someone money or overdrafted or something like that. The really incredible thing about this kind of conversion rate of active to transacting is we have the data of before and after where like it was really just a money out before and now it's a real social product where people are doing all these things with money. And we're seeing way higher initial user conversion rate to actually doing something on chain. That's awesome. Because now it's not like, oh, you have to go like make an investment decision. It's like you can like your friends post. Yeah. Right?

Starting point is 02:34:43 Or you can like, you know, have to pay at a store that you're going to. And that transition from a speculative thing, which is so important and such a big important part of base in the crypto economy to a daily thing that also has like way bigger TAM in terms of the number of people who do social products and other things on a daily basis. That's that's really a shift we're trying to make. Yeah. It's super cool. Like the like the, like we know that the everything apps work elsewhere. Yeah. They don't work in America for some reason. This feels like an end run around, a different strategy to try and make it work here. But it's an everything app for a specific subculture, right? Right now. But for now, but you know, billion people is that a subculture?

Starting point is 02:35:18 And what I'll say is we were just out of Airwine. I just spent two hours at Airwant. And we were giving out free juice. I was filming content. We got some good content coming later. And I was just talking to people about the juice in the app. And when people hear especially like millennial Gen Z creators, when they hear, do you want to join a free social network where you can get paid to post? And do you want a free juice? They're like, hell yeah.

Starting point is 02:35:42 Like I've been posting on other social networks for a long time. And I've never gotten paid. Okay. And so that hook of your content is valuable. You get paid. And you can get paid for it because that value is going to get unleashed by the free market. Yeah. And then we're going to distribute it on crypto economic rails that is next generation internet platform. I'm fired up. I'm fired up. Well, good luck. We got to get to our next guest. Thank you so much

Starting point is 02:36:05 to stop and buy. Thank you for me. Can we get a gong hit on the way out here? Goong hit. Get the base app at base dot app. You can get on the wait list from me letting more people on every single day. There we go. Founder mode. Talk to you seen. Have a good one. Great chatting. Do it. Talks. Next, we have Higgsfield, Alex, from Higgsfield AI.

Starting point is 02:36:27 We've been using Higgsfield AI to generate crazy photos. Jordy posted one of himself for a horse. The horse had a little bit of a long neck, but otherwise, photo reel. And the facial reconstruction, like it nails the face really, really well. I'm not exactly sure how they do that. I want to talk to it. Welcome to the team. Welcome to the stream.

Starting point is 02:36:50 Yeah, Higgs field. We have a lot of amazing people on today. I have been incredibly excited to speak with you. We were just talking. I don't know if you caught it, but I posted an image generated by Higgsfield a couple days ago. I think most people still think it was real. It was a completely ridiculous image.

Starting point is 02:37:11 It was like this cinematic upshot of me on a horse that had, I don't know why people didn't catch that the neck was quite a bit bigger than a regular horse's neck. It was great. Clearly a breakthrough. Yeah. It was the first time that I've seen an image generation product and realized that I think you guys have probably already started to take over Instagram style content, but really

Starting point is 02:37:37 disrupting the potentially disrupting the Instagram boyfriend market. If people can just generate infinite images of themselves, Instagram boyfriends will have nothing to do. But great to have you on the show. It would be great to get a quick background. on yourself and the company and then we'll get into a bunch of other stuff. Great. Thank you very much for the kind words.

Starting point is 02:38:01 Definitely. It's great to be here. I thank you for the invitation. So quickly about myself, I'm a veteran in the video generative AI space. I built maybe some of the most iconic products in that space. Maybe you remember Snap, face filters.

Starting point is 02:38:15 Oh, really? Which, yeah, totally. That's amazing. Yeah, there were like billion people throughout the world who played with that. Just a billion. was pre-gen AI, right? And the models were like thousand times smaller than they are today. And although this product was specifically targeting like augmented reality use case, so basically overlay on top of the existing camera. With all these learnings, which I got from

Starting point is 02:38:45 my exciting times at Snapchat, we started Higgsfield with a way bigger ambition, is to creates camera of the next generation. Yep. Even today we can see that there is an emergence of UGC content. The quality unfortunately goes down. And with the end with Hicksfield we create a new ways to tell stories, it's helping creators and brands to get attention on social media. What's the key insight? Do you think, how important is it to create a great image generation model? Just scale. You just need to throw a ton of compute versus changes in the algorithm.

Starting point is 02:39:31 We've heard rumors that images in CHAP-CH-G-T isn't pure diffusion. There's some transformer architecture in there, some different layers. I've started to suspect that there are different layers in some of these where there might be a different neural network or different system to put text on top so the text is really clean. Are we kind of recreating Photoshop at a certain level? Talk to me about the actual technical infrastructure to the degree that you can. Totally. First of all, I think it is important to admit that today we are early in our journey with video AI technology.

Starting point is 02:40:10 I think Hicksfield is probably the first example of the technology platform, which helps to create compelling content for social media. The next step is going to be to build a reasoning engine on top of that. So let's say you post a video. The system is going to suggest you and analyze your accounts and actually provides you with more suggestions what to post. And eventually, I think we are going to find ourselves in two years in a new version of the world, where most of the content out there is a AI.

Starting point is 02:40:50 generator. There is no way to stop that. And on our side, we do our best to provide a simple interface so that non-AI users, like let's think about market of social media professionals of tens of millions of people, so that this broader user base can actually tap into the power of generative AI technology. And I think we will see that, the models are going to get way better than they are today. It is true that various research labs, they do experiment with various architectures. We found that our post-training techniques

Starting point is 02:41:32 and allow us to substantially differentiate from the competition. And we do believe that the general quality of the technology is already there to surpass average human-produced comments. What, uh, I have this one, eval for AI image generators where I ask it to create a where's Waldo and and no system's been able to crack it and I think it has something to do with the density of information in a proper

Starting point is 02:42:03 where's Waldo you'll typically see hundreds of little characters doing very intricate things and so it's very clear that the artists that create the actual where Waldo's work at a very small scale and they actually piece together the full image like it's a puzzle. That's something that I feel like could be solved with a reasoning layer on top. You understand that you're trying to create something that's really, really layered. And so you need to kind of create tiles and then blend them together. Is that, but this gets into the question of like, how are we going to generalize and scale reinforcement learning in LLMs and agentic workflows? Is there a similar path that we're going down in terms of, image generation. Yeah, this is a great question, by the way. I think this is a trillion-dollar question, which you just asked, which is we all saw the power of reasoning engines with maybe models like 03, and then we see that at GROC 4, they actually spends on reinforcement learning more than they spent on the pre-training stage,

Starting point is 02:43:08 right? Where is the market right now in terms of pre-training, post-training in images in your estimation? Yeah, I do believe that in the video AI space, we are relatively early. It's probably we still see that the post-training stage in the core video AI models can be maybe 20, 50 times lower compared to the pre-training stage. Wow, that's really low. Yeah, but we are really just crashing the surface there. I think then the future is building the video reasoning engine, and this is a trillion-dollar question. because this will, because think about the brands out there.

Starting point is 02:43:49 Today, what we are seeing is that brands start to, and agencies, they start to actually experiment with various models. And primarily, we all rely on our stereotypical understanding of the customers and some maybe qualitative data which is available out there. I strongly believe that with this video reasoning, engine, the way how the stories are told is going to be completely different. Instead of just running one video, we can run hundreds of the videos out there and A-B test them and see which one performs the best. And today, there are only a few top creators who actually do that. If you

Starting point is 02:44:35 look at Mr. Beast and similar size of the creators, they do A-B-test thumbnails very aggressively. We all know that. They actually beat as the hooks. So far, this privilege is only available for larger teams who can actually do that, who have the manpower to do that. And the next generation VJ reasoning engine will empower everyone to do that, which is going to boom to the boom of AI generate funds. What you're talking about is basically like RLing on humanity with human graders, which is the algorithm and likes. And that's effectively what the, like, the Mr. Beast algorithm is doing. My question is like is there a way that we can bring that into the data center because if it stays on Instagram if it stays on YouTube it's probably

Starting point is 02:45:20 pretty rough for you because Google and meta are going to have an advantage there but if you can figure out how to do RL with verifiable rewards or something that looks like a rubric for grading you know and finding errors are we going back to the generative adversarial network where you'll have two competing models to to determine, like, whenever I generated something with VO, it's always like, cars driving, looks amazing. Then all of a sudden, I'm looking at the front of the car and the car's driving backwards.

Starting point is 02:45:48 And clearly the model is getting confused, but we need maybe like a detector for that. But how are we actually gonna do RL at scale and imagery? This is a great question. So I think the first step is exactly, I mean, the first step is going to be reinforcement learning with AI feedback. Part of that, we already do at Hicksfield.

Starting point is 02:46:06 Obviously at the post-training stage, not yet at the interim stage, just because of the cost associated with that. That makes sense. So we have to train video generation model in a way that it's sort of competing with video understanding model. And like at Higgs fields, we cannot go and label millions of the videos. That's why we have a set of powerful agents for video understanding,

Starting point is 02:46:33 which help to tune the video generation process. But this is the process in the vacuum itself. itself. What's going to be powerful is when we condition the outputs of the model and train the model based on the engagement data from the social media. First, it's going to be a number of likes, a number of comments. Although if we look at meta ads, we see that they provide a very detailed breakdown and drop off second by seconds. So training the models based on these outputs is going to lead to complete the next level of reasoning. can success rates for their end customers.

Starting point is 02:47:12 Cool. We have another one in person. I know we're totally over. What happens to the legacy Instagram influencer that has built a business basically on MV? They're constantly traveling around the world, at the best hotels, on boats and private jets. What happens when anybody can be on a private jet

Starting point is 02:47:36 on a platform like Instagram or on a yacht somewhere? Have you thought through kind of the implications for the tech across different categories of content creators? Hopefully. So first of all, I need to admit that we're a technology platform first and foremost. Although we think about ourselves as a scientist and we are constantly monitoring and listening to the creators and how they use the technology. And I cannot bring up the names out here, although like some of the top 50 YouTubers in the world, they actually want to get reads of the team and build all agency of AI influencers, to be honest. That's they actually, they have a bunch of ideas which they want to tell and they don't want to condition their existence on the social media just to their likeness today because people just are getting older and sometimes they get irrelevant.

Starting point is 02:48:32 We have seen many examples of that on social media and creators actually want to create those digital agencies. where they can express all their ideas through various synthetic AI influencers. And they can use Higgsaw platform to do that. Yeah, it's very, very wild time. Let's have you back on again soon. This is fascinating. Yeah, we'll talk to you soon. Thanks for helping on, Alex.

Starting point is 02:48:56 Thank you. We'll talk you soon. Let me quickly tell you about public.com investing for those who take it seriously. They have multi-asset investing, industry leading yields, and they're trusted by millions. We have Billy from Regent, also calling in from re-industrialized, Sorry to keep you waiting, Billy. Great to see you. Every time we talk to Billy, he's in a far-flung part of the world.

Starting point is 02:49:17 We're working to bring him in from the waiting room. And he is sideways. Can you turn your camera? Yeah, I can go vertical. That is possible. Here we go. There are you. You're in the cabin?

Starting point is 02:49:30 Explain. Join you from the sea glider. Look at that. No way. It's super comfortable. And here we got plenty of leg room. That's incredible. We got internet.

Starting point is 02:49:39 this is the future. This is amazing. Amazing. Wait, wait, so, wait, are you, is this like a demo or are you actually in- I do believe that's a screen behind him? This is the demo, you guys, can't- This is the demo, you guys, okay. Well, we just talked to the AI generated guy who says all this will be fake in two years, so, you know, I don't know what to believe at this point. We're building hardware in the real world guys.

Starting point is 02:50:01 We're at a reindustrialized conference right now. Thank you for bringing me back to the real world. How is reindustrialize? Give us the overall update on Regent. It is awesome. Like hardware is cool again. Like building real stuff is cool. Adams are cool. So it's awesome to be here for this. Yesterday, we just announced the launch of Regent Defense here at Reindustrialize, which is a big step for the business. What is Regent Defense? What's the immediate application of the glider in a defense context? So Regent's actually been doing defense for years. This is sort of a growth of the business we've already been doing. Basically, everything is about moving around island chains as our key conflicts and theaters are in the Indo-Pacific. So it's going back to World War II style tactics.

Starting point is 02:50:49 It's about naval operations. It's about being able to move around island chains logistics from the head of the Marine Corps and throughout the services underwrites of success of that naval campaign. So it turns out that all the things that are commercial customers like about sea gliders, the high speed, low operating costs, you know, ease of operation. and in the DOD space that we're hard to see. We fly really low over the water, so it's hard to pick up on some long range radar systems. It makes it a really perfect fit. So we've been on contract with the Marine Corps for a couple of years now,

Starting point is 02:51:19 and now we're expanding as we get into our full-scale prototyping, expanding that product portfolio to meet the need. Very cool. Incredible. Anything else? I think that's it. Thank you for this. Thank you for this demo.

Starting point is 02:51:31 This is the best call in, the best guest calling, I think we've had. Guys, you've got to get you on the sea glider in Rhode Island at some point, too. We're doing the next call in from the – we have real hardware. We'll do the whole show. We'll do the whole show from the glider. I hear it's very smooth. You know, it's perfect for live podcasting. Starlink on board yet?

Starting point is 02:51:53 We'll put Starlink on board for a TVPN episode. There we go. I love it. It's there. Awesome. Fantastic. Well, congratulations to you and the team on the progress and say how to everybody at reindustrialize for us.

Starting point is 02:52:04 We'll do. Thanks, guys. Enjoy the rest of the conference. We'll talk to you soon. And in the meantime, let me tell you about 8Sleep.com. Get a pod five, five-year warranty, 30-night risk-free trial, free returns, free shipping. Go to 8Slood.com. And we have our next guest joining the stream. I'm going to guess that it's Jonathan, but I'm going to make sure once he joins. Welcome to the stream. How are you doing? Did I get that right?

Starting point is 02:52:32 Yes, Jonathan. What's happening? What's happening? Hey, a lot of people call me JMO, actually. Oh, JMO. What's up, man? Give us the news. I hear we, you got some good news, break it down for us.

Starting point is 02:52:43 Yes, we are coming out of stealth launching a new company called Confident Security. It's about providing confidential AI. Did you raise any money? We raised 4.2 million with decibel. Congratulations. Of our Commons. Congratulations. There we go.

Starting point is 02:53:01 Why is confidential? AI important. I can imagine a bunch of reasons, but what was the kind of catalyst to start the company? Yeah, I think, you know, unless we do some work, I think we're going to have Cambridge Analytica times like a million, essentially, unless there's just too much incentive for folks to train on data, and people need to care about privacy. So we thought, you know, Apple shouldn't just be the only ones providing privacy. Everyone else should do it too. Give me some concrete examples of how to use the product. What data specifically am I keeping secure?

Starting point is 02:53:38 Is it stuff that's on the web? Because I feel like if I have an air gap data center somewhere with a whole bunch of hard drives, there's no crawlers that are getting to that. That's right. If you are taking that air, if you're using the air gap data and you have your own GPUs and it's completely in there, you might care about various users inside your agape environment, not seeing all the data. you know, a standard like top secret versus secret type of stuff. But this is, you know, you're an employee at Pfizer and you upload a PDF to OpenAI

Starting point is 02:54:07 that describes how you do drug discovery. Maybe you care about that secret. Maybe you care about how it's going to be used. And after the recent Open AI core case where they were forced to retain deleted data, you know, I think people are being a little cautious. So how does your, so how do you actually plan in plugging in to companies? Like, how does the product actually get installed and used on a data? day basis.

Starting point is 02:54:30 So it requires two parties. One party who's running the server to install our wrapper around it and then the other party who's like making requests to that server to use our SDK. And essentially you can think of it as like a very specialized form of encryption where you can only decrypt the data that you've submitted to the server if you've met some constraints like you don't log the data, you don't train on the data and no one has access to the data, all that type of stuff. And is that just like enforced at an engineering level?

Starting point is 02:54:59 or enforced at a legal contractual level? It is a technical level. Great question. And that's what's different, right? You can make promises that say you want to train on the data, but we make it's a technical guarantee, and it's essentially a bunch of fancy cryptography that makes that guarantee.

Starting point is 02:55:15 So we actually offer unlimited liability and indemnification for data breaches and misuse because we're so convinced that we cannot see the data, no third party can see the data. Yeah. So what's the go-to-market like? You mentioned an example of like a big biotech company. Is that the most obvious customer or are there other segments that are logical?

Starting point is 02:55:37 Biotech legal, finance, of course, defense and government. And not just like pure defense, but local jurisdictions, states. They're all trying to figure out how to deal with freedom of information. When have you made something public? Was it too soon? this helps, you know, manage all that stuff. Privilege for lawyers, same problem there. Trying to figure out, well, if I give my day to open air, have I disclosed it and it's no longer subject to privilege?

Starting point is 02:56:06 Yeah. That makes it sense. Super interesting. Close it out with how big is the team? Where is the team size going? What are you hiring for? What comes next? We're about six people right now.

Starting point is 02:56:18 And we've been building. Building, building, building. And now with the launch, we're ready to start selling. And so our focus is bringing on salespeople. I've done two previous companies and every time I've learned that I should spend more on sales. So that's what we're doing. Interesting. Good takeaway.

Starting point is 02:56:34 That's good. That's the environment that a great sales leader or individual contributor wants to join. I mean, it takes a lot of technical CEOs, like a while to actually get through that. So it makes sense that you're in your third company because, like, yeah, a lot of people learn that lesson late, right? Awesome. J-Moh, thank you for joining. Thank you so much for joining. Come back on when you have news.

Starting point is 02:56:56 and thank you for doing this. We'll talk to you soon. Good to me, guys. Cheers. Let me tell you about wander. Find your happy place. Find your happy place. Book of Wander with inspiring views.

Starting point is 02:57:09 Hotel great amenities, dreamy veds, top of your cleaning, and 24-7 concierge service. It's a vacation home, but better folks. And next, we have the founder of contextual AI. Welcome to the studio. How you doing it? What's happening? Also, if I have this correct, the inventor of Rag, correct?

Starting point is 02:57:26 I'm one of the authors of the rag paper. Okay, okay. Yeah, that was a team effort, lots of folks, and it's a long history of, you know, research that has gone into that. Yeah. Give me the state of the union. Very, very humble. Yeah.

Starting point is 02:57:42 No, a man who invented rag by himself in a, in a cave of scraps. Yeah. Give me the state of the union on, on rag. Some people are saying, oh, just use a bigger context window. So like what are what are companies actually using RAG for on a day-to-day basis? What's the state? And then what's the shape of the industry that's popped up around the technology? Yeah.

Starting point is 02:58:06 So the RAG is really a very simple idea, right? It's about having Gen. I work on your data. And you do that through retrieval. That's the R. And you then use your retrieval results to augment. That's the A, your generative AI. That's the G.

Starting point is 02:58:20 So it's a very simple idea. How people do RAG right now is radically different. from what we did in the paper originally. I think the bus word these days is about context engineering. How do you actually give language models the right context so that they can do their job? And as it turns out, all the language models are pretty good these days. And there isn't that much of a difference between your clog and your open AI models or

Starting point is 02:58:44 Gemini. If you give it the right context, then it can solve the problem. If you don't give it the right context, then you can have an amazing language model, but it's going to fail. And so that I think is an operational opportunity that a lot of companies are looking at now. How can we make sure that this context layer really works? Do you feel like there's a RAG step or layer in the typical deep research products, products that I'm using on a day-to-day basis? I feel like it's mostly like going out to the web, searching, and then kind of just summing all this stuff up. But I can't tell if it's actually,

Starting point is 02:59:18 like, the basis of RAG is like embed all of that into weights and then and then search over it. But is that happening at the state of the art right now? I think so. You're right. A lot of it is web search. And if you want to do web search efficiently at scale, then you probably use simpler algorithms. So things that aren't that involved. But even web search very often uses embeddings and very research there. So you could argue that web search is also just... Interesting. Rag. Why is search so bad right now?

Starting point is 02:59:53 I feel like I can't search my email for anything. And Google has like frontier models. But yeah, I feel like search has just become like really hard. But then at the same time, I'm like having my mind blown by LLMs and deep research products. But I don't want to wait 15 minutes just to search my email. But maybe that's what I need to do.

Starting point is 03:00:11 Maybe that's the future looks like. Like what is going on there? Yeah. I mean, that's a hard problem, right? but I mean, AI should be able to search your inbox for you and just give the right answer. That's the goal. I think it's happening. It's just search is a really hard problem to do.

Starting point is 03:00:26 You don't want to really do it in a single step. So the way you do proper retrieval is multi-stage, sort of cascading with smarter and smarter models that look at what might be a relevant result. But you're right. If you retrieve the wrong things, then you can never give the right answer. Yeah. So that's a big open. Where are you guys focused today? I know there's a number of different use cases.

Starting point is 03:00:53 Yeah. So the use cases we're looking at are really your bread and butter use cases for Rags. So answering complex questions on complex documents. In our case, we scale to millions of documents, which is unusual. So one of the most common misconceptions about RG is that people think that it's easy, which is actually probably true if you have like, two or three documents, you understand the use case. It's not that complicated. But when you go to the real world and you talk to some of our enterprise customers, they have very difficult

Starting point is 03:01:24 problems. The data is all over the place. It's very complex data. There are millions of documents that it needs to work on top of and then search breaks down. So you can't give the right answer even if you have a great language model. Yeah, I mean, the scale of data at some enterprises is probably, what, seven orders? I mean, I imagine like the number of emails sitting. in Gmail inboxes across the entire network is way beyond millions. So that is also to your earlier point about like long context models. You can't fit all of that in the context of the language model. You need to do it.

Starting point is 03:01:58 No way. What's the state of the business today? How big are you? What are the new challenges? What are the next milestones? Yeah. So we're 70, 80 people working on a lot of interesting problems. That's kind of nice.

Starting point is 03:02:12 kind of across the board trying to expand into different use cases as well. So beyond the traditional rag use cases, looking at things like root cause analysis, code gen, because code gen is very hard and also often requires technical documentation that you don't want to incorporate in your code gen. So, yeah, making a lot of good progress on very interesting problems. Very cool. Well, thank you so much for stopping by. Love to have you back for a longer conversation.

Starting point is 03:02:41 Yeah, let's do it again soon. Hope you have a great rest of your day. We'll talk to you soon. All right. Thanks for you. Cheers. Bye to meet you. Let me tell you about Bezell.

Starting point is 03:02:48 Go to getbezzled.com. Your Bezell concierge is available now to source you any watch on the planet. If you like enterprise, agentic workflows, you might like fine watches. And up next, we have Open AI. Big launch today. We are going to break it all down. Welcome to the stream. How are you?

Starting point is 03:03:06 I think we got caught up on guests. We have two people. What's happening? Great to have you guys on the show. I will start by saying sorry about the sorry that Coldplay had to have a conference last night like that you know it was hard to you know hard to predict you don't know companies aren't normally launch anything on the internet I know we were just talking about this yeah well we're here no idea he says it should it's okay well normally you don't have to you don't have to plan your

Starting point is 03:03:33 launch schedule based on Coldplay yeah Colplay is out maybe it's something to keep in mind break down the launch what is in which we be focused on now? So I used to work on deep research. You used to work on operator. We both had our launches earlier this year. And I think after our launches, we realized that our products are very complementary. And so we've basically combined the best of both into this new product, chat GPT agent.

Starting point is 03:03:58 So chat GPT agent has access to a virtual computer with a bunch of different tools installed. So it has text browser, visual browser, terminal. And it's able to do a lot of different things that you would do on a computer. So it's just like very flexible and pretty powerful model. We trained at using end-to-end reinforcement learning, like our past reasoning models. And yeah, it can make slides, make spreadsheets. Yeah, what initial use cases are you guys most excited about? Have you been using internally?

Starting point is 03:04:28 What kind of companies should be and just individual should be kind of taking advantage of it as of today? Yeah, I think so deep research, as I was saying, we combined sort of deep research and operator to build this. Deep research was really good at research. Like, I think the best product out there in terms of, or at least that's what I thought, or I still think, in terms of researching. And then operator was there to take actions. Combining these two, you open up a lot of possibilities.

Starting point is 03:04:54 You can do research. You can do actions. You can do research and then actions. And then on top of that, we added APIs. So, like, for example, if you have connectors, which we launched, I think a couple of months ago, you can connect Gmail, Google Drive, linear, and whatnot, all sorts of products. and combined with that, it becomes an extremely powerful

Starting point is 03:05:12 research and action tool. So, for example, like, personally speaking, I've been using it a lot internally just to even talk with the code base. Like, I'm solving a particular problem. I'll connect it with GitHub. I'll ask it like,

Starting point is 03:05:26 hey, can you sort of go figure out what's happening in this code, which is, let's say, a new code base for me. And then the model is also very natively multi-turned, which means that I can just have conversations back and forth with it, which is not true necessarily. wasn't true, for example, for a deep research model, which released earlier this year.

Starting point is 03:05:43 So it's really, really useful specifically from at work when I'm able to connect all these amazing tools and able to just understand what's happening all parts of this decision. Secondly, I was going to quickly, I was going to say, like, personally, I also use it a lot. I think there is a lot of small things or big things that I have to do. I can do them myself, give an example. My wife and I have a date night every Thursday. I forget to book it most of the time. And then I get in trouble.

Starting point is 03:06:12 And then now with agent, like you can schedule tests, etc. You can just say like, look, every Thursday, just go ahead and figure out, give me five recommendations in the morning, which are available. And I can just do it, show up on Thursday morning, just click it, and it's done. So things like that. That's amazing. Once deep research came out, it felt like there was this little bit of a meme about like we have 15-minute AGI. And I'm trying to understand, I could imagine this new product stretching that out to let it run, if it's building a spreadsheet and scraping data from different sources and putting a whole bunch of

Starting point is 03:06:49 different things together. I could imagine letting it run for like an hour and coming back. But you said it's multi-turn. So what does the typical interaction look like? Is there a wider variance? I noticed in the latest revision of the chatchip-tie app, there's now a little like 15-minute UI element next to deep research to kind of hint that, hey, you're getting yourself into a 15-minute cycle. Obviously, there's lots of efforts to speed all of those processes up and that'll come.

Starting point is 03:07:15 But what is the typical interaction time look like? Is this more asynchronous or synchronous or kind of you can do either? Like, how do you think about those tradeoffs? Yeah, I think our team in particular is really focused on solving harder and harder tasks that take people more and more time. So a lot of the agent tasks can take anywhere from like around five minutes to, I've seen it take over an hour. Wow. But a lot of times these are tasks that would have taken humans through many, many hours. Yeah. So I think like as our agents get better, probably the length of time they'll take to solve tasks will also get longer because just the task will become so much more complex. Like imagine a task that takes a human like many days. Yeah.

Starting point is 03:08:02 Yeah, so maybe, so a question I have is right now the agent can browse the web for me, do research, take action. I imagine the next step would be an agent, like some, like a voice extension to it where as an example, I might say, hey, I use GEICO right now. I want to potentially switch. Can you go out and do research? Here's my, here's the cards that I have, try to find a cheaper price. and then or even call Geico and negotiate a lower rate or, you know, you're traveling, let's say, and you're to cancel an old internet line on spectrum right now, and they only allow me to call during weekdays and I'm busy. Yeah, the other example, you're on a vacation and you want to change your flight. And you know, you're going to maybe have to like call and sit on a wait.

Starting point is 03:08:48 So is that the direction that you think we're going towards where it can not just browse the web, but then actually start to. It's funny to think about because. the voice agent would then just call and maybe it's talking to another agent on the other side. But, yeah, where are we going? Yeah, voice is definitely an interesting form factor. I think the way to think about where we're going is twofold. The first is what I said is I just mentioned.

Starting point is 03:09:14 I think we want to continue to solve harder and harder and longer and longer tasks. I think today we can solve, let's say, an hour or so of tasks. Hopefully in future we can solve multi-day tasks and it might take longer. It might take shorter depending on how we're doing. but continuing to improve on reliability and complexity of tasks that we can continue to solve. So that's sort of a core part of what we want to continue to improve on.

Starting point is 03:09:38 Second part, the Geico example, for example, the one you gave, technically you can do it today, not with voice, obviously. You can just type it in an agent and it'll be able to essentially answer the query. You can log in with the virtual browser that we have an agent with your whatever internet provider you use, and I think it'll be able to tell you things

Starting point is 03:09:58 about what's happening there, what's not happening, what other alternatives might there be, and all of those things. But then at the same time, you bring up a really good point, which is like voice might be a very natural sort of way to do this in future. And that's a form factor. Yeah, and other companies introduce voice as an intentional point of friction, like the sort of call to cancel. Spectrum would never do that to me. Yeah, they would never do that to you, John. And so that, that to me feels like this, you know, if I can go into a web app and just click cancel or do something like that.

Starting point is 03:10:29 My question is about like the user experience of like having something that could take five minutes or an hour. How predictable is that? I've gotten in a great pattern where I expect deep research to take 15 minutes. And so I know when to go to deep research and I'm going to come back later and it's great. But if there's some variability there, is it going to send me a push notification when that's done? Is that how that works?

Starting point is 03:10:52 Like how do you train the user to get the best experience? Yeah, I will send you a notification. I think actually the fact that deep research always takes the same length of time is probably I don't want to say a bug, but I think of it is a feature. It's not the end state. I think of it. I should think that, yeah. I should think for as long as it needs to think.

Starting point is 03:11:12 Totally. But I think for deep research, it always just thinks for a really long time, even if you ask what the weather is. Yeah, that's true. So I think that's a better middle state. And I think that this model is like a step towards that, but I think it still will think for too long on like really simple queries. Yeah.

Starting point is 03:11:26 Yeah, you're totally selling deep research short. Sometimes I ask you what the weather is, and I want the history of weather from start to finish and what it is tomorrow and yesterday, and I want the history of meteorology and how the Doppler 3,000 works. I want everything. And I love that about deep research.

Starting point is 03:11:41 It's my favorite. The use case that I'm sure that will immediately start happening that is pretty hilarious to think about is a student that just says, hey, these are the three websites that host the homework that they have to do. proactively go to the website and figure out the homework created and submit it. Yeah, but teachers are going to be like check this homework.

Starting point is 03:12:04 If somebody wants to say that's not that's not AGI, I don't know what to tell them. Yeah. I don't know. What about other tool use? Jordy mentioned. Jordy mentioned phone usage. You mentioned spreadsheet integration. What's kind of further down the stack of integrations that you've already announced that

Starting point is 03:12:24 might be kind of under explored or under appreciated at this point in time. To me, I think that the tool that we've given the agent is very general and powerful. Like you can almost do anything that you need to do on a computer with this tool because it's browser and terminal, which you can do most things. It might not be the most readable to a human. So I think that now it's about pushing the capabilities. Like you can ask it to do anything in theory. It's just the agent won't be good enough to do everything you ask it to do.

Starting point is 03:12:56 So I think that we just need to make it better and better using the tool it has. Yeah, I think the frontier continues. I think we, as these have said, the tool is extremely general. It can, it has access to a browser. It obviously has access to a terminal. And we can give it access to as many APIs as possible. Sure. That should be, that should allow you to build whatever you want to do, generally speaking.

Starting point is 03:13:17 Like, for example, you can totally imagine in future it has access to a voice API or what not, whether it's internal, external, depending on, like, how things go. Right? Like, you can have access to everything and build everything, but we still need to push. Like, it's still early. Like, we've not solved everything. It's still early. We still want to make sure that we can solve use cases with really, really high reliability,

Starting point is 03:13:39 and that continues to be a pretty large focus. Yeah. Well, I'm excited. I mean, think about a world where you can give it access to your password manager and things like that, so that it just immediately can act. We're just API integrations, right? So then the passwords don't even need to pass back and forth. That makes a ton of sense.

Starting point is 03:13:55 Yeah, I'm excited for, I feel like deep research maybe doesn't have access to images in Chatchabit yet, but I could imagine those being way, like the reports being way richer if you can define them. And then sometimes when I'm just generating like a general chart, I actually want to use like a visualization library and Python and kind of going back and forth. So very cool to see it all kind of come together and very excited for where this is going. What's the rollout strategy? When can people actually start using this stuff? People, everyone on pro plan should be able to use it by end of day to day. Let's go. And we'll get it to plus users over the coming days and then enterprise over the coming weeks. Very exciting. All right. Well, congratulations on the launch. Super exciting.

Starting point is 03:14:43 We're going to turn this day around. It's now just about opening I agents. Ignore all cold play memes going forward. Well, thank you guys for joining. Thank you so much. Thank you for having us. We'll talk to you soon. Cheers. Bye.

Starting point is 03:14:57 Up next we have Dan Shipper, friend of the show, over at every. Who got early access. And he's been using it. And we're going to get some feedback from him. John just spilled his base juice all over the table, all over the FT. That's really disappointing. You're not going to be able to read that on the way home. I have lots of papers that I can.

Starting point is 03:15:16 Do we have Dan in the waiting room? There he is. How's it going, guys? Great to see you. Every time you're on, you're in a different place. Thank you. You finally caught me when I'm not traveling. Yeah, yeah, yeah.

Starting point is 03:15:30 Yeah, this is the first time. I like that light up logo in the background. Very nice. Very nice. Settled, subtle kind of in-home, out-of-home advertisement. That's what we're going for, exactly. What's happening? What's on your mind today?

Starting point is 03:15:47 There's a lot going on. There's a lot going on. We're here to do a vibe check. Vib check of chat Chb-T agent. So I was lucky enough to hang out with it and work with it for the last couple days before it got launched. And I have a bunch of things to tell you about how it works. Incredible. So as your previous guests who are amazing told you, it's sort of like deep research and operator had a baby.

Starting point is 03:16:13 And it does some really cool things. So one of the first things that I had to do is I had to go through all of our support, emails and all of our feedback forum posts for the last like two months. So it's like about 50,100 support emails and maybe like 500 posts on our forum to gather for Cora, which is our email management AI app to gather all of the customer archetypes of like, okay, who's posting, who's a promoter, and then going and looking on their LinkedIn to be like, what's their job? You know, where do they go to school, all that kind of stuff and put together a like long research

Starting point is 03:16:50 report of who our promoters are, what the archetypes are, and who our detractors are, and why they don't like us. So that's the kind of task that, like, obviously, like, I could have done or someone on the team could have done, but it would take it so long. Yeah, it takes a long time. And it's the kind of thing that you almost want on, like, a recurring schedule. Like, you just kind of want to see, like, once a month, but no one wants to do that once a month.

Starting point is 03:17:13 And you can schedule with chat DBT agent, you can schedule it to run. So I can just say, like, every month, I want you to just send me your, you just send me a or on the first of the month, which it's really freaking cool. That's wild. That's amazing. I wonder how compute intensive that's going to be because if I know anything about building dashboards and building like these reports, there's always like an intense amount of like, oh, we got to have this dashboard. And then you check the analytics and it's like, oh, it turns out the team just said that for a week and then like stopped watching it.

Starting point is 03:17:43 And if it's running, you're like burning. Dan is every foundation lab's worst nightmare because he gets on the most expensive plan and uses his. hundred times more than anyone else you're single-handedly gonna bankrupt a lab there's other there's other users that are probably higher margin but but yeah talk to me about that that that actual experience did you have to oh off with any different services did you have to share any API keys did you have to export any data or is it really as simple as just a prompt it's basically a prompt what happens is you type in you type in a prompt you say I want you to check out

Starting point is 03:18:16 Cora I want you to check out our emails I got I want you to check out our support forum. So chatchipi has connectors. So I previously already connected my Gmail. So you just like log in on the Oa. And then what it will do is it spins up its own computer on the cloud, its own virtual machine. It goes in the browser and starts browsing the web. It also then connects to connect to Gmail. If it hits a login, so for example, when it hit LinkedIn, it like couldn't log in and you can take over the browser in the virtual machine and type your password in, which is like a little a little janky, but it works pretty well. I think the interesting thing about this, though, is it seems like there's two main approaches to agents, and Open AI and Anthropic are taking

Starting point is 03:19:00 very different paths, and they have very different tradeoffs. So the really cool thing about Chatshibati agent is they're essentially abstracting away the browser and the computer. So all you're doing is you're interacting with Chachbtbt. And on the back end, all this other stuff is happening. So it doesn't matter if you're on your phone, if you're on a crappy computer or whatever. They have this whole virtual environment set up. It spins up. It does the task and it spins down. So it's like it's a very good consumer experience. Cloud code for example from Anthropic, which is I think cloud code is way more for developers. Chattiebtee agent is a way more I think for consumers. Cloud code is all on your computer. It's all in the terminal and and you have access.

Starting point is 03:19:43 It has access to all of your files and you have the ability to use it wherever and whatever and however you want. So it's much more customizable and much more composable. So I find that Claude code is much more powerful, but it's much more intimidating. And it's just not something that a consumer can use. And I think that... But still the crazy thing there is doesn't Claude Code have more downloads right now

Starting point is 03:20:10 than the actual Claude mobile app? It's something crazy like that. I honestly think people are sleeping on cloud code. Like I use it all the time for non-programming tasks. And I think most people think they can't use it because it's in the terminal. And the terminal is really intimidating. But it's an incredible product. So, yeah, how would you solve this problem if you were to do the same eval of like generate your net detractors, net promoters using cloud code?

Starting point is 03:20:36 You know just open up the terminal on your laptop. You wouldn't be able to do on your phone. But you'd you'd just engineer a prompt that told it to do that. and it would just write all the code that it needed to do exactly the same thing. Do you think it could hit that? Do you think it could do that for sure? And I think the nice thing about cloud code is you get many bytes at the Apple. And you can like, so for example, with cloud code, what you can do is you can have it make a full plan.

Starting point is 03:21:05 So it can output like a full markdown document with like a, you know, 300 or 500 or 1,000 word plan. You can modify it and go back and forth with it. and then have it execute it. I think it would be more complicated. Like, yeah, it would probably write some code to hit the Gmail API. And I'd have to, like, think about that as opposed to just, like, clicking the connectors button. Or it does have a web research tool, so it would be able to go to, like, our

Starting point is 03:21:27 feedback forum and do all that stuff. And it would be able to save all the data so I could kind of watch what it was doing as it was doing it. But so I think you can get basically the same experience. I think CloudCode is a little bit more controllable and therefore a little bit more powerful, but chatypt is just much easier to use. It makes sense. What use cases do you expect chat chatt chaptiPT agents to have the most PMF around?

Starting point is 03:21:55 I was imagining the student use case, which is just like monitor the homework that I have do across, you know, I remember in high school, even teachers would host their homework on websites. You could basically run something that was like, monitor the homework assignments that I receive, and then take a preliminary pass at doing the assignments. and then give me a draft that I can review and sign off on or tweak and then automatically submit my college applications attend college write my college essay just deposit the money that you make as an engineer at multiple companies into my bank account and then also plan a trip to

Starting point is 03:22:29 Europe because I'm retiring uh watch out clearly uh chat jbt agents coming for you um guy boom breaking news breaking news um that's funny no but it but like give it this powerful of a tool to everybody immediately, not everybody's going to realize it, adopt it right away, but you can imagine like a few use cases just spreading like wildfire. Totally. Yeah, I mean, I think what are the things that you would immediately do if you had an assistant? Like if anyone, if someone just dropped an assistant into anyone's lap, like what was the person that they would do?

Starting point is 03:23:07 I don't know. Help me book a vacation. Help me like figure out how to order groceries. help me like another one another thing I use it for is like research the web about all the topics that I care about and every day give me a report on all the things that happened in the last 24 hours and it just does that and it's incredibly well and I can go into you know go behind login walls and paywall and all that kind of stuff so I think those kinds of use cases are gonna be the gonna be the the most

Starting point is 03:23:34 interesting ones but I honestly think right now for most of my consumer use cases 4-0 or really 03 is the best it's much faster I mostly don't need it to use a full computer to spin it up so I see chatagasy agent as being something that you use every once in a while rather than saying you're using every day sure yeah so so we're we're increasing the level of like complexity like 4-0 is kind of a Google search replacement for me now I just kind of hit it with like when was this person born how old this you know what's the state of this what's the capital of this state or expect a really quick answer then go 03 if I'm willing to wait a couple minutes

Starting point is 03:24:15 want something that's a little more thoughtful maybe some search results from the web then deep research if I'm actually trying to understand the full story read a whole report agent if I think it's going to need to use a computer actually take some actions pull some things together how was the actual interaction of the like the back and forth this is something we talked about with the open AI folks was like if it gets stuck it it pings you I like that deep research, yeah, it takes 15 minutes, but I've trained myself to just be like, forget about that until tomorrow. And then when I have time to sit down and read the full deep

Starting point is 03:24:50 research report, which is going to take me a couple of minutes, like then I'll come back to it. I know it'll cook and it'll be done. It would be kind of annoying if deep research came back after two minutes and said, hey, I'm going to pause all that while I ask you for an update. Like it feels like there's a little bit more. I got to be on answering questions, but push notifications maybe solve that walk me through like how in how involved you were how active over process it is I mostly was not involved every once in a while like it does have a like a stop you can help a stop and like change what it's doing which is nice because like if it goes off the rails that's helpful but I think and it has a push notification thing

Starting point is 03:25:30 but I think this is an interesting problem with agents where I can't stop watching them and so I spend a lot of my day just like watching the agent doing something and chagipkee agent It has its own like cool UI where you can kind of like see interesting animations of what it's, what it's researching or which websites is using and stuff like that. So I find myself actually glued to it to it a little bit. And I just don't think that's a very good way to spend time. I think it's I think it's mostly solved by having push notifications. But like there's a sort of emotional process of training yourself to be like, it'll let me know when it's done.

Starting point is 03:26:03 And I don't have to like wash over the shoulder. Like we design a lot of assets here at TBPN and that like even if I trust the person creating it to do a great job, there's still this tendency to want to like hover and be like, okay, tweak this, tweak that. Oh, let's do it this way in real time versus like waiting. But it's the same thing with 4-0. Like, you know, early on I would kind of be in this loop of like, okay, I got a result. I still got to go fact check this and check the underlying links because hallucinations are a big problem. And well, now that they beefed up search so much and they're referencing direct quotes, like I feel like I'm much less like the anxiety level around hallucinations is a lot lower,

Starting point is 03:26:44 just in general queries. Totally. I think these things are tools. Anytime you're delegating to something, whether it's a human or an AI, there's a learning process you have to go through. Like human managers go through this with employees all the time. Like if you're a new manager, you have to like decide, okay, am I going to delegate this? or am I going to micromanage? If I delegate, like, I get more leverage, but it might not come back the way I wanted to.

Starting point is 03:27:10 And good managers know how to split up a task or communicate it to their employees or figure out who's good at doing what and know when to get into the details and when not to. And I think we're going through the same curve with models. So we're becoming model managers. And everyone is learning how to how to solve the same problems that human managers have solved. And so the more experience you have with the tool, the more you know, like, okay, I don't have to check this 4-0 answer, or this looks a little fishy. Same thing with Chachibhia agent.

Starting point is 03:27:38 I think will be much better at using it in three or six months than we are today. Cool. Makes sense. Dan, always great to chat. I do want to have you back on very soon to talk about LLM-induced psychosis. I think it's important to talk about. Let's talk about it. But we'll need a lot more time.

Starting point is 03:27:56 Thank you for the vibe check and everybody listening. Go subscribe to every right now. Do it. Awesome. Talk soon. Talk soon. Bye. Cheers. Up next, we have Chris Best from Substack, the best CEO. Substack's ever had, arguably. Undoubtedly. He's in a conversation.

Starting point is 03:28:14 Welcome to the stream, Chris. How are you doing? Congratulations. You got some news for us. You got some numbers. Get it ready, John. What's going on? What are we got? What's going on in your world? Please tell me at least nine figures. Do we good. We've raised a hundred million dollars series C.

Starting point is 03:28:28 Woo! Congratulations. I was hoping you guys would ring that thing. Of course. You guys are incredibly back. Yes. It's been a journey since I believe you raised something. It was like 75 on 700 back in, what was it, 2021?

Starting point is 03:28:50 2021. Those were different times. I don't know if you guys remember 2021. I do. Oh, I do. I remember it fondly. You were at the center of the storm and I was in the depths of a slog basically. Anyway, give us.

Starting point is 03:29:05 the update how the round come together what is the plan going forward i heard you were it was the reporting accidentally profitable going back into burn mode what's the money for what are you thinking yeah i like that um yeah we're you know partnering with mood rogani at bond um i love that guy consummate bro joining the board um the big thing is look the the big thing that's happened is like substacks gone from being like a rinky dink email newsletter company to a proper sort of network that's taken over the world. And we kind of want to look, basically kind of like switch into a mode

Starting point is 03:29:41 of thinking about long term ambition, long term, like how do we actually make the big fucking version of this thing? What investments do we need to make? How do we focus on the things that actually matter for like the long term flywheel growth of the network? And this just gives us like a total free hand

Starting point is 03:29:57 to do that thing and build the best possible version of it. Okay. Give me the pitch for, I think of substact. as the no-brainer place to launch a newsletter. You have a lot of other products. Talk to me about what the future of the substack creator or someone who has substack as like their primary out. It's the main engine of their creator economy business, for example.

Starting point is 03:30:25 What does that look like over the long term? I imagine that people are still doing top of funnel stuff on TikTok, Instagram, other places, but you're adding more and more features. What does a well-run substack business look like? Yeah, you know, the core of substack is the direct connection with your audience, right? So people subscribe. You get their email.

Starting point is 03:30:47 You get the ability to reach them. You can get the ability to like leave substack and take your list with you, which is a big deal. People can pay you directly so you get recurring revenue. I don't know if you guys have had this, but recurring revenue hits different. It does. The sponsorship business is a great business. I think that thing matters. Lots of people on Substack have sponsors.

Starting point is 03:31:07 We love it. But that thing's like very cyclical. It's boom and bust. It's like, you know, whereas you have recurring subscribers, these diehards, that's sort of like it funds sort of like you to be creative. It funds you to take risks. And so Subtack's the place where you sort of like your hardcore people are. That's where you have a real connection to your audience.

Starting point is 03:31:25 You can write. You can post short form. You can post video. You can do live video now. You can have a community. We're kind of like building more and more stuff. center is not any one format. It's like the relationship with the subscribers. And then Substack is just becoming, you know, you said you do Top of Funnel on TikTok and LinkedIn and

Starting point is 03:31:43 YouTube and everywhere else. Sure, keep doing that. That's great. Those are massive platforms. But increasingly you can do that stuff on Substack too. And because you have such a dense audience of like smart people, the quality of growth you can get there is already very high. Interesting. That makes a ton of sense. Jordi. I think the, I think the magic thing. that you guys tapped into that I end up find myself I find myself explaining to other people is there's this beautiful like like economy of people on substack that just want to support people that are nerding out about a specific topic just want to give them money and so it's almost

Starting point is 03:32:22 like this there's like there's this like exchange of like yes I want the content but it's also enabling somebody to live a life that allows them to just just obsess over one thing or just explore a series of topics or just be who they are and be entertainment through that. I mean, we've had Emily, Emily Sundberg has come on the show a bunch of times and it's just like it's so awesome to see what she's built and the kind of creator, writer that she's able to be unshackled from being at a specific, you know, platform, legacy media company. So it's just, it's so, it's so awesome to see. Talk about the use of funds.

Starting point is 03:33:05 You said you can afford one AI researcher now. I imagine that won't be how you spend it. Never know. Yolo. Yolo. Start coaching from Mark. Concentrated bets, man. That's how I'm working out.

Starting point is 03:33:17 Concentrated bets. But I mean, concentrated bets, it's not the craziest idea to go give a bunch of money to, you know, creators to kind of pull forward the leap from what they're doing currently to get on substack. There's different incentive models, kickstart ad businesses, just higher engineers that can build new tools and new features and just chop wood and advance the ball down the field. What are you most excited about to put that money to work over the next 12 to 18 months? But you probably think in like decades now at this point, right? Yeah, I mean, that's the big thing, right?

Starting point is 03:33:52 It's like what's the this lets us have that longer horizon. You can still have all of the same math, but you can just like put the planning horizon further in the future and look for something really big. Listen, all that stuff you said, the stuff that I'm really excited about is like making the product fucking great. Right? I wanted to feel my joke is substack does everything for you except the hard part, right? You are the talent. You got to figure out how to write something that's worth reading, how to have a conversation that's worth listening to. If you can do that thing, though, we should just build this magic machine that takes everything else off your hands and makes it dead simple, makes it just like this magical thing where anybody who has something worthwhile to say can make something.

Starting point is 03:34:31 We're starting with that. We have a bunch of little bits of that are kind of working that we're really proud of that are exciting. But I just think the new technology coming online is going to make us like the version of that magical sort of like media studio personal media empire in a box that we can build now is going to be so much more powerful. And then building up like the network, right? The fact that we're getting, you know, not just political commentators, but politicians. I think we get not just sports commentators but athletes. I think we can start to build up kind of like this network and this. ecosystem that winds up being this positive sum game, right, where everybody that's on

Starting point is 03:35:04 substack benefits from this growing network. Yeah, I've certainly never subscribed to anyone on substack and been like, ah, I didn't get my money's worth. I always have a good time. What are the different strategies for substack writers? I know in like the Patreon podcast world, the people do like, one is free, one's behind the paywall. I've also seen substacks where there's like a fold and you get every email, but you get half for free and then you at some point there's a call to action to go and subscribe and and finish reading essentially but you get every email what works what are the different strategies what are some of the weird stuff that you've seen around the way people are using

Starting point is 03:35:44 sub stack today there's a pretty big mix right some people make almost everything free and it's just basically like you know if you want to comment or if you want to get the occasional thing that's what you're paying for so people do really low margin for chris over here you just give everything away from free. He doesn't make any money. That's the beauty of your system. Nobody's paying to get more, to get more things to read, to get more email, to have more seconds of audio

Starting point is 03:36:08 in their inbox. They're paying for perspective. Interesting. Right? Yeah. That's the thing. Like, you know, even if you have a magical LLM that can spit out media in any format, you care about who it's aligned with. You care about like what version, you know,

Starting point is 03:36:23 what worldview you're getting. Is this something I trust? Is this something I want to be culturally and aesthetically a part of. You talked about people paying because they want to support people. The other way to say that is it gives you agency. When you choose who to subscribe to, you're choosing what part of the culture you want to live in. You're choosing what gets created. In a world where people, I think, feel like a lot of the media they consume is kind of like stuff down their throats, getting to kind of like exert a voice and say, I'm causing this thing to exist that I think is great,

Starting point is 03:36:54 is really powerful. And a lot of different ways. It's like paying with your, paying with dollars versus attention is super powerful, right? Because if you're paying with attention, it's like, well, then everybody's focused on the cold play, the cold play debacle last night, which is trying to steal thunder from your fundraising announcement, but we're not, we're not letting it. And being intentional about like, I want to pay for this because I want more of it to exist. I want it to get better. This is how I want to spend my life. I don't want to be on a platform that just is designed to like suck my time from me. but I want to spend my time and attention better on smart things like TBPN and everything on substack. Yeah, no.

Starting point is 03:37:37 Amazing. And coming soon, hopefully. We will figure it out. We're about to bet big on substack. Yeah, we're very excited. We're riding with you. So congratulations. It's really tremendous to see how far the businesses come since those glory days in 2021 and excited for the next five years.

Starting point is 03:37:57 Thanks, guys. Cheers. Congratulations. We'll talk to you soon, Chris. Cheers. Awesome. Next. One last guest.

Starting point is 03:38:07 Deccart.AI. That's right. The first ever world transformation model turning any video game or camera feed into a new digital world in real time. Very, very cool.

Starting point is 03:38:18 I played around with a lot of this stuff, not this in particular. Very excited to talk to the founder. Welcome to the stream. How you doing, Dean. Welcome. Sorry for the chaos. We are so happy to chat with you.

Starting point is 03:38:27 Super nice to meet. both. Super nice to meet you both. Good to meet you. Why don't you start with an introduction? I already have questions, but just give me a little background on yourself and the company. Okay. So, you know, the card, the car's a very young company. We're less than two years old. We're a research lab and our goal is to build a consumer company. Okay. Okay. And what we just launched today is the only real-time video model ever. I can just show it to you. Yeah, please. Let's do it. Share screen somehow here. Yes, but you are live.

Starting point is 03:38:58 So whatever you share on that screen is baked into the internet forever. Amazing. We just have to make sure that we don't really leak anything. Just do the right tab, not the API keys. I wasn't sure we've never met. I mean, we've met over DMs, but not face-to-face. I wasn't sure if this was your real face or just a character that you're playing. This is definitely not me.

Starting point is 03:39:22 You realize that, right? Yeah, in the real world, he's an anime. But he's using a transformer model to appear like a human. That's right. He's in fact a Minecraft character. I have so many questions about this model. I want to jump into how you built it. We could also potentially have the team pull up the demo video on your website, mirage.

Starting point is 03:39:44 Descartes. com. And just kind of show folks what that looks like. Whatever would be helpful. It is currently. I'm trying to get the Zoom call to be able to share this. But if the team could do it from your side as well, that could work too. Yeah, why don't we just have the team pull up the core website?

Starting point is 03:40:01 And we can just jump into questions. So real time, what's the secret sauce? Is it a condensed, distilled model? Are you using a special chip to inference this stuff? So real time, let's just talk about the use cases. The obvious use case would be like real time video calls. You're dropping into a Zoom call, for example, and instead of yourself, There he is. There's D. Here we go. Do you guys see me? Yeah, we see you. Looking

Starting point is 03:40:28 Okay. Okay. Okay. Okay. Now we got it working. There we go. There we go. What is the use case here again? Who are you? We're just cycling through. We're just cycling through everything. What do you guys like? What do you guys like? Anime? Not big into the anime world, but what about like Legos? Anything there? Let's let's put in Legos. We can just type it. Oh, you can just type it and it'll exactly okay. Let me make this full screen. You can just type in Lego and It'll just turn everything into Lego. Wow. Okay, there we go. This is our house that Wow, you see the real stream and the zoom yep you can see you know this will be the house office thing All the Lego characters walking around here. This is insane. Oh and then and then you can decide that you

Starting point is 03:41:24 you're into Christmas and so everything becomes very Christmassy and your house is decorated. Oh, wow, yeah. Wow. I feel, uh, what's the word for a moment like this? This is actually feels like enter the metaverse. Indeed. I was, I was hoping you wouldn't say that because that become like, you know, became a cursed word. Yeah, but like it's, it's a good thing. It should happen at one point, right? Because this is a this is a big transition from the moment where Zuck was you know said you know did his like hello from Horizon's world There's wee graphics was there's we graphics and this feels like being in a video game What it what is that is that a wand this this is this is this is you can see it's straw on the regular stream right? Yep, yep, and inside the wizards prompt it becomes a wand and if you flick it hard enough sometimes it does magic spells

Starting point is 03:42:17 It throws things out. I love how much fun. Guys, which one, last the team, the teams are playing with the prompts here all day. Like, which one is the, Galactic Wars the one you like? Oh, that's the lightsaber one. Okay. No, you had one that you shoot guns. Here.

Starting point is 03:42:35 Here. So is this running locally on your computer? So this, no, this is running on the server. It's running on the server. It's going from your webcam to the server, back to us. back to us over Zoom. So how quickly is, like, Kai Sinat going to be running this, like, half the time that he's streaming? This is, this is insane.

Starting point is 03:42:56 This is fun. It's pretty fun. Now, no, no, no, no. The big question is what you do with it, right? Yeah, yeah. This is a bold demo, too, because the variation in the different prompts and how, how well it's working is absolutely insane. Totally, totally. I have forgotten what you look like in the real world entirely.

Starting point is 03:43:15 Here's a question. Would you guys run an entire show through this? We can try. Probably not. We might put Tyler on the intern cam. Yeah, yeah, that's a good place to start. So we have an intern cam over here. We can get Tyler running on this.

Starting point is 03:43:27 Oh, my gosh. The way that you're just cycling through these prompts is actually. Tyler can be on wizard cam for a little bit for one stream. That's pretty crazy. This is insane. Something cool, let me show you this. Something cool that we found out is that it's really fun watching YouTube through this. Okay.

Starting point is 03:43:46 Watching YouTube. Tell me, give me a YouTube show that you like, something you like. I mean, just do that Mr. Beast video. Whatever's pulled up right there. Okay. So you can just look at this Mr. Beast video. We can look at this ad. And look at the ad.

Starting point is 03:43:58 And we can share the Mr. Beast video. I somehow have on the- You guys hear the audio? No. Yeah, we do, actually. You hear the audio? I don't know if we should, but. Okay.

Starting point is 03:44:10 Okay. And then it's being transformed now? Yep. This is the original Mr. Wow. Yeah, even trippier. You can, here's, like, Mr. Beast videos worked really well with cosmic medieval golden. We iterated through so many problems that we found the ones that worked really well.

Starting point is 03:44:28 Yeah, yeah, yeah. This is Mr. Beast videos in the cosmic medieval golden world. Okay, cosmic medieval golden. I mean, they're already pretty stimulating. This is even more now. Wild. Very wild stuff. This is insane.

Starting point is 03:44:42 Where do you think this lives? Does this live with the consumer and they decide to put it? on the rose-colored glasses or do you think the creator decides. Somebody's joining a stand-up tomorrow. Can they just automatically put the entire team into whatever character they

Starting point is 03:44:58 want? Let's see. What stand-up comedies do you like? I was talking about an actual like engineering stand-up. Oh, an engineering stand-up. Okay, hey, it can definitely do that. I'm more asking about you. I'm asking you, where do you think it winds

Starting point is 03:45:16 up living? Like, do you think Here's what I think is cool about this. You know, I think it's the first time that we have a new new kind of consumer interaction with video AI. Because so far I know video AI was just okay, let's create AI slot, put it on existing platforms, Facebook, Instagram, TikTok, whatever. Here for the first time you can actually do something that's slightly different. We're showing it to a bunch of kids and what they ended up doing was for a few hours,

Starting point is 03:45:43 they just fought each other with sticks and they started doing like TikTok dances in front of this. And you have a new kind of consumer experience here. Yeah, yeah. It's like the original in the, what was it, Steve Jobs did that demo of the Mac where it like warped the image. And it was like a crazy historical photo in the original like Mac OS launch of that. How expensive is this for you guys to run if somebody just starts streaming this in real time? Are they paying for it?

Starting point is 03:46:18 Are you guys just eating it? It's we're doing it efficiently enough. Like we had to write like all the low level assembly code for GPUs to get this to be both real time and super cheap for us. It's at a point that we can actually provide this for free. Like it can actually be monetized without like subscriptions. There are ways. It's it's cheap enough to actually be able to build a platform on top of this. Interesting.

Starting point is 03:46:40 Do you think, yeah, I mean, do you think one of the use cases will just be who create video content, running their video through this to kind of as like a previs for what they ultimately want to build? So that could be very interesting. Well, the other thing you can imagine it actually in YouTube. So like a creator, like Miss Rachel, for example, popular kids creator, could basically say like, do you want the Lego version of this video? Do you want the pirate version?

Starting point is 03:47:11 About where this lives? Because I can go and put my phone in gray scale. and I can view everything on my phone without color. And that is effectively a style transfer. And that's something that I, as the consumer, decide. Or, you know, Mr. Beast can turn up the saturation in an, or, you know, if you're watching a Hollywood movie, they might turn up the teal and orange. Because that is a traditional Hollywood color grade. If you're watching the Matrix, they might color grade at green when they're in the Matrix.

Starting point is 03:47:40 They might color it blue when they're out of the Matrix, right? And so it will be interesting to see where this lands, whether it's on the consumer. I like watching Lego version of YouTube. I like watching Lego version of my conference calls or it lives on the producer of the I want to show up as Lego. And then that can be transformed as well. Can you quickly do a Gigacad filter? Oh yeah.

Starting point is 03:48:05 Do you have a gigacad filter? Let's check that out. Let's see what Gigacad does. I will say that the model is really, in the current version of the model, is getting changing the entire style. Okay, yeah, yeah, not just the first. It's not, oh, turn me into Trump. Okay, you're pretty orange, yeah.

Starting point is 03:48:20 But it is doing something to the cheeks and the jaw, which is textbook gigacad. There we go, there we go. Very funny. There we go. Yeah. This is wild. Maybe you need to prompt like.

Starting point is 03:48:32 So is this already fully open access? This is open. We launched it literally 30 minutes before the podcast. We're waiting for you guys. Congratulations. Thank you. Yeah, yeah, yeah. This is the first time it's used on a video call.

Starting point is 03:48:47 Amazing. Insane. Well, thank you so much for joining. Yeah, this is a lot of fun. We will get access to it. We'll set it up on the intern cam tomorrow. For sure. Wait, you already have it set up?

Starting point is 03:48:57 Okay, we're going to hop off with you, and we are going to check it out with Tyler, get some more feedback. Thanks so much for hopping by. Congratulations to you and the whole team. Insane. And is there anyone else in the waiting room or are we good? I think we're good. Let's, can we go over to the... Let's take it over to Tyler.

Starting point is 03:49:13 The intern can and see if that works. I don't even know if he's there. Oh, oh, technical difficulties, of course. Anything else you want to cover in terms of news or... Yeah, there's some new news. Oh, wow. That's Tyler there. Wow.

Starting point is 03:49:26 Really? Whoa. Yeah, you just did it. Yeah, this is me like 30 seconds ago. Okay, wow, yeah. It looks better without the zoom compression. It's a pretty high fidelity model. Wow.

Starting point is 03:49:38 Very cool. Yeah, so we were talking. in the background so you hear our talking and then Tyler's in the corner too we're getting recursive not that we should use that word now it's the M dash or the modern era interesting pretty good and these were preloaded prompts that you were just clicking through Tyler yeah just like selecting you didn't generate any of your own prompts I didn't but I think you might be able to yeah yeah that seemed pretty cool oh some of this seems fun I imagine that people will be having fun on this on the internet very very very soon very very quickly I always just wonder

Starting point is 03:50:08 with this stuff like what the like the novelty is like you have to it's like wow it's incredible technology but then people have to actually figure out like a real use case for it like like well so here's a use case streamers yeah having some type of like basically like if you tip a certain amount or hit a certain button it puts the streamer into that's cool yeah yeah I mean a lot of streamers stream with with green screen backgrounds and they already drop out the background and put themselves in an environment that matches the game that they're playing we're just clean up the messy background. And so, yeah, you can imagine this being like basically in the VFX, in the VFX pipeline for, for streamers. I think if most people showed up to a, to a Fortune 500 Zoom call with this,

Starting point is 03:50:53 it might not go over too well. Well, they can always make everybody on their screen look a certain way and not make themselves look normal to everyone else, right? Yeah, yeah, yeah, yeah. If your boss is yelling at you and you put them in Lego mode, it's probably going to hit a little, bit differently. It's a good shield. More tolerant. Yeah. Turn the volume down. Lego mode. Lego mode. Oh, sorry, Lego boss. Block man.

Starting point is 03:51:17 You expect me to be a lot. A couple more headlines since we're here. Perplexity apparently just closed a new round at 18 billion. 18 billion. Wow. So perplexity has around 2% of queries according to semi-analysis right now. I use it regularly. We talk to Chris, who was on the show. 7 million DAUs, 30 million daily queries, 4.3 queries per user per day, a 2% share of users, and a 2% share of queries.

Starting point is 03:51:49 Solid. That's the perplexion stats. Solid numbers. We got their new browser. They're going to be investing heavily. And then outside of that, lovable, just raised $200 million at a $1.8 billion dollar valuation led by Excel. So lots of heat on the timeline. Lots of big rounds getting done.

Starting point is 03:52:07 Everyone decided to launch today, I feel like. And then they got steamrolled a little bit. But that's why we're here. We didn't spend too much time on the concert fiasco. Instead, can move on and talk to you about the news. Also, we want to send our best wishes to Tyler. He says, Hi, crew, I had surgery to remove lots of infected fluid for my chest

Starting point is 03:52:29 and a big lung abscess with infected tissue. Pneumonia, sepsis, antibiotics weren't working in the ICU. Now. Surgery went well. Surgeons are heroes. I'm so grateful recovering. Thank you for your prayers. I'll get. So I hope he's doing well. Let's send some prayers. Yes. Yes. What else is in the timeline that would be worth to close on? I have a good closing post. The lads over at Reindustrialize are having a lot of fun.

Starting point is 03:52:59 If you can pull this up in the bangers tab, you can see quite a number of former TBPN guests all in the back of a pickup truck. Fantastic. If you can pull this up, team. We're working on reducing the time between talking about a post and getting it live on the screen. There we go. Oh. All the boys.

Starting point is 03:53:23 We got Augustus, Packy, Steinman, everybody all in one place. Love to see it. So anyways, thank you for tuning in today. Super fun show. And I can't wait for tomorrow, John. Can't wait for tomorrow. Leave us five stars. Apple Podcasts and Spotify and we will see you tomorrow. Have a good day. We love you.

Starting point is 03:53:39 Bye. Bye.

TBPN - Astronomer CEO Affair at Coldplay Concert, JUUL Approved by the FDA, Is AI Causing Historic Crashouts? | Chris Power, Jeremie Eliahou, Chris Best, Alex Mashrabov, Jesse Pollak, Billy Thalheimer, Jonathan Mortensen, Douwe Kiela & More

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.