TBPN Live - Suno Sparks Music Rights Firestorm, Travis Kelce’s Six Flags Play | Philip Johnston, Justin Murphy, Darren Rovell, Guillermo Rauch, Brendan Foody

Episode Date: October 27, 2025

(01:11) - Suno App Stirs Debate Over Music Rights (39:23) - Travis Kelce Joins Six Flags Stake (49:05) - 𝕏 Timeline Reactions (01:00:09) - Philip Johnston, founder of Star Cloud, discusses the company's upcoming launch of a satellite equipped with an Nvidia H100 GPU, aiming to demonstrate the feasibility of running terrestrial-grade GPUs in space despite challenges like radiation and heat dissipation. He outlines a roadmap that includes progressively more powerful satellites, leveraging advancements like SpaceX's Starship to reduce launch costs and increase payload capacity, with the goal of making space-based data centers economically viable. Johnston also addresses skepticism from industry experts, emphasizing that the primary challenges are engineering and manufacturing-related, rather than scientific, and highlights the potential for significant disruption in the telecommunications industry through technologies like Starlink's direct-to-cell capabilities. (01:30:33) - Justin Murphy, a former political science professor, left academia in 2019 to pursue independent scholarship online, focusing on philosophy, social science, and technology. In the conversation, he discusses his transition from academia to the internet, the challenges and opportunities of being an independent scholar, and the importance of adapting to technological acceleration. He also highlights his forthcoming book, "The Independent Scholar," which aims to guide others in navigating this path. (02:06:04) - Darren Rovell, a sports business analyst and avid collector, discusses the recent sale of a 1914 Baltimore News Babe Ruth rookie card, which sold for $3 million less than its 2023 purchase price, marking the largest card loss in history. He explores the card's obscurity, noting its late recognition in the 1980s and its distribution as a newspaper insert featuring a schedule on the back, suggesting that its lack of widespread recognition contributed to the significant loss. Rovell also touches on the broader sports memorabilia market, mentioning his own successful sales and the importance of understanding market dynamics beyond player performance. (02:26:16) - Guillermo Rauch, CEO of Vercel, discusses the company's recent $9.3 billion valuation and its strategic shift towards building an AI cloud to support the next generation of software agents. He highlights the development of Vercel's AI SDK, likened to the "React of AI," which is rapidly gaining popularity among developers for creating AI agents. Rauch also emphasizes the importance of integrating AI services into the cloud, noting that Vercel is becoming the default platform for this new agentic workflow. (02:40:13) - Brendan Foody, CEO and co-founder of Mercor, discusses the company's rapid growth, including paying out over $1.5 million daily to experts and securing a $350 million Series C funding round at a $10 billion valuation. He emphasizes Mercor's focus on increasing platform candidates, improving matching, and accelerating delivery to enhance client hiring processes. Foody also highlights the trend of businesses training AI agents to automate redundant workflows, positioning Mercor at the forefront of this transformation. (02:51:06) - 𝕏 Timeline Reactions TBPN.com is made possible by: Ramp - https://ramp.comFigma - https://figma.comVanta - https://vanta.comLinear - https://linear.appEight Sleep - https://eightsleep.com/tbpnWander - https://wander.com/tbpnPublic - https://public.comAdQuick - https://adquick.comBezel - https://getbezel.com Numeral - https://www.numeralhq.comPolymarket - https://polymarket.comAttio - https://attio.com/tbpnFin - https://fin.ai/tbpnGraphite - https://graphite.devRestream - https://restream.ioProfound - https://tryprofound.comJulius AI - https://julius.aiturbopuffer - https://turbopuffer.comfal - https://fal.aiPrivy - https://www.privy.ioCognition - https://cognition.aiGemini - https://gemini.google.comFollow TBPN: https://TBPN.comhttps://x.com/tbpnhttps://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231https://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://www.youtube.com/@TBPNLive

You're watching TVP. Today is Monday, October 27, 2025. We are live from the TBPN Ultrudeau. The Temple of Technology, the Fortress of Finance, the capital of capital. We have breaking news. Ramp.com. Time is money. Save both. Easy use corporate cards, bill pay accounting, hold out more. We know we actually have TPN breaking news, which is that tomorrow. Just to give some backstory here on Friday, we said this week was going to be magnificent. Yes. That was our teaser. Here's why. Tell us, John. Tomorrow, we will be interviewing Satcha Nadella, the CEO of Microsoft, live on TBPN. Tune in, get ready.

It's going to be great. And to be clear, that is a real picture that you just saw on your screens. It is. It is. It's a real JPEG. It's a real JPEG. It's not anything but a JPEG. That's right.

It's real. That's right. Super excited for that. We will be up in San Francisco tomorrow. Yes. We will live. I cannot wait. Yes.

And we will be, of course, leveraging re-stream. One live stream, 30 plus destinations, multi-stream, reach your audience, wherever they are. We take re-stream with us on the road when we go to do our big interviews, big interviews and small. We're always streaming. So over the weekend, there was some debate over Suno. Is anybody paying for AI music? Who's paying for this thing? It was the question of the day. And this was on the back of reporting from the information that says music app Suno nearly quadruples annual recurring revenue to $150 million. They were doing 40.

Now they're doing 150. And Michael Rosenfeld asks, can someone explain where this revenue comes from? Who is paying? And Harun says me. Suno music is one of the most magical products I've ever used. Harun is the founder and CEO of Rocket Money, the app, which was acquired by Rocket Mortgage. fantastic outcome rocket money if you're not familiar used to be true bill i think it was a billion dollar acquisition really cool and uh and haroon is enjoying it and if you scroll the replies in

this viral post that got over a million views and four thousand five hundred likes uh palmer lucky says he's using it uh gabriel the uh the open ai uh saura he's not the sora lead he's saw a research. That's right. He says, I'm paying. It's an incredible product. Dwight over at, at captions, I guess now Mirage, is saying it's such a great product, super fun if you have kids too.

And there's a couple of people that laid out specific ways that they use Suno. And so I wanted to kind of revisit this, noodle on this, like understand like what's actually going on, what's driving the revenue, because there's a whole bunch of different buckets of value that you can be extracting if you're Suno the company. So the worst possible scenario is that 100% of your revenue comes from someone who's just scraping your API basically and trying to distill your model. Like if we went back to what happened with Deepseek, Deepseek was paying for GPT4 tokens, but GPT4 obviously was generating money from all over the place and had people paying for knowledge retrieval and a whole bunch of different use cases.

coding tokens, all sorts of stuff. And so it wasn't exclusively that. But you could, in theory, like the most circular problem, you know, the one that would be really thrown out the red flags would be, is it all coming from one customer? Some other lab or someone who's just trying to like, you know, basically like distill everything that you've done. That's clearly not what's happening. You can see that so many people love it.

And if you add up the number of reviews on the iOS and Android app store, it's over a million. Wow. So they have over a million reviews. And I mean, I don't even know how many people leave reviews. I don't even leave reviews. Yeah, yeah, yeah, exactly. And so you're probably looking at, you know, even if they get, what, 10% of people who use the thing to leave a review, you know, you're looking at, you know, millions and millions of

people that are using this. Lots that are paying. And it just kind of adds up pretty quickly. I think they have an $8 a month plan and then like $20 or $5 a month plan. And the thing I've been noticing recently on social media is it seems like they've had some type of, I wouldn't call it specifically a Studio Ghibli moment, but people are realizing that they can go to their favorite artists, like a rapper, let's say, make this, make a, make a, make a, make this future song, make the jazz version of this future song. I saw DMX, X going give it to you,

but in 1960s jazz rendition, and that's that sort of, it's almost just style transfer. It's really just filtering. It is, I mean, that's the beauty of the studio Ghibli is that you don't have to come to it with something that's completely a blank canvas. You take something that you like, a picture of your dog, a picture of your family, a funny meme, an image from a movie, and you just say, hey, re-style this in the style of Ghibli, and you're good. They're certainly having that moment, and it's doing very well, and I do see those viral things. Good thing about that is that I don't, I don't think the jazz community is up in arms. I think they might be.

Maybe, maybe, maybe, but they don't have a strong case to be like, you are infringing on X, Y, Z, IEP, right? Like, there's no company called... I think that they might. I think they'll definitely be a lawsuit. Well, so, of course, like, individual artists might try to figure out, okay, are they training on my music? And then in that case, did they buy the music?

Because we now have precedent that shows, if you purchase the songs, can you train on them? The question is, like, if you purchase Spotify, if, you know, Suno is, like, getting, like, streaming? Yeah, we have a $20 a month plan. Yeah, we streamed it. to get inspired.

Yeah, that is funny. But in general, I think this kind of use case is probably, like, good for the underlying artists. It's just like marketing. It's fan engagement. And then they're just doing kind of like broad style transfer from a category of music onto this new artist. Yeah.

I wanted to go a layer deeper on this. But first, let me tell you about Privy wallet infrastructure for every bank. Privy makes it easy to build on crypto rail. securely spent up white label wallet, sign transactions, and integrate on-chain infrastructure all through one simple API. So Chris Dixon, who actually was on the show last week, he, 10 years ago, he posted a blog post called Come for the Tool, stay for the network. And the thesis was that if you use the example of Instagram, he also uses Delicious and a couple other examples. But basically, you're trying to start a social network. Why would someone come?

well, they came to Instagram because if you took a photo with your iPhone back then, the cameras were terrible and adding a filter to it made it look cool and grungy in old school instead of just actually terrible. And so photo filtering apps were very popular. There was one called Hipstimatic,

but it cost money. And so Instagram was able to give you a free tool. You came for the tool. Hey, I just want to filter my photos. Maybe upload them. I'll just have a nice little online photo album, essentially. And then over time, you stayed for the network. And then eventually the Algo feed and eventually the multi-million dollar professional content producing cohorts that are producing like the greatest content at like Netflix level that will get fed to you perfectly.

And so we're in this era where there's a lot of tools right now. There's a lot of AI tools. Suno feels like it could be an AI tool. And I was debating this with you earlier this morning. of the $150 million that are you know that's their ARR how much of that is people showing up for instead of Spotify I'm just going to listen to Suno and I don't really care about the prompt I'm not really viewing it as like a challenge or game or anything how many of them are just like I need to listen to music and Suno sounds good enough for me so I'm doing that versus the next bucket is I like the activity of coming up with creative ideas and figuring out, oh, I'm, you know, normally I would just be sitting here noodling on my, on my guitar or my bongo drums or something like that. And I enjoy the process of making, but I'm not, I'm not a, I'm not a, I'm not a full-time creator. I'm not trying to monetize this. I'm just having fun with the expression of like

actually making music, right? That's the second. That's going to piss some, some, some hardcore traditional musicians off. You're going to say, you're not making music, but objectively, if you're playing around with that you are making music. Yeah, yeah, yeah. And then the third bucket is like your job is to create content. You need music for your content. Maybe you understand that if you take a popular song, you take the latest Taylor Swift and you make it 1950s theme,

that will go viral and you will get a lot of impressions on Instagram, on TikTok, on YouTube, and you will just make money. Or you're producing a documentary. And you know what? You need some sort of background music. Eric startup making a launch video. Yeah, exactly, exactly. Historically, they would be like,

okay, we're going to use stock music. Are we going to use, like, copyrighted music and just wing it? And if it gets taken down at some point, you know, whatever, but there's tons of different. So there's those three buckets. And my question is, like, in the early

Instagram days, no one was going, no one was going to Instagram being like, yes, I'm a professional photographer. I did a wedding shoot, and I need to filter the photos. I'm going to use Instagram. They were like, no, I'm good with Photoshop. Photoshop's fine. And no one was going there being like, yeah, like I just want to look at something great. I'm not into this whole like watching friends on Netflix thing. I'd rather just scroll Instagram. No one was into that because

there was nothing there. And it was just like a couple filtered photos from your friends. You'd see like one or two updates a day in the early stage in the first few months. And so nearly all of the early Instagram usage was this middle case of coming for the tool. You want to just filter a photo on your phone. Instagram lets you do that. And I'm wondering for Suno, what's the, what's the percentage now? Are most of the people there because they want to be full-time content creators? They really need it as a professional tool. How many of them are viewing it just as like a video game essentially? And then what will it be over time? Will it be something where you've prompted it enough that when you get in your car or you get to the gym, you just say you open

Suno and you just say play because you've prompted it enough. And I feel like it's very much in that second bucket of video game like, it's fun, it's entertaining. And that's certainly where SORA sits right now. And the question is like, if you come for the tool, will you be staying for the tool? Will you just, will you just live in tool land forever? Or will you eventually, will this become a network. I mean, yeah, so my lens on it right now is that it is such a magic, it is still, even though we've had this technology for a couple years now, broadly available, it's one, it's gotten a lot better, and it's just so mad, undeniably magical to be able

to generate a song. I mean, I remember growing up, being into music, it was hilarious. And my dad is a musician, and my dad would be like, I'm going to make a song about the last week of school. And he'd play a song and he'd sing it and it would be entertaining for the whole family. And so to take something that's that entertaining and then condense it down into a prompt is like so magical and it reduces the just energy needed to create that magical experience down to 10 seconds that there is a massive amount of people that will just pay for that magic

experience today. They'll come and they'll maybe do one free prompt. I don't know how their pricing works, but then they'll actually willing to pay because they're going to make one for their grandma or make one for their kid or make one for their job or whatever. And at least there's a high volume of people that are happy to pay

to have that magical experience today. And so I would, my view is like, there's a lot of people that are, I'm sure, generating these and not even sharing them that broadly, right? They're just sharing them with friends, family, et cetera. We're just listening to themselves. Yeah, so that's the thing.

I don't, what I really don't, what I'm, what I, again, we should have, have the founder on, but I don't believe that these songs are for listening to yet. I feel like they're for entertainment, but not passive listening. I might be wrong, but that's my current, that's my current sense. Yeah, I, like looking on their website right now, it's make a country song about just being late, make a jazz song about watering my plants. I think, yeah, I mean, I, I think you're seeing something similar to, like, the dog of electronic music.

Like, when I was a kid, I was able to actually steal, but, or Torrent, a product called Fruity Loops, which was a digital audio workstation. So there was Ableton, and there were a few others of these, like, more advanced, what's pro tools. But Fruity Loops was very intuitive. like a really, really simple interface, and you didn't need a lot of extra gear to actually make music with it. Kind of like garage band-esque. Yeah.

And it basically, like, lowered the bar. Yeah, FL Studio, Joe knows. FL Studio, it lowered the bar because you used to need to go and buy every instrument. And then all of a sudden, it was like, you could pull a piano off of a menu bar and just say, I want a piano right now. Yeah.

And what does that do? Well, that lowers the... that lowers the barrier to entry. And so you get a lot more product, a lot more creativity. And obviously a big debate over whether digital instruments can hold a candle to traditional analog instruments. We kind of went through all of that and realized that, you know, the people who are going to go see marshmallow live do not care or the people that go and see, you know, anyone at Coachella or something. They don't, they don't, some of them care. some of the bands make their whole

shtick, well, we use real instruments to make this music or we still use real instruments or we have a throwback song and we went back and pulled real instruments off for this. There's a couple DJs that will be like, oh, for this next one, we have someone coming out and playing

like a real guitar or we have someone on the drums live with this. But overall, you're just dropping the cost to create and so more people are going to do it. It creates a steeper power loss so that you have more competitive than ever because the people that are truly talented are going to be able to rise up. There's no one, if you're truly talented,

there's never going to be the person who's really, really talented, but doesn't have the money to enter the market. So you're not keeping them out. And so all of a sudden those people are going to run up the score with cool creative things. The question is like, is yeah, does this become a more of a video game market where people are playing soon? like a game for themselves and maybe for a few friends like a fortnight like a like a like a like a Minecraft like you build in Minecraft you are creative in Minecraft but you don't often sell what you built like you're you don't often directly monetize yeah and so my my question becomes you know if I've seen these like reels go viral on Instagram going back to this like future as jazz

music yep or codeine crazy by future as in jazz yes and the question is yes that is like a hit like social media asset that is getting millions and millions of views and listens and things like that. The question is are people finding that song or making that song themselves and listening to it on a repeated basis? And my guess is as somebody who grew up listening

to Future, I love Future, I have no desire, I was entertained by that, but I have no desire to like go find the song and save it and put it and listen to it on the drive home, right? And the examples they give on the website of like make a song about Jess being late for work, that feels like more like a meme than something that's meant to be come back and listen to. Yep. Now, I can't imagine the use case as a parent of like generating a song, like the cleanup song and putting your kids names in it. And you make, and it's like, hey, every day after dinner, we're going to play this. And you're going to put, you know, whatever.

That's actually hilarious. I like that. That's very foolish. So that's something that I could come back to multiple times. But a lot of the use case make a country song about Jess being late. I'm sorry, that's like a funny moment and maybe you send it again when Jess is late. But it's not,

like nobody's listening to that. Yeah. Tyler, what do you think about this AI music nonsense, what these kids are up to T-Day? Yeah, I mean, I think current songs are like quite gimmicky. So it's like you have a country song, it's about AI data. centers in space or something. I saw that too. Oh, I saw that one, yeah. And it's like,

it's fun, but it's very much like a gimmick. It's not like, I'm not going to actually listen to this. Yeah, yeah. It always almost collapses down to just like, you could just tell me the prompt. And I'd be like, oh, yeah, that would be funny if there was a country song about AI data centers in space. I don't actually need to listen to the instantiation. Yeah, I think the quality bar for music is actually like quite high to get people to actually start listening. Totally. Yeah, could you, could you actually play the bongo drums while you talk? Well, I need to unmute. I can only play with one hand. So, so, because of this but oh you you have to press a button to be unmuted he's truly in the there we go okay we have

producer uh okay so i think this is like two art yeah but uh but um okay if you look at like short form uh like video they're only like 10 seconds long right so it's like very low i can't believe you guys got him like a dead man switch that he has to hold down i thought he had i thought he could just flip the switch and then the channel's open that's ridiculous he has a dead man switch for tyler Tyler, say something while holding up both hands. Oh, wait, you can't. Oh, no, he's muted. That's very funny.

What a quirk of our production set up. I want to continue, but let me tell you about cognition. They're the makers of Devin, the AI software engineer. Crush your backlog with your personal AI engineering team. So the reason that I think it's important to understand is Suno a network? Are you coming for the tool and staying for the network? Will it be a place that is it a competitor to Spotify long term? Is it a competitor of Fruitilups long term or Ableton long term?

Or is it a competitor of Fortnite long term? Is because of how resilient and what the market structure will be long term based on the actual usage. And so if it's like a video game and if people come there to do some sort of like self-expression, they're trying to get a high score, they're trying to like, They're trying to create something, and it's more of like this creative tool that they'll use elsewhere, that feels slightly less defensible than, okay, we're building out a proper network of, of, you know, connections between people that you won't want to leave. And so video games can be hugely profitable. They can be great. But the power law in video games is just a lot less steep than the power law in social networks.

Yeah. There is GTA. There is Red Dead Redemption and, you know, Madden and stuff, and they're Call of Duty. Like, there are some huge franchises, but there are way more, like, base hits in video games than in social networking. In social networking there, you have, like, you know, the meta properties, which are in the trillions, then you have, you know, YouTube, LinkedIn, X, TikTok, you know, all the smaller kind of mid-caps, but then you really don't have anything down in the lower end.

Whereas with a lot of video games, like you can just have like base hits. And so. Yeah. So one thing I've been thinking about is where does Suno need to go to maintain their current, I would say, dominance in the sort of AI music category? Yeah. And because we just figured out that Open AI is working on an AI music product.

Yes. And I would expect that Open AI eats the category of music as memes, which I'm talking about, which is like. think that. We had a funny thing happen at work and we're going to generate a song and we're going to share it in the team chat. I think that that use case for people that already subscribe to various language models that I'm sure will all over time add the ability to generate audio. And I think that Suno will still be fine, but they're going to have to really focus on other more integrated opportunities,

stuff in the enterprise, stuff in kind of pro-sumer, working with creators, et cetera. Actual musicians, too, is the other one where I'm sure when we get the founder on, I was just texting with Michael from Lightspeed. I'm sure there's a bunch of musicians, like real musicians that also use it. They're, I'm sure, a much smaller percentage

of the overall user base, but they just use it to get ideas. Yeah. Yeah. So if OpenAI builds like a great music generation API and it's, you know, similar to their coding API for GPT5, like it's just, it's just, you know, frontier level, maybe that gets baked into Ableton and garage band and other. I was just thinking for this category of what I'm trying to understand is what percentage

of songs are just jokes. Yep. And I think that Chad GPT will ultimately. do that pretty well. Yes. So, yeah, I'm trying to, I'm trying to think about the images in Chachapiti versus Mid Journey. Like, Mid Journey has been able to, because images in Chachapitee did not kill Mid Journey by any means. Like, the Mid Journey community is still going. They have something special there. They've created, like, their own special vibe and then sold it to meta, to create vibes.

Like, there's something there where it's not just frontier. It's like the specific UI, the specific community. what they want, like the choices that David Holes made, it's possible that Suno winds up in a similar situation, but when most people are in a flow of, you know, they're doing some research. Yeah, so what you're saying is Mid Journey basically created a genre, whereas it seems like Suno's product market fit is not creating entirely new styles,

but it's like make this country art, country music artists do a house song. It's style transfer versus genre. I think that might come over to chat GPT as well. I'm just thinking about the professional versus prosumer versus casual. Like when I think of the casual AI image generator person, they're not like I got to go to Mid Journey and get my S-Refs prop, like in shape. What they do is they just say like, oh, I'm already in ChatGBT, BT. Like, let me just see if ChachyB can do it.

And like even if Mid-Journey would, was a little bit better, the casual user is just like, yeah, chat GPT is good enough or Gemini is good enough for them. Like whatever their daily driver AI app is can just bolt that on. And so if you're, if you're daily driving in a chat GPT or a Gemini and you have an idea for an AI song, you're just going to do it there for the casuals. But then for the prosumers, for the people that are really into the actual community, like they're in mid-jurney, the Discord, getting ideas, sharing S.RFs, really pushing it to its limits. And I think you'll probably see the similar thing in Suno, where you have, where you have like this community

that really loves it. And like, a mid-journey creator will laugh at what comes out of images in chat GPT. And they will say, uh, it looks bland. It looks, it looks sloppy. Like ours looks better. And I imagine that that that's the way it'll play out too. But it's clearly just like an important feature to have because just like when you're, if you're developing a grocery list in Chachapiti and you want to say like, okay, generate an image of the grocery list so I can just print it nicely and give it a little design. Like it should just be able to do that. And you should also be able to say, generate a song of this. Like sing it to me and it should just be able to do that. Like that's what a multimodal model should do. So anyway, it's just interesting

that like the, we were coming out of the browser war yesterday or last week and there really is going to be a war in every single vertical here every company is going to do everything everyone's going well yeah opening i certainly going to open up a front of a new front in in every new war new war don't mind if i do yeah browser war there's going to be the commerce war there's going to be the uh certainly the coding agent war yeah and then now the music opening i never never met a war it didn't it wasn't interested fighting uh well let me tell you about figma dot com think bigger build fast Sir Figma helps design and development teams built great products together. You can get started for free.

Let's go. Well, people are really having fun in here. AI generated music from Suno V5 is now nearly indistinguishable from human-made songs. In blind tests, listeners guessed wrong as often as they guessed right. I wonder if images are at a similar level. Is music ahead of images? What do you think, so I saw someone, I forget exactly who it was, but they were doing a quote to it on this.

And what the paper actually says is people who have never heard AI music before couldn't tell. But people who have heard a lot of it can tell very easily. You see the same thing in text, where people who don't read a lot of AI-generated text have a very hard time distinguishing them. But like, you or I, we write a lot of... Clock it all the time. It's very easy to tell. So I think it's a similar thing there.

The first big AI music moment for me was people that know me, know I love God. Yes. And two years ago, somebody made an AI album for Gunna called Slime Pharaoh that was my, was the first moment where I was like, okay, if I was maybe distracted, somebody put this on and I was doing email or something else, I was busy, and they were like, hey, Gunna just released a new album, I'm playing, you know, like, listen to this. I wouldn't have immediately clocked it as AI. But that is like a distinct type of music that, you know, he's not always like, you know, enunciating words, you know, super, super, super precisely, right? Yeah, yeah. If you're, if the underlying inspiration is, is a little bit sloppy, then AI slop on top is just fine.

And you just has. Yeah, I, I, I, I'm actually kind of excited. I know we were, we were a little bit bearish on that YC company that, does AI generated launch videos. But I do like the idea of an AI agent that moves across all the different domains. So you come up with a concept, it can build out a script, and then it can build out a song, and then it can build out imagery for all of those and mash those all up and just like kind of puppeteer and let the different systems cook for like two hours to get you a result.

At the same time, it's really frustrating to let something cook for two hours and it comes back and it's like sort of a mess. I kind of did the manual version of that with Tyler, my personal agent. I had Tyler generate TBPN live streaming from every different historical innovation moment. So the steam engine, the light bulb, the discovery of DNA, and he did a lot of hard work on it. And it probably took, I don't know, an hour of rendering across all of them, something like that. What do you think? No, it was definitely faster than that.

Because you can, I actually used foul. Oh, you can run them in parallel. Okay, so you ran everything in parallel. Yeah. And it was like 16 videos. Yeah. It still took like 10 minutes maybe. Okay.

I have a song. Okay. Gunna F you mean, 1982. Soul Funk. 1982. AI cover. Okay, let's play this.

Let's play this. I think we won't get just demonetized because of this. That's a beauty. Oh, this is, okay. This is gonna. Pretty good. Production team. Everyone's jamming.

Everyone's jamming. Watch me jump right off the curve. Finley spur fly like a bird. Spinning first and the third. Solid I'm keeping my word. the supporting vocals in the back I haven't heard the original song I can't immediately clock the original song

All right, we can pause it But this is pretty good That's good I like that And I mean if we're going to be able to play That level of audio on the stream Without getting demonetized That's actually like a huge unlock

I feel like there's a kind of fun That we could have with it Yeah, it's interesting because It doesn't immediately clock as slop No in the same way and of course like a a producer probably can clock at a stop but to somebody that doesn't work in music isn't necessarily like I don't consider myself the biggest music fan yeah it's like at the very least like better than elevator music yeah yeah I mean it is it is just a simpler it feels simpler than getting video right like video if if there's six fingers you're going to notice but in music music, like if you're sticking to general chord progressions and you're sticking to a general key and, you know, not like completely, you know, slopping up the basic rules of how to generate music, you should be able to get something that sounds like at least reasonable. But I don't know. What do you think about this AI music stuff? Do you think you'll be listening to a lot of it in the future? I think there's something interesting on the consumption, like, streaming side almost, where

if you think of, like, there's, like, the Spotify DJ, and it'll kind of update a little bit if you, like, depending on how long you listen to the song. DJX, yeah, sometimes I'll skip, like, five in a row, and he'll be like, I get it, I get it. So it'll update. And basically, like, you can imagine it's, like, trying to find the next group of songs, whatever, in latent space, there's, like, some area where it thinks that, like, you might like that song, but then it turns out maybe there's actually no song in that point, and Layton Space or whatever. Then you can just generate it.

Interesting. So if you have like a generative playlist, I think that's kind of interesting. I feel like Spotify's got to do something here. That DJX is like two years old at this point. That was like an early, that's just recommendation algorithms. That's not like the AI boom. That's not generative AI. But at the same time, Spotify,

if they launched some generative AI thing, like a lot of the artists will probably be pretty upset, right? Well, yeah. I mean, it's like kicking the industry while it's down. It's like, hey, we know that streaming already doesn't generate like enough revenue at least in their view right a lot of the artists will post like i've seen artists like share their like payouts yep and it's and and it basically shows like you need to be constantly touring if you want to make money as an artist yep and so if you introduce AI versions of their music or or music that is just decent music yeah it's just

going to hammer their earnings even more. What should Spotify do with regard to AI music? I want your answer. Jordy, pitch me on a five or 10 year strategy. You're the CEO of Spotify for the next five or 10 years. What do you do in AI music? But while you're thinking about it, let me tell you about Vanta. Automate compliance, manage risk, improve trust continuously. Vantage's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation, whether you're pursuing your first framework or managing a complex program. Well said, John. My immediate thought, and I haven't thought this through.

I gave you an entire ad read to think through. You should be ready to step into the boardroom. Be a boardroom general. So the obvious, the thing that feels most obvious, and I'm sure there's some potential negative externalities to something like this, but if you just make it so that if I generate a song that's like an AI gun a song. that we just listened to if every person like if you can figure out a revenue split that basically takes care of the people that created that the track that inspired it yep in that case it's

extremely obvious because it's an existing gun a song that's converted into into a new song if you're just paying out effectively the same amount to the producer and the writers and all that stuff that feels like an easy solution yep now the problem is if you're just generating entirely net new music, which is like the prompt would be like make me a funk song, then you're still eating into the potential earnings of like real artists on the platform. And I don't know how you solve that. I mean, it's potentially a pull case for Spotify because they can take out. Why did Spotify like lean so heavily into podcasts? One, because people wanted to listen to podcasts. And two, because they didn't have the same dynamic where they're

paying out at such a large amount of their revenue to artists on the platform. So they could go out and pay to acquire, you know, Joe Rogan or whatever was originally for like a year or two years because you were acquiring users that would presumably listen to podcasts and you're not paying out the record label and the artists and all the underlying. That is a good point. Yeah, but yeah, the monetization has to be wildly different between. Yeah. So if I'm an art, if I'm, if I'm, If I'm gonna and somebody's generating AI gunna, I'm cool with that if I'm getting paid like that's a normal song. Yes.

But it starts to get really complicated when it's like generate a gunna song that sounds like this. And there's no underlying song. So how do the how do the royalties work on just like a generic kind of song? Yes. I don't know. I have a rebuttal. But first let me tell you about graphite. Dotter for the age of AI.

Graphite helps teams on GitHub ship higher quality software faster. So I know exactly how you do it. Tyler, you know how you do it. You take the song and you do the algorithm backwards and you basically reverse engineer the prompt from the final MP3. And so if someone went to one of these AI music generators and said, I want Taylor Swift singing a gunna song

when and then it generates that music, you can use the same, basically the same system to reverse engineer and just say, based on what this MP3 sounds like, what do you think the prompt was? And if you can figure out the correlation between the prompt and the actual output, you can figure out how to do attribution. And I think in general, Spotify, I would come out and say, hey, look, we're not going to get into the game of generating audio directly, but we are happy to let our creators use any tool that they want to generate music and come on our platform, whether that's a microphone and an acoustic guitar or Ableton with a bunch of digital drums or an AI

music generator like Suno. You're welcome on Spotify, but we're going to take 50% basically because we take our take our take rate, whatever the take rate is. YouTube's around 45, 55%. And then how are we going to How are we going to actually parcel out the royalty? Well, we're going to look at what does our AI system tell us about what the prompt was. And if it's if it's gunna singing Taylor Swift or Taylor Swift singing Gunna, well then they're both getting 50% of that new generation. A bunch of ideas from the chat. Taylor says the future is exclusive listening parties, zero streaming presence.

So artists just charge, not even for a concert, just show up in a room, $10,000 a pop. You can be one of a small number of people to ever hear the song, and then they just retire it. I like that. That's good. I think there's something there. Ryan says, I will cancel Spotify if they start pushing AI music. Pushing. What is pushing me?

That's the question. It's exactly what Tyler just outlined, right? They're DJing for you, and they're like, hey, we made this song for you. We thought you'd like it. So I think this is kind of potentially a stated preference versus revealed preference. I think that actual broad user feedback would be like, that's entertaining, that's kind of fun, that's cool. And I say that because people like doing this sooner stuff. If I'm scrolling my X feed and I see Harry Potter Balenciaga, which is AI generated, is X pushing that on me?

Or is that just a good MP4 that day? people liked it and so it rose at the top of the feed I would say the latter I would say that it's not like there was any but Spotify is historically

less algorithm right it's like we have this but they have this like basically billboard and the home page and we're just gonna put stuff there but they have an algorithm

and the question is is if there's I think there's a big difference between them actually choosing to generate music and then allowing anyone to come with generated music

but they're already doing that yeah yeah there are already A.V. says Spotify should launch a standalone audio SORA or vibes. That could be interesting. I feel like they're so any, any business that is reliant on Hollywood or the music industry is just so sensitive. Like it actually has this at scale business doing that. Yeah. Is so sensitive. And they're yeah. Spotify's in a way, way different place than open AI, which doesn't really have any Hollywood connections. And even YouTube, which YouTube's like the, vast majority of their ad dollars and creators like do not come from any sort of Hollywood organization, any sort of large organization. Spotify does really have like a few key

massive partnerships that they need to keep happy. John Philippine says Spotify buys soon now. I could actually see that. Well, uh, if you need to generate some media, head over to fall the world's best gender of image, video and audio models all in one place, develop and fine-tune models with serverless GPUs and on-demand clusters. Speaking of Taylor Swift, Travis Kelsey is here to save six flags. And the Wall Street Journal has a fantastic, look at that header image. Like, what is going on with that collage? Is that AI?

I don't know. It looks like some sort of illustration, but it's a lot of fun. Have you been tracking what Travis Kelsey is doing with six flags? Just this year, six flags, the stock has dropped from $50 a share essentially to almost $20 a share. So massive sell-off in the company. It is now popping on news that Travis Kelsey is investing. But the Travis Kelsey effect, it's up 22% over the last five days. Let's go.

The stars of the Super Bowl always celebrate in the same way. They bask in confetti, hoist the Lombardi trophy, and then proclaim where they plan to celebrate their epic victory. I'm going to Disney World. If Travis Kelsey wins a fourth Super Bowl this NFL season, there are hundreds of millions of reasons why he might pick a different amusement park. The Kansas City Chief's tight end just joined an activist investor campaign that's targeting six flags. Kelsey, New York-based Jana Partners and other investors have put around $200 million to buy roughly 9% of the theme park operator.

That means Kelsey is now part owner trying to sell America on stomach churning drops, head spinning turns and dip in dots, just what he loved as a kid and still does. On its surface, a 36-year-old football player teaming up with rabble-rousing investors might seem as unlikely as getting engaged to the world's most famous pop star after talking about her on a podcast. But as the two sides got to know one another, they realized the partnership was Kismet and a ripe opportunity to revitalize an American icon. Kelsey in recent years has developed a wide-ranging investment portfolio, mostly in business he's passionate about beer, games, and clothing. and he's often done it alongside institutional investors. Kelsey's business team had Jana on its list of potential partners. What would your turnaround plan for six-flex-flex-d? This is the classic investment strategy,

just buy the stocks of the companies that you spend money with, right? Do what you love. Help's just doing that. Make your find a job you love and you'll never work a day in your life. How would you turn around six-flags? How would you increase the revenues, increase the profit, increase the attendance? How would you make six flags? Great.

And before you answer, let me tell you about Julius.a.i. Julius, what analysis do you want to run chat with your data and get expert level insights in seconds? What's the first thing you're doing? Get on Julie. You actually probably should, that would be a good first step. I know it's a sponsor, but you actually should take all of six flags data. It's notable looking at NFL athletes, looking at their investments.

strategies. Kelsey, obviously, going the activism route. Sequan going the Fifth Avenue route. He has to buy Fifth Avenue. He likes co-investing with the founders' funds of the world, with the thrives of the world. It's interesting to see these two strategies play out in the market. I don't actually know that I've ever been to Six Flags, which is... So I don't feel... And I'm not a fan of theme parks. in general. So I don't feel like it's super qualified to, to, what are your ideas, though?

I, I feel like there is, there is potential for a comeback. I, I am, I was convinced by Brian Chesky when he came on and said, you know how you never. Long the real world. Yeah, long the real world. You know how you never see your phone and dreams. That hit, that hit hard. I know it sounds like something you see on a Pinterest live laugh love board, but it hit with me. I'm serious.

I've been thinking about this when I was looking at the sphere in Las Vegas. I thought the sphere in Las Vegas was awesome on day one. I thought it looked really cool. And then I sent, I got tickets for my wife and her friends to go see the backstreet boys there. They had a great time and I was further confirmed. I still haven't even been to it, but I'm still just like extreme. bullish on the sphere. And so I do feel like

what you probably need is more differentiation across different Six Flags theme parks. What makes the sphere so unique is that is the only sphere. They couldn't even build one in London. They tried. And the Londoner said, no, thanks.

And so I would be focused on, if you look at the roller coaster database, RCDB, this is a real thing, like IMDB, you can pull up which theme park has the tallest ride, Which one has the fastest ride? And if you're a real crazy theme park fan, you can go around and I want to ride the tallest one here, the fastest one here, the biggest wooden roller coaster, and the fastest steel roller coaster. There's all these different permutations and you can go collect all the badges, but it's not meaningful.

It's like five feet taller over there. I have two ideas for you. Please. So one, we talked about this a little bit last week, which is just raising the stakes, right? I think even as a kid, I understood that statistically, even though. know the roller coaster was going really fast, and it was really high, and then it was going to go really low, and then I was going to go upside down. Statistically, I knew the odds were in my favor, right? The odds that I was going to make it to the end in one piece, maybe a little sick

to my stomach, whatever, but I knew I was going to get to the end. Just I had a good feeling, you know, statistically. So raise the stakes, right? If it was like one out of a thousand rides just comes to like a devastating halt, right? Okay. Like upside down. And so you just don't know. am I in that batch, right? So that would be exhilarating, right? And we saw there was... There could be a mandatory chili eating contest. No, no, no.

That's my next idea. So America loves to gamble, right? And so what if they required every ride? You had to eat a bowl of chili before going on it. And then there was like a draft king's like betting booth nearby. So you could see the people walking onto the ride and you could bet every seat go long or short on if they were going to hurl at any point on the ride. I like that. So you could be, they could eat the bowl of chili.

Yes. They're walking out. They go on, you know, there's maybe like glass separating or something like that, but you can see them walking out and you're like, oh, that guy. That doesn't look too good. Oh, that chili's not sitting well because especially throughout the day, people have had more and more chili because every time they have to do a ride, they have to eat a bowl chili. And so you'd have a huge new influx of guests that aren't there to do the rides.

They're just there to campbell on will people make it to the end. So by the end of the day, it's just, it is just a mess out there. People are in hazmat suits. You're walking around, you know, goggles and everything. Because there's just, the whole park is just grow up going everywhere. I'm a firm believer in adding variable rewards to Six Flags. Generally, I would be very interested in when you get on the ride, there's a 1% chance that it goes 1% the speed.

And so you get on and you're like, yeah, you got unlucky and you're going to be on this for two hours now. And it's just going to be clinking along, just one inch at a time. And you're just like, yeah, I caught the raw into the stick. Yeah. I got the short stick. So another one, like lottery upsells for every ride. So we're basically just going to cure gambling right now. I'm just trying to juice their margins.

Yeah. You got people going. Yeah. How do we get more dollars out of every guest? Yes. And so I think, you know, if you have an option, you know, you're waiting in these long lines, right? If you have an option to buy, people walk around. with an iPad and say, like, would you like to do digital slots? Would you like to buy a lottery

ticket and kind of have fast feedback loops so that, you know, they can say, like, somebody in the crowd right now is going to win $1,000. You can do the numbers. You can run the numbers. There's, there's only a few hundred people online. Yep. You got a good, you got a, not a decent chance, but you got a chance. Yeah. And so I think it's just about getting that incremental dollar. I have a different idea. I think you're getting a little D-Gen. Let's let's go to Tyler. But first, let me tell you about turbopuffer, search every bite, serverless vector in full-text search, built from first principles and object storage, fast, 10x cheaper, and extremely scalable. Tyler, what's your idea?

I think I like the, you know, what weather than line doing stuff. So I think data labeling would be good. That's actually great. Yes. So it's kind of a scale AI route. I would take it a step further. You know there's like big, like massive land parties with just tons of monitors. Like the JP Morgan, I would make the JP Morgan experience with a bunch of quad monitors.

And you could go to Six Flags and you just lock in and they just have Google AI Studio cooking on every monitor and you just go to the, you get in the line to go vibe code, you chat with models, you vibe code, you monitor usage. I think that bringing in the corporate big tech dollars, either through branded integrations, rides, so you can ride the Google AI Studio coaster.

Another thing, that could be a way to upgrade. Another opportunity, wellness has, has been booming, right, while theme parks have been, uh, have been in relative decline. How do you get? And so, um, so the obvious thing is, is, you know, people have been promoting, uh, that intravenous use of like ketamine, right? So if you had ketamine mental health stations throughout the, what about facelifts?

Did you see this article in the Wall Street Journal? Why tech bros are getting facelifts now? Yeah, so it needs to turn into a wellness destination. you can get a facelift. I'm not familiar with this at all, this whole story, but people were having time. To pull up this picture, I think they're trying, I mean. Whoever this dude is is getting Dr. Ben Talley.

He's a 60-year-old who runs a solar and tech business. He had an advanced deep plane facelift and necklift with SMAS optimization as well as other procedures. Well, I mean, he looks. I mean, he looks, he's looking great. I think this is a, this looks like a good ad for facelifts, if you ask me. Yeah. Well, Keith Rboy, friend of the show, got a, got a nod in this article. Really? Tech is a young person's game. In a 2024 talk, Keith Rubei recounted advice his fellow investor Peter Thiel once gave him. You can't hire anyone over 30. It's unsurprising then that men in

tech are increasingly spending thousands of dollars on procedures such as mini face lifts, necklifts, and eyelid lifts to beat the signs of aging according to plastic surgeons. Yes, the latest addition to the tech bro look is a brand new face. Tyler, maybe you should go 10 years younger. Maybe you should try and look like a five-year-old. Just maybe you should shave your head and get like that one curly cute, like the baby, you know, the baby with a single, single hair. I need the cartoon baby.

Yeah, the cartoon, cartoon baby. That's the way to. Yeah, you would have to get a much, you could get a phase lift, but then put it, make your face a lot wider. Yeah. Like, round it out. You need to get on some sort of diet that gets the baby fat back on instead of working. Oh, that's just filler.

Tyler just gets so much filler. Oh, you just have a baby face. That'd be ridiculous. I'm amazed that people were quick to volunteer, volunteer their own personal stories for this article. Who did? But I guess, I mean, all the people in the photos. Yeah, but are these people people know in tech?

I don't know, but they're tech on the nerves. If you scroll down. I mean, you need to ID these people. Oh, unnamed patient. Men used to wait in 2016. But they're saying it's a 46-year-old tech executive. Tyler, figure out every 46-year-old tech. Who this is.

Anyways, I think it's cool. I don't have a problem with it. Yeah. Well, if you're trying to do facelifts, you need to get your brand mentioned on chat chabit. That's right. Get your plastic surgery firm mentioned it on chat chabit with profound. Reach millions of consumers who use AI to discover new products and brands.

The White House has posted that the console wars are over. They shared an image of Donald Trump donning master chiefs battle arts. armor with the plasma sword. I'm really not that sharp on Halo terminology. But it is big news because apparently Halo is coming to PS5. Is that correct? Who's familiar with this? Break this down for me.

Are the console wars over? It does feel sort of over. I think this was a quote on GameStop's post where they said, you know, we're announcing the console wars are over. yeah um the halo is coming to playstation that's exciting halo's coming to playstation it does feel like the the console wars are over in the sense that um the the two companies have just diverged so thoroughly in strategy where you know Microsoft owns Xbox but is very much like console agnostic and Sony is very much like all in on PS5 as like it the core strategy that's like it the core strategy

and that's like, I don't know, they're still duking it out a little bit, but it feels very much like PS5 is like a real legitimate console. And then Microsoft and Xbox and Xbox Cloud are like a strategy. And they would be okay with you playing Xbox games on your PS5 because they feel like they are definitely pulling back from like the hardware wars. What do you think? I was just going to say it's pretty funny that Ryan Cohen is so anti, like violent video game. It's wild.

I mean, like all the old images of like GameStop right the day before the new Call of Duty comes out. It's like, you know, it must be, I mean, I would assume it was a big part of the revenue, but maybe not anymore. Who knows? I wish, I miss when Q4

meant that there was a, all I was thinking about was a new cod. That was a simpler time. I played a little bit of the new battlefield. It was not anything super special. Yeah, I think you got to set your whole life up to really dominate cod plus i feel like i got to have mouse and keyboard and it's just not in the cards for me yeah i can't i can't do the controller stuff seriously it's just too

i don't know it's just not for me um anyway the the console wars um i was i was trying to get up to speed on like how do you tell the story of the of the last 30 years of the console wars uh going back to the atari then the nes arrives in the united states in the in 85 89 Sega Genesis in 1989, Super NES, then PlayStation and N64, go back and forth. The Sega Dreamcast comes in,

apparently one of the earlier online playable consoles. DVD playback of movies, moves units into living rooms. People start buying consoles because, oh, you can also play DVDs. And then eventually, you can also play HD DVDs and Blu-Rays.

There was a battle between HD DVD and Blu-Rae. anyway we'll see how much it's over it feels it feels pretty over um but they're not to be clear they're not going to allow like cross-platform live play between playstation and xbox oh they definitely do that in cod oh they do all right for sure yeah is there any advantage to playing on one or another like as like the hardcore players do they do they pick aside uh i mean the i believe the hardest score players always pick mouse and keyboard but they might okay but hardcore so so PC would be the most hardcore got it because the reaction time i mean i'm sure that there's plenty of people to debate this but uh in general that the the the muscle like because if you're

trying to pick click on a head like moving your arm yeah you playing on a pc would smoke you playing on a console usually usually like there are there are clearly some people who are phenomenal with the sticks but in general this much movement is going to be more accurate than this much movement because this is just not that much it doesn't it doesn't leave that much for like precision now some of the games have worked to crude to kind of level the playing field by having different levels of auto aim exactly different levels of auto aim and so you're in this weird realm where if you play on consoles you do get it's not auto aim a console player would be very upset with you calling it auto aim

because what it actually is is that when you're over a target, it slows down the speed at which the stick moves the cursory. Exactly. And so it's not snapping to the target. It's just letting you slow down. Using auto aim with this laser pointer on the team. Don't shine. I don't put it in people's eye.

I don't put the laser pointer. You don't know where you're going with that. It's risky. It's risky. is. Should we talk about take him on why AI is underhyped

and isn't a bubble yet? Sure. Let's do it. Take him. Whatever happened, Tay was supposed to be on Friday? Yeah,

but it was like a 2 p.m. last minute booking and it wasn't, there wasn't anything in the news right then, so we're going to have them on later. I wanted to have them on the show for a long time.

In the LinkedIn chat says, I'll take it that Jordy wasn't exactly a trick-shoter. Not on, I've had my, I've, I've, I like that Joseph is like, as a PC player, it is auto aim.

I agree. It does feel like there's a big difference between the computer is you're setting your sensitivity once and then you go into the game and you interact with the game and there's nothing that's like context aware. As soon as you add something that's context aware, it feels like that's auto aim and that's cheating and that would certainly be an advantage. And it is designed to be an advantage because you have a lot of disadvantages when you're using a console with a controller that has small stakes.

So, Take Kim says, here's why AI isn't a bubble today. He's distilled data points and ideas from previous columns and coverage. If you prefer facts and evidence-based reality over vibes and conflated narratives, enjoy. Big tech valuations are reasonable, leverage is low. We're at the beginning of multiple AI super product cycles in the year ahead. We are in the early innings of a technology computing shift to AI, the largest in decades. think 1994 versus 1999. Yeah, that's a good question.

When does the dot-com bubble start? Everyone says like 99 to 2000, but was it not a bubble in 94? Because like it can be a bubble, like even big bubbles have small beginnings, right? Like they start as small bubbles and then they get bigger and bigger as they inflate. Remember, Amazon went public May 15th, 1997.

So, yeah. And so you could easily claim that we're in, we're in the very early stages of a bubble. My personal timeline was that, like, the further you went in the 90s, the crazier it got. Like, it was, like, heating up. Yep. And there was plenty of people that made insane amounts of money, like, prior to the bubble, right? Totally.

And a lot of them probably wish they held a bit longer, but... No, no. I think a lot of them were happy that they got out when they did. Well, sure, but, but again, like, a lot of, like, the euphoria... I'm talking about, like, I've talked to some folks who are, like, they founded a company in 1994, they sold in 1998, it does not exist, it didn't exist, it does not exist anymore, it didn't exist in 2002, but they wound up with like $100 million liquid, and they're like, set for life. Yeah. Because they like nailed that. That's enough to retire? Yeah.

I'm kidding. I always, I always, I feel like the debate comes up, like, you still have to work as Metro Capital. Is 10 million enough to retire? You still have to. You still have to You still have to raise a VC fund and LARP is VC for a little bit. Continue. Well, we actually have our first guest to the show. Okay, we can come back. Let's go to space. Let's go to space data centers.

Our first guest of the show is Philip from Star Cloud. Burning up the timeline, Philip, how are you doing? Good to see you. The current thing himself. The current thing himself. Did this all kick off with Nvidia posting about you? Was that when the kerfuffle began? I think that was the kind of.

I think that was the catalyst, yeah. Okay, so thank so much for having me on, by the way. Of course. I've been a few fans in big ones. I'm glad to have you. This is why the show exists. We address every timeline that is in turmoil. So break it down for me.

When did you start the company? What's the pitch? And then we'll go into some of the like, how are the haters wrong? Yeah, yeah. But just give me like the recent back story. Yeah, so we started about a year and a half ago at the beginning of last year. we basically in that time built a satellite

we've got the first launch coming up in a week and that's why Nvidia just posted their first thing because we're going to be launching the first H100 to space about a hundred times more powerful GPU compute than there's been a space before. Okay. And yeah, that's the sort of brief history. What were you doing before this?

So I started my career as an engineer for five years, started applied math and theoretic physics undergrad and masters and then I spent a few of the McKinsey working with the space agencies of the UA and Saudi government. And then I found it and sold another company not related to this and then moved my way into this. There you are. Harvard, Wharton, Columbia. He's been to every academic institution.

Collecting them all. What is the strategy for this first launch, this first satellite? Like what are you trying to figure out with this launch? What is the value of putting this H-100 into space? And then kind of like, how do you think about the timeline? because I think like the I would say like overarching criticism of data centers in space has been like people I don't think people generally believe that this will never happen it's more of like a debate around like the timeline on which it happens yeah yeah which is a fair I mean it's a fair debate to have I think so the first satellite the reason we're launching this lots of people believe that you can't run terrestrial data center grade GPUs in space because of high high radiation environment and you need to dissipate the heat in a vacuum. So this is really a demonstrator.

It's a 60 kilogram small set. It's launching on a Falcon 9, a quarter plate there. And so we'll be running high pad. We'll do a whole bunch of first. We'll be the first to run high-powered inference in space on imagery from working with various DOD customers. We'll announce that soon. We'll be the first to train in modeling space.

We're training mini GPT from Carpathie. We'll be the first to do, we're running a version of Gemini called Gemma, which is like a cut-down version. working with Google Cloud on that. So we'll do some fun demos. Like, for example, you can SSH into it and then just ask it. Things like, where are you now and how does it feel to be a satellite? And it will give you a kind of coherent answer.

But yeah, this is really a demo. The next satellite launching in October next year is 10 times of power will have by far the largest commercial radiator in space, a 30-meter wingspan, solar panel. And then we'll have very, very sophisticated optical and connectivity options on there. So customers, including DOD and other satellites, will be able to send us raw imagery. We process it on the edge, and then we just down the insight.

And that's the roadmap. How far are you, like, there's a launch that's happening very soon. And then there's the renders, which are like, it looks like sci-fi Star Wars. Like, oh, yes, there's like 25 shipping containers that are descending. Like, what's the delta there? And what was the decision between, like, kind of showing more of a vision document?

because I think that's what triggered a lot of people was like you were showing an amount of mass in orbit that feels like beyond the ISS like it felt feel it looks huge right yeah yeah no to be fair it's it's extremely difficult to imagine what's about to come down the line with starship and so like the launch cost might come down by between 10 and 100 x launch capacity is set to go up by a thousand X or more and the reason is if you build a new starship every day for a year which is like a very conservative investment for what they're trying to do. At the end of the year, you have 365 starships. It has five times payload capacity. You can launch 10 times as frequently. So we're really talking about a thousand times like more tarnish per year that we can get to space. And that's coming like three to five years, which is really not that far away. So the idea of the render, like you can launch the whole ISS in two starship launches. So the render is to show, look, there is going to be a lot more mass in space. But yeah, I mean, I almost feel like you should have done the Andrel thing where it's like anime if it's like sci-fi where the vision's

going and then it's like render for like what you're actually sending because like the render that I saw of like the multiple shipping containers putting together like that's not what's actually launching next week so I think that's where like a lot of the cognitive dissonance is coming from right but yeah I mean you'll iterate all this it's fine yeah I recently still comment talking about that and I was like man that's pretty cool yeah yeah it's nice to have because there is a there is a value for like the vision document like what do you where do you want to go in 10 years but it just hits different when it's like the official nVIDia account is sharing this render and it's like okay like that feels like it's happening soon it's not like they don't do their

research like they have a whole team of like PhDs that go and look at this shit and like they have they have a board that approves that this can get shown that it's like feasibly possible yes so talk about like yeah i guess i think it's a science problem yeah i guess i think it's about so this first launch is meant to be like a demo basically like correct me if I'm wrong but effectively a science project to start to show capabilities in space to start to prove that you guys can actually put GPUs in space that can run workloads like what is the actual timeline so and then it sounds like maybe defense is like an early place where there might be some commercial applications but then what is what's your timeline on on when uh if or when you know

i would be you know prompting a model like chat tvt or jemini or whatever and and uh it's like it's just cheaper to do it in space because that that that's the long-term vision right is like it's just cheaper it's not anything special or like oh we need to be there for some geopolitical or defense reasons it's just like the energy is free and the heat dissipation is free, so it's just cheaper, so that's where you go, even though the launch costs are expensive. Yes, yeah. So maybe I'll start with the second part. So when will it be cheaper, that's coming with Starship. And so we're looking at the first commercial launches of Starship in 18 months, but we're probably going to be seeing, you know, mass volume production of Starship on a three to

five-year time frame. So, yeah, the roadmap is next year in October we're launching our second satellite. That's the one. It's about a 10 times the computer of the first. one, but much more capable in terms of continuous power generation and solar panels. That one is the first commercial self-service. That will produce more cash than it costs to design, build, and launch. I mean, that satellite pays for itself just on hosted payloads, because that satellite has such high throughput and processing capabilities, there's a whole bunch of people that just want to stick sensors on that.

It's not the end-state business model, but that's a nice revenue stream in the meantime. time. Then we're launching a satellite after that, which will be, I mean, yeah, I can't talk too much about it, but you can see this Pez dispenser form factor that's coming out of Starship. Oh, yeah. Yeah, I probably shouldn't say too much. And that was successfully tested on the last Starship launch, right? Where there was a whole bunch of like debris blowing around. It was a crazy, crazy launch, but it did come back. It landed and the Pez dispenser thing shot out the the fake Starlink or like the dummy. Yeah, exactly. So that's about a four ton, at least 100 kilowatt satellite. The second one's 10 kilowatts. First one's one's one kilowatts. So we're like

going up in order magnitude each time. That one could feasibly launch end of 2027. And then from there, we basically will launch lots of small versions of, or lots of continuous versions of that, because that will have very low launch cost. And it will be a while before we start launching the full Starship payload base size. I mean, for one, we need this clamshell door to be ready before. we can do that. Yeah, but to be frank, it's quite a while until we start docking things in space. The only reason you would dock things in space

like in the concert video is if you want to train a large model because then you need the whole data sensor to be internally networked. And so that we're probably looking at, you know, early 2030s before we start docking. I mean, I don't have a problem with thinking in decades. Like that's like what we say in tech we want people to do. And so I don't have any, I don't think there's any problem with saying

like, yeah, like there's some crazy stuff that needs to happen with launch costs and, like, it's going to take a while until this really works out. Like, that's actually exactly what people want from, from technology founders, ideally. I guess there is the question of like, you know, how do you frame it as like how you don't want people to think. Yeah, if you just needed to generate a lot of revenue quickly in space, you could do the first space casino where people could beam up, tap into it, all ages, all ages. All ages. People, they hit me up. Oh, really? Yeah, I mean, I guess space, space token. You got a bunch of, What, how do you think about, you know, some of the, we, we, we'd ask Dylan Patel at one point about GPUs and space. And I don't think he mentioned, like, the challenges around heat and heat dissipation, but he was talking about how sometimes you just need to, like, unplug the server and plug it back in again or switch out individual GPUs. But, like, what's the solve on that?

that long term, are you thinking about, you know, these being like fully robotic systems that allow you to make changes on the fly? Like, what's the solution there? Yeah, I think probably two points on that. I mean, the first is for these massive data centers where you've got like, you know, a kilometer long or a mile long row of racks, when the people I speak to you now are saying, look, if an H100 breaks in that, you just leave it in there. Nobody's going down there and unplugging and putting that back in there, just writing software around it, and they'll just legal, basically. So that's number one. But secondly, for the first few iterations of the satellite, it will be exactly like Starlink, and we won't be able to touch it. And so we'll have

to have redundancy on the critical systems, and then we over-provision things like solar panels and radiators, which degrade over time. Over time, the whole industry is, well, space industry in general is moving towards robotic maintenance, but also terrestrial data center industry is moving towards robotic maintenance. And that's where on a sort of five-year time frame, you know, some of the people we speak to in data centers are saying, we are fully expecting data centers to be maintained by humanoid robots in a five-year time frame. And, you know, this is where it starts to sound a little bit wacky and sci-fi. But, I mean, humanoid robots do not require too much modification to work in space in the

sense that a human space who would take care of radiation and thermals, you would need to retrain it slightly. But, you know, that maybe sounds a little bit wacky. Well, yeah, it doesn't have to be a humanoid, right? Like, there's a bunch of other forms factors. Yeah, it could be any form factor, but humanides will be mass produced. so probably the cheapest. Hmm.

That's interesting. I want you to react to Andrew McAllip's viral post. He gets 7,000 likes, breaking down a bunch of different things. The core claims I see in here are questions about what if demand for payload to orbit doesn't decline. He does talk about the difficulties of heat dissipation. He says, you just finished watching a Scott Manley video. on radiative heat transfer, and now you think you're going to disrupt AWS with a few solar

panels and a ride chair slot. You're going to believe that right up until the next month when you crack open DeWitt. And he goes into a lot of the technical level. But I was wondering if there were any key points in here that you think were easily debunked from actually being in the trenches. You know, obviously he's, you know, he does work in space and does something somewhat similar and does seem like an expert who'd be equipped to know this stuff, but you've obviously focused on this case specifically. So what was your reaction when you saw his sort of comedic breakdown? Yeah, I mean, I like the person.

I mean, I'm a fan of Andrew. I've been following his stuff and he's a great engineer. Same with Kellyan, by the way. I'm very impressive what they're doing of Arroy. Yeah, totally. Yeah. So, I mean, on the thermos, the criticism usually is, in order to dissipate that heat,

you need a large surface area, and they think for some reason that that's super impractical. I mean, my co-founder, Ezra has a PhD in engineering, spent 10 years designing and building large deployable structures, solar panels and radiators. And, I mean, you just have to build a large surface area, and that's what we're doing. So half our engineering team is building a very large,

low-cost and low-mass deployable radiator. So that is the core IP of our company is that. So that's number one. On the launch cost side of things, so sometimes people say, if there's no competitors to Starship, which it doesn't look like there will be for at least five years, then will there be any incentive?

for Starship to, for SpaceX to drop the pricing. Now, I have a bit of a different opinion than most people. Most people say, well, no, they'll just keep it high. I don't think so, and it's the same reason that they keep, that the Tesla is sold at under the price. They could, they could potentially sell Tesla at a higher price, but in order to reach a much larger market, you need to sell at a lower price. So, for example, they're going to have all of these starships sitting there.

They can only launch once every two years to Mars. Their marginal cost of launch on each Starship, there's two or three million dollars. Now, either they can leave it there because there's no demand at the, you know, $200 million for launch price, or they can say to us people like us, hey, you can launch it for $20, $30 million. They still make sure a lot of money. That's still very profitable for us. And so it's like a win-win. So my expectation is that actually the launch price will come down even if they don't have a huge number of competitors. Yeah, yeah, like the elasticity of demand there.

Like they might just be able to sell more volume at lower prices, so they might wind up lowering the cost. Or, yeah, kind of giving you like a bulk discount, if you want to put it in that terms. Interesting. On the heat dissipation question, how much do you think that that is in the domain of science? And we need to have a breakthrough discovery. You need to discover the solution to that problem versus an engineering problem where maybe it just comes down to, can you make, manufacture something that is already follows the laws of physics and all the physicists agree it works if you can just build it profitably. Yeah, I mean, from what we're seeing, it's very much an engineering challenge. It's

really like a manufacturing. So we know that it works because the International Space Station has a radiator that dissipates 70 kilowatts. The problem is it's ridiculously expensive. So our radiator on our second satellite is 10 times lower mass per watt of dissipation and 100 times lower price per watt of dissipation than the radiator on the International Space station. That's the whole game is making this radio to cheap and light. And yeah, we've got some very innovative

or very unusual manufacturing techniques that we are working on basically. Fun. You posted a few days ago, people don't understand Starlink is going to just be the internet direct to sell is going to hit hard. Basically all telco

is doomed. Let's check in this time next year. Expand on that. That was at 1 a.m. I had a very heated argument with somebody. I was in the cabin. I was like, I'm just going to post this. And then like 300,000 views later. I was like, okay, this was probably like a little bit too aggressive. Well, yeah, obviously, if you want to get attention on acts, just say the most inflammatory thing possible, say that, you know, all these multi-hundred. No, but, but what's what's your view on how quickly, you know, we saw news on Apple, boring something with a Starlink on direct-to-sell. It seems like every major manufacturer is going to be interested in this. And then there's also that other company that that's like a competitor, right? That what is it?

That is Verizon propping them up? They don't have any satellite. A-S-T-S. Yeah, yeah. No, I don't think they have any hope and hell. And the reason they don't is, sorry. There's a whole bunch of people that have. they have a retail

the retail traders are going to come for you dude good luck what do you know about space you got you you got the private markets they're doing fine you got the private markets Varda guys coming for you now you got the retail

traders and ASTS coming for you you got no friends in this foxhole dude you gotta get out of it AS ASTS is up 7% just today oh wow that's amazing I mean okay I want them to succeed and I'm like

oh okay well in that case I reverse my entire position. Seven percent move. Let's go. That's very funny. Yeah, I don't know. It's a launch cost thing. It's a launch cost thing.

They can't compete with Starlink on the launch cost. So, yeah. Yeah. I mean, that argument can be leveled against us, by the way, as well. So on what are the,

what are the telco companies going to do because they're multi-100 billion dollar companies. They're not just going to roll over. Is it is like, how do you? you see them kind of reacting? I mean, it feels like there's lots of multi-hundred billion dollar companies that just get

backs into a corner and die. I mean, blockbuster is a hundred billion dollar company. You are such a space, Maxie. Well, well, no, I guess might be right over, like, you know, it just depends on the time frame, but, like, I mean, it just doesn't seem like, like, internet backhaul is not going to move to Starlink in the next, like, five years, you know. This is like not going to happen. Like, Akamai.

That would be honestly cable for sure. That would be honestly cable. Yeah. The things that people are underappreciating is the amount of capacity they can launch per starship is absolutely mind-blowing. And that's coming in 18 months. So like, and that's like proper hardcore, you know, very high bandwidth direct-to-cell in buildings. In buildings. That's what people don't appreciate it.

In buildings. That's pretty crazy. Yeah. Yeah. Yeah. Yeah. I mean, I, I love a starling. I guess they, generally.

the telecom companies will could end up being like Starlink resellers yeah right is that is that one scenario I mean they could be yeah doesn't sound like the best business to be in but yeah yeah anyway uh thank you so much for hopping on and defending your honor I think you did a great job defending your honor I think you're thinking in decades I'm excited for you to continue working on this how you're you're in the Washington area right yes Washington state how how how how how how What's the scene like in Redmond before you go? Good? How's the energy? I mean, there's a lot of, it's where Kuiper and Stalin cars are all of the people. I think 95% of satellites launched in the last two years were designed in Boat in Redmond. So there's a lot of space town out there.

Yeah, it's not the liveliest of cities in the world. Well, you guys have work to do. So thank you. Thank you so much for joining. Fun conversation. And we're rooting for you. I really appreciate. We'll talk to you, too. Let me tell you about linear. Linear is a perfect. purpose-built tool for planning and building products. Need the system for modern software development. Streamline issues, projects, and product roadmaps. You can start building.

Do you want to go into Bub Talk? Are you still interested in talking bubbles? Or do you want to move on to something else? We can move on to something else. There's a bunch of Bub Talk in here. Maybe more importantly. The maker of Oreos. We need it hammer Bub Talk more.

I mean, it's been a lot of bubble talk lately. But I feel like we've hit it a bunch. There's a bunch of charts. Yeah. If you're trying to figure out if we're in a bubble or not, I recommend going outside, picking a daisy and just picking the petals off. That does feel like where we are right now. It's a bubble.

It's not. It's not. It's not. And then whatever you end up landing on, that's what's going to happen. So good luck. And so in Oreo world, Mondalaz, the owner of Oreo has trained their own video model for television advertising. They invested $40 million.

And they say it cuts production costs by 30 to 50%. I don't know how that's possible. Like, it should cut costs by 99%. Like, do you know what it takes to do a video shoot? But I still think there's a huge amount of budget goes to coming up with good ideas. Sure. And then the editing and then making infinite variations of it, which I'm sure AI can already do well in some cases.

so um but i still think you pay you pay people to come up with great ideas and then execute but the the idea is is like the actually the highest margin part right and so the ad some of the ad agencies i think will do well um apparently the oreo's parent company only spent a hundred million on digital print and national tv advertising in the last year wait so they spent a hundred million on advertising and then they spent 40 million on the top okay What's your take on Oreo Net? So this was my idea when we were talking about SORA, where I said that Coke should, or Coke Cola should release a video model open source, where you can do anything,

but every single video has a Coke. Oh, yeah, yeah, yeah. So they should release this model, make it open source, but there's just Oreos everywhere in the video. You just, like, can't do anything else. You ask you to, like, make a picture of your dog, and your dog just is an Oreo for a head or something like that. It just gets crazy. Yeah.

Yeah, that would be, that's a brilliant idea that would be abused so badly that it would be shut down with 24 hours. I mean, does this math make any sense? Like, where did the 40 million go? Like, what were they, were they fine-tuning? What is it? That 40 million is ours now. That is ours now.

Yeah. I'm not sure it was like, I really want to know, I really want to know the breakdown of that. Was it actually like running the fine-tuning? Was it generating, you know, training data? Like, what was the actual project that got them there? I mean, I could imagine, you know. it was it was uh somebody took the 40 million dollars and they made it an app layer on top of sora too yeah also if they have a whole bunch of footage of oreos throughout the 30 years that they've been

advertising and doing video ads maybe 50 years uh and they load those in can they count that as like depreciation i it's hard not to see this being like a huge waste of money that somebody somebody high up said, like, I'm going to be the AI guy at Mondalez, right? Because you have to imagine whatever they were able to do today by investing $40 million, $40% of their ad. So again, like production cost, which is where they're saving money, is not advertising budget. I would look at the way they're talking about their ad budget as like what they spent on ads plus the production cost. Like that's the ad budget. It's very possible that the production cost was like 10% of that. Yeah. So you spent 40%

percent of your overall advertising budget to slash 10, potentially 10 percent of your budget by 30 to 50 percent, meaning that the payback. Yeah, I mean, it's a, it's a, it's a, it's not the craziest thing. I don't know. Yeah, but, but the models are going to, the models will just be one-shotting. I promise you they will be one-shotting people eating Oreos better than your model, your custom model by now. Yeah. Like, by the time they finish, by the time they finished it's it's like they get on sore two and they do it and they're like oh no why do we fine tune like some you know oh they did it with Accenture I called it no way I knew I knew it I knew it I knew this was just like yeah like big consulting you know big strategy firm says yeah we're

we're happy to give you if you want it what's your budget okay 40 million dollars we're happy to give you an AI play here you're going to be you're going to be you You might just be the open AI of food. You might become, we can help you be the anthropic of cookies. Of cookies. And then, and then it's just immediately a write-off. It is so crazy that they created their own model. How did this leak?

It's so bad. I think, no, I think they're bragging, John. They're bragging? They're bragging. It's like, why are they telling on themselves? Why are they telling them? They're not, John. They're bragging.

Wait. Yeah, they said, The, yeah, the head of the global senior vice president of consumer experience said they're going to use it in the 2027 Super Bowl. I don't know. Is there any reason why you'd want a three-year-old fine tune? You could have just run eight Super Bowl ads. You could have run eight. That is actually crazy.

That's actually an under. This is so inspiring. If you want to make money in the AI boom, go and get $40 million for. a project like this. It's genius. Like, you just made so much money. Yeah.

Coup says, sounds bubbly to me. Totally agree. There's really so much money. Nobody has applied. So the friend.com billboard strategy is potentially a new strategy. Just like buy so much advertising

that people can't stop talking about it. Why has nobody done that for the Super Bowl? Like, has anybody bought, like if you really wanted to break through with the Super Bowl, would buy 10 ads or you'd buy, I mean, friend tried to buy every single ad. Can you imagine doing a whole Super Bowl buyout? It's just, it's just four. How many minutes of

Super Bowl ads are there? Could you do 40 minutes of ads? The, we, we, we, we, we, we can't get away from the, from obvious billboards. We drove by today, uh, this one that just says, uh, when you're on the street, it just says, end, end dot com. Um, is at the end. But yeah, I would have, I definitely think that Oreos just being like, we want to be drilled into the mind of every American. We're going to buy 20 Super Bowl ads this year. That's our budget for the year. Yeah. I mean, there is like some sort of bowl case where it's like you need to generate a whole bunch of Oreo creative across Facebook and, you know, Instagram ads and whatever. But at the same time, it's like, I don't know why you need a custom model. The $40 million custom model is crazy. You have to assume that every.

model is training on every Oreo ad ever in history. Am I crazy here? Is there any reason why you'd want to fine tune for $40 million? I would go farther. I would train a language model. Oh, yes, from the ground out. Yeah, because then you have the... Do a full pre-train.

Do a pre-train. Do a pre-trained. Do a pre-trained. Do a full-based model. Yep. You know, what's it going to score on? You mean, he's last exam. What's it's going to score on? And let's let's assume I'm a Fortune 500 buyer and you're an Accenture consultant. How much is that going to cost me, Tyler?

a couple billion. And you're telling me that that's worth it, right? Of course, yeah. We've got to get this guy a job in Accenture. He's going to 100 X revenues. Yeah, I mean, just to close it out, this is publicist group, French multinational advertising company, one of the largest advertising

agency conglomerates in the world, and Accenture serving up a farm-to-table AI strategy for an executive at Oreo who wants to be able to say we're not just using generative AI we are generative AI we are the future of food

I hope they need a vertically they need to vertically integrate they need their own data center they need their own chips they need to bake this Oreo algorithm into an ASIC they need to go to Broadcom they need to go to TSMC they need line time at TSMC

because I mean it's possible that they need to go to ASML and get an entirely new lithography machine. They need to get to rare earth. They need to get into rare earth. They need to go all the way down to... Maybe acquire a desert just to get enough sand. They need to teach sand to make Oreo hats.

That's what they need to do. They need to acquire sand. Brutal. Best of luck out there, Oreo. Best of luck. We're rooting for you. Well, you know what Oreo should do. They want to use AI.

They got to use numerally. H.Q.com. Put your sales tax on autopilot. Spend less than five minutes per month on sales tax compliance. If you're selling Oreos online, you've got to get numeralhq.com. Lulu says it's the year 2025. You wake up in a progressive white billionaire with a blackface NFT profile picture is saying GM to you, saying wag me, friend. This is a crazy choice. Why did he pick this one? This is like the how do you do fellow kid? kids meme with the... It's such a crazy choice

on a million different levels, even the smoking, even the NFT. Maybe Reed secretly likes heaters. That would be sick, I guess. Jay said this GM is proof that LinkedIn is four years behind on every new cycle.

Pretty remarkable. How did the... I think the crypto community loved it, to be honest. Really? I think they're pretty fired up. If you're one of the 10,000 people that own Cryptopunk? A punk.

How are the punks doing? Are they still hanging out? Hang it around. Well, before we dive into that, we have Justin Murphy in the Restream waiting room. Let's bring him into the TVP and Ultradome. Justin, how are you doing? I'm good. What's up, guys?

Good to see you. You're looking fantastic. You might be the sharpest guests we've ever had. I love this suit. You're looking great. Thank you. You know, you're the reason, you're the reason I use one password. you you had some sort of course on like creativity or something and the main thing I took away was use a password manager because you gave such an emphatic endorsement of password management and I adopted it and it changed my life and I love it and then we interviewed the the CEO of one password and I was like I love this I learned this from Justin Murphy wow that's a really interesting to hear I'm an avid user of Lucy I'll throw in as well oh let's go there we how people don't use password managers is beyond me I mean even with

With password managers, it's living hell, but I'm glad you, glad you solved that. Yeah. Maybe for the folks who aren't super familiar, what's the shape of your business? What's the shape of your day to day? I know you have a number of different outlets. I wanted to invite you on the show because I noticed you were live streaming and you actually mentioned us. And I was like, oh, that's so cool. Because I love when we like create some format and then other people can kind of spin off of it and whatnot. But how do you describe like your whole life like yeah your whole person your cinematic universe i suppose yeah so i mean i left my very nice cushy academic position about six years ago and i left at that time with a very clear mission

which was to basically reinvent the scholarly life for the internet era um i had i was watching sort of academia go downhill and all the creator economy stuff was sort of kicking up kicking off and it clear that there would be ways to essentially port everything I was doing as a professor onto the internet. And I would just have to experiment and test a lot of things out and hustle and try to figure it out. So that was my mission six years ago. And it still is. I mean, that's my calling. That's my vocation. It always has been. And I basically built a couple businesses, tried a few different models and had moderate success with a few of them. And then, you know, I started having kids and kind of took a step back to kind of, you know, think about what this stuff

really looks like, you know, over the long term. I had a lot of successes, had some failures as well. And that's when I decided what I really needed to do is sit down and write this book called the Independent Scholar to really map out the whole thing and really put this kind of new lifestyle and business model on the map. So basically the past couple years, past two or three years, I've kind of been in hiding, honestly. I've been having kids figuring out how to do that and writing this book to try to basically teach everything I've learned. about this lifestyle and business model that I've built, and also to ground it in a kind of longer-running history

because a lot of the greatest thinkers and philosophers and what I call independent scholars throughout history were actually free-floating individuals, hacking it in weird entrepreneurial ways. And most people don't know about this history, and it was largely what inspired my kind of mission. So I decided I needed to do it right, sit down, take the time required, and put it down all in one place.

So that's finishing up. And then I also, I occasionally work with startups if something actually really interests me and seems really exciting and asymmetric and I like the people. So my life's really crazy right now just because I, a startup I started working with like two years ago. It started. I was just sort of like consulting, kind of advising, helping out with the branding and messaging. They just keep winning. And this thing is like kicking off. So we're in the thick of it right now.

That's called knock chain. It's a layer one blockchain. So that's kind of consuming everything at the moment. Apologies to the poor people who pre-ordered my book. A lot of people are waiting on it. It will come out by the end of this year. Mark my words. But right now I'm in the weeds with knock chain because this thing I think is going off.

So that's a summary. It's a little messy. But I basically have accomplished my mission that I set out on six years ago. And I still haven't come up for enough air to tell the story fully, but I will. And I think, you know, I'm pretty proud of everything I've accomplished. Yeah. How do you think about the, just like, the basic economic forces around like long time horizon content,

which is how I would put investigative journalism and academia. It feels like social media and algorithmic feeds are like fantastically aligned with like news and kind of fast, high throughput creativity. Like just because if you're producing a lot of content regularly, there's a lot of ad inventory space, there's a lot of room for a little side partnership here. And it just kind of all flows together very easily. feel like substack and the social media platforms have done a great job of providing an outlet for someone who would effectively be like a beat reporter or someone who's just a commentator,

an analyst, somebody with a column, a weekly column in a newspaper, you can take that weekly column and you can go and develop an audience and monetize effectively. But if you're Seymour Hirsch and you're going to get like one scoop a decade and it's going to be something that reshapes the global political world order, really hard to underwrite that against like $10 a month on substack. And so how are you thinking about like the the long term monetization summary like like forces that allow for like long time horizon work? Is that solved? Is there a good pattern there yet? I have a very strong and clear answer to this after a tremendous this amount of experimentation and also this kind of prolonged engagement with the history of

independent scholars. I believe the answer to your question, John, is you need to stage one is you have to get out there and you have to write a lot in stage one. So you don't necessarily have to be chasing clicks or playing the engagement baiting game or whatever, but you do have to get out there. And inevitably there's going to be some kind of volume game where like you have to get in the mix. But I don't think you have to do that actually for too long because if you can just get out on places like X and have some kind of blog or newsletter or videos or whatever your medium is, if you can get out there and just put out interesting stuff that's intelligent and thoughtful and build even a small audience, nothing that

you would be able to live on. But if you could just do that stage one of showing that you understand content, that you have real ideas, that you are really interesting and worth listening to, and that you can execute on a basic kind of modern content operation, if you can do that for even a little bit, what you then do, stage two, is you get exposure to tech companies. That's really the missing link that people have not fully figured out. A lot of people have sort of accidentally figured this out. But the fact is that Silicon Valley and technology is obviously the engine of all kind of radical wealth creation moving forward. And these companies need really educated, thoughtful, creative people who understand how to read real book.

and how to write really high-quality copy and very thoughtful essays. This whole new world of sort of high-brow thought leadership is incredibly valuable to almost every company today. And the fact is, it's incredibly hard to hire for. Tech companies really struggle to find sort of PhD-tier intelligence for hire and who also have any kind of understanding or experience

with actually running modern content operations online. And so what I have found, and this I think is really the tip. ticket is that if you can just simply prove that you know what you're doing online with content and you can build any kind of audience and just put out stuff that's worth reading that someone smart could look at and be like, oh, this person's smart. He might not have a million followers. He might not have many followers at all.

But clearly he publishes consistently. He's smart. He's got alpha. It's, I think, actually increasingly easy to find some sort of creative position with a startup or a tech company of some kind. And that's going to be your ticket to actual sustainable wealth and income. on which you can then do the really long-form thought leadership.

I've seen enough. Basically, use like consulting fees from a tech company to bootstrap like a independent media platform. Absolutely. And maybe some people don't want to build their own platform necessarily, like someone like Seymour or these kind of older guys,

like often they don't want to build a business, right? They don't want to build a whole thing. And for them, that's kind of a non-starter. But if you can get some, if you can finagle some kind of situation where you're adding value to, you know, a growing tech company or even just a funded startup for a couple of years, even if they fail. Like, you can drive real value that they have a hard time hiring for. And in an ideal world, the company wins. And then you really have, you know,

economic security to do your intellectual work over the long term. But even if, even if you have to hustle this way for a few years in different projects, you can get a very decent paycheck adding real value in a way that is actually quite conducive to your intellectual, you know, orientation and goals and lifestyle that you're building. So you don't necessarily need to build a company or a big media operation. But on that economic basis, you can definitely grow your audience in that time. You can write, you know, a really heavy-hitting long-form essay once every month or two months or three months or whatever the case might be. And in this way, and, you know, you start publishing your own books. And there's so many different ways to

monetize. But the tech partnership, I think, is the big opportunity that a lot of people are sleeping on. People have done it. There are enough examples in addition to me to make me feel like this is obviously the future. But people haven't really systematized to get or kind of like written down a reproducible playbook yet. Maybe it was, I don't want to get the person's name wrong, but there was someone who did like a philosopher in residence at a venture capital firm. And there's been a few people who have maybe partnered with like striped press in one way or another and kind of like pulled forward some income in various ways. There's definitely, yeah, different ways to like slice it. But I think you're, I think you're right that there's just like

There's a pool of capital over there, and if you're able to provide some interesting service, I hadn't thought about that. So basically, if you want to be an independent scholar, get ready to hustle. Kind of, yeah. Well, but look, it's a hustle to get a job as a professor. It's a really bad hustle. It's absolutely grinding competition that is as brutal and as intellectually exhausting and suffocating as grinding on a startup for a few years, honestly.

The only difference is you don't get paid very much at all, and all. Also, there's no upside. So, I mean, when you compare working for a startup or, you know, being creative and hustling with different kinds of creative positions in tech startups, you compare that to what it takes to get a full-time tenured position in academia, it's actually, like, more liberating and mentally kind of freeing and more conducive to a kind of creative intellectualized. I had lunch with an academic couple months ago, and he was like, yeah, I'm having a kid in like six months. And I was like, go get a job at a lab right now. And he actually did. Oh, he did? No way.

That's awesome. And, you know, again, I think he'll be able to probably do more interesting work at the frontier with more resources. And I guess is part of this like sad that like what's the downside here that the techno capital institutions are going to end up financing? scholarly research and academia like is there is it is it basically lindy like this has been happening for like hundreds and hundreds of years and this is like not so different is that kind of what you're getting at with the book yeah exactly it's not sad i mean contemporary academia is very sad living a life as a kind of cloistered professor today is incredibly sad i can tell you from from firsthand experience and what we're talking about here with sort of the the new breed of independent

scholars joining forces with Silicon Valley. It is clearly part and parcel with a long-running continuous history because when you look at some of the most interesting and impactful thinkers and philosophers and creatives throughout all history, what you find with many of them is that they lived on the margins of institutions, both economically but also socially. And in multiple cases, what's really going on is they're looking for any kind of of unexploited economic niche where they can just make some money in a way that is maximally conducive to a certain state of mind and a certain kind of lifestyle. And they are really kind of opportunistic. Like they'll, you know, Spinoza was famously a lens grinder, right? This is at a time

when being a lens grinder, it's not as manual as it sounds. He was a kind of scientific and technical expert of optics. And to grind lenses was actually an interesting kind of philosophically and scientifically sophisticated industry, but he also involved manual labor, and he sold his lenses to leading scientists of the era. And that's actually how he networked with some of the brightest lights in European science and philosophy. That's how he got on people's radar, and that's how he earned their respect, actually, is from the quality of his lenses that he was grinding. And this is in my book. And so Samuel Johnson is another interesting example. He's a case study in the book. Samuel Johnson, people don't know this, in the early 18th century, he basically

kind of invented the substack model. And this is before there was even email. I kid you not. So he did paid subscription newsletters by print well before email, obviously, and he would just write essays twice a week, and he would walk them to the printer's office and where the mail goes out or whatever, and he would send them off twice a week by post. And And he made a living doing this for quite a few years. He was at the end of the patronage era. So he was well known in elite circles as one of the most brilliant genius thinkers and writers and conversationalists of the time.

And he never got a patron. He never got the bag. He was always forced to hustle. He was always doing creative things. And that was one of them. And he managed to do it and ended up changing the English language in many ways, one of the most impactful kind of scholars of his lifetime. And so, to answer your question, Silicon Valley and tech startups today are just the current

manifestation of this. Like, independent scholars always have to look, where is the money, where's the opportunity, where can I use my brain power in the way that is maximally adding value, but also maximally conducive to me, having freedom and time and being able to think clearly and provocatively. And it's obvious now, if you, when you look at this, that in the United States, perhaps in the West, more broadly, Silicon Valley is now the site of that. Why do you think Silicon Valley is obsessed with Nick Land? There's still a tremendous amount of alpha there that people have not fully been able to grok because it's, you know, it is dense and a little bit difficult to read.

But that's the reason is because there's real alpha there and people know it and no one really has the time or the patients to kind of work through it. But they know the alpha is there and that's why they are obsessed with it. What is the nature of the alpha? just investing theses or like lifestyle, how you live your life, or predicting the future and what the shape of the political economy will be? I would say, I mean, there are many specific things you could look at. If you look at his writing on AI back in the 90s, it's incredibly prescient and much more so than even, even you compare Nick Land's early writings to someone like Nick Bostrum, writing about AI much more recently.

And a lot of Nick Lance texts actually match reality much more. I would say. But the reason, so you could double-click on many sort of specific ideas, I guess. But the high-level answer to your question, John, is that there is this kind of ideological problem in Silicon Valley across the board. I mean, the big intellectual problem with Silicon Valley is that everyone is a backpatter. Everyone is mutually invested in each other, so everyone is back-patting everyone else. I mean, just structurally, right? It's not even not throwing shade at any individual, but just literally, everyone is,

invested in each other. And so pretty much all public content that you see is, you know, it's someone speaking their book. And that's fine. And that's America. And that's cool. Like, that's the world that we live in. But there is just a big, sort of one big drawback to that, which is that people are too, like, pro-social is the way that I would put it. In Silicon Valley, you know, no one wants to be, you know, anti-human or antisocial, right? Because it, it, it, it's not it doesn't sound very nice it sounds a little scary maybe you don't want to invest in that we had this we had this this morning as a team john tyler and i were sitting around we had a we had a post that will not be posted that was like genuinely had us all in tears but there's one group

that would be uh you know taking the the brunt of the joke and we like them we're friendly with them and we just wasn't it wasn't something we were going to share um and so you yeah, there's this pressure to be, pressure to be social, which in some ways is healthy, like in some ways that stems from like, this is a culture that somebody can move to San Francisco and be a nobody and in two years be an icon. And that is like, that's positive, but it also means that the culture is like, I'm not going to do anything antisocial here because two years from now, this person, I might be trying to lead their round, right? Exactly. And so the issue with technological acceleration is that it is a brutal kind of machinic

global process that's been happening for hundreds of years, and we know a lot about it, and a lot of it does not look very pretty for a lot of humans, not all humans, but for a lot of humans. And so when you really try to model this really rigorously without any kind of pro-social biases and without any of the kind of human ideological filters that, frankly, most Silicon Valley players are forced to kind of lay over their thinking and their publishing, you see, like, no one in Silicon Valley really is able to talk, frankly, about kind of the underlying brutality

and the really kind of scary structural elements involved with capital acceleration. But Nick Land does that in a very aggressive way, and that's that's the source of the alpha basically well and dario's hasn't been i mean i'm sure that the anthropic comms team uh freaks out every time he goes and gives a talk and says you know 80% of or whatever the statistic he's saying it's some some large percentage of white collar work is going to disappear and then they kind of like walk it back a little bit but clearly yeah like the message is being received well but notice notice that whenever you do hear some silicon valley thought

leader or big player come out and say something that's kind of antisocial or bad or scary. There's always kind of immediately behind it the solution that they're providing that will make all okay. You know what I mean? So the thing with Nick Land is you've got a model, you've got a really hardcore, I think very rigorous model of modernity and capital acceleration, which has no such convenient solution hiding in the background to sell you. And when you look at that stone cold in the eyes, most Silicon Valley people don't want to hear it. But on the other hand, they know there's alpha there, so they kind of can't look away from it either. In shape of desire, there's this concept of like artificial intelligence being this time traveling force in invasion from the future.

As a human that makes me feel like I want to fight it. Is that the right desire? Is that the human desire to counter the machinic desire? So land does introduce the site. Well, he doesn't introduce it, but he very much plays up this mental model that you get from the Terminator movies that, yes, Capital is this kind of time-traveling entity that comes back and not only kind of composes itself from us, but just as the Terminator movies suggest, comes back and kind of composes itself out of enemy resources.

So there is in that kind of model, an implicit kind of antagonistic structure, but I'm not so convinced of that, frankly. I think the fact is, capital acceleration will be quite brutal for some people, for many people. I mean, it clearly shreds apart sort of modernity and technology shred apart traditional values and morals. I mean, even Marx knew that. I mean, Marx perhaps knew that better than anyone. That has been clear, that has been clear for a very long time. And whatever the errors of Marx were and the terrible

kind of tragedies that came downstream of his ideas, you know, the critics of capitalism in the early modern period were structurally on something. Okay. And there's no doubt about that. I mean, it absolutely shreds any vestige of any kind of traditional morality, basically. And that is sociologically true beyond a question, in my opinion. And so play that forward. Play that forward. You know, it's going to be ugly for a lot of people, I think. But I think all we can do really is try to basically, you know, try to stay one step ahead, be honest and rigorous and aggressive. and take the risks necessary to bet on the truth and try to just stay ahead of the machines and try to take with you all of your friends and family and everyone in your tribe. There's really no other solution. And I think personally that it actually goes in a more positive direction than some of Nick Land's suggests, you know, Texas seem to suggest, I think it's actually, it actually matches much more

clearly to something like the city of God by St. Augustine where, you know, basically the good people and the bad people increasingly kind of diverge over time. So technological acceleration kind of pulls these people apart what he calls the city of God on the one hand and the city of man on the other. What technology really does is it accentuates the differences between people who are oriented to heaven and the truth and the good and people who are oriented towards anything else. So if you're oriented towards anything else, you're going to go down fast. You're going to get wrecked, I think, and increasingly fast. Thanks to technological acceleration. How do you summarize the current discourse,

because I would loosely group it into general fear among the broader white-collar class and potentially even blue-collar because you have everybody, you know, people outside of tech are worried about automation and losing their jobs. And it sounds like you think that's a very, maybe the correct concern to have. Inside of tech is kind of like bifurcated into people that are thinking like, okay, Open AI just is starting to look more and more like Facebook every single

day. They have SORA. They're going to introduce ads. They're introducing commerce. This just looks like a big tech company. And then you have people at labs that just believe like, you know, just wait, like you know, just one more training run, bro. AGI is right around the corner.

but it's hard to like actually read too much into what they're saying because they obviously have you know tens of millions to hundreds of millions to billions of dollars of like you know equity tied up in in these narratives that are kind of dependent on on those scenarios playing out to some degree but how would you kind of summarize it and where do you think it is like even a year from today yeah so again i think a lot of these categories that people use to parse this this phenomenon are just totally wrong and they're totally ideological. So, for instance, the jobs story, like, that's all people can come up with to think about the devastating impacts of technological acceleration. We're going to lose our jobs. We're going to lose our jobs.

So, like, everything potentially bad about AI gets shoehorned into this, like, one topic of, like, are people going to lose their jobs? Like, it's much more complicated and in some ways worse, but in some ways better. So, like, the better way to think about it is really just that with all technological acceleration, you have to, everyone has to hustle harder, basically. Like, you have to work to stay ahead, to succeed. You have to work really, really hard and increasingly hard. And yes, there is more surplus. People are wealthier on average.

That is true. So, you know, a kind of poor person in America has exposure to, like, a lot of really nice things that, obviously, you know, only kings could have in the past. So that story is true. But nonetheless, to actually build a life that is healthy and happy and sustainable, you You have to be incredibly smart and sharp and judge and making good decisions about a number of different things in all aspects of your lifestyle and you have to be working super hard and you have to be dodging this existential threat. Like nowadays, I mean, think about how much research you have to do on like food just to stay healthy, right? Like you have to be incredibly smart and sharp and you have to hustle increasingly hard just to survive the chaos of technological acceleration.

And so AI is just a kind of a further acceleration of that. It's going to get more and more insane. The payoff will be bigger for people who are right and who keep their head on their shoulders. For those people, the payoffs are going to be bigger and bigger. But for everyone else who can't keep up with this kind of shredder of traditional norms, you do fall behind. And it's not a matter of losing your job. It's a matter of like your whole family is, you know, opiate addicts because they got one-shotted by complex processes that they had no chance of, you know, even understanding. When does this start? If we take the time travel analogy, should we start

this whole story at 1860 when GDP really starts to accelerate? Is that the beginning? Do we go back earlier? Is this latest like AI boom, anything special? Carpathie on Dorcasch was kind of saying, no, like, this is exactly, this is just an extension of computing. So this is an interesting research question about how do you really mark that breaking point, um, associated? with modernity and technological acceleration. People have very different answers. I think it's an interesting kind of empirical debate. I've always kind of impartial to Heidegger's answer.

If you read Heidegger's question concerning technology. He seems to think that the key break point was when we started storing energy, basically. So he says the windmill, for instance, was fine because the windmill just sits there. Nature kind of operates on it and it can create energy. And he seems to think, that's my reading, is that it's fine. but when we start storing it, when we're putting it in batteries and we're putting it into these different containers, he seems to think, he seems to suggest that that's kind of the breaking point because that's where you can start to sort of run much farther ahead of the future

because you can store these things up, right? I have a stockpile of coal and I have a steam engine and I can build up a larger pile of energy and then I can deploy it all in one jolt to jump forward. Right. And then the Marxists have this kind of theory that they like to say that it's the enclosure of the commons. So like in the, it really starts in England. There were for, for, you know, aons, these traditional norms of anyone could kind of graze on the commons. And much of the land was commons. And then at a certain point, there's a specific kind of decision made where they start enclosing it in order to capitalize on it and privatizing it. And some people point to that as the kind of the breaking point. You can get many examples. And people kind of debate that. that. I think it's sort of, it's sort of very hard to answer conclusively, but those are some, those are some possible answers that I think are interesting. Do you think, yeah, do you think the escape the permanent underclass meme is actually healthy then? Because it's basically is encouraging people to like, you know, sharpen up, avoid the, avoid the pressure to just quit your job and

gamble on meme coins, like avoid, you know, becoming immersed in an infinite slot machine. I have a completely different take on this. And I want your reaction, too. is I feel like the future things playing out, like kind of the darker scenario is like there's a massive underclass, but there's no permanence whatsoever. And on any day to day basis, somebody could go a thousand X up and become extremely wealthy and then everything collapse and go back down to the underclass. And I feel like the volatility is actually increasing way higher. And we're actually getting more underclass but also less permanence. But what's your take on the permanent underclass meme? So that meme, I actually think, is a rare good one that is directionally correct in an important way. And that actually gets more to the heart of the threat and the brutality of economic acceleration much more so than the jobs meme. That is the kind of anxiety, that is the general diffuse anxiety and fear that everyone feels and is right to feel. You are forced to feel that with capital acceleration and its current inflection with the AI revolution. So I think that's basically,

basically that that is spot on in terms of the generalized and problem it feels like durable because everyone feels it and it feels like yeah like you're saying like it's somewhat real right it's like the techno capital acceleration people feel it they realize that making the 20 million dollars in three years is a lot better than making the 20 million dollars in 25 years or whatever historical time works look like. It's worth noting that it's kind of self-enforcing as well, because if that's a meme and you know other people feel that way, then you really have to trust it and you better get after it because even if you think it's false in stage one, right, if you think other people are buying into it, you think, okay, everyone else is going to be hustling harder,

they're going to be getting the bag. And shoot, even though I thought this is false, now they're all getting it, I better get after it or I'm going to get crushed and left behind. And so, yeah, I mean, what a lot of people say is that capitalism produces so much surplus that you could just choose to not participate in the rat race and, you know, just enjoy the surplus. But that's not totally true because the things that we really want in life and that really matter to us are often relative, things like recognition and identity and, you know, standing in a community. These things are relative. And so it doesn't really help that much that you can, you know, have, you know, a big screen TV for really cheap. You can have many other things. It's clear, like, everyone will be entertained in the future.

You will not need to be elite to be entertained. Like, you're going to be entertained, you're going to have the screen. You're going to be able to put on the head set. You're going to be able to take any range of telemedicine drugs to make you feel anything. You will be entertained. You'll maybe have one in a 10,000 chance every week to, like, escape or whatever. But I do think it's notable that we transition from a pull yourself up. by your bootstraps culture, which was like, if you just do anything in America in the early

19th century, you can get a house, you can, you know, you can live kind of, you can pull yourself up by the bootstraps to live the American dream. And now the American dream is on the other side of escaping the permanent underclass, at least the American dream that you were sold as a kid of having a home in a desirable area and sending your kids to college and raising a family. The American dream was the middle class. It was the default. Right. I mean, frankly, I think the big antidote to all of this, the other poll of this entire problem that we're talking about is, of course, crypto, I believe. I mean, in so many ways, it's sort of diametrically kind of opposed to AI, or it's like perfectly poised to be kind of the solution to a lot of this. Because one way,

to understand technological acceleration is, especially in kind of human societies with all of the faulty human wetware that we have, is that it becomes this like big lying game. Like everything gets inflated. Like, you know, the money supply is really just kind of a metaphor for many other things where there's a ton of inflation, right? Like verbal inflation, you know, credential inflation. In a way, like modernity is this mad race to kind of inflate all kinds of things. And we get ahead of our skis in many ways. And that's why people have to, that's why it's so cutthroat because other people are inflating, so you have to inflate or you die. And so, like, human society and modernity is totally torn asunder from all of its traditional

values and traditional, you know, anchorings. Everyone is competing in this mad, materialistic race to stay float, which is memetic. And even if you don't want to do it, you kind of have to do it. Meanwhile, crypto is this kind of, I mean, cryptographic protocols are truth machines, basically. They kind of restore integrity and kind of, like, primordial. almost physical truths to this kind of chaotic, constantly inflating kind of material in modern world. And so my mental model is very much that the reason you're seeing, the reason the early stages of crypto have been so chaotic and crazy with, you know, so many stories of big crashes

and crimes and so many kind of scammers and the reason it's so crazy like that is because people intuitively understand that this crypto stuff like is the next. big vector through which humanity will be able to secure its future. And so the people who rush into that first are a wild grab bag of sketchy people, I think. But it's going to be the thing that, like, it's going to be the only thing that puts kind of grounding on all of modern chaos. And you see this like, that's what Bitcoin does, right? The Bitcoin is like literally just regrounds money and like physical resource kind of

anchoring. One suggestion. crazy idea for you before you jump and then we'll have to do this again soon. It's been really fun. Thank you so much for out. Instead of a book, how about you buy a Lambo and you sell a course? I will teach you to escape the permanent, the permanent underclass by becoming an independent scholar. I mean, that's not. That's one option, but I don't think so, man.

That's the role of knock chain. We say escape the leveling process. I don't think you should worry about escape. I don't like the phrase, you know, escape the permanent underclass because it's much more general. You have to escape the leveling process. There is a long running gradual historical gradient that is trying to sort of destroy everything of traditional value. Everything that is permanent and timeless is getting sort of eroded or hidden and kind of taken away from you. And like you want to escape the leveling process.

And that's why we're building a blockchain. That's what it's all about. Escape the leveling process. That's very cool. Well, thank you so much. Super fun. This is a great conversation. We'd love to have you back on and go deeper.

Everyone really enjoyed this. But thank you so much for hopping on. We'll talk to you. I appreciate it. Cheers. We'll talk to you soon. We have our next guest in the Restream Rating Room. But first, let me tell you about Fid.A.I, the number one AI agent for customer service.

Number one in performance benchmarks, number one in competitive bakeoffs, number one ranking on G2. Let's bring in our guest from the Restream. Let's do it. waiting room. Darren, what's going on? What's up? How are we? Second time on the show.

Thank you so much for helping on. I'll let you take this. Sweet. Great to have you back. What's happening in your world? Here you've been flipping, flipping pretty hard. So break it down for us.

So this weekend, a 1914 Baltimore Bay Bruce, Baltimore News, Babe Ruth, sold for $3 million less than the guy bought it in 2023.

Brutal. Making it the biggest card loss of all time. No way. Now, I thought they only went up. Yes. So did Adam Smith never says that. No, never. So, you know, it's very interesting to me, because I don't like to admit when I don't know something.

But when this card, when I first heard of this card was when collectible, which was one of those fractional share companies that went bust, when they said they were buying a piece of it, I think it turned out to be like a 1% piece for a value of $6 million. And I had said, I've never heard of the 1914 Baltimore News Bay, Bruce. So I did a little research, newspapers.com is probably the best value that I have. I mean, $150 a year for all the newspaper stuff.

I mean, wow. That's crazy. I don't actually, I'm actually not familiar. This sounds incredible because we're always, I mean, we have a little recurring segment on social. We talk about this day in history. I'm sure we could get a bunch of, bunch of alpha. Oh, you can get. Now, hopefully they didn't see this because I've been subscribing for a long time. and they're not too good at business, because I would really pay for it. Yeah, exactly. So anyway, I looked and tried to see when the first mention of the 1914 Baltimore newspaper with this, again, because I didn't know, and I think I know everything. The 2006 Durham Herald Sun, that was the first mention, at least on what newspapers, and they have a lot of newspapers in there, that was the first mention.

of the 1914 Baltimore News, Babe Ruth. So I kind of have this theory that people don't think it was a card. Interesting. On the back, on the back is a schedule. So it's Babe Ruth in the minor leagues the year before his rookie year. And it has a schedule of who they're playing. And it came in a newspaper and it was a 10 card set. So one of the reasons why I think this guy lost was because

there's only so many people that know this. A lot of people buy cards and big cards so that they can say they have the big card. And when someone says I have the big card, oh, what card is that? I don't even, I've never heard of it. It's a whole lot different to have a T-206 Honus Wagner or a 1952 Mickey Mantle than to have the 1914 Baltimore News Bay Bruce. And so I think that's why the guy got it handed to him. Okay, and why would, do you think the person that just paid $4 million for it knows that it might not be a real card?

No, I'd like to know who it was that bought it, but I don't think I'm going to find that out. Again, it's a real card by a lot of standards. So after I looked at the 2005-2007 Durham Sun, I did more research. had mentioned it in the early 80s. So they've mentioned it in passing. But to me, it doesn't inspire a whole lot of confidence that something that was beginning in 1914, until 1980, no one talks about it. I'm just not comfortable. I'm not comfortable. So that's my take. Might not be able to think, but it's my take. What is it say about, do you think other people that own, you know, own some of these like flagship cards are a little bit concerned now about the like real market

value of whatever they're holding or is this like a one-off like no the market the market is as hot as ever um i sold 31 things uh this weekend on heritage auctions and some of the prices that i got were were amazing um some are are painful in terms of you know really really able to do this. Like, you know, I just, uh, it's hard to sell. It's always hard to sell. But I got some, some really awesome prices too. So this is, the question is, did you buy 31 times two to just even it out? Well, you tell the wife you didn't. You say I made so much money. That's how it works. And then, you know, there's a different bank account that then goes out. It's all about trust. Because in the end, I mean, I'm married 17 years. It's all about trust. It's all about

we are going to win anyway. We're doing it for our family. So I did buy a massive, massive, massive piece about two weeks ago, and I had to clear some room. Did it clear some room? But yeah, this is investing, right? So we made a lot of money. We're going to invest some of it again.

That's how it goes. Talk to me about Otani on the Dodgers. Because my family, I'm not super into baseball, but I do go to the occasional game. I often get Otani bobbleheads. I was so confused, by the way, I had a Dodgers hat on, not because I'm a fan of the Dodgers. It's just a green hat, and I like New York City.

I've always, I've always, I was wearing, New York City? They made a green Dodgers hat. I was wearing a green Dodgers hat. And it was for the Brooklyn Dodgers? On Saturday, so many people were telling me, we're yelling baseball things. I'm like, I wear this hat all the time. Why is everyone yelling?

And then I found out from the security. my neighborhood that the World Series was happening. That's how I learned the world. So we're out of loop here. Wow. That is. That is what you guys have to pay a little bit more attention. We do. Well, no, I'm more interested in the market. I'm more interested in the market than the just alt's market in general than the actual game being played. Right. For sure. Yeah, but my question, yeah, my question about Otani is that I've been getting bobbleheads when I go to the game, but it feels like there's the, uh, the market, like the equilibrium of the market is beating expectations, right? Because if there's really high expectations, a lot of products are

produced, right? And so, where are we on, like, clearing prices around Otani? Is he blowing out expectations? And so market prices are rising, or is there a different dynamic? Yeah, absolutely everywhere. So Fanatics has a deal with Otani, and they're buying up all the Babe Ruth balls because they want to have Otani sign the Bay Ruth balls and then, you know, sell them. So if you're looking for a single sign, what do you think about that? Is that, is that, Is that not, like... I don't want it. I want Babe.

Now, Babe Ruth signed plenty. He signed a tremendous amount. But to me, so there's... So the value of Babe Ruth's single-signed balls are going up because they are buying in the back room. They're buying as many so that they put a tony on it. The bobbleheads, I mean, the bobbleheads since 1997, I mean, it's amazing the life of bobbleheads and the lines of the bobbleheads. The Dodgers did a really good job game.

So you have to gamify it for the degenerates and the youth every we live in a degenerate economy And so you have to have ten that are golden and on the couple with that come with the dog And you know, and then that really kind of you know jams up eBay. So the Dodgers have done a good job But bobbleheads I can't believe it. Why does it jam up eBay? You do by jamming up you mean just makes it popular? Yes. Yeah. Yeah. People people people you know people really like to go to buy their bobbleheads on on on eBay It's amazing to me that that's been going for 30 years. I'm surprised there's not another thing. I mean, the market is so crazy.

I don't know if it's always like this, but when I go to Dodger Stadium and they give me a bobblehead, there will always be someone inside the stadium who paid for a ticket. Who offers? And you think about, okay, what is the sunk cost of that ticket? They're not actually watching the game. They're there just to acquire the actual merchandise. By the way, I got to jump in.

I misspoke, I meant the, it was a, it's a green Yankees. I was about to say, yeah, everyone is what are you talking about? I was a Yankees hat. I'm not going to come on anymore if you guys don't know the difference between a Dodgers and a Yankees. I do know, but I have both, I have a, I have a Yankees hat and a Dodgers hat, and I wear them interchangeably. Okay. And so I just like to confuse the people in my neighborhood. They don't, they don't know.

I mean, the Dodgers were in New York at one point. They were in the Brooklyn, but, yeah, you were aware. They were. But then why were people talking to you about the, Were you wearing a Yankees hat? I was wearing a baseball hat, and I didn't know as a World Series, so I didn't know why people kept yelling like random baseball-related things at me. Like, you know, people were just cheering for the-

Yankees haven't won the World Series since 2009, so you should know that. So they're in a 16-year drought right now. Brutal. What, yeah, so why would this person that just sold for, Do you think this was a for-seller, like somebody that just had to get out? I think otherwise you would just sit on it. Yeah, I think so. I think so for sure.

I mean, I always say that the assets that are in this space are not it's. They are what's. And, you know, if you, there's no absolute here, right? It is, it is when do you sell it? What auction house are you selling it? What's the description in the auction house? what's the publicity what's the marketing these all play it's like when when someone says to me hey i'm going to invest in otani because he's playing well i said this isn't collecting isn't

fantasy collecting isn't fantasy baseball collecting is a amalgamation of all the factors that go in yeah right it's it's how many people collect otani what is the worth of the people who collect otani what is their ability when they when it's going down to stay and hold And if you're a 35-year-old man with money and you collect what people who are 16 to 20 collect, you better know because they don't have the wherewithal to hold on if it goes down and then that affects you. Or you better know if there's a guy who owns 600 of them because, again, those are the market factors at play. It's not as simple as a game of fantasy. Do you have any insight into what's going on at GameStop? We had the CEO of GameStop, Ryan Cohen, on the show, and he was talking about.

about prize packs or what was it power packs let me let me this is this is this is a fun one yes so repacks are like the new thing there are if digital refax are even crazier now people who are watching the show I'm now going to blow your mind okay uh digital repacks this is how digital repacks work you open a pack digitally that card really exists in the company's coffers or vaults okay you open it you don't like it. So you pay $25 for it as $8 in value. Then they immediately offer, do you want to exchange it at 80% of the value? And then you're like, okay, I'm comparing this to gambling where it's a zero-sum game. If I lose, I'm zero. And they're like, wow, I can, I only

lose 20%. And so these companies like courtyard and arena club who do these digital, by the way, then it goes back into the pack. They could sell the same card 30 times a day. Wow. And they show you the checklist. They show you, yeah, the checklists. So, so courtyard did 70 million this past month. Yep. Uh, they, it is, it is going nuts. Anyway, GameStop, what GameStop has done is it has a relationship with PSA, uh,

where people can grade their cards. And a normal company, and since GameStop is always going to be to me, a maim stock, uh, a normal company would be very worried about taking submissions from, from, from, from, from people holding the card and then sending it to PSA, right? If this was another company that was a Fortune 100 company, you would be really scared about the liability of this. But I feel like GameStop has a little bit more flexibility than most. And so they are the biggest player in this game.

And it is really, we've written about it on Collect a lot. It is really interesting. how are startups faring in all of this you have game stop playing here fanatics a couple that you just mentioned you have heritage i feel like the big the big auction houses have been surprisingly durable against um you know upstarts coming to their markets if you look at the automotive industry uh everyone is like like every industry uh everyone wants the multiple so they're all in tech you know that's what they do right they they they want the sass model that's that's what that's what everyone's after um so you know originally there was card scanning where it told you what the price is a lot

of people were coming out of covid and they're like i have a bucket full of stuff and no one's going to go through it so i'm going to have to scan it who's the best scanner and then we had too many scanners um and so there's there's just a you know there's a lot of information there's a lot of people fighting. I don't know who the winner is. Fanatics is all over the place. It is probably in terms of monopoly, they are probably the number one monopoly in any business today in terms of having a stronghold on so many things. So it makes it hard to get into business. Do you think that'll keep just compounding? It depends. It depends on what teams and leagues. whether they you know i mean teams and leagues want to take the bag and they're going to give the biggest

bag now if you choose to you know build it by yourself then that's a little bit different but as of right now um that's that's what we're looking at um i mean some of the parts of the consulting part of my company it helps teams and leagues and brands get into this and help you know build um by themselves but you know right now fanatics is is is the two million-pound gorilla. It should be interesting to see what they're going to do with gambling because, you know, I just think the one and two are Fandul and Draft Kings, and I don't even think it pays to be three, given how much the business costs and the consumer acquisition. Yeah. What about how much attention have you been paying to prediction markets entering

sports? I mean, obviously they've been at it, they've been at it for a while, but it feels like You mean Polly Market? Polymarket, Kalshi, it's really heating up. As someone who was in gambling for a long time, I'm shocked that Polly Market and Kalshi got to the place that they did. Sure, they always had kind of that like were peer to peer, but I can't believe that, just like prize picks, I can't believe that they got to where they did in their evaluation because I always thought that the lobby of Draft Kings and Fandul and Fanatics and Caesars

and the old casinos, I always thought that they were going to win. The tribes, the tribes, yeah. I mean, you know, the Seminole tribe, I thought the lobbies would win. So I'm actually shocked. I always thought that Pollymarket and Kalshi would be kind of lower end and never really breakthrough, and that is not proven to be the case. What is the secret to fanatic success? Like, it feels like there was a time when every team and maybe every league had their own kind of merchandising operation.

I remember being able to, as a kid, go and buy an NFL jersey. And then at some point, fanatics like finagled the ultimate deal and got everyone in the room together. Well, they came, again, they came with the bag of cash. Is that what it was? Is he just a good, is the founder just a fantastic deals guy, just got all the money? Well, Michael Rubin has a great, you know, he has a great story. You know, if you don't know this story, I'll quickly tell it. But, you know, he owned GSI Commerce, which was essentially around the turn of the century, around 2000.

All these companies were saying, oh, we could do a virtual warehouse. We could do our warehouse virtually. And Michael said, I'll do a physical warehouse. I'll get it to you, I'll get it to your consumer a day faster. and I'll take the risk that 7% of the time Jeremy Lynn is going to get traded from the Knicks to the Rockets and I'll have 9,000 Jeremy Lynn dolls in a room because if I win every business it doesn't matter if I have waste

all these other guys are not taking any sort of risk they're just opening a virtual warehouse and so that was his that was his way to get there and he got everyone's business and then eBay bought it from him for 2.3 billion dollars and then they said oh but you know we don't want to compete with our consumers so we'll give you the sports business and then he turned fanatics into a 20 billion dollar business so you know he he he did he did benefit from uh getting the he did he did benefit from getting the the the biggest guys who you know some of them overvalued the market right

he got big soft bank money you know and and and and and and and and and that helps because money was money and they took whatever evaluation. You know, obviously they had to write down on a lot of businesses, but he got the right money and he has the right connections. You know, he knows the owners well. He knows he knows everyone so well. And so they feel like it's a good relationship, makes it really hard to get into the game.

Do you think it's niche enough that he'll never face, like, anti-monopoly pressure? If he hasn't now, he won't. Last question. Blue Jays are Dodgers tonight. Who you're rooting for? I am rooting for the stars to be the stars, because that's better for my market. So if Vlad Guerrero and Bo Bichet and Otani and all those guys are the stars, that's great. What I don't want is, I don't want a no-name hitting a home run, even though you can say, oh, he's going to grow his own market.

I'm more benefited by the big guys getting bigger. You want the winners to keep winning. Okay, you want to let your winners ride. Thank you for all the context. I feel like I truly... You got love being on your show. You guys are very curious. This is very fun.

I really learn a lot. Thank you so much for coming on. This is really, really great. Let's do it again soon. We'll talk to you soon. Have a great good luck out there. I clearly need to learn a lot more about sports. I've been spending too much time

learning about CRMs. I've been learning about Adio, customer relationship magic. Adio is the native CRM that builds scales and grows your company to the next level. I mean, it is hilariously ironic that we always make jokes about not knowing about sports. And then of course, I try to talk about sports and get brutally mix up a team. Our next guest is Guillermo Rao from Vercell. Welcome to the TBPN Ultradome. Germa, how are you doing? Do you know about sports? I know you know Counter Strike. Do you know sports?

I know a lot about Counter Strike. I know some about soccer being from Argentina. Okay. Right? Yeah. Baseline and Jiu-Jitsu. And Jiu-Jitsu. So I know as much as one class, but I'm ready.

Okay. Okay. Did you get brutally injured in the first class because... No, I brutally overcame because I boxed for many years. Oh, okay. So I had some baseline. There you go. Fight in me.

I got into Jiu-Jitsu briefly, and it's like so... It's an incredible workout. It's so fun. It's like perfectly competitive and skill base. You know, you can like clearly level up. And then I just saw that like every guy in the gym that was over 40 that had been doing it for like 20 years would just be like hobbling in, you know, all these different injuries. And I got, I got a little scared around, you know, destroying all of my joints. But, but I guess if you're in the right gym, it's probably better for longevity. Yeah, we're working down the tech CEO checklist, you know? I love it. AI, Jiu-Jitsu, more to come. Well, we missed you a week ago. We've been bopping around, but give us the latest news in Vursell world. What's the latest on your side?

Well, I mean, the big one is our funding at $9.3 billion valuation. $300. It's awesome. Congratulations. Insane. to miss a gong insane um oh yeah we had a million in primary 300 million secondary led by i don't know if you can hear but this is a good omen the gong is still ringing still ringing still ringing a little bit did it ever stop it never stopped it might never stop yeah yeah you might be back here with

the series g before it stops ringing hopefully uh what's the uh what what what do what do you actually capital constrained on is it just hire more people build a better product are you actually like a big consumer of AI data factories, token factories. Are you spending a lot on inference right now? Clearly not two capital constrained because half of it was secondary. So chapter one of Versailles was providing the developer experience to developers to build pages. I mean, I'm being overly simplistic, right? But the world is going from pages to agents. So it's not that the web services of the internet are going to be irrelevant. But if you think about the cloud, the cloud was born with Amazon Web Services.

The next chapter of the internet is going to be AI services. So Varsel is building the AI cloud. And we felt, and obviously investors felt this way too, that this is worth a very, very big bet. Because every type of software will become AI-native software, and the new primitive of the cloud will be the agent. And this requires new services. It requires new frameworks.

Something that developers and investors are really excited about is our AI SDK. You can think of it as the React of AI. So the most popular package for developers to use to build AI agents, and it's growing basically vertically. Like something we've never seen before at Vosel and we've been doing open source for a long time. Is that driven by new ideas, new companies, startups that are coming up? And instead of spinning up just a website or a web app, they're spinning up an actual agent that then they get new customers for? That's right.

It's mostly net new customers. It's a lot of net new, but also I think a lot of enterprises are going to basically leapfrog their classic digital transformation journey and just go to agents directly. And what is resulting in is a whole new set of demands for the cloud that were never there before. Software is running for a lot longer, which is fascinating from a compute standpoint. Obviously, investors that are excited about GPU compute, but now we're seeing all of these new types of software

that are just running, like, thinking and producing, obviously demanding tokens, but running for minutes, hours, days. We call these workflows. And so we hosted our Versailles-Ship AI conference the other day, and we basically came with an actual definition of when an agent is. And agents basically remind us a lot of the old types of software that people were excited about maybe decades ago, which is like this workflows and business processes

that look like nodes connected in a graph and have all of these steps. It turns out that kind of software, which will now call agent, is actually kind of hard for developers to write, especially the ones that were used to creating the typical SaaS type of software. So Varsall is sort of becoming the default place

where this new kind of agentic workflow is born. How are you thinking about hardware or your own cloud, your own physical infrastructure, infrastructure, it feels like that was crazy to talk about five years ago, but now there's, there are neoclouds. I know, I know people who have just stood up, whole data centers. And so if you came to me and said, yeah, as part of this fundraising round, we're also going to spin up a couple different data centers and we're going to be on our own infrastructure in a few years, I'd be like, yeah, that doesn't sound crazy to me at all. But how are you thinking

about it? Well, Versal is in this incredibly unique position that developers come to us because of the developer experience. Sure. So we can actually automatically make the best infrastructure decisions on their behalf. Okay. And today that's sitting on the both of the hyper-scaters because you want to be multi-clad. So you want to be so much flexibility. Yeah. So for example, a couple of years ago, you could have made the wrong hardware decision

because you were going to say, well, all my CPU workloads are going to be like short-burst, really immediate, CPU-bound, rendering-based. Well, it turns out that a lot of the new workloads are agents. They're spending a lot of time thinking. So you need new kinds of computer, you need new kinds of software infrastructure to support these workloads. And I believe that the hyperscated clouds that still give us the most flexibility there.

Also, keep in mind that a lot of these agents need to be in close proximity to the enterprise data. So, yeah, the neoclows are exciting, but you kind of want to be in the cloud from a security perspective, regulatory, and also latency. I do think there's something to the cloud

that is a bit of a rich, get richer scenario when it comes to latency. Because you need to fetch data. You need to do rag. You need to get actual context into the models, and every millisecond of latency that you waste there is actually very, very costly. So on one hand, you know, proximity to the cloud and being in the cloud gives us a great advantage. On the other hand, because we offer services like our AI gateway, we also make them

compete frenetically. So at NextJ.S.Conf, we announced the NextJAS evils, which is basically what I call sort of the like Oracle of Truth of how good your model is at coding real-world workloads. with things like Next.js. Here's a... Oh, sorry. Yeah, I was going to say, our developers really benefit from the fact that we let them choose.

Yeah. They're not bound to specific model. How are you thinking about integrations with consumer AI apps? I feel like there's a world where I go to Gemini or I go to ChatchipT and I am iterating through a prompt and kind of laying out the...

Let's say I want to make a landing page for a new start. And I've kind of figured out the value props and the customer testimonials and how I'm going to wireframe everything out. And then I want to actually deploy it right now. I don't think I can do that within any of the consumer grade chat apps. They're going to kick me over and say, hey, go set up this account. But there's a world where that doesn't need to happen, but it's a delicate balance because you don't want them to be like, now we're on top of you and fully. And so how are you thinking about like onboarding, like almost like prosumer creators? Yeah.

So on my blog, I drew this parallels between the Cloud 1.0 and the AI cloud. And so you have, you know, the poster child of frameworks that React and XAS, now it's going to be the AISK. Before we had CDN for delivering pages and images, now we have the CDN of tokens, the AI gateway. And one of the other big transitions that's going to happen is, article. So we're going from HDB to things like MCB. MCB is really exciting because if you look at how ChatGPD is now allowing you to embed apps inside of ChatGPD that also just came out a couple weeks ago. I went to NextJAD as I was able to deploy Doom inside of ChatGPD.

Why? Because we have an MCB integration that allows our customers to embed themselves inside of ChatGPD. This is really exciting. If this takes off, this is the world that you just described. But right now is there a app? Yeah, does this open app have like a review process for that? Because I haven't just like run into the, I do run into those like hydrated UI elements every once in a while. I'm asking for like movie reviews.

It'll show me some pictures or like links. But it's all kind of like pre-baked UI. Yeah, you need to be working on creating your own studio Ghibli moment around that. Yeah, this feels really, really big. I feel like it's right for it. So yeah. Yeah. Tell me more about like how this actually rolls out.

Yeah. So on a pericle level is MCB. So as long as you have. something like for sale, you deploy your thing, it can speak MCB, you're embedded into chat GPD. I deploy my app in developer mode, which means like the lady who basically embed everything. It's going to be a little bit like an Apple App Store where you're going to have to get approved. But it seems like, you know, their guidelines are pretty open. So I think a lot of it will fly. And to your point, this allows ejection points from chat GPD into other kinds of

platforms. So you could imagine one where if you start up cooking up an idea, it recommends V0 because Versel registered an application with them, and this is the right application for prototyping and vibe coding. In fact, the other day, while I was in developer mode, it started recommending things to me. So I guess some of the early adopters of their apps platform. I believe MCB will continue to enable this sort of like agent-to-agent communication. This is already very big when you're coding with agents, because in order to, you know, for example, a good example would be Figma converted to V-Zero, so you can convert from Figma to NXHS app. that's going to speak the MCP protocol.

So I think as far as the AI cloud goes, enterprises will be wrestling with these questions, right? Like the most common question I get is, how do I not get this intermediated by Google AI overviews, right? Which is sort of summarizing all of the results of the search pages, and how do I not get this intermediated by chatyPD? And I think they're really good answers. How do you think about M&A?

You don't have to talk about M&A interest in Vurcell because I'm sure you're turning down stuff all the time. but how are you thinking about, you know, MNA in terms of expanding and building on the existing platform? Yeah, there were so many investments being put into the AI categories, right? And vertical agents, coding agents, IDs, things like this. I think, you know, there's going to be more consolidation in the coming months and years. There's like 10, 20 different options for anything you can conceive AI-related, AI for QA, AI for testing, AI for this.

So I think creates a lot of opportunities for Versel, which is a very broad platform with lots of users to sort of make some strategic MNA. But I'm even more excited about our marketplace. So we announced at Vercelt Ship AI our marketplace for one-click install of AI agents and AI infrastructure services. So I mentioned if you're going to sit down

and create your own agent, it's very likely we're going to need some services that Vercel doesn't offer. One example is browsers on them. I think you guys have had a couple browser CEOs here on the show. And so for sale doesn't offer browsers and service built in, but now we offer it one click through the marketplace. Sure. Another example is security audits that AI agents can help with.

One click install services like Corridor. Because for sale contains the deployment URL and we have access to the code and we have access to the production data, it gives us a very special place so that we can grow a massive ecosystem of agents. that automate the cloud. We're going to create some of those agents. We might acquire a few. But overall, I'm very excited about becoming sort of the nexus where all of the agents are collaborating together.

So they take the burden off of maintaining the cloud. We also, the outage the other day with US East 1, the internet melted down. You know, all of those pagers or mattresses were waking up people. Our vision for the future of the cloud is that they wake up agents. The agent has the first pass of like, oh my God, anomaly alert, 500 errors.

All right, let's have the Vercel agent do the first pass. But this is also true for, you know, you vibe coded something and you want a security engineer to review it for you. Well, now that's an agent. You can one-click install through the Vercel agent marketplace, which is very, very exciting. Yeah, that's awesome. I mean, what an interesting place to be. You have so many different pieces of the business.

And I imagine that a lot of them are, you know, growing. business lines, I can imagine the series F process was a lot of fun. Yeah. Given all the booms. Jordy, anything else? No, this is great. Great to get the update. Yeah, thanks so much for hopping on.

Have a great rest of your week. We'll talk to you soon, Germa. Cheers. See it. Bye. He mentioned eight sleep. Let's talk about our sleep scores. How'd you sleep?

We need to... We need to talk to Mateo, and I don't want this... I don't want the John handicap. You have a conspiracy theory that, like, for some reason it just gives me better results. I got a 98 last night. Let's go. Seven hours, 44 minutes. Our next, our next guest, Brendan from Mercor is in the restream waiting room. Now he's in the TVP and Ultronome. Welcome to the stream, Brendan. Congratulations. A $10 billion man. How are you doing? Called his shot. He's the biggest shot caller in all of Silicon Valley.

Yeah, we got to get some new shots. Yeah, I was telling him this. I was like, you need to call the shot. You're holding up the global economy. within five years. Every company is dependent on you. Like, there's no, there's no shot. This man won't call. But give us the news. Break down exactly what happened. Tell us what is new today. Yeah, I mean, the company's been growing like crazy. We're now paying out over one and a half million dollars a day to experts in our marketplace. Hit the gone for that. And what else? And we've raised our Series C of $350 million at a $10 billion value. Massive.

$10 billion value. Yeah. Not bad for your first startup. The gong is really going. Something's up with the gong these days. I don't know if you can hear it. It's ringing. Let's get right in.

You had a, I feel like I remember you had a blog. Was it blog post Friday? Do I have that right? Is it worth getting into that? About the big things. Yeah, the big things. Let's talk about the big things. Yeah, it's really starting with the question of what are the things that we know are going to be the same about our business over the next 10 years?

And that there's a lot that's going to change over time. But we know that the three main areas that our customers will always want more improvement and progress are in more candidates on our platform, better matching and faster delivery. of all the candidates that we're hiring for them to improve frontier models. And so those are really how we- Bezos inspired? It is, exactly, yeah. Similar to how Amazon said

that there are three big things were more products, better prices, faster delivery. For us, it's a little bit different because no two people are the same, but inspired by it very significantly. Totally. Let's go through them, I think.

And then I guess, like, want to understand like want to get kind of more into the into the future too i think i think people uh some people would have a sense that like there could just be this you know incredible uh demand over the next five years and then you know does that demand evolve evaporate but i i want to i want to hear how how you view it maybe more specifically totally what we've been seeing is that there is an enormous trend towards humans training agents how to automate redundant workflows that they would do in their jobs, because it's structurally more efficient to do so. Instead of a customer support representative, redundantly responding to hundreds of tickets, they create an e-vel to train an agent one time how to do that thing.

Instead of a banker redundantly analyzing data rooms, they create an e-vel to train an agent how to do that thing. And we believe that a huge portion of the broader knowledge work in the economy is going to converge on training agents, how to do these activities. activities and our marketplace is really at that frontier of creating this new job category and training agents. How does this look long term if one of the big AI platforms becomes really, really dominant understands what everyone does and can just ask me to do a task within the app? Is there any sort of disintermediation risk?

Or do you think that what you're building with Mercor is special? enough that even if, you know, the open AI chat GPT app can detect that, oh, well, you've talked to me enough. I know you're an investment banker. I'd love to do an RL environment with you. Here's an offer directly to come and do some training data. How are you thinking about counter positioning against that risk or is that just like not a thing at all? Well, one of the most important things in building out these environments to train agents is that they need to be very consistently reliable in that you need to put a lot of work into them. And if there's noise and that there's some mistakes or issues, then it's very difficult

for a model to learn from. And so as a result, it's really difficult to ask your users to say, put in 10 hours to helping to see if we made a mistake in preparing this financial model for you. And those users might flag a mistake, but that's incorrect half the time. And so what they instead need to do is build out armies. of experts that are able to diligently analyze model performance, go through strict review processes to ensure consistency, and build these RL environments to adopt models for any specific use case. So you obviously have massive diversity on the talent side. But how do you think about customer

concentration over the next few years? Do you think that, like there are clearly power law winners in AI. There are trillion-dollar labs, you know, Google and opening eyes, half a trillion dollars. There's huge winners. And then there's a ton of startups. How do you see the shape of your business? Do you think you'll have a lot of medium-sized companies or just tons and tons of small companies? And every individual company will need to come to you for talent to optimize whatever little thing they're doing or will you be doing more work with the really, really big labs? How do you think that all shakes out? I think it'll be some combination. Definitely near-term, it looks similar to NVIDIA.

And so far as working with a dozen of the hyperscalers that are investing huge amounts in this. But over time, what we're seeing is that every enterprise wants to train agents to customize their specific workflows. And we have the expertise to help them do that in building out all of the e-bell sets and environments that correspond to every workflow in their businesses that they want to automate. what any any I had an eye-opening experience yesterday I have a friend who's a lawyer and about a year ago I started I started asking him like what kind of AI tools they're using and he was like broadly like not impressed he was like we're like signing up for them they're not getting a lot of value but it's interesting I'm keeping an eye on it kind of thing and then yesterday he said to me I've seen the future. Harvey isn't perfect, but has better attention to detail and is more thoughtful than almost any junior person at our firm. I've watched to do 100K of associate level work in 10 minutes.

And that felt like just extremely notable to me to just see how hardy, like, 180 on that is like how, like, what is the structure? And you can talk at a high level to not give away specifics, but like how much investment is like going into just that category alone from your view? I mean, we're seeing a huge amount of investment in law, certainly. I would say the top categories are probably software engineering, then finance, then law, maybe close between medicine and law. But I think that the models are very quickly going to be able to do even partner-level work at a lot of these firms. And so that transformation and how it impacts businesses and the economy is going to be really profound.

and obviously all of the data and e-vals to enable that is one of the primary blockers to getting there. We have a shout out to one of your employees, Varat Talwar. So just wanted to say, congratulations to him. How did you do, Varan? The team. The chat is just shouting him out. Obviously, it's a big milestone. So we want to celebrate everyone on the team.

What, uh, how in conversations for this last round, uh, obviously the business has so much momentum. I'm sure people are just like, take my money at all caught, you know, whatever. I just take it. I'll send the money and we'll figure out the docs later. But how big can Mercor get? Like what is like, if somebody asks like, okay, if everything goes right, what is his business look like? And does it, does it look like, you know, you're paying out on the levels of like an ADP, right? Is you become like one of the largest pseudo-employers? in history, like, what's like the, what's the craziest outcome? The way I think about it is that right now, businesses are spending about $40 trillion a year on

knowledge work, all for people doing these largely monotonous things that have similarities from task to task. But all of that is going to be transformed to training agents to automate workflows. Instead of doing it ourselves, we'll have agents do those things. And Mercor is building the infrastructure to enable humans, to engage with models, to teach them, to fit into the AI economy. And I think there's an opportunity worth tens of trillions of dollars a year to help that transformation happen over the coming decade. And so we're only 1% of the way there, if that, and very exciting for everything to come. Are you a foodie? We have another question from the chat. Do you like fine cuisine? I have a foodie. You are very

foodie. It's very fourth show coincidence, yeah. That's good nominative determinism. We're very happy to hear that. Last question. When IPO? We don't have a specific date in mind yet, potentially on the horizon, but we'll be sure to keep you out guys updated.

Let's go. Potentially. Let's do an air horn for potentially on the horizon. Somebody's probably going to write an article on that, by the way, so I'm sorry. This is a new thing. I see it all the time. every time they interview Palmer Lucky and Blonde Bloomberg, they ask him, like, when are you IPI? And I'm like, you know that he would tell you if he had news there. Like, why are you even asking?

But I guess it's just fun to ask. And then in a couple of years, we can look back. Potentially on the horizon is much different than we have no plans, though. Yeah, yeah. I mean, you could have said absolutely not. I would never, I will never take this company public in my life. You didn't say that. So at least we got something. So there you go. Progress. Anyway, massive congratulations on all the progress. really, really remarkable growth. I mean, what a fun time. What a fun business. And congratulations to everyone on the team. I mean, absolute rocket ships. Cheers. Have a great rest of your day. Later. We'll talk to you soon.

Let's get back to the timeline a little bit. Let's get back to public.com investing for those who take it seriously. They got multi-ass investing, industry leading yields, and they're trusted by millions. Yes, let's go back to the timeline. A lot of people have been focused on this. this chart overlying stock market performance over job openings and how it basically diverges. At the introduction of ChagipT. Basically. That's where people like, which is also the introduction of high rates.

Yes. And I think, I think we need to put Tyler on this. Is this a chart crime, Tyler? Do you think that there's some funny business going on? This is a pretty simple just, I don't think that, I don't think that it's literally like the lines are wrong.

Mark at A16Z broke this down. He said this is particularly trippy chart crime. The break point was exactly when the COVID stimmy bubble pop and the Fed hiked interest rates to the moon. He says, The great forgetting that COVID ever happened is wild. I can't tell if it's adaptive or suicidal or both. But yeah, I think everyone wants to memory hole this one.

And so what is exactly on this chart? I don't even have it pulled up. It's a number of jobs versus financial performance or something like that. So Gavin Baker here says, really interesting chart from Derek Thompson surfaced it, former guest on the show. Stock market and job openings were highly correlated for decades. No longer, stock market is up 75%,

and job openings are down 33%. Would think at least partially due to AI, Chimoth Polyhapatia chimes in and says, it could also be that the plurality of the stock market is now highly concentrated across seven to eight companies who on a relative basis just don't employ a lot of people. So the stock market goes up, won't pull up employment along with it. I still think what has been happening to date is CEOs are desperate for every possible AI narrative. And if they need to do layoffs because they got bloated during COVID, they will say that we're getting efficiency out of AI.

And that's a much better reason than like we overhired. Remember, Microsoft has done a number of different layoffs this year. There's a layoff that got announced today. Well, we should get Derek Thompson back on the show. We should also get a couple other economists. Maybe Tyler Allen would be there. Amazon is targeting as many as 30,000 corporate jobs beginning tomorrow. So probably not a fun, not a fun day over at Amazon HQ.

Some more positive news. Timothy Mellon, a banking error, is said to be the anonymous donor, not so anonymous now, I guess, according to the Times, who gave $130 million to the U.S. government to help pay troops during the shutdown. This guy doesn't, according to Wilmonitis, worth $14 billion personally guaranteed payroll for the armed services, and the only and most recent photo of him is from 1981. I mean, this photo looks so cool that I kind of understand how he was like, all right, that's it. yes i took the only photo i need for life yep i won't do any photos because this looks i mean suit the tie with the train in the background yeah you know with the with the the hair the stare

i mean yeah he he just he one-shoted photography he did it's remarkable because when you look at the photo and then you hear the headline it's like meet the man who backstopped the uh and paid for the army's payroll in a time of crisis with his fortune you're like Oh, yeah, like this guy was probably crucial in World War II based on the image. And you're like, no, this happened like last week. He's talking about the 2025 government shutdown. That is what they're talking about, which, and I just keep, I just, every time I see this image, I think, like, oh, this is a... Mike in the chat is asking, $130 million is what percentage of monthly payroll?

And that is a good. Yeah, I have no idea. Well, maybe he levered it up. The fiscal year, 24, military personnel category of the DOD budget, which covers pay and retirement benefits, was about $192 billion. Hmm. So, yeah, I don't think the $130 million is going to... It's a good start.

We've been on gambling this entire show, but I'm just thinking about, like, what if instead of government bonds where you earn like 4%, the government just released, you know, a one massive lottery ticket or something that would be... It costs a billion dollars,

but there's a 1% chance that it paid, pays out a hundred billion dollars. And if it does, then the U.S. government is indebted to you for like hundreds of years. But if it doesn't, if it doesn't pay out, then the U.S. just gets a hundred million dollars. I think you basically need like seven-ish billion dollars every two weeks to keep the payroll going. So what does $130 million do? It's like a day. Yeah. Mike, Mike didn't.

8.8 billion per month. So I guess, I mean, it's a good start. I mean, there are some divisions of the government that have basically savings that they can kind of tap into. And then also there's a big question about how much are you actually going to, like, do you need to bring everyone back on payroll? Or can you get by with 10% or 20%? And maybe this is something that he's backstopping and saying, hey, let's pay 20% of people instead of 10% of people. what does polymarket say about the government shutdown let let's pull that up uh government shut down uh will the government shutdown end by um what is the most likely November 30th by December 31st

it's a 93% chance uh November 15th is like a 50 50 mark basically so expect another two weeks maybe another three weeks according to polymarket so uh there's your there's your polymarket update People are still talking about double speed. The, uh, lets you control thousands of social media accounts with AI, ensuring they look as human as possible. Never pay a human again. Control is all you need. Uh, Jeremiah Johnson says, every startup is like, we found a cool way to monetize undermining the social contract with, uh, 30,000 likes. This is, I mean, we follow, we follow this person or something like that. Um, this person's like a liberal, they should be pro tech and they're, it's not like this person is

like by default, some like crazy anti-tech person. And so, you mean by, you, Jeremiah Johnson. Like, Jeremiah Johnson is, is not like some like tear it all down anti-business, you know, and, and he's making a good point getting 30,000 likes. And so I think this is a harbinger of like what the shape of the tech lash will look like. Well, I think in this case, even tech was not excited about this one. Tech was lashing itself. Self-flage-lating. Self-harm. No, but tech, tech is saying, like, you know, broadly, this doesn't seem like it needs to exist and probably is a net negative, even if there's some, some legitimate use cases. Anyway, you know what is a casino? Last, last, last, last, last, last, last, last, last last last, last last last last last last last last last last last last last last. running an out-of-home campaign with adquick.com,

out-of-home advertising made easy and measurable. Say goodbye to the headaches of home, out-of-home advertising. Only ad-quick combines technology, out-of-home expertise, and data to enable efficient, seamless ad buying across the globe. Zumer went viral again. This guy is just harvesting hatred

from the entire world at this point. He said on a few days ago, at this point, I've lost all motivation to post on X. The new algorithm is so demoralizing. I have 30,000 followers in him getting 500 views in an hour. I'm done with this place. Well done, Elon Musk. And everyone and their mother was dunking on him and saying, you just got paid $10,000 a week ago to post a picture of Steve Jobs's daughter. He didn't take the photo. It's not like he's a photographer. In this post by ADIS,

Elon, Elon, I need my dopamine dollar-redues. I posted another hot girl. Where's my engagement? Elon, Elon. I'm a content creator. So anyways, not a lot of support for Zoomer, but I guess, I'm sure he'll be back. I doubt he, I doubt he. Okay, we need, we need Tyler to give us a review of chadlabs.a.I. Apparently, I didn't even see this, but there's a YC company that is a brain rot IDE, something like that. No way, this is real. You can, you can, is this real, is this actually real? And a cashback IDE. It seems like a parody. This is for sure somebody just making it.

This seems like a joke. But we need you to get to the bottom of what's actually going on. It's in, I mean, it's a real YCE company. Yes, but is it a stunt? Or are they actually trying to bring Tinder into your IDE?

Which is like, I mean, I'm on the website. So for everyone commenting, learn what I'm on the Y Combinator website and Cloud Labs says it's the brain rot. IDE. It's a fall 2025 bat. Oh, it's clad labs. I thought it was Chad.

No, it's clad labs. Founder of Chad IDE here, cladlabs.aI. So it's not Chad Labs. It's clad labs. Yeah, I mean, I think this is just called Zoomer. This is just Zoomer marketing. They do call it the Chad IDE, the first brain rot IDE. They say, sup giga chads. This is on the YC websites. Subgigga Chats, we're Richard and Kevin, founders of clad labs.

TLDR as highly requested we built the Chad IDE Chad integrates your brain rot vices with your agenda coding workflow Chad increases productivity and reduces friction of context switching so Kevin Kevin is kind of saying like I thought the problem and solution was pretty straightforward crying emoji not sure what people can't understand about that so he's kind of like leaning into this it feels like this might be rage bait the product branding is an epic rage bait

but this is on the YC website they have a screenshot of the product where they have Tinder in one panel and then the IDE and the other panel. And so this is the modern thesis. They're saying you can you can rage bait your way to... Yeah, so you can rage bait your way to actual usage. I'm surprised that YC is putting this gambling IDE on their website, I will say.

But maybe they're just not going to form an opinion. No, no, I think there is a open debate amongst a lot of people like respectable respectable venture capitalists around can you launch with something that's rage baity and then pivot into something

that's serious? Can you build a serious business with a rage baity go to market? And I know you I know you're against this. I've been working on my next essay. Yes. Rage bait marketing is for losers. I like it. Which is that I think when you

do rage bait marketing, It obviously gets a lot of attention, but it makes it so that you make the entire world prey on your downfall. And I think that that is spiritually a bad strategy. So I'm still working this one out. But I don't, you know, I don't think that, I don't think that rage bait marketing helped clearly win the enterprise, right? Yep. But there were a lot of VCs who invested. Maybe it worked at your students.

And the story's not over. But there were a lot of VCs who invest in and said, hey, look, I mean, we talked to Brian Kim from Andreessen on the show about this. Like, why did you do the deal? And he was like, I like that they were able to get a lot of attention. And that gives them an opportunity to deliver a product. That gives them the opportunity to go build something. And he was like, look, like, we don't know if they have necessarily.

But at the very least, it's gotten attention. They would do the pictures of like, just spent $10,000 on this suit. And you could tell that it was off the rack or whatever. Yeah, yeah, tons of stuff. I just don't think it's. I don't think it's a good strategy, and I think we're going to be, that's going to be proven correct. I think, I think you're correct in the, in the spiritually risky strategy, because how much better is, is it to just go and, and actually build a beautiful, optimistic brand and

pay for some billboard ads or, you know, even, you know, direct response ads? Beautiful thing. You don't even need advertising if your product is, is, you know, Advertising is an accelerate. Let's say you do need advertising because you've built a product but no one knows about it and you have two options. One, like the dark castle and the bright, you know, the fork in the road meme. One of them is you rage bait everyone, you make your product annoying and then everyone will have to check it out and the product's actually good versus, yeah, you have to shell out

some cash to run some ads, but you slowly compound and you slowly grow. After a decade, you have a product where everyone's like, wow, that came out of nowhere. Those guys have just been running great ads for years, you know? Yeah, I just think you're ultimately, Chad IDE is competing with Cursor, Winsurf, GitHub, co-pilot. They're competing with serious companies and what kind of, like, you're going to, you're going to attract unsurious developers. This is the lack of, this is the lack of the April. Fool's Day joke. This company could potentially have just been this should have been an April Fool's feature or WinSurf's April Fool's joke and it should

have just been maybe I don't know exactly what it is we got to dig in and understand what this is but there's a whole world where the idea of a of a brain ratty IDE is something like Google does as a side project as a joke right yeah like they used to do stuff like that and some of it like they had a whole they had a whole didn't they have a whole feature that was just like talk to your dog or something I don't know there's a so many. There used to be a rich culture of April Fool's jokes where you could effectively rage bait

the world, but everyone was in on the joke, and now we're in this world of like you rage bait everyone. Rage bait is your whole brand strategy. It's your whole brand strategy. That's where... And it's really hard to pivot away from. I don't think anyone's done it successfully. Yeah. It's very, very hard. Well,

on that note, we have to get on with the telephone. Get bezel.com. Your bezel concierge is available now to source you any watch on the planet. seriously any watch and also wander find your happy place book and wander with inspiring views hotel great amenities dreamy beds top tier cleaning and 24-7 concierge service it's a vacation home but better play the double-kill sound effect for two sponsors thank you to everyone who supports the show thank you love you to everyone in the chat thank you to everyone who's listening to

the show we will see you tomorrow live with satchan adela the CEO of microsoft cannot wait can't wait have a great rest of your day see you then goodbye cheers folks