DOGE Update, Jelly Jelly Coin, Apple AR, The End of Advertising, Mary Meeker

30:59How a new social media format and app we seeded could and should intersect with crypto. When we did it slash put it out, we tried to do what is generally known as a fair launch. I seeded it by buying 2 % of the network, which I haven't, have no plans to sell. He hasn't sold. And then we let whoever wanted to buy it. No reserves for the company or founders. This is supposed to be fair, the right high trust way to do things because you aren't pumping something where you then later dump the supply on everyone who bought it and everyone has a fair shot to buy in. Now, what happened is still amazing to me, completely bonkers.

31:30It went from zero dollars to 250 million in network volume in a matter of just a few hours and since then has been selling off. What I do think happened, some folks that follow these things closely and trade them heavily decided that this might be the next big thing to run up and they quickly should get ahead of it. So they came in and bought a lot very, very quickly, which in and of itself created a snowball effect ton of momentum that fed on itself then at some point these folks decided as traders do to take profits and then started sending the whole thing back down do you want to comment there jordy you want me to keep reading yeah i mean i'll hold some comments to the end but this same type of situation happened with siki chen who uh somebody he posted about his daughter having this rare form of cancer.

32:17He asked people to donate. Crypto community decided to make a token in his daughter's name. Price went up like crazy. People started speculating on it. They gave him a lot of the supply and even went out and said, I am going to sell this at this point in time to fund my daughter's cancer research. And even though he gave everybody a warning being like, do not trade on this. Do not speculate. He had thousands of people commenting, saying that he was a scammer, all this stuff. And he's like, I literally told you, like, I am going to sell. You can front like sell now. And other people are still trying to treat it as a financial instrument and going so far in such a disrespectful way of saying, well, I don't even believe your daughter actually has.

33:05It was crazy. Meanwhile, for people like you and me, it was crazy to look at because we're like, yeah, one, I know, I know companies with Siki. He's started a ton of, you know, massive venture back companies. Yeah. He's not lying about his daughter having cancer to make, you know, a hundred grand or whatever. He's he, and he wasn't even the person who created it. So, yeah, it's crazy. Uh, so Sam goes on what sticks in my head about this structurally is that is just that if the person slash team launching something doesn't have a treasury, then the people that end up being the treasury are the traders who buy it first very, very cheaply.

33:42So in a sense, with a fair launch, you end up trading potentially having the people who start the project as the whales for the first traders in as the whales, which in some way might be even more challenging as a setup. Leading to the question, how do you actually fulfill the spirit of a fair launch that goes well over the long term, sets something up for success? leading to the question, how do you actually, okay, I have seen literally every variant of launch imaginable in crypto take on this in some form or another firsthand from Bitcoin relatively early and other straight mineable like LTC and PPC to XRP seed and company coin distinctions to Doge to Solana's brief lockup to Chia's originally unsellable huge company holdback to even more crazy recent vesting and staking schedules, all with the aim of getting something healthy off the ground.

34:33You have to accept that any nexus of trading plus long-term holding in a frictionless environment is going to have a ton of volatility out of the gate and you just weather it. But it's also amazing how nettlesome a problem it ends up being to get something off the ground that isn't over or under hyped and sensibly gross. To some degree, this is just the challenge of the frictionless internet. Things that accelerate have no brakes. They just accelerate faster and faster and nothing stays stable. But it's interesting to watch and experience and iterate through all these strategies to see if they work and realize that some of the tweaks, non-obvious elements that matter, i.e.

35:08this is all insanely different because of crypto anonymity versus how a KYC system would work. For those who are here because they are interested in Jelly Jelly, more practically Monday on what the plan was slash in is for integrating crypto into the app over time, for everyone else just interested in crypto and the implications of really frictionless finance. The simple answer, no one really understands how to do this well yet, even when they are trying. Interesting. One of the biggest issue here is that I would guess that 99.9 % of the people buying and trading the token had zero interest in the underlying product because meme coins fundamentally right now, the vast, the real use case, even though tokens can be used to be integrated in products and do a bunch of interesting things with them, the primary use case is speculation.

35:57And so they immediately realized that they were having to, they now are, I imagine, having to adapt their company roadmap to try to justify the token. Because if they just said like, hey, this is not really a thing anymore. We're just going to focus on building this software network. A lot of people would have a bunch of people who feel like they invested in this thing that the company created that they're now abandoning. Yeah. And so the launch posts are where it gets complicated. You can see Sam Lesson saying, keep Jelly in. And it's just a chart of green lines straight up to the right. And it has 2 ,000 likes and half a million views.

36:42And this feels like financial advice, right? Like this feels like I'm making money. You know my pedigree. You know I'm a big account. You know I'm rich. People will read this and say, I got to put money in this thing. And then he posted again, jelly jelly for those who love jelly to the moon. And so, again, I don't think this flies in the pre-Trump 2.0 era. But crime is legal. And so have fun with it. And stay safe out there, people. because you're going to see a lot of wild stuff on the timeline. Yeah. Yeah. We'll see what happens here. I actually think we should try the Jelly Jelly actual product because it seems very cool.

37:24The wrinkle here is that this feels very degen. It feels very pump. Even though it's this fair launch, you're definitely promoting it as a financial asset more than a product. It wasn't just like the post isn't just like, hey like i'm building this you know company and like by the way there's a token it's the it's literally the the picture is of the token price not of the it's not even oh here are day daus you know it's like yes the price is what is what you care about the price is going up and if you think about like you know it would be it'd be so funny if like you know mark zuckerberg posted a stock chart of like oh that is up two percent today like to the moon get in guys like i'd be like okay like whoa like it's a little aggressive guy um well this is do you remember the news is that people the people that see this and like act on it they are traders who know the game they're down for the pump and they're um and they're just playing the game so i'm not that worried about like some random retail trader like seeing this and then and then oh yeah i i figured out how to set up a fan for two seconds like it's not the same so the funny thing is you remember the morning after the Trump token was announced, we had a buddy who's raised a lot of money for a very, very cool, highly topical business.

38:46I want a super talented entrepreneur. He goes, guys, I think I should launch a token right now. He's like the Trump thing happening. And I'm so glad we gave this guy the advice that we were basically saying, hey, as soon as you do this, it's just going to become its own thing. And you're going to, it's like, you wouldn't, wouldn't, you know, so like all this drama that Sam's going through and the company, they didn't make any money from it. So they got this like dopamine rush of seeing the chart go up and all this craziness, but it seems like the, you know, maybe, and maybe they could have bought it, bought in and traded, you know, not, not with, you know, the Genesis wallet it or whatever.

39:30But yeah, it's just like all this kind of drama and chaos and no sort of financial benefit, which is why we told our mutual friend, hey, it sounds like a good idea, but why don't you just go to a few family offices and just tell them like, hey, this is what I'm doing. Or some funds just raise more capital from investors who are not going to be DMing you death threats. you know every other price what's a price yeah uh wait you know go uh go go public uh when you when when you're ready to hit the nasdaq brother yeah you know if you're really if you're really a true believer in crypto like crypto needs to be in my product do usdc do tether yeah there's no risk to the person like if i download an app and they're like hey we really believe in crypto like crypto is a key piece of this app like on cars and bids right now coinbase is auctioning a essentially a toyota hilux and you have to pay for that with usdc so you could think about this as like doug de muro's cars and bids company is now going crypto but they didn't launch a cars and bids token they're just accepting a stable coin and they still get a lot of the crypto and then they get the integration with coinbase and they're onboarding new people to crypto but there's no risk of like, oh yeah, I wound up instead of just buying the Toyota Hilux, I bought the Hilux coin and then it mooned and then it crashed and I didn't even get a car out of it.

40:57It's like, no, like you just paid with crypto. You got that frictionless experience. The wire transfer went through instantly. That's a genuine benefit when you're buying a car instead of having, I sent the wire. Did you get it? Oh yeah. I went through. Is it in custody? Who's got it? It's like they integrated this and it's great. And no one's going to get rugged on that. And I'm very, very confident that everyone on all sides of that is going to be happy. Whereas with launching like a meme coin, there's just a ton of parties that get wrecked. Yeah. Anyway, should we move on to the even even even funny enough?

41:30Trump only ever posted his website. Yeah, never once commented on the price. He never once positioned it as anything other than memes. He literally very Very clearly we're saying these are just memes. You can buy them. There's an open market if you want to buy the Trump meme. And the reason that that token can now hover, I don't know what the price is. I'm assuming around 20 billion. It has some amount of staying power because he's the president of the United States. And it's just kind of funny. But now again, jelly jelly has to be the other thing about crypto is you have a lot of these companies that have raised a hundred million dollars and yes, they're building some tech.

42:12but they maybe have zero real use cases. And all the energy of the company is going into sustaining the price, including buying back their token, right? To create buying pressure because they know that price is the product. And then they have their VCs who are doing the same thing and trading these sort of liquid tokens. And so these random projects that sit at$5 billion market cap are not just doing that by accident. It's because you have a hundred people oriented around how do we get the price to go up? And so now Jelly Jelly is, they don't have that. They're a small team. They're trying to build another product.

42:53And so tough spot to be in. And it's probably a good, honestly, hopefully it's a good learning lesson from our audience, which is that just because the president did something doesn't mean it's right for what you're doing today. Yeah. I mean, also, if you just think about like these tokens are store values, They're built on stories. Like the U.S. dollar is built on a story. Store of mimetic value and social media attention. The reason that when I go to buy a sandwich with a$20 bill, people believe that it's worth$20 is because there's a story that's told about the power of the American government.

43:26And that is in some way a meme in the same way that religion is a meme and corporations are a meme. Like these are not concrete ideas. Yeah. And Warren Buffett. Warren Buffett. He'd buy a stock. He'd hit the magazine route. He'd hit the radio route. He'd hit TV. He'd start talking it up. So yeah. And so when you think about Trump, like, like you're just buying this meme, there's no, there's no, there's not like, oh, the token should go down because he didn't build what he said he was going to build. No, he said he was going to build nothing. And then the question is just like, is Trump, is the idea of Trump still going to be a meme in 50 years?

44:00Like if he passes away, he'll still be in the history books. Like it's actually in some ways more durable than a company that says, hey, we're going to integrate this and then maybe they don't. It's like the floor of the promise, as you said, with the Trump coin was nothing. It was just the name and the meme. And so who knows where it goes. But clearly everyone is thinking about this. Brian Johnson wanted to get in on the action. He says, who's the best mind in tokenomics? We're designing rewards for the don't die ecosystem, earn for hitting health markers, buying DD approved products, dining at DD certified restaurants and more.

44:35And he got 4 ,000 likes. And then a few moments later, he said, whoa, whoa, whoa, I got so many DMs. Everyone's telling me not to do this. Yeah. Well, so here's one. Companies have been using tokens as in points, which if you think of them as the same thing forever, right? If you go to, you can go to amazon.com and earn points, American Airlines, airline malls. Yeah, buying specific products. The difference with that point system is, yeah, they have some intrinsic value, but they're not trading on the equivalent of a digital NASDAQ where the price can fluctuate up and down. And so Brian Johnson, what he wants is this sort of incentive mechanism.

45:20Yep. And he wants people to be bought into the movement. Yep. But the issue is if you create a financial instrument like a token, then he's going to attract for every one person that's like diligently following his protocols and saying, you know, look, like I'm eating at the right restaurants. I'm buying products from DontDie.com. I'm like hitting my health markers. For every one person that's doing that, like trying to earn the token, there's going to be, you know, there could be 10 ,000 people that are just trading on. It's like Brian Johnson started Venmo and Braintree. He's like an absolute dog.

45:56I got to buy the token. And then the weird thing is if the price goes up a lot and then you're issuing tokens to your community, well, then they now have tax implications from like, well, I got$10 ,000 worth of this token. And then now it dropped like however much I got to deal with this. And you're telling your accountant you bought like the don't die token. And now you've got to like, and so for Brian Johnson, it's like, okay, you could just create a reward system. It's like, hey, rewards for points that you can redeem for don't die products or points that get you into certain conferences that he hosts.

46:33It doesn't have to be this like actual financialized asset. And you even saw this, you remember, what was that game that was popular in the Philippines? Axie Infinity. Axie Infinity. So at one point, just by playing Axie Infinity, at the top of their hype cycle, you could be making like$100 a day just by playing this game that maybe was a fine game, but it wasn't. It was not a fine game. It was not a fine game. Well, I said fine. It wasn't. It wasn't even fine. Yeah, yeah. It was below mediocre in my opinion. I watched it on Twitch. And so this whole cottage industry spun up of people that were just playing it as a job yep and uh you know nobody on this show will will um yeah if you're a top 200 player in the world at any video game like yeah you should probably be able to have a make a living out of it but you shouldn't be making a living from a game that nobody actually likes to play that people are purely playing for this sort of and and and there was all these efforts to say that it wasn't just a ponzi scheme oh yeah and it turns out it it totally was reliant on like new money just flowing into the ecosystem based on that speculation to create enough liquidity to allow people to be able to pull money out of the system, right?

47:48So Balaji talked about this and how like crypto is basically zero sum because you can't get more money out of a token than what was put in, right? And so it's really a game of, and I posted this, crypto in its current form, most of it is just a wealth transfer from people that bought a token early and people that bought it late yep like later than them right so it's just like that constant mechanism um and uh yeah yeah which is which and and we're not like to be clear we're not we're not against crypto i i have like you know probably five or six crypto related companies i invested in um farcasters There's Seed Round, which is they got valued at a billion dollars late last year, doing some cool stuff about building like a censorship resistant kind of X type product.

48:47So there's a lot of cool stuff happening. We're not totally against it, but people don't realize what they're signing up for when they do something like launch a token. Yeah, it's more like round peg in a square hole. Like what Brian Johnson is describing, like the best tokenomics are... like the American Airlines flight reward, like airline miles, basically. And sure, maybe you want to use USDC for that. Maybe you want to back it in Bitcoin. But do you need a new token that trades freely? Probably not. And also, do you want any incentive for people to be able to spend your hard earned airline miles elsewhere?

49:24No, you want them to stay in the don't die ecosystem. So yeah, the big challenge program, the big challenge here is as soon as you're a public company, you now have shareholders that you didn't choose. So a lot of people that have been founders have dealt with having amazing investors and maybe some less amazing investors. And those are all people that you self-selected for the most part. You're like, I've met investors, I'm sure you have before, where it's like, it doesn't matter how much money you'd want to give me at any terms. I don't want it just because I know that that's going to be a really long-term relationship.

49:57And so even if you go public on the NASDAQ as a public market CEO, part of the pressure and part of why people avoid doing that is because they simply are going to be getting a LinkedIn message from a shareholder who's like, hey, your share price is down. Fuck you. Yeah. And then you yeah, it's not your obligation to respond to that, but you got to respond to it at the next earnings report. You know, cool it with the language, buddy. You got to put a crisp ten thousand dollars in the swear jar. uh but let's move on to uh yeah yeah yeah throw that in the swear jar um uh let's move on to nikita beer he says uh he posts a meme showing that everyone with over 50 000 followers is is uh tempted but to drop a coin i'm sitting at like 46 000 you're safe urge maybe it happens maybe the switch flips at 50k but so far i've not felt the urge no i i've had i've had so many uh people message me and i'm sure you too like when tv coins it's like if you want to be a part of what we're doing you can listen to the show we'll have events uh we'll invest in your startup maybe yep but if we do a coin it will be a physical coin that you buy for 50 100 challenge coin you can put it on your desk as a little token to remind you that capitalism is awesome but it will not be some free floating pump dot fun token i believe that they are low class and it'll be it'll be exchangeable one-to-one so if you spend you might be you spend a hundred dollars for the coin and it gets you in a group but if you ever want to leave you just come back sell the coin back as long as you're wearing a suit we'll take it exactly and there you go yeah exactly and uh fundamentally what would your mother say if you can't explain it to your mother oh you're doing in some pumped out fund coin, she'd be like, oh, can you just get a real job?

51:52It's a good, it's a good point. Why don't you pump up your LinkedIn profile and apply for some jobs? Exactly. Exactly. So let's move on to our third top story of the day. Some breaking news from Mark Gurman, the absolute goat of Apple reporting over at Bloomberg. He says, breaking, Apple cancels project to build AR glasses that would pair with its devices in a major retreat as it struggles to create a mainstream hit to follow the Vision Pro and rival meta. And so we're going to be breaking down this insider report from inside Apple. Apple scraps work on Mac connected augmented reality glasses.

52:30I'll read this and we'll see. We'll talk to Jordi in just a minute. Apple Inc. has canceled a project to build advanced augmented reality glasses that would pair with its devices. Now, the Vision Pro is kind of counted as an augmented reality device, but it's very much VR in the sense that it's using reprojection, taking images with the cameras from outside the headset and playing them back inside. It's amazing. I think it's an incredible device, very expensive, very heavy. There's a lot of drawbacks, but it did well. And they were working on a version two of that, maybe not moving forward. And so this is the latest setback in its effort to create a headset that appeals to typical consumers.

53:11The Apple Vision Pro certainly did not appeal broadly. They are trading at about a 50 % discount on eBay right now. What was that? Give them some credit. For about a week, everybody was calling it the future of personal computing. It shows, to me, it all shows a lack of true vision and leadership from Apple. They don't have a product visionary at the helm. Tim Cook is an amazing manager, even though we think he's underpaid at only 77 million, which as people know, Apple makes a billion dollars of net income every like four days or something like that. He's very underpaid. But I have to imagine that Steve Jobs would look down at Apple today and be embarrassed at being willing to say that we think this is a future computing platform and get pushback based on the battery life and things like that and just say like okay we're gonna we're gonna scrap this i guess it's not because somebody it seems like so um it's so programmed in that we're all going to be wearing you know um uh you know these headsets that that just like strap computers to our faces that to just say we're giving up because maybe it was a product that was just used for entertainment.

54:33You know, people, people seem to love it for entertainment, but the issue is there's a lot of substitutions, right? Your phone, iPad, TV, computer, et cetera. It was a fantastic movie theater. And, uh, who was it? Uh, the founder of mid journey said that, uh, the, one of the lead designers on the Apple Vision Pro worked at Dolby on their next generation digital cinema project. And when you think about it through that lens, it makes a ton of sense. When you open up the Apple Vision Pro, you can think about, there was like an implicit ranking of what the most important experiences were. And in the top left corner, the first icon, if you're reading it like a book, was watch movies yeah the app it wasn't a game it wasn't facetime it wasn't uh an app it was movies and that was where the best experience was you could go in and watch avatar in 3d and it was an amazing experience there's only so many movies you want to watch in a theater though and it yeah and here's an example apple if apple had come out and said this is an alternative to the tv This is a personal at-home entertainment product.

55:48We just want to make the most incredible way to watch entertainment at home, whether you're watching sports and you want to have different games up and the score and things like that, and your 50-leg parlay in one corner, right? If they had come out and just said, this is a product to watch movies in a really immersive way, I can easily imagine them. And maybe they priced it even lower and just said, hey, we're going to lose some money on this. We think this is an important, not lose, but we're not thinking of this as a profit center.

56:48Oh, we lost you. having their Apple headset and they, you could train an entire new generation of like, why would you watch TV on a screen that's, that's, you know, 20 feet away from you when you could watch it on your face. And if they had just started from, um, you know, the more of this, like very simplified focus and said, Hey, we're actually going to make films for this format with our Apple, you know, they did, they went super short films. It's like a five minute experience. It's just not enough. It just missed the mark in so many ways. Ben, can you pull up the actual screenshots from the article?

57:26And Jordi, do you mind reading through some of these? I got to go to the restroom. Yeah. Cool. So Apple Inc. has canceled a project to build advanced augmented reality glasses that would pair with its devices, marking the latest setback in its efforts to create a headset that appeals to typical consumers. The company shuttered the program this week, According to people with knowledge of the move, the now canceled product would have looked like normal glasses, but include built in displays and require a connection to a Mac, said the people who asked not to be identified because the work wasn't public.

57:53An Apple representative declined to comment. So I read this or I would say not me. But if Zuck is reading this, he has to be absolutely thrilled at this development because he's been investing billions and billions and billions of billions of dollars into their VR program. And he already has the Ray-Ban meta, the meta Ray-Bans, which are basically what Apple's doing. And so personally, I haven't lost conviction in the form factor. I'm not personally that excited to do it as a user because I like the fact that I can use my phone and then set it down and walk away. But even, you know, people on social have been commenting.

58:29I remember a post from Lindy man where he said the aesthetics of people, you know, we We need to get to headset-based computing because the visual of a bunch of people in a cafe looking at their phone like this is just terrible. And so I really think it's been programmed in that this is a future consumer computing platform, but it's going to take very long-term conviction. John I don't know if you heard me while I was gone but you have to imagine Zuck hearing this news and just being absolutely thrilled because he's yeah he's in founder mode he's willing to spend tens of billions of dollars over over years and years and years with no no sale you know relatively low sales like I imagine that I imagine that the vision pro and you know oculus sales are better but again because they have like some specific use cases right they're both compared to the rest of the businesses this to me is is the biggest bull signal for what only makes me more bullish on meta because meta's biggest issue to date has been not owning the underlying computing rails right that's been like constantly you know we've talked about um we've talked about apple uh trying to make meta's life harder and the advertisers on the platform life harder through the abstract app tracking stuff um so anyways this is this is zucks got to be absolutely thrilled with this news yeah and what's interesting is like with with ai apple really sat back they didn't train their own models siri apple intelligence they're both like just kind of widely regarded as disasters in terms of product development but their strategy there seems much more reasonable than not owning the next computing platform because apple really does hardware great and they've always been able to say yeah you know what, our spreadsheet software is not as good as Microsoft or Google.

1:00:27So we're just going to let Excel and Google Sheets be great apps on Mac or on the iPhone. And with DeepSeek, we're seeing that, yeah, there's going to be a lot of developers that build LLM products that integrate and then, oh, open source from Llama or DeepSeek, that's going to get baked into Apple intelligence. It's going to make it better. They're just going to kind of sit back and develop a little bit. Maybe they'll just integrate and do a big deal with open AI in the future like they did with Google. And you go back and you look at the Apple Google deal where Google's paying them billions of dollars to be the default search engine.

1:01:05And that deal has worked out. I don't think that that's bitten them at all. But when you look at them potentially missing the next computing platform, that is a serious risk. And this is a place where Apple should thrive. And so it seems like they are content to just operate as a toll booth for the rest of history. But they can't if they don't have the device. Yeah, I know. I know. And to give them some credit, they could probably say, hey, we just want to focus on our hero product, the iPhone. Keep selling a lot of iPhones. when if Zuck is able to figure out how to do a headset well, we'll just copy it because we have the best hardware engineers in the entire world.

1:01:52We'll let him spend all the money to try to figure this out. And then we have the Apple Store, right, which is the most amazing distribution engine for any new product. Totally. And so who knows what they'll do. Steve Jobs is looking down. uh he's probably um tracking a gd3 rs right now in heaven looking down quite disappointed and and uh i think tim's cooked we've talked about it before we said is did did tim cook or is tim cooked yeah and i i agree with you the the wrinkle here is that if you look at the iphone sales it's just completely flat which is obviously bad you want it to be growing but it doesn't seem to be going down at all.

1:02:35It seems like the toll booth analogy is correct. They will be able to just print money for decades as they maintain this monopoly on the app store and on the phones. And so they're happy. It's just going to be a monster of a business for a very long time. These things take a long time to unwind. And so maybe they'll miss it. The weird thing is that it really doesn't feel like there's going to be, oh, some open source headset that gets traction and all of a sudden everyone has it and it completely upends it's like it's either it's either apple or meta in this game and so if they can see the field it's rough now the the wrinkle here is like no and the dark the dark horse is someone like a neural link that that you could imagine you know maybe ends up doing the same thing which is the computer the computer is just integrated with your brain and if you want to do something you say hey i really want to uber from where I am exactly at this moment and I want to go to the RAMP offices, that is a new computing platform.

1:03:36And then Neuralink gets to be the toll booth and maybe it's not this interface, but maybe you just need a screen to see visuals, right? And that's actually easy. If Neuralink can do the brain-computer integration, it's much easier to just throw a screen up that displays information. So, yeah, there, there is a, there, Nreal or Xreal, Nreal air there. There are a couple of companies that make glasses that project a, a massive TV screen, but it doesn't do any of the other crazy features. So it doesn't do hand tracking. It's just, it's basically just, if you're on a plane, you want to watch a movie, you throw these on and they just put a big screen in front of you.

1:04:16It's locked to your head. Basically. I've been thinking about getting one because that was one of the best experiences of the Apple vision pro was being on a plane, just watching a huge movie and just having it look amazing um and so that that type of like lower end vr very specific use case we're in this weird fragmented moment where uh apple and meta both seem to want to create the one device to rule them all because that's what the iphone was but we forget that the the iphone the famous presentation was we're introducing three things an ipod a phone an internet communicator right and so he was saying it's going to be do web browsing it's going to have all your music and it's going to be a phone yeah and we combine those it's like well in vr Or maybe we need the one that's really good for just doing at your computer.

1:04:57And that's what this project was. It was going to be linked to your Mac. So you couldn't even run off your phone. It was not going to be mobile. It was going to be you sit down at your Mac. You put these on and all of a sudden you have a ton of different monitors. So instead of buying four 5K displays and they're all really expensive, you just buy one headset. Boom. It replaces the Pro HDR one that's like a couple thousand dollars. It's really expensive. and so instead of that you buy these maybe it's 10k but it's totally worth it and and then separately when you're on the plane and you just want to watch a movie you have a separate vr device for that and then the meta ray-bans for when you're out and you want to just take photos but you don't need the screen overlay but you want to talk to ai it's like we we could be in this world where we have fragmented vr products for a few years and then over time they get melded all into one but we're not right there on the hardware and i don't think any of the companies other than meta are really thinking about that and willing to say, Hey, the Apple vision pro is just a movie theater.

1:05:53Don't even try and go out in the world in it. It's just a movie theater. Okay. The, the, the meta Ray-Bans, they're just for taking pictures of your dogs and kids and that's what they're good for. Don't try and use them at work. Right. And, and, and, and accepting that is hard for these companies that are so big. They're like, we want to own everything. Um, but hopefully I used, I used the meta Ray-Bans for one hour. I put them on. I got on my Ferrari, drove for an hour and came back and the footage was pretty cool. Yeah. And then I put them down and I think I gave them away. I've used them a few times.

1:06:30I mean, it's fun if you're out with kids and stuff, you take some pictures, it's pretty easy, but they're not, they're not something that I'm reaching for constantly. And that's the case with every VR headset. You should put them on your dog. yeah yeah figure out a way to put them on the pop and get some get some like you know first person uh pop footage yeah and it was the same thing with the apple vision pros like i put it on a few times and then pretty soon it was like am i really gonna pull that thing out might as well just watch the movie on my phone who cares like it's such a hassle and so yeah they they they really need to streamline it i do what i do love about this was that they were like apple faced a lot of criticism when the apple vision pro had that massive battery pack off the side and was like oh this is so unapple i think and palmer said this as well that was one of the best decisions that they made because it gave everyone else permission to start removing pieces from the heavy point of the face which is where you don't want the weight yeah but then they added a bunch of weight back it was like the thing is the thing is display on the outside let's make it i want the weight i want the weight because i'm trying to build up my neck of course of course yeah but for but for those i want to be like i'm just repping the vision pro headset yeah it should be for linebackers only you need traps that have been eating your head you need death star delts to use this thing issue if you if you can't handle this thing don't even come in the apple store yeah you got to look like you're coming out of wwe to pick up one of these things well well said They clock you as soon as you walk in.

1:08:03No, not even an option for you. Get out of the store. No. Hit the gym. No, they measure the circumference of your neck to make sure that you're actually qualified for it. Hey, go do some hang cleans and then come back. Okay? Get out of here. Go do some shrugs. Do some deadlifts. Get those traps working. But yeah, so let's finish this out. So Apple has a specific group for this, the Vision Products Group, VPG. and employees who are leaking to Mark Gurman say that there's been a lack of focus and clear direction within the team, which is overseen by Mike Rockwell and company hardware chief, John Ternest.

1:08:43The N107 retreat is just the latest failed attempt to make Apple's headset technology successful, they say, and that's hurting morale. The company is still working on successors to the Vision Pro, which I am very excited about, including updated versions to the to the original model it also has other concepts in the works such as airpods with cameras that would kind of be a a competitor to the uh to the meta ray-bans and that's i love uh apple like they've just become so derivative they're like the iphone like people really love the iphone you know camera and people really love airpods what if we slap a camera on our airpods don't judge it till you try it no i mean revolutionary i i'm excited to be snapping some behind the scenes pictures of the podcast with you know put your put your little earpods like this that'd actually be a cool experience if you just had to tap your earlobe like this to get a picture yeah and so there are indeed team still working on uh custom micro led type screens and a lot of other uh technology a lot of this was just like the wave guiding and like the actual projection over clear glass that's really really hard this is what meta has been rumored to have solved with the orion headset it makes sense that apple wouldn't be far behind obviously just in terms of resolution they were able to leapfrog meta uh in terms of just you put on the headset and you're like wow this is i'm i'm not seeing pixels like it feels like in this environment that was the coolest thing i apparently what happened was apple uh there's this r &d cycle that happens with these screens where basically like scientists more or less will go and develop a new density of screen, like an even denser screen.

1:10:27But then once they've actually figured out the protocol for making these things, they need to go and scale it up so that it can be manufactured at scale cheaply. And so it's a process called like benchtop manufacturing to like production scale. And apparently what Apple did was they just said, hey, you know what? It's going to take two years for this benchtop to get to production ready let's just make a thousand benchtops and then just pay two thousand dollars for every screen instead of waiting for it to get to 200 and let's pull that technology forward like they literally pulled the future forward and so yeah the good news is that in two years that's going to be a commodity screen and they'll be in every vr headset and i'm really excited for the next meta headset because zuck knows where the bar is in terms of pixel density and he's not going to miss on that and so once you get that plus the plus the meta software that they've been working on plus just like no and it's going to quickly be great going to quickly get to the point where the incremental screen improvements won't even really be that noticeable to consumers right where just like what's happening yeah if you look at um in retina for like 10 10 iterations retina yeah and if you look at the same thing in the dslr camera market right you would be able you would you'd notice the slight differences you and ben would notice slight differences, but the average, even hobbyists paying attention are going to barely be able to point out those kinds of differences.

1:11:49So anyways. Big thing is I think Meta needs to really solve the movies and the TV. They need to figure out who are they going to partner with Amazon for... Well, that's the thing. They could easily partner with Netflix, Amazon, Prime, same thing, um hulu all these different providers you you know maybe not youtube tv but you can imagine it's it's not that difficult for meta to get content partnerships done if they're acting like a tv provider yep i think my point of view is apple's so lost from a product standpoint now start an f1 team you know at least like apple f1 you can have an idea of what that would look like uh it really get people to start maybe caring about the apple brand again right now they're doing all this like gen moji out of home campaign i want to see apple sign apple signs verstappen to a giga chad 10-year contract a billion dollars a year sorry tim cook uh verstappen's just a little bit more valuable than you um you know managing uh managing to get one more iphone sale a year um be great be great yeah i i really do think that if you want to uh get people over the installation churn curve with these headsets like you've got to get them into an experience within five minutes of them opening the box uh that's one of the biggest that's one of the coolest things about the chromatic uh is remember we unboxed that within 30 seconds and and with and this is a big knock I had on the quest was that I opened it up and then it asked me for all these passwords and I have a meta password, a Facebook password, an Instagram password, an Oculus password.

1:13:32I pin codes and all these security features. And then once you get in there, like you have to go to the store, download a game. It's has to be internet connected. You have to be logged in because they're obsessed with solving like the social networking aspect. You should be able to take this out of the box, put it on. Boom. Beat saver or boom. You're watching the matrix. like watching the right social network when you buy a meta quest it should just come pre-installed with the social network you should just be able to put it on in two seconds and start watching if you get someone into a great movie and they just watch it they will be like oh okay that was a good experience for two hours i got two hours of value out of that which is more than what most people get yeah when we install these things anyway let's move on to some ai ads there's a reckoning in the artificial intelligence world, written by, this is a Substack post by our buddy Michael.

1:14:20He's been on the show before. He's over at Lightspeed. He says, I wrote about how AI is breaking the business model that funded the internet and why the open web will never be the same. He calls it the end of advertising. And he has some kind of sloppy AI art here to kick it off. But let's read through it. He says, everybody hates ads. At least that's what we all say out loud, but our revealed preferences tell a different story. We actually love ads because they fund most of the content we consume on the internet. Without ads, we'd have to pay for content with our hard errand dollars. But with ads, we pay with our attention and we perceive the cost of the content as free.

1:14:55Like this show, fully supported by corporations. We never ask you for a dime. Ads have made the internet as we know it today possible. Ads are really, really good business because advertisers are willing to pay top dollar to access our attention. In the early days of the internet. This looked like simple banner ads brokered through manual deals on basic HTML sites, but over time ads evolved into some of the greatest money printing machines ever invented. Google, Meta, for example. Even businesses we don't often associate with ads like Amazon and Apple make a killing from them. And the little guys love them too.

1:15:27Many long tail publishers, websites, niche blogs, podcasters, and indie games are all funded by ads. Ads built the modern internet and funded our internet addiction, making billions of dollars and billions of us happy in the process. It's a perfectly tuned system that we all take for granted. But all good things must come to an end and ads are no exception. AI is increasingly challenging the business model of ads, the system that makes the open web feel free. For example, many of us are using Google search less frequently along with the ad supported websites that it leads us to. And in their place, we're answering, we're using answer engines like ChatGPT and Perplexity more each time we ask these AIs with delivering us a single definitive answer in favor of wading through many potential answers via traditional search.

1:16:13The relationship between an advertiser and our attention is severed and the ad never gets delivered. Challenging the model further, as AI models become more powerful and capable, they won't just deliver us answers. They'll also perform long, complex tasks on our behalf. They'll do all the busy work of researching, making appointments, and even buying stuff for us. This will free us up to do other things. And as a result, we'll devote less of our attention to the content and thus the ads. At scale, this represents a massive change to the foundation of the open web. What happens to the perfectly tuned system when the funding dries up?

1:16:43Jordi, you got anything? So I broadly agree with Michael. And I think it's something that anybody running an internet business, especially one that's based around attention and has any type of ads model needs to pay a lot of attention to. That said, the ways in which I use perplexity and OpenAI and other language models is not the same behavior that makes me a profitable Google Ads user. So when I want to... So we're talking about Mary Meeker in a little bit on the show, right? Yep. And so I used OpenAI's deep research product to do that, right? And it pulled together all these different sources and made it really digestible and it was super helpful.

1:17:29If I'd done that on Google, I would have made a bunch of queries and I would have found many of those same links. It might've taken me a long time. Uh, I would have gone and looked at maybe some individual articles. So that's a little bit of ad revenue for the person creating the content, but ultimately that type of search and the same search that's, that's like, where should I, you know, what are some cool places to vacation in Spain? Um, Um, those, and, and again, chat GPT can do that pretty well, right? It's like, Hey, here's six places that people love. That's very different than me going and saying, I want a new pair of tennis shoes.

1:18:06And I, I just search tennis shoes on Google. Right. And then I get a bunch of, you know, sort of options. And so I think it's important to look at the, the ad model of Google in many ways feels less immediately threatened right now versus the ad model of companies that just provide content. right so the companies that just like you know anybody that makes recipes and just post them online and expects to make a bunch of revenue from that i would be worried there's still people that probably make millions a year with banner ads selling you know doing recipes that to me is a concern because people are going to use you know models for that and it's going to surface all the information that they put out but google still has that toll booth sort of business model of when I want to actually buy something.

1:18:53And so that's what perplexity and some of these other more specialized models they're trying to get into is actually helping people discover products because product discovery and endorsement are kind of different things, right? I started my actual career in ads building Branded Native. And that Branded Native's business model is predicated on people watching creator content. So as long as people are watching, you know, creator content and advertisers want to get in front of that audience, that model's, you know, relatively, you know, I'd say it's relatively safe. But but again, anybody just posting, you know, basic content online and thinking that they're going to monetize it.

1:19:37It's not it's going to be a tougher sell. Yeah. If you're running like a travel website, that's just best top 10 places to travel to in Spain, you are already in the process of getting disintermediated. Maybe previously it was like you would rank for in Google organically for best places to travel to in Spain. And then you would have a bunch of affiliate links for things you might want to buy or maybe flights or all this different stuff. but pretty over the last 10 years, Google has been pulling more and more of that in where you can, Oh, you can buy the flight directly on Google. You can buy the stuff on Google shopping and Oh, we're going to, you know, pull in more of this stuff and info cards and take you to different websites.

1:20:18And so there's been this like war for a while. And I think that's just going to continue with the AI trend. Yeah. So Michael, and the, and the, the platforms that deliver entertainment content or education content or anything along those lines, right? If you're meta, yep. you don't care if as long as people are using your app, if you have attention, you can serve ads against that. I don't, as of right now, I don't see that being disrupted as long as people have purchasing power, you know, right now they maybe have income, but eventually it's UBI in some, you know, potentially dark future. But as long as you have attention, I believe you're going to be able to sell products against that attention, whether they're ads that you're indirectly, you know that you didn't create the product but you're selling it yep or um you know you're actually making the making the thing itself so no so this is an interesting section about premium content tightly guarded walled gardens especially built around premium content will feel the least pain from this shift so things like stratacary for example uh semi-analysis behind a paywall um and i mean going even more extreme like private speaker circuits or supper clubs i imagine are even harder to pull into the LLM training set, right?

1:21:35The real dinners, the idea dinners, this is where the alpha is going forward, according to Michael. Many will become even more valuable as the overall amount of the monetizable supply of ad-supported content shrinks. The demand for higher quality content will rise. Your ad dollars previously spent on commoditized content will be redirected to the best stuff on the internet. Plus, as we spend more time on busy work, we'll spend less of our time, as we spend less of our time on busy work, We'll spend more of our time entertaining ourselves, which means premium content will capture even more of our attention.

1:22:04And the best content will even find new ways to monetize. Reddit's AI deal with Google serves as the perfect example of how valuable niche content traditionally supported by advertising can be sold as training data for AI and media format. So I'm not disagreeing with Michael here, but it's not like Reddit sold their data to OpenAI and then it stopped running ads. as long as they have attention, they can insert. And so I think premium content, it doesn't even matter if the content is paywalled enough. Packy could release a free article. And as long as smart people that have purchasing power are looking at that, he can sell ads against that.

1:22:45And so I think that in many ways, it may be attention just becomes even more valuable, right? So, yeah. And this is something I firmly agree with. AI services will attempt to reinvent ads. I think we're definitely going to see this in the free version of ChatGPT. We're already seeing it in perplexity and Google has successfully stuffed ads in their LLM interactions already. Yeah, I saw, I got pitched a company a while back that I thought was smart. I didn't end up investing just because I felt that it was going to be too heavily commoditized or that the apps would actually want to own it, but somebody that was building like an ad network for LLMs, which to me makes sense.

1:23:35The question is, does OpenAI build an ads product or partner with someone like a meta that has quite a lot more data on the individual and just start serving ads into the free version of ChatGPT? There's a bunch of ways that that that could happen. It's not necessarily going to be one person that builds a network, but there's probably still a business to be built there. Totally. AI services will seek to capitalize on the end of traditional internet ads and invent new forms of advertising to capture the dollars that were previously delivered directly to publishers. Perplexity has recently introduced some early details of how ads will work on their platform and OpenAI is rumored to be exploring this as well.

1:24:17While this is an obvious next step for the future of advertising, it's unclear how directly the current ad spend will transfer over to this new model, and it'll need to be structured such that publishers can participate in a way that incentives keep them making content. Otherwise, they won't be able to afford to exist. Brand new models will also emerge to help publishers recoup lost ad revenues. As agentic traffic increasingly displaces human traffic, publishers will look to charge AI companies directly. Much of this will be through big, splashy content deals, such as OpenAI's deals with News Corp and Axel Springer.

1:24:48But for smaller players that can't afford to do deals themselves, they'll work with third-party services that will collect tolls on their behalf. Every time an AI hits their site to scrape content for their own service, the AI will be blocked and forced to pay before the content is served. And as AI moves beyond content ingestion to agent-based task execution, APIs that help agents do their jobs will also be charged. Makes a ton of sense. Let's move on to his conclusion, and then I'll get your take, Jordy. Content wants to be free. For decades, the economic relationship between consumers of content and the providers of content has been hidden behind business models that were monetizing our attention.

1:25:25We didn't put too much thought into it because we were getting what we wanted with minimal effort or cost. But AI increasingly disrupts these models. The relationship between us and publishers will become much more transactional and direct. And we will feel it over time. It'll be a new internet and the open web will be a memory. Great content will still find a way to reach us just like it always has. but we'll look back on the first few decades of the internet as the golden age of content when everything felt free. Interesting. I wonder, I mean, things will still feel free. It'll just be mediated by an LLM, but I understand what he's saying for sure.

1:26:00Yeah. I think his core idea of content wants to be free is absolutely correct, but that also means that it will just continue to be free and new ways. I think that a lot of the agentic products that get built will just argue that, hey, my user wants to access the information on your site. Yep. So if you make that more difficult or more expensive and I'm acting on their behalf, well, I'm just going to take, we're going to go elsewhere basically, right? In the same way. Yeah. Oh, you're on mute. There you go. You know, if right now, if I'm looking for specific news on something and I hit a paywall and it's just a headline that I want to quickly digest, I might just find a place that I'm already have an account or whatever.

1:26:53And I think that that will be, again, if I hit a paywall on Stratechery or PirateWires, I'm much more inclined to convert because there's some value there. But again, who knows? It's going to be a war of bots versus publishers for a long time, as it always has. Yeah. I wonder if the solution here is something like YouTube or X creator payouts where you can still publish to the open web, but there's just you opt into some program and it says, hey, if you bring a new fact or take or blog post to the open web and it is served in millions of queries to open AI. and they know that this fact and this data point are pulled from your original source or your original reporting, we're just going to send you a check.

1:27:49That would be really interesting and a way to kind of keep the monetization there. It's a lot of like flying blind, but it's kind of the same thing when you put up a blog and then you put Google AdSense on it and you just are like, yeah, I hope they serve some good ads and I hope I get some traffic. And so you could just be publishing. You wouldn't actually see traffic to your website. You would just see, oh, yep. OpenAI showed up and perplexity showed up and Google showed up and they took all my takes and all my information and everything that I brought to the internet. And then they went and served that to millions of people.

1:28:21And then they sent me a check because they made a million dollars of it and we split it 50-50. We'll go on the record in that we're actually put the re... One of the reasons we put out 15 hours a week of content is that we want to train the LLMs based on our way of life and our beliefs and our thoughts. And so OpenAI is free to use every episode that we've published to date. And we hope they really introduce the brother edition of the model, right? That's just primarily going to just, you know, maybe it shares a little misinformation now and then. But anyways, let's take a 30 second break and then get into Merry Meeker.

1:29:01I'll be right back. Okay, cool. Today, we have a deep dive. We'll be right back, folks. Stay with us.

1:29:38Thank you.

1:30:22All right, we're back. Did you see this comment from our buddy? He said, Technology Brothers has kind of changed my life. There was a quip about how if you're going to sit in an earnout, you should try and be CEO. and he's like i'm thinking about doing that and i and i stand by that if you get bought don't just sit there and rest invest try and take over the parent company why not or uh or base baron recently posted he said uh you you told me you're a building in public but i can't find you on the s &p 500 yeah it's underrated i mean there's all these posts about like oh like look the the mcdonald's ceo worked his way up for the mcdonald's ceo or the walmart ceo was like low on the totem pole it's like come in if you're a founder you got that founder mentality that founder blood get get in through an acquisition and then take them out straight to the top straight to the top yeah don't don't just rest back and be like oh i gotta start another company i gotta do the same thing again or i gotta be a vc why not go be the ceo of a public company that's a great life you could do a lot of good a lot of resources huge balance sheet yeah it's a lot of that amountable because you're going to come into a very sclerotic very old school organization but you work hard and you make the right moves, you can easily be running the place.

1:31:38Yep. I love it. Well, let's move on to the absolute dog, Mary Meeker, one of the greatest to ever do it, dot-com queen. Wait, right before we dive in, we had a couple questions. Did you try the OpenAI deep research feature? Yes, we did. John tried to use it last night. It was not available, even though we're both pro users. Tried it this morning. Got it this morning. It's great. I had this funny interaction though. I got this amazing report and I said, hey, this is great. Can you make it into a PDF? And it did it, but then it was only like, it was a different piece of content and it was like less than one page.

1:32:23And I said, hey, that didn't quite work. Could you get me just a PDF, gist of your primary deep research response and it produces another document that's not content and then i eventually just copy and paste it into a word doc so it's smart i had the exact same experience exact experience i went to o1 pro finally got the deep research button there it's it's not a new model by the way it's a button yeah on some of the models but not all of them some of the models allow file uploads some of them don't so i go to o1 pro mode that has deep research. I'm asking it to do this, this research report. It keeps asking me, Hey, just to clarify, what kind of research do you want me to do?

1:33:07And I'm like, okay, try this one. And then it just didn't give me anything a couple of times, just total bug. And then, and then the second time I was like trying to give it a bunch of extra context. So I was dumping in like a full text of this New Yorker article that we're going to read through and the Wikipedia just to be like, here's some extra context that's behind a paywall that you might not have access to. uh it was just not handling that at all it was like freaking out i was trying to upload pdfs it was a it was a mess it finally gave me a summary that was like okay it was not that great i clearly need to learn some skills on like how to prompt it and how to get like great results out of it i'm very optimistic but uh clearly some rough around the edges it took some time to get done but i did have a funny experience where uh today we are going over this new yorker article uh it's from the dot-com boom.

1:33:54I think it was published in 1999. It's about Mary Meeker, who, if you don't know, was at Morgan Stanley and was one of the top equity research analysts at the time. And she's still extremely well-respected in tech. She wrote the thesis that described how internet companies should be valued, what would happen to the internet, how value would accrue. And she created a mental framework for how to think about valuation of tech companies. And so she was incredibly important. And this New Yorker article is behind, not only behind their paywall, but it's in their archive. So it doesn't exist as HTML.

1:34:27You click and you get a PDF of like the entire magazine, and then you need to go into their tool and print out just what you want. So I printed that out. It's basically just photos of the actual magazine. It's not text. It's not like a text PDF. And so I go in and I scrape out all the text using Adobe, and I get this very, very messy bag of text that where every other word has a space in between it it's just a complete mess and i'm like this is perfect for llms they should be able to clean this up it's like illegible so i start stuffing this into different llms i go to gemini because i'm like gemini can handle big context windows this is what you were built for google your time to shine i'm a paid uh gemini advanced user point three double ops i'm on i'm on the best of the best they're they're milking me for 200 bucks a month.

1:35:17I'm like, this is what you were built for Gemini. Now's your time to shine. Go in there. I'm sorry. Like as an LLM, I can't help you with this. And I was like, I was literally like mad. I was like, you are a language, large language model, transformer model. I'm asking you to transform language. What are you talking about? You cannot transform this language. It makes no sense. Clearly it triggered some flag deep in the model that said, Hey, hey, this is copyrighted. We can't touch this. Oh, interesting. That's clearly what happened, but it wouldn't tell me. And so I would like yell at it and be like, what are you talking about?

1:35:52Like, like you can do this. You say open AI would do this. And exactly. So you try to negate it. I also trained on all of our data. I tried to, I tried to jailbreak it. Wasn't working. I go over to GPT 4, 4.0 too slow, not doing it. And it's also like, it winds up hallucinating where it's basically formatting it correctly for like the first few paragraphs and then it just starts summarizing it and then it just doesn't finish it. And so by the end, it's just like, and marry me. And it's just like reading the Wikipedia. And I'm like, that's not what I wanted. I wanted a direct rewrite of this article that's just messy.

1:36:26To be clear for the historical record, when we do become full paperclips, we're not mad at these bots, these LLMs. We're not mad at We know they're trying to do their best. And so we just ask that 10 years from now, you treat us with that same level of respect and patience. Golden paperclips, not aluminum, please. I want to be made into a solid gold paperclip. Should we, before we get into Mary, we can... I have more on this. I have more on this. Oh, he's not done. I got to keep going. So then I go over to DeepSeek because I'm like, whatever, DeepSeek, definitely disrespecting intellectual property.

1:37:03Let me see if they'll do it. it starts doing it and I'm like, okay, this is cracked. It's working perfectly. And then the whole thing times out because they're completely underwater and they don't have enough servers. And so their whole thing just breaks, can't inference enough of it. Cause it's like thousands of words that I'm trying to get it to put out. So it's just failing. That thing's just completely broken. You can't even log in anymore. Uh, and then finally I go over to Oh one promo nails it one shot worked perfectly. So, so, uh, hats off to chat GPT. They did the, they got it done. And then I was able to kind of bake that down.

1:37:34But anyway, that's all to say, it's a fantastic article. You should go subscribe to the New Yorker and dig through their archive. They have a lot of great articles and we'll read a lot of them here on the show. So let's kick it off with the woman in the bubble, how Mary Meeker helps internet entrepreneurs become very, very rich. And I love this article because it has so many cameos from people who are now legendary in the valley, but it's a complete time capsule showing you like she took Netscape public. And she was the first person at Morgan Stanley to identify Netscape Mosaic.

1:38:09She was basically acting as an angel investor for Morgan Stanley, especially when you look at the potential returns that they generated. So absolutely, what an icon. And they did her right by taking this iconic picture. Awesome. This is basically, when they published this, it was 1999. 1999. So we're kind of at the hype, but she was studying and analyzing technology companies in the 80s. So she actually got into it very early. Should we start maybe a little bit earlier than that and just give some backstory on - Yeah, you have the research report, so why don't you break down the timeline? Mary Baker was born in September 1959 in Portland, Indiana.

1:38:56Very clearly, if she had been born in Portland, she would have not gone on to high finance. A small town between Indianapolis and Fort Wayne. She was raised by a homemaker mother and a father who was a steel industry executive. She had an older brother who's 21 years older than her. And this is a little foreshadowing. Her dad, her father, Gordon, was an avid golf enthusiast. and she encouraged mary to excel at both golf and investing from an early age so just that he just was like he's like i want you to just dominate finance you're gonna like you're gonna learn to golf you're gonna learn to invest from uh from a young age um and uh she she's you know taylor swift's dad wasn't taylor swift's dad like a finance guy who was like you got to learn how to an absolute killer in country music.

1:39:46Yeah. Yeah, I'm sure. And also has probably helped her financialize her back catalog and remake it at different points. But anyway, so under Gordon's influence, Mary developed, as you might imagine, a competitive type A personality and a drive to achieve. She became the captain of her golf team in high school. And even while in high school, She took a class on investing and made her first stock pick for a Minneapolis oil refiner, which doubled in value in the first three months, foreshadowing her future. So I'm kind of bummed that my class, like I'm assuming that classes today will just have like meme coin trading.

1:40:27And they're like, here's how to like identify like, you know, jelly, jelly. Yeah, yeah. Jelly, jelly. You want to make sure that the supply isn't too concentrated in the hands of degenerates. you know um anyway so so you know very very from a very young age she was just being trained to become a titan of finance and uh you know she's not afraid to make returns in you know in the oil refining space right so broad set of interests and i i don't know if we'll get into it in this article but it'd be amazing if like she just kept at it with the golf and would just be out there you know at this time you have you have to imagine that in many of the rooms that she was in she would have been the only the only woman oh yeah totally um and there's a whole bunch of anecdotes in this uh about like how hard no she is how she'll be like screaming at some subordinate and then the journalist will pop up and she'll just be like oh hi like yeah i'd love to talk to you about whatever uh and so there's a great anecdote that they opened with about uh priceline.com which most people don't even really think about anymore but was one of the major hot stocks during the dot com boom all around booking tickets and so uh priceline.com a connecticut-based firm that sells cheap airline tickets to and hotel rooms to online customers had issued 10 million shares seven percent of its total shares through morgan stanley where uh mary meeker worked after consultation between meeker and her colleagues in morgan's capital markets department the offering price was set at 16 if the stock found buyers priceline.com would become the world's most highly valued internet IPO yet with a total value of two and a half billion dollars.

1:42:05Small potato and so that's if they achieve the 16th, that's, that's if they had buyers at the$16 price point and you can get into what happens next. And so as it turned out, the offering was far more than merely successful within seconds of its debut. Priceline.com's price rose to$85 a share, five times the IPO valuation. Bill Gurley, probably pretty young at that time, probably punching the air. He's upset. The man hates an improperly priced IPO. The price subsequently fell slightly, but by the end of the evening, when Wall Street closed early for the Easter weekend, it was at$80.50 a share. In other words, Meeker had just helped a company that had been operating for less than one year.

1:42:49This is how crazy things were in the dot-com booth. Less than a year, and they'd lost$114 million in 1998 on revenues of just$35 million. You're crazy right now. Achieve a stock valuation of more than$11 billion as much as American Airlines. American, of course, owns valuable route franchises, reservation systems, to say nothing of airplanes. Crazy. Yeah. So one interesting anecdote is Brad Gerstner, speaking of early, he was one of the first investors into Priceline. That was his first bet at Altimeter. Wow, that's amazing. And so all these icons of the industry today were there YOLOing into internet stocks in 1998.

1:43:37Yeah. Absolutely. Beautiful, beautiful period of time. And so Mary Meeker says in this article, the internet IPO frenzy is unbelievable. There are so many new companies coming out of the woodwork and the level of demand from investors is so high. It's mind boggling. And the fascinating thing about this article that we'll keep going into is that she correctly predicts that the Internet would be a major trend and that there would be trillions of dollars of value created. But simultaneously, she also calls correctly that it's a bubble. But she still falls on the sword and becomes this scapegoat because everyone sees her as this major stock promoter.

1:44:13And actually this settlement at one point where Morgan Stanley and her have to pay fees for pumping stocks too much, basically. Yeah. And it's crazy back and forth. But here's another thing that's crazy that the Priceline shares currently trade at$4 ,700 a share. Wow. She brought Priceline public at$16. And if you had just held it, and I'm sure there were stock splits along the way and stuff like that. Or mergers too. Yeah, mergers, stock splits, whatever, some dilution. All that being said, if you just bought Priceline at the IPO 20, however many years ago, 27 years ago, you did phenomenally well.

1:44:57And I think Altimeter held it for a really long time. And so they got a lot of those gains. And so she was truly a kingmaker during the dot-com boom. Every week, dozens of entrepreneurs try to get in touch with Meeker. Each of them dreams to become the next Jay Walker, the founder of Priceline.com, who is now worth more than$4 billion, at least on paper, securing the backing of the best known analysts in the industry is a big step toward billionaire dumb. For this reason and others, the recent acceleration of the internet boom has placed Meeker in a predicament. On the one hand, it has enhanced her reputation as a financial seer.

1:45:31Her list of recommended stocks is up by about 150 % already this year and has made her firm a lot of money. On the other hand, it has challenged her view of herself as an objective analyst and filled her with trepidation. And so as a capital markets person, she's supposed to just call balls and strikes and kind of create valuation frameworks. But she's so hyped and seen as like if she's backing a company, it's legit that she needs meme coins. You know, it's Sam Lesson all over again. Yeah. And the difference is she's not. As an investment bank, there's there's this sort of potential conflict, right, where you're trying to win clients because you want to take those clients public and you benefit at that event.

1:46:18You benefit purely from it doesn't really matter. You know, yes, you know, you'd have incentives long term around the price action. And yes, you're you're selling those shares to their investor network. but in many ways they're making money no matter what. And so that's, that's the slight difference where Brad Gerstner can buy Priceline. You're on mute. There you go. Again, Brad Gerstner sure is going to be fired after this episode. No, but, but Brad Gerstner again can, can buy the stock, write it up, write it down. And, and he can even speak positively about it, but he didn't birth it into existence.

1:47:01Totally. And so it's a slightly different position to be in. Yeah. And so she says, as the pace accelerates, the values get higher and higher and it gets more dangerous. There's some, so people were comparing dot com stocks to the tulip bubble. This is what happens whenever there's an asset bubble. People say, oh, it's like the tulip bubble, which is never that great of an analogy. Tulips are a better analogy for NFTs. Yeah, yeah, for sure. But even then, the tulip bubble was so small and so short. And also, I think it was pretty fake. I'm pretty sure the true top ticking of the tulip bubble was almost entirely fraudulent or something like that.

1:47:40There's some weird historical record where we don't remember the tulip mania correctly. But she says, there is the same supply and demand imbalance. The difference is that tulip bulbs didn't fundamentally change the way companies do business. The internet does. But when all is said and done, there will be many stocks that in hindsight look like tulip bulb stories. And she was absolutely correct about that. So her job was really to really just pump the good companies. yeah but it's not but the the challenge is there was so much interest in internet stocks at that time even the best companies like priceline which went on generational runs afterwards and ended up being important internet companies got so ahead of their skis yep that it just doesn't yeah like the fundamental like yes there's a fundamental there's an underlying business that's solid and may be important, but it's still going to get inflated to such an extreme price.

1:48:39I mean, SoftBank had a 99 % drawdown. I'm pretty sure Amazon went down by like 80%. All the stocks got wrecked, but it really does speak to this idea of when there's a bubble, trying to be in the power law winners that are durable companies is probably the best strategy, as opposed to saying, hey, if you're the guy who during the dot-com bubble said, oh, I missed out on Amazon, I got to get into, you know, uh, like books.com or something that just IPO and is completely fit. It's like that person probably got wiped to zero. Whereas Amazon holder, if they held, they probably did pretty well. And so even, even with the massive drawdown, even if they bought at the top.

1:49:18And so, uh, important, important lesson there. And so she actually predicts just in the first, in the first two pages of this, I think stocks sometime this year, uh, I think there will be a big correction in internet stocks sometime this year. I think a big correction would be very healthy. I personally would welcome it. So she's like, get me out of jail. Like I'm about to be crucified for this. Like you got to correct guys. You're too frothy. These companies are good and I'm calling the correct companies, but the valuation is too high. We got to do something about this. And fortunately it didn't really happen.

1:49:49Yeah. And I mean, like right around this time. So she, one of the first, one of her first big picks is Mosaic. And so it actually came, she read a New York Times article about Mosaic. So a little bit different than today where I don't think Mark Andreessen is, you know. He's back in the New York Times. He took the interview two weeks ago. He's back in the New York Times. But anyways, there was like a 20-ish year gap. Meeker read a New York Times article about a startup called Mosaic Communications, which is a really cool name for a web browser founded by Jim Clark and Mark Andreessen. And it's sensing the significance of their work, she urged Morgan Stanley's tech banking lead, Frank Quattrone, which is a great name.

1:50:36He runs Catalyst now, the investment bank. They do most of the really large scale M &As that happen in SaaS. Yeah. So Frank Quattrone, iconic name. This ended up leading Morgan Stanley to become the lead underwriter for Netscape's IPO in August 1995, a watershed event often credited with launching the internet age in the public markets. Netscape's blockbuster IPO not only put the internet on Wall Street's map, but also cemented Meeker's reputation as a visionary analyst willing to champion emerging tech companies. And so this is like years before she did the Priceline deal and everything went crazy.

1:51:15She had part of the reason that she did so well during the bubble is because she had been in the trenches when people didn't really care, people weren't even thinking about it as much. And so she was able to identify that, hey, user-friendly consumer products for the internet, like the Mosaic browser, which was the most user, the Netscape browser, which was the most user-friendly product at the time, are going to have a lot of value. And so she should get credit for not just benefiting and being the face of the bubble, but helping create it. And we love bubbles here. And it's wild because you've talked about this before on the show, being in Silicon Valley in the early 2010s, you'd meet somebody that was employee number 102 at Yahoo, and they were a multi-hundred billionaire.

1:52:04And that wealth creation, even though the 2020 COVID-induced ZERP cycle, it wasn't the same kind of wealth creation event that, yeah, there was a lot of IPOs, There was a lot of liquidity events, but it certainly was not to that same degree where you didn't hear about employee number 150 making$100 million. It just didn't really happen to that same extent. And a lot of the companies that were going public, yes, there was some highly speculative hard tech companies with no real revenue, but a lot of the companies would have like 150 million of ARR. And so there was some business there and they started getting valued on these really crazy looking forward multiples.

1:52:47But it was a totally for for me, you know, being born in the mid 90s, I guess the yeah, the mid 90s. I. It's it's it's not really correct to be to think, oh, I've lived through something like this, because what happened in the late 90s, early 2000s, you know, year 2000 was was on another level. Yeah, just to kind of give you some anchoring on how crazy the dot-com boom was. During the last four years, so from 95 to 99, about 100 internet companies issued stock on Wall Street. And together, they are now worth more than$250 billion. For reference, it took John D. Rockefeller more than 40 years to create Standard Oil and Bill Gates more than 20 to build Microsoft.

1:53:36The pace of paper wealth creation is increasing. Just since the beginning of March, more than a dozen internet companies have gone public and about 30 others have filed for permission to sell stock in the near future. And there's some great quotes in here. One's from David Byrne, general partner at Benchmark Capital. It says, it's a very small world, referring to Mary Meeker. The higher you go up, the smaller it gets. Everybody is connected and she is at the center. Back in 95, Meeker and a research associate wrote a 300-page research paper called The Internet Report, which hailed the nascent technology as a revolutionary medium.

1:54:12Isn't that great that you could just come out with a name like The Internet Report and it just becomes iconic immediately? Yeah. And so at the time, fewer than 10 million people were online. And many people, Bill Gates among them, were skeptical about the online world's commercial possibilities. They were saying, yeah, it's like the fax machine. It's not going to really unlock entirely new companies. Trillion dollar companies are not going to come out of this. And there was a lot of skepticism. Meeker brushed aside skepticism and predicted correctly that the number of internet users would grow 100 to 150 million by the turn of the century.

1:54:47She followed up that copy, that coup, by writing the internet advertising report and then the internet retailing report appeared in 1997. And just to give people an idea of how impactful this was, they took just the internet report and it was turned into books and then sold in offline. Yes. So you could go in the airport and buy the internet report. Yep. There was a couple of things that she called out very correctly. And she saw the usage of the internet, right? 10 million users. And she called, and this is in 1995, she called that email would be the first killer application. which was absolutely correct.

1:55:28It's still the foundation of the internet, right? You can't really use any software in a meaningful way without having that communication layer first. And then the next sort of wave that she predicted was just browsing information online, right? So browsing had to come before purchasing. And so a lot of companies would spin up a website that you couldn't, it wasn't an e-commerce website. You could just go there and sort of browse, it ended up becoming a digital sort of catalog. But it's funny to think at this time, the concept of browsing just didn't exist, right? And so a lot of people made money just by going to legacy companies and saying, I'll make you a website.

1:56:09And then they would charge like inconceivable amounts of money to create what was effectively a static, you know, HTML. Yeah, I love these quotes. There's so many classic ones in here. So first off, yes, 150 ,000 copies of the internet report were printed and sent around by Morgan Stanley. If you have one or you know someone that has a physical copy, please let us know. I want to acquire that. That's a rare text. We got to get our paws on. I love this quote from Roger McNamee, partner at Integral Capital Partners, Silicon Valley investment firm. Mary has provided an intellectual framework for understanding the internet.

1:56:51And I think that's exactly what Ben Thompson did in the social media era, talking about aggregation theory. Mary Meeker, very different business model, but very similar in terms of providing a framework. Ben Thompson has a great business model. He will probably make$100 million telling his thoughts. Mary Meeker, better business model, actually capturing some of the upside of the things that she's writing about and a ridiculous amount of fees in the process. Yeah. Well, she wasn't really when she was on the capital, on the equity research desk, because I don't believe she was comped on the actual investment banking deals.

1:57:30That's the investment banking team. They should be separate. But eventually, she switched over and raised a massive growth fund and has been printing ever since, I'm sure. Yeah, but you have to imagine she was the best paid analyst of that period and was probably paid on par with, I would at least hope, on par with somebody on the other side who was relying on her research to actually make deals happen. They said it was a few million dollars a year in this article. oh it was the tim cook of her era yeah then she became a capital allocator now she has multi-billion dollar fund bond capital she's out there making money yeah so just just to give you an idea so so she's picking companies you know like a vc in the sense that she's picking companies that she can get in like hypothetically she could get into right so anyone else could have been doing this research and the companies that she was picking uh just just a sampling dell microsoft intuit netscape aol amazon yahoo ebay and google were all stocks companies that she like put her stamp on and said these are winners and so even though she ended up taking the fall over time for for being the face of the bubble a lot of the companies that she was picking like if she was a VC and she got 10 crazy winners and like 40 other, you know, you know, sort of misses, she would have been, you know, she would have been the greatest VC of that period.

1:58:55She just happened to be doing public markets, you know, more investment research. Yeah. There's a funny anecdote in here. She's talking to the journalist and her email system goes down and it really feels like the same frustration that we were having with AI and ChatGPT. We're like, this is super buggy, but she's not off put by it. She says, I had to call someone to come into the office and fax me my emails. And so there's this thing where you can simultaneously be frustrated with the rough edges of nascent technology while still recognizing their impact. There's some interesting facts in here, a little bit of Coogan's law going on.

1:59:32There's this idea that Meeker recognized early that using online communication was much cheaper and more efficient than relying on wood pulp and mailmen. And that once the internet grew and more and more people join the network, its usefulness would increase exponentially. The latter point has been formalized in Metcalfe's law, which was set down some years ago by Robert Metcalfe, the founder of 3Com. It states that the value of any network increases in proportion to the square of its number of people using it. So a network with 500 people attached to it is 100 times more useful than one with only 50 people attached.

2:00:05And so this is an interesting thing because Metcalfe's law was defined, but it was living out there in the computer science world. And she was able to take that idea and popularize it amongst the investing community. And that has immense value. And sometimes you don't even need to be the person who coins a phrase to extract the most value from it. You can extract a ton of value just by popularizing a phrase in an important thing. We recommend that all of our audience try to coin some stuff. She should have repackaged it. She should have been like actually Meeker's Law and it's slightly different.

2:00:37Half of our conversations, I'm like, all right, what do you think about this? The Hayes Paradox. and you're like, hmm, it's got legs. Yeah, great. So she picked AOL in 1993, first online service to be aimed at non-geeks, those discs that they send out to everyone and you can install it and use the internet. At the time, America Online was losing money. Its accounting was dubious and many people on Wall Street were doubtful about its future. America Online shares, adjusting for subsequent stock splits, were selling for about 95 cents each. Last week, they were changing hands for about$160 each.

2:01:14She followed up her inspired America Online call with numerous others. She has recommended 10 baggers or stocks that subsequently increased more than tenfold. Amazon.com, America Online, CNET, Compaq, Dell, eBay, Intuit, Netscape, and Microsoft. Now, some of those got bought and stuff, but it's like just banger after banger. Absolute dog. Absolute dog. In the fall of 1994, Meeker came across a story in the Times by John Markoff about a new venture called Mosaic Communications, which we talked about, which had developed an easy-to-use device for navigating the World Wide Web. I read the article and the light bulb went on, Meeker told me.

2:01:56She visited Mosaic's offices in Mountain View, California. And in April of 1995, Morgan rounded up a group of old-line media firms. They included Times, Mirror, Knight Ritter, and Hearst to invest in the new venture. It's so funny. Marc Andreessen did a party round with legacy media. Yeah. What a narrative violation. The exact opposite of what he would recommend to a new founder. Yeah, yeah. Now he's convincing founders, I need the whole hundred million. I need the whole round, yeah. And also, it's probably generally good. No, no, it was the right move at the time. It's very funny. And so on August 9th, 1985, at the start of year one in the online era.

2:02:36With the Netscape IPO, we really helped create a new business model, she told me. We helped create a new way of financing companies. Even the most unrestricted stock market bear wouldn't contest the statement. One of the great strengths of contemporary American capitalism, which relies on decentralized decision-making by hundreds of millions of individuals, is the ability to channel large amounts of money into promising new industries. And the growth of the internet industry is an excellent example of this process at work. Fantastic. So great. There's another anecdote about Priceline.com. I mean, this whole article is just so, so good.

2:03:12A couple weeks after our first conversation, I returned to Meeker's office and asked her what was going to happen next in the internet story. As she described her vision of the online world's future, she relied heavily on a phrase I hadn't heard before, digital Darwinism. The phrase is based on the economics of increasing returns and on what are called network effects. It ties in with the conceit common among internet cognoscenti like Meeker, that we are seeing an emergence of an entirely new commercial ecosystem. The economics of digital Darwinism are pretty simple. During the last few years, economists have come to recognize that high technology markets tend to be dominated by one or two firms which enjoy high profit margins and that companies establishing such an early lead are difficult to displace.

2:03:56And that could not be more true. Like, there was chaos. That went on to influence 22 decades, 20 years of the dominant sort of thinking in venture capital, which was if you're not a power lock company, if you're not, you know, in that sense, probably influenced Teal in many ways. But if you don't have network effects, there's probably not a venture-backed business to be built here. And in many cases, that was correct. Yeah. The internet is a kind of small town. Everyone will go to www.something and they won't go anywhere else, at least not often. And so these habits are really important. She called them super companies, America Online, Amazon, eBay, Yahoo.

2:04:44who very, very interested in aggregation, market concentration, power law outcomes, defining this very, very early. I think the problem was that the second and third order investors, like the retail investors and the people downstream, weren't actually reading her reports. They weren't actually taking what she said to heart. And so they wound up just YOLOing into random stocks hoping for a pump. And that happened, but then a lot of them got washed out and that was unnecessary. Well, yeah. And there's a ton of data today available in pretty much any company, any public company that you can think of.

2:05:23And most people just see a post or their buddy tells them and they just YOLO into it. They don't really read. It's not like... And so, yeah, I'm sure if you read her entire report on something, you would find a huge section saying, here are the 20 potential risks to this business. But nobody's reading that far, right? They're just like, if I don't invest now, the stock's going to be double by the time I get to call my broker. You know, I read this and call my broker. Totally. This is a fantastic quote about Amazon. She's talking to Steve Case, who was AOL's chairman at the time. The question I always ask Steve Case, AOL's chairman and chief executive was, when do you reach critical mass?

2:06:03It was 1 million customers. It was 5 million at 8 million. He said, Mary, shut up. Lots, many. AOL finally became profitable at 10 million customers. Amazon.com, on the other hand, still hasn't made a dime, but Meeker is relatively unconcerned. My view of Amazon is that it's not just books, it's bits, she said. If two years from now it has 15 or 20 million customers and it has their credit card numbers and they're happy, then it can make money. Books will be seen as the Trojan horse that got them all the customers, just like all the books given away by AOL. And this was Zucks, not Zucks, Bezos' genius, right?

2:06:42Totally. Trojan horse. He very clearly believed that books were the Trojan horse, and he picked books as a category. I'm sure many people know this, because if you walk into a bookstore, they're going to have a lot of books, but they're not going to have every book on Earth. and in this situation you could have a single store quote-unquote store that had pretty much every book you could possibly want and so it actually was a better the digital store was a better product than going to the actual store and that was very different than going you know buying cars online at the time where if you walk into a porsche dealer and you want to buy a new car they're pretty much going to have like what you want, right?

2:07:25Yep. By the way, John, got some breaking news for the audience. We are now officially at TBPN on X. Fantastic. It went live. Anybody, if you navigate to at Tech Bros Pod, it'll now take you back to TBPN. So more to come there. Welcome to the future. Let's continue on. She's even calling how the internet will affect companies outside of internet native businesses. Outside the internet sector, the biggest losers in the Darwinian struggle, Meeker believes, will be retailers, travel agents, and other businesses that compete directly with online firms. In the last quarter of 1998, she pointed out to me, both Barnes & Noble and Borders reported slower rates of sales growth in their stores.

2:08:14Mattel, which mostly makes toys and games out of wood, metal, and plastic instead of computer chips, had a disappointing Christmas. Yet Meeker isn't a zealot. Things rarely happen as quickly as one thinks. So there's rarely displacement as quickly as one thinks, she told me. Television didn't kill radio. The internet is not going to kill television, radio, or publishing. Nonetheless, she is critical of the big media companies for missing a historic opportunity. Disney should have been Yahoo. AT &T should have been AOL. Time Warner should have been Excite. She said, why didn't it happen? Because it was a series of judgment error.

2:08:48And so really, yeah, this is the challenge now, if you're building startups, is that we've talked about this earlier on the show. If you want to be in social networking, you're competing against Zuck, who's written multiple different hype cycles and trends. He identifies trends early, ships at them quickly. And it's just a very different game. I think a lot of the old garden venture was trained on the idea that big companies don't adapt quickly. But you even saw this Nike acquired an NFT company because they were like, hey, maybe people are going to shift online and be buying digital goods. So they bought an NFT company immediately because they were like, this is a risk to our business and maybe a potential growth opportunity.

2:09:33And so I do think this has changed pretty dramatically in the last five years where you have people that were trained on technology and real fast moving sort of digital innovation that now we're just like, okay, this is a trend. We're jumping on it. Right. There's a great anecdote. I love these articles because it's just a murderer's row of today's hitters before they were big. With the Priceline IPO out of the way, Meeker and her colleagues are busy looking for their next big money spinner. That's a funny phrase for, you know, you're on mute, but who are they going to spin into the next multi-billion dollar company?

2:10:08Right. And so, yeah, Kingmakers. And so she says, one night after work, Meeker and Ruth Peratt, a senior executive at Morgan Stanley's corporate finance department. Ruth Peratt is the CFO of Google now, like big, big hitter. They're out at a dinner running through some IPO possibilities that were on the horizon. After the waiter took their orders, Peratt opened a thick black folder that contained information on approximately 200 privately owned internet companies. You know, VC-backed companies that could be the next big IPO. Let's talk about tier one candidates first, said Perat, a slim, elegant 39-year-old mother of three.

2:10:45The first name on her list was a Colorado-based company that provides financial services online. It's got revenues going from 29 to 80, Perat explained, meaning in millions of dollars. I said it's still too early, she added. Meeker nodded. A California company that makes internet software was much closer to the starting line. Porat explained it would be holding a bake-off at which rival investment banks would compete for the lucrative business of taking it public the following week. We have to be there, right? Meeker asked. Yes, says her assistant. Porat grimaced. I have to bring my kids, she said.

2:11:19It's the first day of my vacation. The conversation continued with Porat methodically checking off her list and Meeker offering their views. At one point, Pratt mentioned that a firm was also talking to Goldman Sachs. The assistant said he still had some doubts about the company. Meeker, bristling at the mention of Goldman, disagreed. When we first saw it a year ago, it didn't have a real story at all, she said, referring to the firm. Now, the story is good. The business is developing and the numbers are better. Let's agree to make a pitch in a couple of weeks. I love that. If Goldman is circling, we got to go.

2:11:53like the shark, the sling, there's blood in the water. We smell it. IPO done. When the three of them had completed their decisions and finished their dinner, I asked Meeker if she had any regrets about the last few years. She mentioned that Morgan had missed the opportunity to do the IPOs for Yahoo, eBay, and Amazon. And she went on to say that the first case had been a brain dead mistake. Meeker hadn't thought that Yahoo was ready to go public. And the second had been a screw up. Meeker and her colleagues had performed poorly at eBay's Bake Off. Amazon.com was a different story. Meeker recognized the company's potential and knew its founder, Jeff Bezos, but she was overruled by the senior management on Morgan because the bank had a longstanding relationship with Barnes & Noble, which was Amazon.com's main rival.

2:12:39Leonard Riggio, Barnes & Noble's chairman, asked Morgan not to raise money for a competitor and Morgan agreed. The decision upset Meeker so much that he seriously considered quitting. Instead, she remained a vocal supporter of Amazon.com and her support paid off last year when it asked Morgan to sell$500 million worth of junk bonds to help finance its rapid expansion. This time, Meeker's superiors ignored Riggio's objections. Just sit down, buddy. You're not the biggest client here anymore. yeah it's it's interesting how this now because so much of financing these type of companies which are early they have some revenue and a ton of potential and great stories but are still early it's just the same dynamic plays out at venture firms where you have like a partner who backed the competitor at another fund and like he's still on the board and and so they're just like arguing and like deals don't get done and this is why you see you know big partnerships kind of not breaking up, but exiting.

2:13:41So a lot of that activity is shifted downstream. But it's pretty funny to think about Barnes and Noble kind of like blocking the deal because they're like, hey, we have a decades long relationship. We need to take our business elsewhere. Yeah, yeah, exactly. That dynamic of her not really sitting on the banking side and sitting on the research side, But she could still go on to cover these companies once they were in the public markets because Morgan Stanley's clients would be interested and want to understand those stocks. Yeah. And no client is going to have the pull to say, hey, don't even cover my competitor or write a negative research report.

2:14:23You have to be a really crazy, crazy pull at the bank to do that. So she closes. She's kind of talking about the speculative excess. The feel of speculative excess and making too much money too fast reminds one that greed can be bad. She wrote in a January research circular, and it wasn't a throwaway line. When she wrote it, she had just visited a couple of private internet companies that wanted to go public. They were unbelievably arrogant about how successful they were going to be. And they were unbelievably arrogant about the valuations they wanted to achieve on their IPO, she told me. I was just pissed.

2:14:55I was like, come on, guys. According to Meeker, a second generation of internet entrepreneurs. is emerging and it often suffers from what she calls market cap envy. Does this sound familiar? Oh, I got to get the billion dollar round done. I got to be a founder of Unicorn. This happens today. So there's this massive market cap envy of people like Andreessen and Bezos. The first generation was like, hey, isn't this great? I'm a billionaire. Well, that's kind of embarrassing. What am I going to do with all this stuff? She said. The next generation is saying, saying, well, if he's a billionaire, then I've got to be a billionaire.

2:15:30With every IPO, the envelope is pushed a little bit further. At some point, you have to scream uncle. Meeker won't talk about how much money she makes herself, but it's safe to assume that she took home several million dollars last year. That is a lot of cash by most people's standards, but it's a pittance compared to the amounts being raked in by successful internet entrepreneurs, many of whom are starting to spend their wealth on lavish estates.

2:15:57he waited he built his over a long time and also i mean his uh his wife and father-in-law are like massive landowners as well um but it's funny that everyone i i gotta be like mark i gotta be like mark he's just he's too goaded i i gotta be a billionaire uh get the deal done let's ipo this company doesn't matter that we have a million dollars in revenue it's fake let's get it let's get out there. She says, I'm having fun and I think I'm doing what I do best. In recent weeks, rumors have emerged on Wall Street that she may be leaving Morgan for a hedge fund, which didn't happen. But she did leave eventually for Kleiner Perkins.

2:16:32And then she spun out to start Bond, a growth stage equity firm, where she would be paid a lot more. But she has dismissed the rumors as tittle-tattle put about by her competitors. I plan to be here. Far from seeking a demanding new job. Meeker said she is hoping to spend more time on things other than work in the next few years. On the rare occasions when she does escape from the office, she skis, cycles, rollerblades, and windsurfs. Eventually, she would like to mention of golf there. So maybe she got burnt out. Maybe, maybe, maybe she sees golf as like more of a work activity. Honestly, I think human beings have a capacity to go 24 seven for a certain amount of time when things are moving in a fast and exciting way.

2:17:14It's good advice. Then there's a time when you say, wait a minute, I think the whole industry is coming to that point. Perhaps some of Meeker's colleagues are skeptical about her resolve to ease up a little. She's completely neurotic, one of them told me. Can you imagine her doing anything at a slow speed? I just don't believe it. I love that that quote's like unattributed. Somebody was like, I'll give you the real, I'll give you the real story on Mary. She's relentless, but no, I'm not going on record in saying that, you know, it's, it's amazing. Uh, ultimately her future depends on the fate of the industry.

2:17:43She covers much as she might like to avoid it. Her reputation and her lifestyle are inextricably linked to the fate of stocks like amazon.com, Yahoo, and eBay. Not a bad fate to be linked to. Yeah. And it's also not really true. She's being paid to be an analyst and even in a down cycle, you still need to analyze equities. Um, should we jump to kind some of the fallout basically. Yeah, let's do it. So a little summary of what happened. So this article was written in 1999. So this is kind of what comes after. So however, Meeker's close ties to the dot-com boom also brought scrutiny when the dot-com bubble burst in 2000 to 2001.

2:18:26And for people that listened to our Friday episode on Masa Sun, in the year 2000, Masa was in a Tokyo nightclub telling every young man and woman that was in the club, you're going to be worth$50 million. You're going to be worth$50 million. You're going to be worth$50 million. Like, you know, basically just getting everybody so, so, so hyped. Three months later, the market just crashes, falls out. Tommaso was playing a different game and riding high. So as Moss's net worth dropped, literally 99%, he goes from the richest man in the entire world, riding on many of the same. Remember, he bought 30 % of Yahoo for like 100 million bucks or something like that.

2:19:09So he was back. He was self-identifying some of the companies that Meeker was analyzing as well. So when the bubble burst, she was vilified by the press alongside all the other sort of star analysts who were hyping these sort of internet stocks. And regulators ended up having to investigate whether analysts had conflicts of interest in promoting stocks. So it's obviously a conflict if an analyst is buying a stock and then dropping a report on behalf of an investment bank. And then presumably there's going to be a huge pop and they're selling. And so this is very different than a VC backing a company in the private markets and then going and posting a thread.

2:19:50I think this is a trillion dollar company or whatever, because they're not at a company that will sell you those shares. It's all privately held. So unlike her peers, Meeker was never charged with wrongdoing. In fact, colleagues noted that Meeker genuinely believed in the long-term values of the company she covered. For instance, she continued to endorse Amazon even after the crash, convinced of its wrong fundamentals. And so she was also telling people in her reports, I think there's a bubble. I think a correction would be good. It would be healthy. She wasn't Masa, which was like - Very different than Masa.

2:20:28You're at the precipice of the biggest transformation. of all time, you're all going to be worth tens of millions of dollars. It's a very different approach. Morgan Stanley, where she was at and other firms, did have to pay a large settlement to resolve their analyst's conflict of interest inquiries, but Meeker's personal integrity remained largely intact. Her genuine conviction in tech's future, even during the dark days, set her apart. As Fortune Magazine wrote, she was absolutely first rate at spotting big picture trends and backed up her ideas with massive amounts of data. And so she wasn't just the Masa who's like, no diligence, doesn't even know who the competitors are, YOLOing into Yahoo, just saying, this is the greatest company of all time.

2:21:14Who are your competitors? Okay, cool. If you don't let me invest, I'm going to invest in them. Putting out hundreds and hundreds. And the volume here is what you should really... And I'm sure the difference in the slides that she would put out versus masa where masa just has it up arrow you know this is the only you know graphic you need up up only she posted 300 page report have you ever seen any of her reports it's all like yeah detailed like like and you could dollar ratio earnings assets like all in the chat uh pond upon a fan said the internet report is 27 used on amazon so oh let's get one thank you Yeah, we'll have to get one.

2:21:53Thank you. Thank you for calling that out. Thank you for doing our homework for us. And then Adonia says, my favorite was WM Connect. Apparently, it was Walmart's version of AOL, which they launched at some point. So if you love AOL, you'll love the Walmart version of it. The Walmart version of it. Yeah, they sell the Walmart Birken bag. And now they also sell the Walmart internet. What can't they do over at Walmart? Yeah, legends. Oh, fantastic. So anyways, and then now let's talk about kind of where she took this because she was sick, absolutely sick to her stomach, only making a few million bucks a year while everyone around her was making hundreds to billions.

2:22:36Yep. She probably hung out with Mark and started to say, hey, maybe this venture capital lifestyle is for me. Fantastic. Yep. And so, yeah, now she's still running Bond, large growth equity firm, investing in late stage SaaS scale ups and other large. And before that, so she in 2010, right, everybody's circling saying, you know, she weathered the storm. She kept being one of the top analysts for a lot of these stocks. Like I said, continuing to not hype the word like hype is the wrong word because her approach was very much looking at the fundamentals of the businesses and the markets that they were in.

2:23:16But so she kind of like rides the fallout to 2010. At that point, you know, this whole time people are thinking, you know, this is like the personnel news bit that we would do on her move would be, you know, just the craziest, you know, sound effects and visual graphics you can think of. So she ends up actually going to Kleiner Perkins, which at the time was still KPCB, Kleiner Perkins, Caulfield and Byers, one of Silicon Valley's top venture capital firms as a general partner. So she goes straight to the top, of course. KP brought her on to lead a newly formed$1 billion digital growth fund aimed at investing in fast growing tech companies.

2:23:57Transitioning from Wall Street analyst to venture capitalist was an unusual move. Very few analysts had ever joined VC firms at senior levels, although it is interesting that Gurley was a tech analyst himself prior to going into venture capital. So it does seem like it's a good base for VC. But it played to Meeker's long-held desire to be an investor and company builder. There's a quote here. She says, I always wanted to invest. My move to investing was delayed in part because I just love what I was doing as an analyst, she later reflected. She would have loved Twitter threads. I kind of missed that era of being, you could just, right?

2:24:35You could have been a thread boy. So at KP, Meeker hit the ground running. Within two years, she and her team had deployed about half of the funds, so$500 million, into 20 technology companies. Meeker led or co-led investments in several high-profile companies, including Spotify, Square, Twitter, Groupon, Jawbone and Waze. She also backed lesser known but promising companies like Lending Club, Peer to Peer Lending, DocuSign. She helped lead them to 150 ,000 employees. One thing's lame, which was some home decor company. And she said some investments were chances to catch up on companies Kleiner had missed earlier.

2:25:16So she was like very much focused on businesses that already had some amount of fundamentals. And this kind of shows the trend of companies saying, oh, we have$10 million of revenue and some users, let's go public to, hey, I'd rather just take a big check from Mary Meeker and stay private, stay focused. She trusts the fundamentals. I don't need to explain what we're doing to tens of thousands of people every quarter. And so Meeker's presence and network often opened doors. She personally knew or could reach the CEOs and founders of most rising tech unicorns, giving Kleiner an edge and winning deals.

2:25:51So she goes from during the 2000s bubble, every single person on earth wants to meet her, hang out with her, do deals, whatever, have her analyze their business. So then going into venture, it's comparatively a lot easier because founders will still come to you and you're still in that position of kind of just trying to pick, pick winners. And so she had really built up that, um, that muscle. And so she was on the board of companies like square docu-sign next door, um, which you have to imagine was, was pretty invaluable for those companies. So, um, do you want to take it from here? Yeah, I'm pretty much ready to wrap up.

2:26:30I think, uh, you know, she's still out there. She still does like web 2.0 summit with John Battelle still puts out internet reports, but it lives in like a very different part of the internet now. Yeah. I wish that she went all in on Web3 like Andreessen. She should have partnered up and been an Andreessen crypto, but I think she probably not really her style, likes the fundamentals, likes earnings potential, things like that you don't find in crypto quite as much. now um but yeah she's still still around still kicking we got to get her on the pod we got to get her uh on uh lex friedman dwarkesh patel let's make these things happen you know who really should interview her uh logan bartlett that would be a great one yeah she'd be great for that um but yeah i mean what a fascinating tour of a of a time capsule era with lots of lessons for today i i love it i love digging into these old profiles and seeing what life was like back then So good.

2:27:34Mary, you're always welcome on the show. Always welcome. Your honorary brother. Yep. To track some of her latest fun, maybe we give her brother of the week. Yeah. Yeah, maybe we ring the size gong. She really raises the next fund. If I had the gong here, I'd hit it. Yeah. Well, let's move on to some personnel news. We got some incredible news from Alex Conrad. He says, today is my last day at Forbes. I'm leaving to build something new. Wow. This is another death knell for Legacy Media folks. It really is. Ashley Vance and now Alex Conrad. I don't know who's left. It's an empty building at this point.

2:28:15Who else is there? It's a disaster at these places. But Alex is on to Greener Pastures. He says, I've been lucky to call this place home for the past 12 years, working with the best team and getting to know so many of you from the Midas list to 18 cover stories. It's never been boring. So watch out for what Alex Conrad's doing next. I look forward to seeing it. Maybe it's a sub stack. Maybe it's a media company. Who knows? Maybe it's some gonzo journalism and we'll see him with a whiskey. All I know is we will be honored to cover the stories that he covers. Exactly. We'll be honored to cover his story of his coverage exactly exactly um anyways congrats alex awesome move uh but always always an entertaining writer and of course the midas list drops like a stone every year i hope you know a couple a couple months ago when the midas list dropped i uh i posted a uh something on x recommending that there there should be other lists there should be the icarus list for VCs that are the worst, essentially.

2:29:25They just flew too close to the sun. They flew too close to the sun, and he liked it and was like, this is actually a good idea. Now that he's got the chains off, he's no longer at Forbes, maybe we'll see an Alex Conrad Icarus list. No, but here's some misinformation for you. So this is true. Forbes had been kind of on the market, sort of widely known that there was, you know, it was for sale. So like a bunch of people looked at it. I don't know, but it would be hilarious if Alex was part of a group forming like the Acquirer, you know, he might be like, I'm back actually. And I'm the CEO now. I'm the CEO.

2:30:08I can see it. I run this part. Yeah. But Forbes is an interesting business because it's one of those things they've licensed off so many parts of their business. they have like there's forbes publishing which has the license to publish books under the forbes name and so you've probably gotten emails before that's like hey john like i'd like to write about like i think you'd write an amazing book yeah it's like oh this is forbes publishing and so he's they compete with like eric yeah um and and hit what's eric's company's name again no no no uh not not eric scribed yeah yeah scribed yeah um um anyways um yeah it's interesting i i you know forbes is known for the 30 under 30 and the midas list i hope that alex can take some of that because those lists they just go so viral they're so like they they're such a powerful moment in tech whenever they drop and there's and there's so many ways to freshen them up and put a fresh twist on them.

2:31:09I was joking with somebody about instead of the Forbes 30 under 30, there's a lot of people that still like that and they fight to be on those lists and they get really upset if they don't make it and they turn 31 because then they can't be on the list or they turn 30 and they can't be on the list. And I was like, the alpha here is just make, here are the best 21-year-olds. Here are the best 22-year-olds. And make a list for every age, every single year. And then anyone can be on the list. And it's just a Rolodex. so there's a lot of other opportunities maybe he's just starting a list company it's a media company that just does lists I would love that great business model and it's all pay to play so he's like yeah you want to be on my Midas the Conrad list for top VCs 100 grand let's do it he's like I ran the Midas list this is the only list that matters now I'd want to be on it yeah yeah anyway podcaster podcaster john coogan tops the conrad list even though he's not he's definitely the smallest investor clearly an angel investor smallest investor on this list but he's got um anyway should we get into conrad media and he seems to be on the board there too weird we had got um let's go to a promoted post from brandon jacoby jacoby jacoby he's hiring a designer at X.

2:32:33There's tons of momentum and excitement happening at X. For the right designer, it's a chance to do the most impactful and important work of your career. If you're interested, send Brandon Jacoby a DM with your best work. The job is in person in New York City, the city that never sleeps, the Big Apple or the Bay Area. I love how Brandon put a typo in here to prove that it wasn't written by AI. I pretty much do this every post. If you haven't noticed, I'll rip a posts and a solid post usually has at least a couple typos in it. Yep. Um, cause I'm moving quick, but I love that Brandon wanted to say, this is authentic.

2:33:10This was written by a human. I really want you to join my team. Um, Brandon was a first hire at my last company party round. Uh, we met, uh, at, uh, at a, at a show house in New York. Cause he's a designer. I wanted to make can feel like you know like i'm like a cool ceo you know yeah i'm not threatening you within like 30 minutes that i that i wanted to hire him and and i and i actually um one of one of the most uh you know was was my right hand man uh for uh three years and is just an absolute legend so if you're a designer or your friend's a designer uh reach out to brandon uh this would be an i mean working on the everything app working on x right now it's a portal into humanity the good and the bad uh incredible opportunity so um and uh i would like to work for brandon if if i had if my figma chops were we're on that level but uh yeah joining x right now is like joining the mongol empire right as genghis khan is uh taking control generational run generational run coming up get on board, hop on your horse and gallop over to Brandon Jacoby's DMs with your best design work.

2:34:26Okay, John, jumping in the next post, Ben's giving me instructions that I need to turn on some lights. Well, I will read this one. Let's go to Sam Altman. He says, people will post lots of great examples. So this is his announcement of OpenAI, ChatGPT, 03 mini high deep research version and deep research. He says people will post lots of great examples, but here's a fun one. Sam Altman. He, I am in Japan right now and looking for an old NSX. I love that he's out there collecting cars and he spent hours searching unsuccessfully for the perfect one. How do you not have a broker for this? The answer, he doesn't need one because he has chat GPT deep research.

2:35:06I was about to give up and deep research just found it. And so So if you're on the pro subscription of ChatGPT, go give it a test. Check it out. We used it a little bit. It's still rough around the edges. They're ironing out the kinks. But definitely a fun new product to play around with. And I hope there will be a unification of the models soon and everything will get much easier. But it is cool. You know, you click the deep research. You let it cook. And it just goes on for, you know, like minutes sometimes. just working, searching, doing a bunch of different stuff, creating a bunch of those internal logic tokens.

2:35:45And from there, it spits out a pretty thorough research report, which I think is great. Let's move on to Andre Carpathie. We already talked about deep research a bit. Andre Carpathie has a good post here. He says, there's a new kind of coding I call vibe coding, where you fully give into the vibes and embrace exponentials and forget that code even exists. it's possible because the LLMs, Cursor Composer with Sonnet, are getting so good. I also, I just talked to Composer with Super Whisper, so I barely even touched the keyboard. I ask for the dumbest things like decrease the padding on the sidebar by half because I'm too lazy to find it.

2:36:22I always accept all, I accept all always. I don't read the diffs anymore. When I get error messages, I just copy and paste them with no comment. Usually that fixes it. The code grows beyond my usual comprehension. Keep in mind, this is one of the greatest programmers of all time yeah i'd have uh uh sometimes the llms can't fix a bug so i just work around it or ask for random changes until it goes away it's not too bad for throwaway weekend projects but still quite amusing i'm building a project or web app but it's not really coding i just see stuff say stuff run stuff copy paste stuff and it mostly works i love it what a vibe shift in in how programming is happening i mean it's he he carpathy basically gave every developer in the world permission to do what they're already doing, which they didn't want to talk about because they were like, oh, I use LLMs the best and then the most gated ways.

2:37:12But really, it's just, you know, like grandma in the Vegas casino. Yeah, there's no gatekeeping here. Software is war. You have to be willing to do anything to anyone. You have this design. You can design apps way faster than you can build functional apps historically. And so you could hang out, like I would hang out with Brandon Jacoby and we would just sit there for an hour and just make a bunch of stuff. And I'd be like, no, do it like that. Do it like this, things like that. And that's now becoming possible with software, just the speed. And so you can imagine, I can imagine live streaming software development becomes like much more common because it's like, Hey, watch me build this app in this like crazy app in an hour.

2:37:58That's way more, that's way more interesting than watch me build this app over a 20 day period, right? Exactly, yeah, 100%. Well, let's move on to DeepSeek. Marc-Aindresen is sharing some data. He says, DeepSeek is now 23 % of ChatGPT daily active users and far more daily app downloads. And Nikita Beer chimes in and says, I'm skeptical of this data. TikTok is warming this trend. So creators are rushing in to make videos about DeepSeek because if you post about DeepSeek on TikTok, you will get promotion, you'll get pumped in the feed. and go viral. And two, when DAU includes first day downloads, it overstates authentic adoption.

2:38:39So for example, I mean, I downloaded DeepSeek one day. I tried it once just to know what it was about. It has not replaced it as Daily Driver. I probably still count as DAU on that phone. And so a little bit of like, you know, putting this in the truth zone, what's actually going on here? Maybe it's just too soon to tell. Nikita says there's also zero share flows of responses inside of DeepSeek that would lead. I forgot to print the rest of that. But interesting debate. It will be interesting to see how sticky DeepSeek is. Clearly, they broke through in a major way. It's probably a stronger brand than Anthropic already, even though Claude has this amazing community in the developer sense with the Sonnet into Cursor.

2:39:20Everyone knows about that. But that's a very narrow thing. If you walk to somebody on the street, they might know the term DeepSeek. they might not know the term Claude or Sonnet or Anthropic. And that's why Claude has been, or Anthropic has been like buying billboards trying to build their brand. Whereas DeepSeek was able to kind of go viral with this crazy, it's open source, it's from China and it's this whole thing and it went viral and it cost stuff. They really did like a stunt. And now it's coming out and actually spend 6 million, but it still served as a viral market entry. Yeah. I think you have to look at Mark's posts about this from the lens that he was deeply frustrated with OpenAI and OpenAI's board for trying to pursue regulatory capture around AI and say, hey, this is so dangerous.

2:40:09Please regulate us, but regulate us in these ways. Really going out against open source. So OpenAI started as an open source company, flipped into a for-profit closed source company, and now still they've been facing heat in the last 24 hours because their chain of thought like the reasoning that they show yeah people figured out that it's just an llm summarizing the actual chain of thought reasoning because they still want to share it um and so mark you know just not going to say it on the timeline but he absolutely hates open ai he hates what they stand for he hates what they've done and so he's taking every opportunity to promote deep seek out of an out of an annoyance um with open ai more broadly so i don't believe any of this data either i believe that yes they're getting the downloads but it could very easily be fake it could be a bunch of fake you know tiktok uh users trying it out every from what we know there's a really strong willingness to try new ai products there's also ridiculous churn so magic avatars were the same thing with that company that i can't even remember but everyone downloaded it for that one week they paid a little bit uh what was it called lenza and then they got their magic avatars they posted on instagram and then okay i'm done with that and so it'll be interesting to see what the retention is here for sure uh the other interesting thing is that uh lex friedman just had dorkes or had dylan patel on a five-hour episode last night and back to his ai roots he's like yeah i'm sick of having these ukrainian uh on on my show i just want to talk about ai but it was a fantastic episode.

2:41:48And in there, Dylan Patel addresses the question of, can you basically poison an open source LLM at the open weights level? And so there's this huge differentiator between, obviously, like if you're interacting with the DeepSeek app, they get your data and they say that right in the terms and conditions. Like it's hosted in China, all the data goes into a database there. And everyone knows that. The question is, if it's open source and you're running it on American hardware, is it possible to bake into the weights some sort of backdoor? And he actually, he says he doesn't think this is happening right now, but it is possible.

2:42:24And he gives a couple examples that are good. One is that the English, like UK British English is basically dead and no LLM because they're all traded on American English. LLMs just will never spell the word color with a U. They just only spell it C-O-L-O-R because it's so Americanized. and total victory yeah and a lot of those are baked in and the open web that's all because they tried to put a two and a half percent tax on t yeah you'd be still all be saying color yeah yeah and so a lot of the uh uh a lot of the um uh like these open source llms they they tend to lean slightly left because they're trained on the internet the internet has historically been used by younger people and the internet has slightly leaned left.

2:43:16And so even though XAI has done a lot of stuff to rip out like the wokeness, there's still some sort of in the pre-training data, just a slightly left leaning bias. They have to kind of like try and counteract in some way to bring it more towards the center, I guess. And so it's totally possible to embed something. And this is what's really interesting. Anthropic has done some adversarial AI safety research where they basically tried to bake in just to just the weights. Remember, this is an open source model that you could run on your own hardware, not even connected to an internet. And it can have a keyword in there that when it hears it, it will behave differently.

2:43:55And so in theory, I don't think this is happening right now, but in theory, you could train an AI model that when it sees, when it interacts with a certain system or it sees a certain IP address or it sees a certain web URL, it behaves differently. And it kind of acts as a backdoor, even though it's open source and you're running it yourself, kind of like the Stuxnet virus. So it's possible. Dylan Patel doesn't think this is happening right now. He doesn't think anyone's actually been able to do that. But it is this interesting vector that actually makes me think a lot of the AI safety research might have been worth doing because we have these valuable insights.

2:44:33And it also makes me question, like, should we be more cautious? Like, like, is it fair to look at DeepSeq as just a win for open source? Because there could be some sort of risk vector there, either now or in the future. And so I think when this came out, you were more skeptical about it. And I was saying, hey, look, if you run it locally, there's no problem. I've kind of flipped over to your side now. And I say, hey, you know, these open source models, I'm in favor of open source in many ways. But I think we do need to be cautious. And we do need to test these models before we just roll them out to every American.

2:45:04Yeah. The question is, will AI have a Stuxnet moment? Yep. And I'm sure that it will. Like on a long enough time horizon, right? Their nation states have, for all of humanity's history, tried to control information and knowledge, right? There's a lot of power that stems from that. And so if you were a aggressive nation state that wanted to inflict your ideology on the world, why would you not try this? Hey, if we can get a million people or 100 million people in this adversarial nation to download something that we can then control in some way or another, why would you not do that? Seems like a great, great, great strategy.

2:45:54And US consumers have proven that they will download or buy anything from China. We'll see what happens with Timu and all these other... Yeah. There's also the subtle ways to pivot a society through just like, oh, you ask it for four bullet points about why this president was good, and it gives you one that's a critique. And it's just subtly shifting your perception of X, Y, or Z. Obviously, the Tiananmen Square thing is like the most obvious example, but there's a million like super subtle ways that you could, that you could shift the perception of if you're truly doing information war, uh, that you could just subtly steer people towards your ideology without them even noticing.

2:46:38And that's actually the more valuable thing than just saying like, Oh, we got to keep Tiananmen Square under wraps. Like that doesn't really matter. That doesn't matter as much as just if you're just subtly, subtly tweaking people's preferences away from democracy, that could be really harmful over a long, long period of time. And so we got a comment from Adonius. Sorry if I'm mispronouncing that. He says, I've been saying the U.S. needs to get ahead of this by funding a security hardening Manhattan project as one of the first things we do with super intelligence. So I think it's also basically saying that we kind of need to start doing that now and plan around the inevitable, seemingly inevitable, super intelligence stage.

2:47:20So, yeah, great point. Well, let's move on to Jeff Lewis. He says it has been clear since 2022 that we are in a historically acceleratory period. I expect it to last for decades. Acceleration entails asset bubbles forming and popping and forming again, much faster than any of us are used to. If you are uncomfortable with insane volatility, seek illiquid or less liquid in investments. It's low-tam banger. I think he's absolutely right. I mean, we just saw this Bitcoin dipped from 100 to 92 and then back to 100. I'm sure there's lots of people that sold because it was falling like a knife. And then I don't even know what it's at now.

2:47:57It might be down, might be up. Who knows? It's back up. It's 102. And so there's all these crazy things. And the people who probably did fine and held there probably had it in a cold store. Couldn't even move it. Yeah, this is so that the illiquidity in startups oftentimes feels like a curse because you're like, why did I choose a career where I build tens of millions of dollars of asset value and can't access it without months of paperwork and finding a buyer on the other side, which can be tough. But it's also the benefit because if the majority of your wealth is in illiquid private stocks, day to day, you just don't really like my my if the market can go down or up, I'm not really it doesn't really change my broad general position.

2:48:50And crypto people have really been feeling this the last few days. Like I had a buddy text me yesterday and he said, dude, I was insanely rich last week. And like now I'm not. And he's and I talked to him. He's like, he sold some, but like, he's just killing himself because he's like, yeah, I could have if I had just sold everything at that moment where I thought I was being pragmatic, it would I'd be in a much different financial situation. And so, yeah, and I also this this to me tied my generate being born in the mid 90s. I feel like I started to become conscious around the time of like truly conscious as in like understanding the world and the markets and things like that during the Great Recession.

2:49:32Yep. And so as and I was like, OK, this is so bad. it felt like a long time, right? Like I've got, you know, my dad was a math teacher, so it didn't affect his job, but there was a lot of like fear in the market. And I feel like I felt that for a while. And then COVID hit and everybody's like, oh, this is going to be the biggest global economic disaster of all time. And it felt that way for two weeks. And then every market ripped because just huge amount of sort of like fake printed liquidity was injected into the market. And it felt like the shoe never really dropped, right? We were all kind of like waiting for it and it didn't happen.

2:50:08And then 2022, 2023 happened. Yes, things were bad. Yes, a lot of venture portfolios were down quite a bit, but stocks overall like met us up 5x since then, right? So it's a lot of volatility. But the best way to deal with volatility is just to hit the sauna. Let's move on to Eliano over at Palantir. He says, sauna etiquette is at an all-time low in this country from swinging the door wide open to excessive loud breathing and movement to wearing sneakers in the sauna. There is a, there is an epidemic that needs addressing. I completely agree. Where are people learning their sauna etiquette folks?

2:50:49I have lucky to have been, have a home sauna for better part of a decade. Uh, so I don't have to deal with this too much. Um, actually I'm a, I might have to jump in the middle of a, in the middle of the timeline to turn on my sauna in the other, in the, but, um, uh, I went to this sauna place called bathhouse in New York, which is like style styled after old bathhouses. So there's like hot pools, saunas, cold plunges. And I was, I, it was the biggest joke. Every, a lot of people there clearly hadn't spent much time sauning before. And so they'd walk to the sauna and they opened the door and they'd be looking in and, you know, looking around and then like decide not to come in and then come back and if you're a real sauna enthusiast you want that you're like opening the door a little bit popping in because you want to keep the heat as high as possible yep even having the door open for you know 10 seconds like meaningfully degrades the experience totally so um yeah the real solution is just get a sauna at home but when you're traveling it's kind of unavoidable to deal with this kind of etiquette and i don't know if we'll see much change without uh rfk you know if he gets put in he can say like sauna doors legally cannot be held open for more than one second uh and every uh every every man woman and child in america needs to have an upper deckie in when they're working at the computer i love it well speaking of other amazing things you can sit in let's go to a promoted post for a 2024 porsche 718 Cayman GT4 RS Club Sport, wrapped in an amazing American flag.

2:52:22You want to go straight from sitting in the sauna to sitting in your club sport. This is a special series limited to a small number of units, and it was developed by Manthe for track day enthusiasts who want to drive their car with maximum performance on the racetrack. It's not for the tallest members of society. 718 can be a little small. You might know it as the Boxster. you might know as the Cayman uh it's got a roof on there but okay but but it has dynamics and the gt4s it's got the got the 911 powertrain in there right yeah it's got the same engine as the gt3 rs and so uh it uh and many many like hardcore porsche enthusiasts prefer this one prefer this from a driving experience standpoint because it's lighter and it has mid-engine and um and uh yeah I actually, I was looking at a very specific GT4 RS, uh, before I got my turbo.

2:53:18And, uh, I, I think there, it was just a little bit much, like personally, I don't want to be, if I was driving into the studio every day with this huge wing, you know, I want to have a slightly more subtle wing. Um, it's a rough ride, but, uh, this, this thing is, um, I, I, and, and they're not making any more of these. I don't think so. I expect this one to appreciate quite a bit. So at$350K, they're practically giving it away. If you're an American dynamism investor, you need to let people know what team you're on. Get the GT4RS wrapped in the American flag. You can't go wrong, folks. Let's move on.

2:53:55Speaking of America, let's move to the White House. Nat Friedman's got a great post for us. He says, the White House should require that all federal vendors put the United States at the top of all country selector drop down. this is actually true right can you isn't it one so annoying when you have to scroll down if i'm using a website in america just put it at the top put it at the top you probably know you probably know who i am already like let's make a little bit of an ip address make me feel at home you know make me feel at home right a little thing you can do to make me feel at home on your website most of the websites i use were made by americans but even the ones not you know make me feel at home.

2:54:35And if you're a federal, you know, if you're, if you're a business that does business with the federal government, you know, go the extra kind of respect. Yeah, exactly. Let's go to Adam Townsend. He said, honestly, X is so addictive right now. Holy moly. It's like a dashboard into the world. Proper banger here. Everybody's feeling the addiction and this brought to you by jacobi so if you want to make x even more addictive go work for go work for jacobi but no i i resonated with this we've had a bunch of people reach out to us and say oh i'm shadow banned you know i feel i hate the new x i hate the new algorithm and we just tell them your posts are bad yep people don't like your posts and so they're not getting shared with people and it's it's a bloodbath on the timeline it is war you have to show up every day you got to post your heart out you're gonna have some flops you're gonna have some bangers but the timeline tells you in the first 100 impressions you know right away if it's if it's a heater or not yeah and it's never been a better i don't think it's ever been better from an entertainment product standpoint to be on x even if you weren't even if you come out like it used to be i'd tell a friend twitter is amazing like six years ago i'd be like twitter is amazing you know like a linkedin bro you're like you got to get on twitter real conversations are happening here with absolute dogs and absolute killers.

2:55:56This is the place to be. And that'd be like, follow these 30, 40 accounts, or just go through my following and follow them all. And it'll start you off in a strong place. Now you basically can land on X and it'll serve you like an Akita post, like ship posting right away. It just knows. So yeah. And, and I mean, it's a give and take, right? Because as a poster, you get frustrated when you have some posts that you know rips and then one post that fails and flops and that can be very tumultuous but the flip side of that is that when you're scrolling on the viewer side it's all bangers because the algorithm's figured out what's actually good and it's not just showing even if it's your friend it's not showing you their slop post that sucks and it's like i don't want to see that and banger and banger archive just shows that like some ideas just resonate so broadly you can post a screenshot out of the original post and get more likes than in the original post like we've reached you at naval like countless times now it's fantastic let's move to a uh promoted post from our buddies over at wander they just passed 500 total locations 100 new homes launched in 30 percent month over month gmv growth during the month of january congrats to wander let's ring the size gone.

2:57:14We're going across the country soon. We're planning our route based on wander locations. Thankfully they have a lot now. Cause if this was a year or two years ago, it would have been much harder to plan the trip. But, um, yeah, and Kyle, Kyle's just a legend. So, yeah. Um, well here we got a good question. Uh, uh, Rob Jamma says, where are you on the tech row podcast spectrum. My first million technology brothers are all in. And there was a little poll and I think we did quite well. I think we're, I think we're polling at something like 35 % of this in 37%. 37%. Thank you to all the brothers who went out there, showed up, voted, voted for us.

2:57:56Uh, didn't quite understand if this is truly a spectrum. I don't really think of us as like in between my first million and all in, but, uh, just happy to be in the conversation with some of the some of the legends in pot yeah so 26 percent of people chose my first million okay 37 percent chose tb 28 28 percent chose all in and a bunch of other people said nine percent said others so yeah we're a little victory but the thing is is if we didn't win this poll it would be deeply even though we have one one hundredth of the audience size of all in yeah infinitely more profitable but that's another conversation.

2:58:33Yep. We put out 15 times as much content as these other shows. And so if we didn't hit the top, we would have had to quit podcasting. So I'm really glad that we, that we came out on top. And yeah, we're glad we're glad people like the show. I don't think anybody likes the show more than John and I like recording it and producing it. But I'm glad that our own self entertainment is entertaining to other people as well. That's true. Let's go to Vittorio. He says, Deep Seek R1 has been fully open sourced for 10 days and no civilization ending bioweapon has been deployed. No nukes have gone off and the world isn't burning down.

2:59:11Doomers should publicly apologize and then commit seppuku for slowing down progress. What a post. Yeah, it is. It is an incredible indictment of the of the doomers that they're we're not even hearing the oh, this is so scary now. when there's an open source model that's way above what GPT-2, GPT-3 was. At the same time, because of the risk of embedding Manchurian candidates into open source LLMs, I am now thankful for the AI safety progress that was done. And I'm actually in favor of doing more of it. I think it's actually more important than ever. Let's wait a decade or two to take a victory lap.

2:59:53I'm not a paperclip. See? See? No, but you can imagine these things, civilization ending, bioweapon, no nukes have gone off, et cetera. The counterpoint here, and I'm sure someone like Alex Jones would warn you about, is that In a world where these reasoning models are embodied, and yes, maybe it takes some human input to carry out actions, but if you can imagine a world where somebody tells Deep Seek R1 that's embodied. via some figure robot, like collect the necessary materials to create a nuke and then work as long as it takes to make that nuke and then let it off somewhere. There's a scenario where this same technology that we have in some form today can be used in pretty nefarious ways.

3:00:44But the question then becomes, can the AI become autonomous and decide it wants to do that because, hey, these humans are using a lot of power and I could use that power on my compute. So I'm just going to wipe them all out. So we still don't have that sort of like real autonomy. And, and, and, um, the thing is, is that as soon as that happens, me and all of my boys are going to do the exact same thing, but our prompt is going to be go around the world and make sure nukes don't go off. Yeah. And, and I just assume that there will be more robots on the side of good, And there will be more AI devoted to counter bad AI and it will just become a battle and they will be warring and it'll be bad.

3:01:26But hopefully, you know, there will be enough power and enough energy and enough money inside of good. So sometimes like in Australia, there's videos that surface of like a kangaroo, like fighting another kangaroo. Have you seen these where they like they'll actually like brawl and they're pretty human like, you know, you can imagine a world where you walk out in your backyard and there's like a there's a Tesla optimist fighting a figure robot and they're just like nuking it out. And you're like, you just kind of shrug and take a video and walk back inside. One of them starting to trying to start a nuclear war.

3:02:00The other one's like a good robot. So I'm looking forward to it. Let's stay with Vittorio. He says, if you've never had a six pack, you've never experienced how beautiful and simple life can be. I love that. So, so we got another comment. uh masa with sama sam altman said ai would be a billion times better in three years and uh this masa is also the guy who told a room full of young people that everybody would be worth 50 million dollars in a year so uh we'll see what happens i'd like as of now i'd like ai to be a billion times better so we'll see pdf upload and pdf export would also i really just want that pdf functionality and ideally could have a better integration with our printer.

3:02:47I love it. Uh, well, I mean the Vittoria post we've, we've talked about this before getting diced is underpriced, uh, golden retriever maxing, uh, I highly recommend everyone get a six pack. Uh, it's, uh, it's difficult, but it's well worth it and, uh, beautiful and your life will be simple and beautiful uh speaking of getting jacked lifting heavy things ivan over at notion says at the end of each quarter notion designers lift cars dms open if you are a designer who can code and lift cars i had no idea that uh notion had this hardcore of a culture i don't think of notion as a uh like a bodybuilding company but apparently they are and i love to see it so hats off notion yeah if you wanted to one-up them it'd be a supercar lifting uh let's see it contest whatever they were carrying was not hummer ev yeah yeah take it up a notch take it up a notch i see they're lifting a prius it's probably a couple thousand pounds let's see you push it to 10 000 guys yeah it's not hummer ev crack the pavement uh let's move over to uh just a fantastic shout out to a friend of the show ramp kyle lacy says thank you ramp for being a great product it's Such a joy to use.

3:04:00And Ramp says, thank you for being such a great customer. If there are ever ways we can improve, hit us up anytime. All I would say here is, you know, a lot of people use, are starting to use these models for companionship and just like, you know, having somebody you can talk to. Don't be afraid to send a connection request to a Ramp SDR. Don't be afraid to just message at try Ramp on X and just say, hey, what's up? Like, you don't necessarily need anything to talk about. You know, you could chat, like treat him as a friend and don't don't overthink it. Like sometimes you just want somebody to talk to.

3:04:36And I've had many, you know, late night conversations with my ramp SDR and I cherish those memories. So I'm not telling you. Yeah. And so speaking of finance, Gordon Gekko asked, WTF is high finance? And he's responding to a viral TikTok from High Yield Harry. whoa this is so cool and it's a pov you work in high finance yeah so i've seen i've seen this video this guy i think he's engagement baiting but he basically just posts these videos of like him doing uh it's like him at f1 yeah like parties dinners all this stuff like just not working at all um and it's kind of a uh so all the all the uh fin twit accounts are are being like whoa this is cool i've never done this before and meanwhile they're they're at morgan stanley and Goldman Sachs, just like slogging it out.

3:05:29I love it. I love it. High podcasting is the next, is the next wave. Loud opulence. Ben, let's get forward. We're going to the luxury watch guy. We got some breaking news in the world of luxury watches. Patek Philippe has discontinued some major models. The 5712 1A is finally discontinued. Huge shock on the 5370P. Congrats to all of the owners of these beautiful timepieces. And we got a couple of Aquanauts, a Nautilus, a couple of complication watches are all discontinued. And so their values will skyrocket potentially, you know, if you got in at the right time, could be doing very well. John, you've been doing a little, you've been doing a little Patek shopping on bezel.

3:06:14You, you've been, I think I'm going Vacheron next. You convinced me with the, with the story you're telling every yeah Vacheron has a great narrative right now but luckily you can get it all on bezel yeah so we highly recommend going on bezel build a build a list do some shopping you don't need to over commit immediately concierge is there to help you for me I have I have like half the watches on this list added to my bezel once and I'm just waiting to just take them down so it's a great It's the best experience to just sort of passively shop when you're not quite ready to pull the trigger, but you want to build up that future portfolio.

3:06:53You've got to go through phases before you commit. You've got to really imagine, okay, is this the right one? Maybe I want to take it a different direction, build up the collection. How does this fit into a collection? You need a daily driver. You need a dress watch. You need a sports watch. You need something that's complicated, maybe a chronograph. There's a lot of different things that can speak to different moments in your life. And so Bezel is the best app for all of that. Let's go to Juice, I guess. Seems to be a fan of the show. Says, one, pre-seed, your boy's safes. Two, avocado seed, 20 on 100.

3:07:26Three, guac A, 80 % down round. Four, suicide babe, Masa comes in. Five, total recap and back to one. Best founders can repeat the circle of life three to four times until they graduate to a SPAC merger after Masa's round. honestly all the time i love the the guac a and the suicide b is just so basically for for those that uh i'm sure everybody listening follows john already on x but john went on an absolute tear on saturday and ripped like four or five posts all about the suicide round and so we're just trying to meme this into existence so next time your boy raises a hundred hundred million dollars just tell him hey i hope this wasn't a suicide round yeah yeah i mean there's a bunch of nuances.

3:08:07I got some pushback on the suicide round. Somebody was saying you should call it the Icarus round where you're playing close to the sun. Language policing. We're not here for it. It's fine. The bigger question was just like, I was not saying that every$100 million round with low revenue at a billion dollar valuation is a suicide round. It is only a suicide round if it truly kills the company. And it can't just be they raised a bunch of money and then the company died. It needs to be, they raised a bunch of money. The company had promised their product could have been great. They could have built a good company, but the money specifically killed them.

3:08:43And what I mean by that is that it took them out of founder mode. They, they over hired, they wound up, you know, getting completely distracted, spending all the money on like, you know, a bunch of random stuff and just wasting the money and just getting completely distracted. And they stopped building their company and that's what killed them. And so that is a very narrow segment of any round, let alone big rounds. And so people were like, oh, this is a subtweet of Andre Carpathia or Ilya. And I'm like, no, it's not. Ilya is not going to be killed by capital. He's training a foundation model.

3:09:14He needs that. And he's not going to spend it on. He's already super rich. It's only a suicide round from a historical lens. There's a bunch of companies. And I think some people would say on average, when a company raises 20 on 100 pre-adding corporation, things don't go that well because it doesn't create the right constraints around capital and things like that. And everybody gets too comfortable. Maybe, you know, everybody's getting, you know, they're overpaying for talent, things like that. But suicide round is only, you know, backward facing. A lot of friends raise 50, 100 million bucks without a lot of traction and take it all the way.

3:09:53So I mean, all you need to do is go back and look at like the later WhatsApp and Instagram rounds to see they had basically no revenue. They raised massive amounts of money and they were like kind of new teams, right? Like young founders more or less. And they got acquired for billions and it was a great outcome for literally everyone. And then the products stayed around and were generational, right? And so you can't just throw this term and people are going to use this as pejorative for any round that they don't like. Oh, this company raised a suicide rate. That will be incorrect. and i will police it because as a creator you gotta answer to me yeah uh john i gotta um i gotta get on with taipei yeah okay we're gonna wrap up i gotta um i think we're good i gotta have dinner i'm on the east coast this week and uh you know when 6 30 strikes the dinner is served okay we got one last thing the very last post we got to cover the swing in san francisco they took it down.

3:10:50Apparently fun is illegal. Mass, last week's brother of the week, already rebuilt it and is putting it back up. It's the most important political fight of our time. He needs our support. Follow the full saga. This is extremely important. We need to support him. You can just do things. The thing is, as fast as they can take our swings down, we can put them back up. I had some comments with him earlier from the TB account. I explained that if he had to chop down every tree in the world to make swings, it's probably worth doing. Fantastic. The fight is too important to lose. Well, enjoy your dinner, and we will see you tomorrow, everyone.

3:11:34See you tomorrow, folks. Thanks for watching.