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

You're watching TVPN. Today's Monday, October 27, 2025. We are live from the TVPN Ultradome, 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, whole lot more. No, 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 Satya Nadella, the CEO of Microsoft, live on tvpn. Tune in, get ready. It's gonna be great.

To be clear, that is a real picture that you just saw on your screens. It is real.

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 doing it live. I cannot wait.

Yes, and we will be of course leveraging restream. One livestream, 30 plus destinations, multi stream. Reach your audience wherever they are. We take Restream 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? 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. And Harun is enjoying it. And if you scroll the replies in this viral post that got over a million views and 4,500 likes, Palmer Luckey says he's using it. Gabriel the. The OpenAI Sora. He's not the Sora lead, he's Sora Research. He says I'm paying. It's an incredible product. Dwight over 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 lay out specific ways that they use Suno. And so I wanted to kind of revisit this noodle on this, understand 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. 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 at 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 that 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. 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 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 artist, like a rapper, let's say, make this, like, make this future song. Make the jazz version of this future song.

I saw DMX X gon give it to you, but in 1960s jazz rendition, and 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, restyle 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 values.

The good thing about that is that 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, ip, right? Like, there's no company called.

I think that they might. I think there'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. The question is, like, if you purchase Spotify, if Suno is getting 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 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 on, on to 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 rails securely spin up white label wallets, sign transactions, and integrate on chain infrastructure all through one simple API. So Chris Dixon, who actually was on the show last week, 10 years ago, he posted a blog post called Come for the tool, Stay for the network Classic. 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 and old school instead of just actually terrible. And so photo filtering apps were very popular. There was one called Hipstamatic, but it, 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 multimillion dollar professional content producing cohorts that are producing like the greatest content at 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.

We have some.

But I'm not a, I'm not a full time creator. I'm not trying to monetize this. I'm just having fun with, with the expression of like actually making music. Right.

That's the second bucket that's gonna piss some hardcore traditional musicians off who are gonna say you're not making music, but.

Effectively, of course, of course, of course.

Playing around that you were making music.

Yeah, 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.

Gary Startup making a launch video.

Yeah, exactly, exactly.

Historically they would be like go to stock music, 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 to Instagram being like, yes, I'm a professional photographer, I did a wedding shoot and I need to filter the photos. I'm gonna use Instagram, they were like, no, I'm good with Photoshop. Photoshop, Photoshop's fine. And no one was going there being like, yeah, I just wanna look at something great. I'm not into this whole 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 percentage now? Are most of the people there because they want to be. 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 live in tool land forever or will you eventually, will this become a network?

And 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 Matt. Undeniably magical to be able to generate a song. I mean, I remember growing up being into music. It was hilarious. And my, and my dad is a musician and my dad would be like, I'm gonna 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, 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 in, 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, etc.

We're just listening to them themselves. They're just.

Yeah, so that's the thing. I don't. What. I really don't. What I'm. What I. Again, we. We should 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 Jess being late. Make a jazz song about watering my plants.

I think, yeah, I mean, I think you're seeing something similar to make a lot of electronic music.

Quitting your job.

When I was a kid, I was able to actually steal or torrent a product called Fruity Loops, which was a digital audio workstation. So there's Ableton and there were a few others of these 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 kind of like GarageBand esque. 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. And what does that do? Well, that lowers the. That lowers the bar, the barrier to entry. And so you get a lot more. 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 the people who are going to go see Marshmallow Live do not care. Or the people that go and see anyone at Coachella or something, 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 gonna be able to rise up. There's no one. If you're truly talented, there's never gonna 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 gonna run up the score with cool, creative things. The question is, like, is, yeah, does this become more of a video game market where people are playing Suno, like a game for themselves and maybe for a few friends, Like a. Like a Minecraft. Like, you build in Minecraft. You are creative in Minecraft, but you don't often sell what you build. Like, you don't often directly monetize.

Yeah. And so 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 or coding crazy by Future as in jazz. 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.

Put it in your heavy rotation and.

Put it and listen to it on the drive home.

Yes.

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. Now, I can imagine the use case as a parent of, like, generating a song, like the cleanup song and putting your kids names in it.

That's cute.

And you make. And it's like, hey, every day after dinner, we're gonna play this and you're gonna put, you know, whatever.

That's actually hilarious. I like that.

That's very.

So that's something that I could come up, come back to multiple times. But a lot of the use case make a country song about Jess being late. Like, I'm sorry, that's Like a funny moment and maybe you send it again when. When Jess is late. Yeah, 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 today?

Yeah, I mean, I think current songs are 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 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 would 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.

The quality bar for music is actually, like, quite high to get people to actually start listening.

Totally.

Yeah. Could you actually play the bongo drums while you talk?

Well, I need to unmute. I can only play with one hand, 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 a producer.

Okay, so I think this is like two or.

Yeah.

Okay.

If you look at, like, short form, like video, they're only like 10 seconds long, right. So it's like very long.

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. That's ridiculous. He has a dead man switch.

A dead man switch for Tyler.

Hey, 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 setup. 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 to Fruity Loops long term or Ableton long term? Or is it a competitor to 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, 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. There is gta, there is Red Dead Redemption and Madden and stuff and there Call of Duty. There are some huge franchises. But there are way more base hits in video games than in social networking. In social networking there you have the meta properties in the trillions. Then you have YouTube, LinkedIn, X, TikTok, 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 you can just have base hits.

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 OpenAI is working on an AI music product.

Yes.

And I would expect that OpenAI eats the category of music as memes, which I'm talking about, which is like you 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 prosumer working with creators, et cetera. Yeah, 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 similar to their coding API for GPT5, like it's just, you know, frontier level, maybe that gets baked into Ableton and, and Garage Band and, and other.

I was just thinking for the category of what I'm trying to understand is what percentage of songs are just jokes.

Yep.

And I think that Chat GBT will ultimately do that pretty well.

Yes. So, yeah, I'm trying to. I'm trying to think about the.

And so I don't.

Images in ChatGPT versus Mid Journey, like, Midjourney has been able to. Images in ChatGPT did not kill Midjourney by any means. Like, the Midjourney 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, the choices that David Hull's 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.

So what you're saying is Midjourney 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 artist do a house song. It's style transfer versus John.

I think that might. I think that might come over to ChatGPT 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 gotta go to Midjourney and get my srefs prop, like in shape. What they do is they just say, like, oh, I'm already in ChatGPT, like, let me just see if ChatGPT can do it. And like, even if Midjourney was a little bit better, the casual user is just like, yeah, ChatGPT 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 daily driving a ChatGPT 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, they're in midjourney, in the discord, getting ideas, sharing srefs, really pushing it to its limits. And I think you'll probably see the similar thing in Suno 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 ChatGPT and they will say, it looks bland, it looks sloppy, like, ours looks better. And I imagine 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 ChatGPT 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 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 new front.

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 certainly the coding agent war. And then now the music.

OpenAI never met a war it wasn't interested in fighting.

Well, let me tell you about figma.com, think bigger, build faster. Figma helps design and development teams built great products together. You can get started for free.

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 tweet 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 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 Gunna.

Yes.

And two years ago somebody made an AI album for Ghana 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 he's not always enunciating words super precisely.

Yeah. If the underlying inspiration is a little bit sloppy, then AI Slop on top is just fine.

Yeah.

I'm actually kind of excited. I know 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 TVPN 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 fal. And you can run them in parallel.

Okay, so you ran everything in parallel.

Yeah.

And it was like 16 videos. It still took, like, 10 minutes maybe.

Okay, I have a. I have a song.

Okay.

Gonna F. You mean 1982 Soul Funk. 1982 version AI cover.

Okay, let's play this.

Let's play this.

I think we won't get demonetized because of this.

That's a beauty.

Oh, this is. Okay. This is gonna.

Me.

It's pretty good.

Production team.

Everyone's jamming.

The supporting vocals in the back.

I even heard the original song. Like, I. I can't immediately clock the original song, so.

All right, we can pause it, but.

This is pretty good.

That's good.

I like. 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 ton 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 producer.

Yeah.

Probably can clock it as slop. 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 there's six fingers, you're going to notice. But in music, if you're sticking to general chord progressions and you're sticking to general key and not completely slopping up the basic rules of how to generate music, you should be able to get something that. That. 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 streaming side almost, where if you think of. There's the Spotify dj and it'll kind of update a little bit if you like, depending on how long you listen to the song.

DJ X. Yeah. Sometimes I'll skip like five in a row and it'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. In that point in latent space.

Oh. And then generates it.

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 has got to do something here. That djx is like 2 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 launch 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 it basically shows, like, you need to be constantly touring if you want to make money as an artist. And so if you introduce AI versions of their music or music that is just decent music, it's just going to hammer. Okay, so 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 ten year strategy. You're the CEO of Spotify for the next five or ten 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. Vanta's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation, whether you're pursuing your first framework or managing a complex program.

Well said, John.

What do you say?

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 Gunna 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 the track that inspired it, in that case it's extremely obvious because it's an existing Ghana 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. 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 real artists on the platform. And I don't know how you solve that. I mean, it's potentially a bull case for Spotify because they take. 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. It 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 artist, if I'm gonna. And somebody's generating AI gonna, I'm cool with that. If you're getting paid, 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 Ghana song that sounds like this and there's no underlying song. So how do the royalties work on just like a generic Gunna song?

Yes.

I don't know.

I have a rebuttal, but first let me tell you about graphite.dev code review 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 and then it generates that music, you can use 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 are 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 cut, our take rate, whatever, 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 Gunna singing Taylor Swift or Taylor Swift singing Gunna, well then they're both getting 50% of that new generation.

Now we just got 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. So pushing.

What does pushing mean though? That's the question.

It's exactly what Tyler just outlined. Right. They're deejaying 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.

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 the 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 to the top of the feed. I would say the latter. I would say that it's not like.

There was any Spotify is historically less algo driven. Right? It's like we have this, we have this like basically Billboard and the homepage and we're just gonna put stuff there.

But they have an algorithm and the question is if there's. I think there's a big difference between them actually choosing to generate music and them allowing anyone to come with generated music.

Well, they're already doing that.

Yeah, yeah.

There are already partners that are. AV says Spotify should launch a standalone audio Sora or Vibes that could be interest. I feel like they're so any, any business that is reliant on Hollywood or the music industry is just so sensitive like actually has this at scale business doing that is so sensitive and they're.

Yeah, yeah. Spotify is in a way way different place than OpenAI which doesn't have 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 happening.

John Philippines says Spotify buys Suno. I could actually see that.

Well, if you need to generate some media, head over to fall the world's best generative image video and audio models all in one place. Develop a 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.

The Travis Kelsey, Travis Kelce 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 Kelce 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 Chiefs tight end just joined an activist investor campaign that's targeting Six Flags Kelsey. New York 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 dippin 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?

This is the classic investment strategy. Just buy buy buy the buy the stocks of the companies that you spend money with. Right.

Do what you love 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. AI Julius, what analysis do you want to run? Chat with your data and get expert level insights in seconds.

What's Kelsey get on Julius?

You actually probably that would be a good first step. I know it's a sponsor but you.

Actually should start take so it's notable. It's notable Looking at NFL athletes looking at their investment strategies. Kelsey obviously going the activism route.

Yes.

Saquon going the fifth Avenue route.

Yes.

To buy fifth Avenue he's he likes co investing with the founders funds of the world with the thrives of the world. So 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 I'm super qualified to. What are your ideas, though I feel.

Like there is potential for a comeback. 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 in dreams 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 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, extremely 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 it's 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. The fastest steel roller coaster. There's all these different permutations. And you can go collect all the badges, but it's not meaningful. Like, it's like five feet taller over.

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 the roller coaster was going really fast and it was really high, and then it was going to go really low, and then it 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.

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

I like it.

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, something like that. But you could 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 of 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 gamble on. Will people make it to the end? And 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 going.

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 gonna be on this for two hours now, and it's just gonna be clinking along just one inch at a time. And you're just like, yeah, I caught the raw end of the stick. I got the short stick.

So another one, like, lottery upsells for every ride. So.

So we're basically just. I'm just trying to Pure gambling right.

Now, just trying to juice their margin. Okay. You got people going, how do we get more dollars out of every rich? 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.

Yep.

You can do the numbers. There's. You can run the numbers. There's. There's only a few hundred people in line.

Yep.

You got a good. You got a not. 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 degen. Let's go to Tyler. But first let me tell you about Turbo Puffer search. Every bite, serverless vector and 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, while they're the line doing stuff. So I think data labeling would be good.

That would, that's actually great. Yes.

This kind of scaly I route, I.

Would take it a step further. You know those like big like massive LAN 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, that could be a.

Way to upgrade another opportunity. Wellness has been booming. Right. While theme parks have been in relative decline. How do you get wellness? And so the obvious thing is people have been promoting the 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.

So you can get a facelift. I'm not familiar with this at all, this whole story, but people were.

Wait, so pull up this picture. I think they're trying to.

I mean whoever this dude is is getting Dr. Ben Tale. He's a 60 year old who runs a solar and tech business. He had an advanced deep plane facelift and neck lift with SMAS optimization as well as other Procedures.

Well, he looks, I mean, he looks great. He's looking great. I think this is a, this looks like a good ad for facelifts if you ask me.

Yeah, well, Keith Raboy, friend of the show, got a nod in the. This article. Really. Tech is a young person's game. In a 2024 talk, Keith Roboy 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 facelifts, neck lifts 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 cue. Like the baby, you know, the baby with the single. Single hair.

I need the cartoon baby.

Yeah, the cartoon cartoon baby. That's the way.

Yeah. You would have to get a much. You could get a face lift, but then put it. Make your face a lot wider, like round.

You need to, you need to get on some sort of diet that, that gets the baby fat back on instead of working.

Oh, that's just filler. Tyler. Just gets so much filler.

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, anyone we know all the people in the photos.

Yeah, but are these people people know in tech?

I don't know, but they're, they're tech on theirs.

If you scroll down, I mean, you need to, you need to ID these people.

Oh, unnamed patient. Okay.

Men. Used to.

They're saying it's a 46 year old tech executive. Tyler. Figure out every 46 year old. 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 ChatGPT.

That's right.

Get your plastic surgery firm. Mention it on ChatGPT. With profound reach, millions of consumers who are using 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 Chief's battle 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.

Yeah.

This is a. I think this was a quote on GameStop's post.

Yeah.

Where they said, you know, we're announcing the console wars are over.

Yeah.

The Halo is coming to PlayStation.

That's exciting. Halo's coming to PlayStation. It does feel like the console wars are over in the sense that the two companies have just diverged so thoroughly in strategy, where Microsoft owns Xbox but is very much like console agnostic and Sony is very much all in on PS5 as 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, 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.

The whole, I mean, like all the old, old images of like GameStop.

Right.

The day before the new Call of Duty comes out. It's famous.

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. All I was thinking about was the 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 gotta have mouse and keyboard and it's just not in the cards for me. I can't do the controller stuff. Seriously. It's just too. I don't know, it's just not for me anyway, the console wars. I was trying to get up to speed on. How do you tell the story of the last 30 years of the console wars Going back to the Atari. Then the NES arrives in the United States 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 Ray. Anyway, we'll see how much it's over. It feels it feels pretty over.

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 a side?

I mean, I believe the hardest score players always pick mouse and keyboard, but.

They might pick okay, but hardcore most.

So PC would be the most hardcore.

Got it.

Because the reaction time. I mean, I'm sure that there are plenty of people that debate this, but in general, the muscle. Because if you're trying to click on a head, like moving your arm, 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 leave that much for like precision. Now some of the games have worked to crew to kind of level the playing field by having different levels of autoimmune.

Auto aim.

Consoles have 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 cursor. Exactly. And so it's not snapping to the target, it's just letting you slow down.

Using auto Aim with this laser pointer.

Don't. Don't shine.

I don't put it in people's eye. Don't put the laser pointer.

You don't know where you're going with that. It's risky.

It's risky. It is. Should we talk about take him on why AI is under hyped 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, it was like a 2pm last minute booking and there wasn't anything in the news right then. So we're gonna have him on later. I wanted to have him on the.

Show in the LinkedIn chat says I'll take it that Jordy wasn't exactly a trick shotter.

I've had.

I've 360, no scope.

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, with a controller that has small stakes. So Tae 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 it can be a bubble. Even big bubbles have small beginnings. They start as small bubbles and then they get bigger and bigger as they inflate.

Yeah. Remember Amazon went public May 15, 1997.

Yeah. So you could easily claim that we're in the very early stages of a bubble.

My personal timeline was that 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. 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 are 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.

Like I've talked to some folks who are like, they founded a company in 1994, they sold it in 1998. It does not exist. It didn't exist. It does not exist anymore. It didn't exist in 2002 and, but they wound up with like $100 million liquid. And they're like set for life.

Yeah.

Because they like nailed it.

That's enough to retire.

Yeah.

I always, I always love, I feel like the debate comes up like, you.

Still have, you still have to.

Is 10 million enough to.

You still have to cut. You still have to raise a VC fund and LARP as a VC for.

All of this continue.

Well, we actually have our first guest.

Okay, we can come back to that.

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? Is that when the kerfuffle begins? That was the main.

I think that was the catalyst, yeah.

Okay, so thanks so much for having.

Me on, by the way. I've been a huge fan since day.

One, so I am 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, why are the haters wrong?

Yeah, yeah. But just give me like the recent back backstory. Yeah.

So we started about a year and a half ago, at the beginning of last year, 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 100 times more powerful GPU compute than has been in space before.

Okay.

And yeah, that's the sort of brief history.

Okay.

What were you doing before this?

So I started my career as an engineer for five years. Studied applied math and theoretical physics, undergrad and masters. And then I spent a few years at McKinsey working with the space agencies of the UA and Saudi government. And then I founded 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, what are you trying to figure out with this launch? What is, what is the value of putting this H100 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, 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 radiation environment and you need to dissipate the heat in a vacuum. So this is really a demonstrator, it's a 60 kilogram small sat, it's launching on a Falcon 9, a quarter plate there. And so we'll be running high power. We'll do a whole bunch of firsts. Will 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 Karpathy. 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 coherent answer. But yeah, this is really a demo. The next satellite launching in October next year is 10 times the 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 downlink 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. It looks huge.

Oh yeah, yeah, right, yeah. To be fair it's extremely difficult to imagine what's about to come down the line with starship. So like the launch, launch cost might come down by between 10 and 100x. Launch capacity is set to go up by 1000x or more. And the reason is if you build a new starship every day for a year, which is like a very conservative estimate 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 1,000 times like more tonnage per year that we can get to space. So 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, 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, yeah, I almost.

Feel like you should have done the anduril 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 saw Palmer talking about that and I was like, man, that's actually pretty cool.

Yeah, yeah, it's nice to have because there is a value for like the vision document. Like what do you, where do you want to go in 10 years? Yeah, 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 they have a board that approves that this can get shown that it's like feasibly possible.

Yes.

Yeah, I guess. Talk about. So this first launch is meant to be like a demo, basically. 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. What is the actual timeline? And then it sounds like maybe defense is like an early place where there might be some commercial applications. But then what's your timeline on when, if or when I would be prompting a model like ChatGPT or Gemini or whatever.

It's just cheaper to do it in space because that's the long term vision. Right. It's just cheaper. It's not anything special or like oh, we need to be there for some geopolitical or defense reasons. 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 mass volume production of Starship on a three to five year time frame. The roadmap is next year. In October, we're launching our second satellite. That's the one is about 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 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, then we're launching a satellite after that which will be. Can't talk too much about it, but you can see this PEZ dispenser form factor that's coming out of Starship. Oh yeah, probably shouldn't say 10.

And that was successfully tested on the last Starship launch. Right where there was a whole bunch of 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. The dummy.

Yeah, exactly. So that's about a 4 ton at least, least 100 kilowatt satellite. Second one's 10 kilowatts, first one's 1 kilowatt. So we're like going up an order magnitude each time. That one could feasibly launch end of 2027. And then from there we basically will launch lots of, lots of small versions 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 concept video, is if you want to train a large model because then you need the whole data center to be internally networked. And so that we're probably looking at early 2000s before we start docking.

I mean, I don't have a problem with thinking in decades like that. That's like what we say in tech. We want people to do. And so I don't, 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. Casino where people could beam up, tap into it all ages. All ages.

Oh really? Yeah.

I mean space token, you got a minus. What, how do you think about, you know, some of the. We, I think we had, we'd asked Dylan Patel at one point about GPUs in 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 unplug the server and plug it back in again or switch out individual GPUs. But what's the solve on that long term? Are you thinking about these being fully robotic systems that allow you to make changes on the fly? What's the solution there?

Yeah, I think probably two points on that. The first is for these massive data centers where you've got, you know, a kilometer long or a mile long row of racks. When the people I speak to 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. They're just writing software around it and they'll just leave it basically. So that's number one. But secondly, for the first few iterations of the satellite, it'll 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 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 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. And in the sense that a human spacesuit would take care of radiation and thermals, you would need to retrain it slightly. But you know, that's maybe sounds a little bit wacky, but robotic.

Yeah. It doesn't have to be a humanoid. Right. Like there's a bunch of other forms.

It could be any form factor but humanoids will be mass produced. So probably the. Probably the cheapest.

It's interesting.

I want you to react to Andrew McCallop's viral post. He got 7,000 likes breaking down a bunch of different things. The core claims icn 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 rideshare sleep. 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. Obviously 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 with when you saw his sort of comedic breakdown?

Yeah, I mean I like the post and I mean I'm a fan of Andrew. I've been following his stuff and he's a great engineer. Same with Gallium by the way. I'm very impressive what they're doing.

Yeah, totally.

Yeah. So I mean on the thermals, 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'll be for at least five years, then will there be an incentive for starship to. For SpaceX to drop the pricing. Yep.

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 teller Tesla is sold under the price. They could, they could potentially sell Tesla a higher price, but in order to reach a much larger market, you need to sell at a lower price. So for example, they can have all of these starships sitting there. They can only launch them two years, once every two years to Mars. Their marginal cost of launch on each starship is two or three million dollars. Now either they can leave it there because there's no Demand at the $200 million for a launch price, or they can say to people like us, hey, you can launch it for 20 or 30 million dollars. They still make shitloads 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 it's from what we're seeing is very much an engineering challenge. It's a, 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.

Sure.

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, the whole game is making this radiator cheap and light. And yeah, we've got very, 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 cell is going to hit hard. Basically all telco is doomed. Let's check in this time next year. Expand on that.

That was at 1am I had a very heated argument with somebody and I got.

I was in the cab and I was like I'm just gonna post this.

And then like 300,000 views later I was like okay, this was probably like.

A little bit too extreme, too aggressive.

Well, no, yeah. Obviously if you want to get attention on Axe just say the most inflammatory thing possible. Say that you know all these multi hundred. No, but what's your view on how quickly you know we saw news on Apple exploring something with Starlink on direct to sell. Seems like every major manufacturer is going to be interested in this. And then there's also that other company that's like a competitor, right? That. What is it? That is Verizon propping them up. They don't have any satellites as.

Yeah, yeah.

No, I don't think they have any hope in hell. And the reason they don't is because. Sorry, I know there's a. There's a whole bunch of people that gotta.

They have a. I mean they have a retail army. Retail traders. The retail traders are gonna come for you, dude. Good luck.

What do you know about space?

Well, you got, you got the private.

They're doing fine.

Private markets Varda guys coming from for you. Now you got the retail traders at ASTS coming for you. You got no friends in this foxhole, dude. You gotta get out of here.

ASTS is up 7% just today. So that's so good.

Oh wow.

That's amazing.

No, I mean. Okay, I. I want them to succeed and I'm.

I'm like okay, well in that case I reverse my entire position. 7%. Move. Let's go. That's very funny. Yeah, 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.

Yeah, well, so so on, on what. What are the. What are the telco companies going to do? Because they're multi hundred billion dollar companies. They're not just going to roll over. Is it is like how do you see them kind of reacting?

I mean it's. It feels like there's lots of multi hundred billion dollar companies that just get backed into a corner and die. I mean Blockbuster is a hundred billion dollar company.

You are space Maxi. You might be right over like you know, it just Depends on the time frame. But like, I mean it just doesn't seem like. 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 for sure there'll be LC cables.

Yeah.

The thing people are underappreciating is the amount of capacity they can launch per starship is absolutely mind blowing. And that is coming in 18 months. So like, and that's like proper hardcore, you know, very high bandwidth director cell in buildings. And that's what I don't appreciate in buildings.

It's pretty crazy. Yeah, yeah, yeah. No, I mean, yeah, I guess, I.

Guess they, the telecom companies could end up being like Starlink resellers, right? Is that one scenario?

I mean they could be.

Doesn't sound like the best business to be in, but yeah.

Anyway, 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.

You're in the Washington area, right? Yes, Washington State. 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 castle, all of the people. I think 95% of satellites launched in the last two years were designed in Milton Redmond. So there's a lot of space talent out there. But 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 it. We'll talk to you soon.

Let me tell you about linear. LINEAR is a purpose built tool for planning and building products. Meet 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 makeovers Hammer bub more.

I mean there'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 leaves. The picking the petals off.

You'll like, where we are right now, it's a bubble.

It's not bubble. It's not. And then whatever you end up landing on, that's what's going to happen. So, yeah, good luck.

And so in Oreo world, Mondelez, 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. It should cut costs by 99%. Do you know?

But I still think there's a huge amount of budget goes to coming up with good ideas and then the, like, editing and then making infinite variations of it, which I'm sure AI can already do well in some cases. So Tyler has a thing. You pay people to come up with great ideas and then execute, but the idea is actually the highest margin part. Right. And so the ad. Some of the ad agencies, I think, will do well. Apparently, the Oreo's parent company only spent 100 million on digital, print, and national TV advertising in the last year.

Wait, so they spent 100 million on advertising, and then they spent 40 million on this? Okay, Tyler, what's your take on Oreonet?

So this was my idea when we were talking about Sora, where I said that Coke should. Or Coca 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. 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 it to, like, make a picture of your dog, and your dog just Is an Oreo for a head or something like that just gets crazy.

Yeah, I. Yeah, that would be. That's a brilliant idea. That would be abused so badly, it would be shut down with. Does this 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?

Like, that 40 million is ours now. That is ours now.

Yeah, 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. Somebody took the $40 million 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, and they Load those in. Can they count that as, like, depreciation?

It's hard not to see this pulled in like a huge waste of money that somebody, somebody high up said, like, I'm going to be the AI guy at Mondelez. Right. Because you have to imagine whatever they were able to do today by investing $40 million, 40% of the, of their ad. So again, like, production cost, which is where they're saving money, is not advertising budget.

Yes.

I would, 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 5, 40% of your overall advertising budget to slash 10, potentially 10% of your budget by 30 to 50%. Meaning that pays back the payback. Yeah, I mean, 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 shot. I promise you. They will be one shotting people eating Oreos better than your model. Your custom. Okay, this will take by now. Yeah, like, by the time they finish. By the time they finish, it's. It's like they, they get on Sora too, and they do it and they're.

Like, oh, no, why did we fine tune, like some, you know.

Oh, they did it with Accenture. I called it.

No way. They did it Accenture.

I knew, I knew, 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. But, okay, $40 million. We're happy to give you an AI play here. You're going to be. 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 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.

Mondelez is doing.

It's like, why are they telling on themselves?

Why are they.

They're not, John. They're bragging.

Wait. Yeah, they said the. Yeah. The head of the Globe, 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 in tune.

You could have just run eight super bowl ads.

You could have run eight. That is actually crazy.

That's actually an under.

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 fun.

Yeah. Coop 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, you would buy 10 ads or you'd buy. Try to. I mean, friend tried to buy.

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?

We drove. We can't get away from Avi's billboards. We drove by today. This one that just says when you're on the street, it just says end end dot com.

Is it the end?

Yeah. 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 bull case where it's like you need to generate a whole bunch of Oreo creative across Facebook and 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. Yeah.

Because 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, I would go farther. I would train a language model.

Oh, yes. From the ground up.

Yeah. Because then you have the.

Do a full pre train. Do a pre train.

Do a pre train. Do a full base model.

Yep.

You know, what's it going to score on? Many's last exam? What's it's going to score on?

And, and let's, 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?

Dialect.

A couple billion.

I mean, and you're telling me that that's worth it, right?

Of course.

Yeah.

We got to get this guy a job at Accenture. He's going to 100x revenues.

Yeah, I mean, this. Just to close it out. This is Publicis 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. They need to vertically integrate. They need to vertically integrate. They need their own data center.

5 Snapchat.

They need their own chips. They need to bake this Oreo algorithm into an asic. They need to go to Broadcom. Then you 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 get into Rare Earth. They need to go all the way down to.

They need maybe acquire a desert just to get enough sand.

They need to teach sand to think. 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 for you.

Well, you know what Oreo should do? They got to use AI. They got to use numeralhq.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 got to get numeralhq.com.

Lulu says it's the year 2025. You wake up and a progressive white billionaire with a blackface NFT profile picture is saying GM to you. Saying, wag me friend.

This is a crazy one of those. Why did he pick this one?

This is like the how do you do, fellow kids meme with the.

It's such a crazy choice on a million different levels. Even the smoking. Even the nft. J.

Maybe he's. Maybe he. Maybe Reed secretly likes heaters.

Jay, that would be sick. I guess Jay said this. This GM is proof that LinkedIn is four years behind on every news cycle. Pretty. Pretty remarkable. How did. How is this.

I think the crypto community loved it, to be honest.

Really?

I think. I think they're pretty fired up if you're one of the 10,000 people that own crypto.

Punk.

A punk. This is.

How are the punks doing? Are they still. Are they still hanging out? Well, before we dive into that, we have Justin Murphy in the restream waiting room. Let's bring him into the TVP and Ultra Dome. Justin, how are you doing?

I'm good. What's up, guys?

Suited up?

You're looking.

You might be the sharpest guest we've ever had. I love this. I love this suit. You're looking great.

Well, thank you.

You know, you're the reason. You're the reason I use One Password. 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 loved it. And then we interviewed the CEO of 1Password, and I was like, I love this. I learned this from Justin Murphy.

Wow, that's really interesting to hear. I'm an avid user of Lucy. I'll throw in as well.

Oh, let's go.

How people don't use password managers is beyond me. I mean, even 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 create some format and then other people can kind of spin off of it and whatnot. But how do you describe 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. I had. I was watching sort of academia go downhill, and all the creator economy stuff was sort of kicking off. And it was 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 I started having kids and kind of took a step back to kind of think about what this stuff really looks like 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 was 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, that I've built. And also to ground it in a kind of longer running history because most of the, 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 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, they just keep winning and this thing is like kicking off. So we're in the thick of it right now. That's called Noctain. It's a layer one blockchain.

Yeah.

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 Noctain because this thing I think is, it's 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, six years ago. And I still haven't come up for enough air to tell the story fully, but I will and I think I'm pretty proud of everything I've accomplished.

Yeah. How do you think about just the basic economic forces around 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. And I 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 gonna get like one scoop a decade and it's gonna 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 long term monetization summary, forces that allow for 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 amount of experimentation and also this 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, you have to write a lot in stage one. So you're not, you don't necessarily have to be chasing, you know, clicks or playing the, the, 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. Yeah. And have some kind of blog or newsletter and just, 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, you know, 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 books and how to write really High quality copy and write thoughtful essays. This whole new world of sort of highbrow thought leadership is incredibly valuable to almost every company today. And the fact is it's incredibly hard to hire for tech companies. You know, really struggle to find sort of PhD tier intelligence for hire and, and who also have any kind of understanding or experience with actually running modern content operations online. And so what I have found, and, and this I think is really the 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, 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 it basically use, use like consulting fees from, 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 Hirsch or these kind of older guys. Like often they don't want to build a business or they don't want to build a whole thing and for them that's kind of a nonstarter. But if you can get some, if you can get any, if you can finagle some kind of situation where you're adding value to, you know, a growing tech company or, 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 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 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, 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 it yet 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 Stripe 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, yeah. Provide some interesting service.

I hadn't thought about, 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 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 intellectual life.

I had lunch with an academic a 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.

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, you know, scholarly research and academia. Like, is there, is, 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 new breed of independent scholars joining forces with Silicon Valley. It's. 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, 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 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. Spinoza was famously a lens grinder. 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, 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 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 conversations 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 scholars of his lifetime. To answer your question, Silicon Valley and tech startups today are just the current manifestation of this. 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 patience to kind of work through it, but they know the alpha is there and that's why they're all 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 you compare Nick Land's early writings to someone like Nick Bostrom writing about AI much more recently. And a lot of Nick Land's texts actually match reality much more, I would say. But the reason you could double click on many sort of specific ideas, I guess. But the high level answer 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 back patter. Everyone is mutually invested in each other. So everyone is back patting everyone else. I mean, just structurally right, I'm, it's not even, not, 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 someone speaking their book. And that's fine and that's America and that's cool, 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 pro social is the way that I would put it. In Silicon Valley. No one wants to Be anti human or antisocial. Right. Because it's not. It doesn't sound very nice. It sounds a little scary. Maybe you don't want to invest in that. Maybe.

We had this this morning as a team. John Tyler and I were sitting around, 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, you know, taking the brunt of the joke. And we, like them, were friendly with them and we just wasn't. It wasn't something we were going to share. And so 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, 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 the source of the alpha, basically.

Well, I mean, Dario's hasn't been. I mean, I'm sure that the anthropic comms team 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 large percentage of white collar work is going to disappear and then they kind of like walk it back a little bit. But clearly 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 it all okay, you know what I mean, so the thing with Nick Land is you get a model, you get 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 can't look away from it either.

In Desire, there's this concept of artificial intelligence being this time traveling force, an 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 lan does introduce this idea. Well, he doesn't introduce it, but he very much plays up this mental model that you get from the Terminator movies. Yes, 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 movie suggests, comes back and kind of composes itself out of enemy resources. So, 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 that capital acceleration will be quite brutal for some people, for, 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, you know, tragedies that came downstream of his ideas, you know, the critics of capitalism in the early modern period like were structurally on something, okay? And, 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, 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, you know, there's, there's really no other, no other solution. And I think personally that it actually goes in a more positive direction than some of Nick Land's suggestion. You know, texts seem to suggest, I think it actually matches much more clearly to something like the City of God by St. Augustine where 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 technological acceleration, you're chopping.

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, you know, everybody, you know, people outside of tech are worried about automation and losing their jobs. And 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, OpenAI 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, just wait, just one more training run, bro. AGI is right around the corner. But it's hard to actually read too much into what they're saying because they obviously have tens of millions to hundreds of millions to billions of dollars of equity tied up 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, 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, 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 a kind of poor person in America has exposure to a lot of really nice things that obviously 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 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 nowadays. I mean, think about how much research you have to do on food just to stay healthy, right? 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 get, you do fall behind. And it's not a matter of losing your job. It's a matter of like, your whole family is opiate addicts because they got one shotted by complex processes that they had no chance of even understanding.

When does this start? Like, 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? Karpathy on Dorkesh was kind of saying, like, 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 associated with modernity and technological experience? Acceleration people have very different answers. I think it's an interesting kind of empirical debate. I've always kind of been partial to Heidegger's answer. If you read Heidegger's the question concerning technology, he seems to think that the, the key breakpoint was when we started storing energy, basically. So like he says the windmill, for instance, was fine because the windmill just sits there. It, nature kind of operates on it and it can create energy. And that's. 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 could store these things.

I have a stockpile of coal and I have a steam engine and I can, I can build up like 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, you know, aeons, 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, 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, you can give many examples and people kind of debate 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 basically is encouraging people to like sharpen up, avoid the pressure to just quit your job and gamble on meme coins, avoid becoming emerging.

I have a completely different take on this and I want your reaction too is I feel like the future things playing out kind of the darker scenarios, there's a massive underclass, but there's no permanence whatsoever. And on any day to day basis somebody could go 1000x 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 mean. 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, with capital acceleration and its current inflection with the AI revolution. So I think that's basically that that is spot on in terms of the generalized. Yeah, it feels threat 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 in three years is a lot better than making the $20 million in 25 years or whatever historical markets 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 was 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, 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, 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.

It's clear, like everyone will be entertained in the future. You will not need, 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 headset, 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 a 1 in a 10,000 chance every week to like escape or whatever. But I do think it's notable that we transition from a pole 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 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 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 pole 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 human society and modernity is totally torn asunder from all of its traditional values and traditional anchorings. Everyone is competing in this mad materialistic race to stay float, which is mimetic. And even if you don't want to do it, you kind of have to do it. Meanwhile, crypto is this kind. 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, 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 so many stories of big crashes and crimes and so many kind of scammers and. 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, like, humanity will be able to secure its future. And so people are. 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? That bitcoin is like literally just regrounds money and physical resource kind of anchoring.

One suggestion, crazy idea for you before you jump. And then we'll have to.

This has been really fun. Thank you so much for helping.

Instead of a book how about you buy a Lambo and you sell a course? I will teach you to escape.

I've paid for his course. He has a course.

Escape the permanent underclass by becoming an independent scholar. You could be the Andrew Tate of independent scholars.

That's one option, but I don't think so, man. That's noctain. That's the role of noctaine. We say escape the leveling process. I don't think you should worry about escape. I don't like the phrase 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 blockchain. That's what it's all about. Escape the leveling process.

That's very cool. Well, thank you so much.

This was 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.

I appreciate it.

Cheers.

We'll talk to you soon. We have our next guest in the Restream waiting room. But first, let me tell you about Fin AI, the number one AI agent for customer service. Number one in performance benchmarks, number one in competitive bake offs, number one ranking on G2. Let's bring in our guest from the Restream.

Let's do it.

Reading room.

Darren, what's going on?

What's up? How are we?

Second time on the show. Thank you so much for hopping on. I'll let you take this.

Sweet. Great to have you back. What's happening in your world here? You've been flipping pretty hard. So break it down for us.

This weekend. A 1914 Baltimore Babe Ruth. 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. I thought they only went up.

Yes. So did that. Adam Smith never said 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 car, 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 Babe Ruth. So I did a little research. Newspapers.com is probably the best value that I have. I mean, $150 a year for all the newsp stuff. I mean, wow. I.

Look, that's crazy. I don't actually, I'm actually not familiar. This sounds incredible because we're always, I mean we, 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 see this because I've been subscribing for a long time and they're not too good at business because I would.

Because you'd pay like 10x more.

Yeah, yeah, exactly. So anyway, I looked and tried to see when the first mention of the 1914 Baltimore newspapers again, because I didn't know and I think I know everything. Yeah, the 2006 Durham Herald. Son, 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 1940, 2014 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, of who they're playing. And it came in a newspaper and it was a ten card set.

Okay.

So one of the reasons, 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 T206 Honus Wagner or a 1952 Mickey Mantle than to have the 1914 Baltimore News Babe Ruth. 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 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 and People had mentioned it in the early 80s, so they've, they'd mention 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. I'm just not comfortable. I'm not comfortable. So that's my take.

Might not be, what is it?

But it's my take.

What does 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. I sold 31 things this weekend on heritage auctions. And some of the prices that I got were, were amazing. Some are painful in terms of, you know, really able to do this. Like, you know, I just. 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.

You buy. Did you buy 31 times 2 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.

You have.

Yeah, talk to me about Ohtani on the Dodgers. Because my family, I'm not super into baseball, but I do go to the occasional game. I often get Ohtani 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.

Wait, what? I was wearing New York City.

They made a green Dodgers hat.

Okay.

So I was wearing a green Dodgers hat, and it was for the Brooklyn Dodgers on Saturday. So many people were, were telling me, we're yelling baseball things.

Okay, okay.

I'm like, I wear this hat all the time. Why is everyone yelling? And then I found out from the security in my neighborhood that the World Series was Happening. That's how I learned the World Series. So we're out of the loop here, but we're interested.

That is quite. You guys have pay a little bit more attention.

We do.

Well, no, I'm more interested in the market than just alts market in general, than the actual game being played.

For sure.

Yeah. But my question. Yeah, my question about Ohtani is that I've been getting bobbleheads when I go to the game, but it feels like there's 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 Ohtani? 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 Ohtani and they're buying up all the Babe Ruth balls because they want to have Ohtani sign the Babe Ruth balls and then, you know, sell them. So if you're.

What do you think? What do you think? What do you think about 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 single sign balls are going up because they are buying in. In the back room. They're buying as many so that they can put a tiny on it. The. The bobbleheads. I mean the bobbleheads since 1997. I mean, it's amazing the life of bobblehead and the lines of the bobbleheads. The Dodgers did a really good job gamifying it. 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 10 that are golden and only a couple with the. 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 Bob, why does it jam up ebay? You by jamming up, you mean just makes it popular?

Yes.

Yeah.

Okay. Got it. Yeah.

People really like to go to buy their bobbleheads 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. It's 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?

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 gotta jump in. I misspoke. I meant the. It was a. Yeah, is a green Yankees.

I was about to say, yeah.

I'm.

Not gonna come on anymore if you guys don't know the difference between a Dodgers and a Yankees.

I do know, but I, but I have both. I have both. I have a, I have a Yankees hat and a Dodgers hat and I wear them in interchangeably and I just like to confuse the people in my neighborhood. They don't, they don't know.

Rogers were in New York at one point. They were in the Brooklyn. But yeah, you were wearing.

They were.

We were thinking why were people talking to you about the. Were you.

Well, no, I, I just was, I just was wearing, I was wearing a base. A baseball hat.

Yeah, yeah, yeah.

And I didn't know it was a World Series, so I didn't know why people kept yelling like yeah, yeah, random baseball related things at me. Like, you know, people were just cheering.

Yankees haven't won the World series since 2009, so you should know that. So they're in a, in a 16 year drought right now.

Brutal.

Okay.

What? Yeah. So why would this person that just sold for. Do you think this was a forced seller? Like somebody that just had to get out? I think otherwise you would just sit on it.

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, and you know, if you. There's no absolute here, right? It is, it is. When do you sell it? What is 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 Ohtani 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 Ohtani? What is the worth of the people who collect Ohtani? 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.

Yeah. 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 prize packs or what was it?

Let me, let me. This is, this is, this is. This is a fun one.

Yes.

So repacks are like the new thing there are. If digital repacks are even crazier now, people who are watching the show, I'm now going to blow your mind.

Okay.

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 there, it's a zero sum game. If I lose, I'm zero. And they're like, wow, I can, 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. They show you. Yeah, the checklist. So. So Courtyard did 70 million this past month.

Yep.

They. It is, it is going nuts. Anyway, GameStop, what GameStop has done is it has a relationship with PSA, where people can grade their cards. And a normal company. And since GameStop is. Is always going to be, to me, a meme stock, a normal company would be very worried about taking submissions 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 GameStop playing here, fanatics, a couple that you just mentioned. You have heritage. I feel like the big auction houses have been surprisingly durable against, you know, upstarts coming to their markets. If you look at the automotive industry.

Everyone is like, like every industry, everyone wants the multiples. So they're all in the tech. You know, that's what they do. Right? They, they want the SaaS model. That's, that's what, that's what everyone's after. 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, through it so I'm going to have to scan it. Who's the best scanner? And then we had too many scanners 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, do you think that'll.

Do you think that'll keep us, do you think that'll keep just compounding.

It. It depends. It depends 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, that's, that's what we're looking at. 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 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 1 and 2 are FanDuel and DraftKings and I don't even think it pays to be 3 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 for a while, but it feels like.

You mean Poly Market.

Polymarket, Kal. She. It's really, really heating up.

I'm, I'm, as someone who was in gambling for a long time, I'm shocked that Polymark and Kalshi got to the place that they did. Sure. They always had kind of that like we're peer to peer but I, I can't believe that just like prize picks, I can't believe that they got to where they did in their valuation Because I always thought that the lobby of DraftKings and FanDuel and fanatics and Caesars and the old casinos. I'm 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 polymarket and Kalshee would be kind of lower end and never really break through. And that has 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, again, they came with the bag of cash.

Is that what it was? Is it just. Is he just a value founder? Just a fantastic deals guy, Just got all the money and.

Well, Michael Rubin has a, you know, he. Great, he's a, he has a great story. Yeah, 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, 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 your consumer a day faster. And I'll take the risk that 7% of the time Jeremy Lin is going to get traded from the Knicks to the Rockets and I'll have 9,000 Jeremy Lin 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. 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 did, he did benefit from getting the, he did, he, he did benefit from getting the, the biggest guys who, you know, some of them overvalued the market. Right. He got big softbank money, you know, and, and, and, and, and that helped because money was money and they took whatever valuation. You know, obviously they had to write down on a lot of businesses, but he, he got the right money and he has the right connections. You know, he knows, he knows the owners well. He knows, he knows everyone so well. And so they feel like it's a, it's a, a good relationship makes it really hard to get into the game.

Do you think it's, 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 are you 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 Bichette and Ohtani 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.

So that's what I'm keep winning. Okay. You want to let your winners ride? Thank you for all the context. I feel like I love being on your show.

You guys are very, you guys are very, very curious.

This is very fun. I really learn a lot. Thank you so much for coming on.

Learn a lot.

Let's do it again soon.

We'll talk to you soon. Have a good day. 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 ADO, customer relationship magic. Adio is the AI native CRM that builds scales and grows your company to the next level.

Jordy. I mean, it is hilariously ironic that we always, we always make jokes about not knowing about sports. And then, of course, I try to talk about sports brutally mix up a team.

Our next guest is Guillermo Rao from Vercel. Welcome to the TVPN Ultradome. Guillermo, 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 baseline. And jiu jitsu, which I just started doing. 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.

So I had some baseline fighting me.

I, I, I, 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 based. 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 all these different injuries. And I got a little scared around destroying all of my joints. 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.

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 Vercel world. What's the latest on your side?

Well, I mean, the big one is our funding at $9.3 billion valuation.

300.

Didn't realize we had that. It was awesome.

Congratulations.

Insane.

Two messaging.

Insane.

Oh, yes.

300 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 stop.

It might never stop.

Yeah, you might be back here with the Series G before it stops ringing, hopefully. What's the. What, what, what are 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?

Well, clearly not too capital constrained because half of it was secondary.

Yeah.

So chapter one of Vercel was providing the developer experience to developers to build pages.

Yeah.

I mean, I'm being overly simplistic. Right. But the world is going from pages to agents.

Yeah.

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 Vercel 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 Vercel 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 is that 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 this 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 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 this workflows. And so we hosted our Vercel Ship AI conference the other day and we basically came with an actual definition of what an agent is. Agents basically remind us a lot of the old types of software that people were excited about maybe decades ago, which is 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 we'll now call agent, is actually kind of hard for developers to write, especially the ones that were used to creating this typical SaaS type of software. So Vercel 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? It feels like that was crazy to talk about five years ago, but now there are neo clouds. 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 fund raising 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, Vercel is in this incredibly unique position that developers come to us because of the developer experience.

Sure.

We can actually automatically make the best infrastructure decisions on their behalf. And today that's sitting on top of the hyperscalers because.

Flexibility.

Yeah.

So for example, a couple 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 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 computing, new kinds of software infrastructure to support these workloads. And I believe that the Hyperscaler clouds 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 NEO clouds are exciting, but you kind of want to be in the cloud from a security perspective, regulatory and also latency. I do think there is 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, 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 Next JS Conf we announced the Next JS evals, which is basically what I call the 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. 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 ChatGPT 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 startup 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 onboarding? Almost like prosumer creators.

Yeah. So on my blog I drew these parallels between the Cloud 1.0 and the AI Cloud. And so you have the poster child of frameworks was React and Next js. Now it's going to be the AI SDK. 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 protocol. So we're Going from HTTP to things like mcp. MCP is really exciting because if you looked at how ChatGPT is now allowing you to embed apps inside of ChatGPT, that also just came out a couple of weeks ago. I went to Next js. I was able to deploy DOOM inside of chatgpt. Why? Because we have an MCB integration that allows our customers to embed themselves inside of ChatGPT. This is really exciting. If this takes off, this could be the world that you just described.

But right.

Anybody can build up an app.

Does OpenAI have like a review process for that? Because I haven't just like run into that. I do run into those like hydrated UI elements every once in a while. If I'm asking for like moving movie reviews, it'll show me some pictures or like links baked in. 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.

This feels really, really big. So yeah, tell me more about how this actually rolls out.

Yeah, so on a protocol level it's mcb. So as long as you have something like Vercel, you deploy your thing, you can speak MCB, you're embedded into ChatGPT. I deploy my app in developer mode which means like the lady basically embed everything. It's going to be a little bit like an Apple App store where you 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 ChatGPT into other kinds of platforms. So you could imagine one where if you start up cooking up an Idea, it recommends V0 because Vercel 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, for example, a good example would be figma converted to v0. So you can convert from figma to next js 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 ChatGPT? And I think there are really good answers.

How do you think about M and A? You don't have to talk about M and A interest in Vercel because I'm sure you're turning down stuff all the time. But how are you thinking about.

M.

And A 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, IDEs, things like this. I think 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 Vercel, which is a very broad platform with lots of, lots of users to sort of make some strategic M and A. But I'm even more excited about our marketplace. So we announced at Vercel 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 you're going to need some services that Vercel doesn't offer. One example is browsers on Demand, I think as I've had a couple browser CEOs here on the show. And so Vercel doesn't offer browsers as a 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 Vercel 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. 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 to do 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.

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 can. You have so many different pieces of the business and I imagine that a lot of them are growing business lines. I can imagine the series F process was a lot of fun 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, Guillermo.

Cheers.

See ya.

Bye. He mentioned aid sleep. Let's talk about our sleep schedule scores. How'd you sleep?

We need to. We need to talk to Mateo and I. I don't want this. I don't want the. The John handicap anymore.

You have. 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. 7 hours 44 minutes. Our next guest, Brendan from Mercur is in the Restream waiting room now. He's in the TVP and ultra jump. Welcome to the stream. Brendan, congratulations.

$10 billion, man.

How are you doing? Called his shot. He's the biggest shot caller in all of Silicon Valley.

Yeah, we gotta get. We gotta get some new shots.

Yeah, I was telling him this. I was like, 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 gong for that. There we go. What else?

And we've raised our series C of $350 million at a 10 billion dollar value.

Massive.

So it's $10 billion value.

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, but it's like it's raining.

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?

Yes.

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 model.

And so those are really 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, 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 people, some people would have a sense that like there could just be this incredible demand over the next five years and then does that demand evolve, evaporate. But I want to hear how you view it maybe more specifically, totally.

What we've been seeing is that there is an enormous trend towards human 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 eval to train an agent one time how to do that thing. Instead of a banker redundantly analyzing data rooms, they create an eval 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. And our marketplace is really at that frontier of creating this new job category and training agents.

How does, 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. Like, is there any sort of like disinterested remediation risk? Or do you think that what you're building with Mercore is special enough that even if you know the OpenAI ChatGPT 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. Like 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 OpenAI is half a trillion dollars. Like there's huge winners and then there's a ton of startups. How do you see the shape of your business, do you think? I 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 insofar 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 eval sets and environments that correspond to every workflow in their businesses that they want to automate.

I had an eye opening experience yesterday. I have a friend who's a lawyer and about a year ago 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. We'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 it do 100k of associate level work in 10 minutes and that felt like just extremely Notable to me to just see how hard he like 180'd on that is like how, like what is the structure? And you can talk at a high level and 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, of investment in law. Certainly I would say the top categories are probably software engineering, than finance, than 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 evals to enable that is one of the primary blockers to getting there.

We have a shout out to one of your employees, Virat Talwar. So just wanted to say congratulations to him. How are you doing, Virat? Part of the team. The chat, the chat is just shouting him out.

Oh, I see.

Obviously it's a big milestone, so we want to celebrate everyone on the team.

How in conversations for this last round, obviously the business has so much momentum. I'm sure people are just like, take my money at all cost, you know, whatever, I just take it. I'll send the money and we'll figure out the docs later. But how big can Mercour get? Like, what is, like if somebody asks, like, okay, if everything goes right, what does his business look like? And does it look like, you know, you're paying out on the levels of like an adp, right? Do 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 Mercur is building the infrastructure to enable humans to engage with models, to teach them to fit into the 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, do you like fine cuisine?

I am a foodie.

You are a foodie.

Very Four 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 guys updated.

Let's go potentially. Let's do an air horn for Air horn.

Potentially on the horizon now.

He'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 Ipoing? 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. I mean you could have said absolutely not. I, 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 thanks for having me on absolute rocket ship.

Cheers.

Have a great rest of your day. We'll talk to you soon.

Let's get back to the timeline for a little bit.

Let's get Back to. To public.com investing for those who take it seriously. They got multi asset 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 chart overlying stock market performance over job openings and how it basically.

Diverges at the introduction of ChatGPT. Basically that's where people.

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 it's literally like the lines are wrong but.

Mark at a 16Z broke this down. He said this is particularly trippy chart crime. The breakpoint 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 memory hole this one.

And so what is exactly on this chart? I don't even have it pulled up. It's number of jobs versus financial performance or something like that. 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 Chamath Polihapitiya 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 imply 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 over hired. 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 on the show. We should also get a couple other economists maybe Tyler.

Amazon is targeting as many as 30,000 corporate jobs beginning tomorrow. So probably not a fun day over at Amazon hq. Some more positive news. Timothy Mellon, a banking heir is said to be the anonymous donor. Not so anonymous now I guess. According to the Times who gave 130 million to the US government to help pay troops during the shutdown, this guy doesn't. According to Willmanidis worth 14 billion personally guaranteed payroll for the armed services. And the only and most recent photo of him is from 1981. I mean he kind of, this photo looks so cool. I kind of understand how he's 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 the suit, the tie with the train in the background.

Yeah.

You know with the, with the hair, the stare. I mean. Yeah, he just, he one shotted 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 and paid for the army's payroll in a time of crisis with his fortune. You're like oh yeah, this guy was probably crucial in World War II based on the image. And you're like no, this happened last week. He's talking about the 2025 government shutdown. That is what they're talking about. And I just keep, every time I see this image I think like, oh.

This is Mike in the Chat is asking 130 million is what percentage of of monthly payroll and that is.

Yeah, I have no idea.

Well, maybe he levered fiscal year 2024. Military personnel category of the DoD budget, which covers pay and retirement benefits was about 192 billion. So yeah, I don't think the 130 million is going to.

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, one massive lottery ticket or something would be. It costs a billion dollars, but there's a 1% chance that it pays out $100 billion. And if it does, then, then the US government is indebted to you for like hundreds of years. But if it doesn't, if it doesn't, then the US just gets 100 by 2024.

Math. I think you basically need like 7ish billion dollars every two weeks to keep the payroll going.

So what does 130 million do? It's like a day.

Yeah, Mike did the same billion. 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 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's pull that up. Government shutdown. Will the government shutdown and by what is the most likely? November 30th by December 31st, it's a 93% chance. November 15th is like the 5050 mark, basically. So expect another two weeks, maybe another three weeks according to Polymarket. So there's your, there's your Polymarket update.

People are folks in the chat still talking about double speed. The 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. Jeremiah Johnson says every startup is like we found a cool way to monetize undermining the social contract with 30,000 likes.

This is, I mean we follow this person or something like that. 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 Jeremiah Johnson like Jeremiah Johnson is. Is not like some like tear it all down anti anti business. You know. And he's making a good point getting 30,000 likes. And so I think this is a harbinger of 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 slashing itself self flagellating self harm self lash.

No, but tech is saying 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 isn't? 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.

Zoomer went viral again. This guy, this guy is just harvesting hatred from the entire world at this point. He said 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 and I'm 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 daughter just.

You didn't take the photo. It's not like he's a photographer.

In this post by Adias. Elon.

Elon.

I need my dopamine dollary dues. I posted another hot girl. Where's my engagement? Elon. Elon. I'm a content creator. So anyways, not. 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 Chad Labs AI 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 rain rot?

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.

I mean it's a real YC company.

Yes, but is it a stunt or are they actually trying to bring Tinder.

Into your ide, which is like I'm.

On the website commenting, learn what's.

I'm on the Y Combinator website and Clad Labs says it's the Brainrot 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 Cloud 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 Brainrot ide. They say. Sup gigachads. This is on the YC websites. Sup gigachads. We're Richard and Kevin, founders of Cloud Labs tldr. As highly requested, we built the Chad ide. Chad integrates your brainrot vices with your agenta coding workflow. Chad increases productivity and reduces friction of context switching.

Yeah. So Kevin is kind of saying, like, I thought the problem 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's brand 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 in the other panel.

And so this is the modern thesis.

They're saying you can shoot, 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're against.

I've been working on my next essay. 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 cluly win the enterprise. Right.

Yep. But there were a lot of VCs who. Maybe it works for students and the story's not over. But there were a lot of VCs who invested and said, hey, look, I mean, we talked to Brian Kim from Andreessen on the show about this. Like, why'd 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, we don't know if they have necessarily, but at the very least. Yeah.

And they were masters of it. 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.

But I just don't think it's a. I don't think it's a good strategy. And I think we're gonna be. That's gonna be proven correct.

I think you're correct in the spiritually risky strategy because how much better it is, is it to just go and actually build a beautiful, optimistic brand and pay for some billboard ads or even direct response ads?

Beautiful thing. You don't even need advertising if your product is advertising is an accelerant.

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 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. And then 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 Chat Ide is competing with Cursor, Windsurf, GitHub, Copilot. They're competing with serious companies. And what kind of like you're going to attract unserious developers.

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.

April fool joke or Windsurf's April Fool's joke. And it should have just been maybe. I don't know exactly what it is. We gotta dig in to understand what this is. But there's a whole world where the idea of a brain rotty ide is something like Google does as a side project, as a joke, right? Like they used to do stuff like that. And some of it, like they had a whole. They had a whole. Didn't have a whole feature that was just like talk to your dog or something. I don't know. There's so many. There used to be a rich community 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.

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.

Your Basil. Concierge is available now to source you any watch on the planet. Seriously, any watch. And also wander find your happy place. Book a wander with inspiring views, hotel grade amenities, dreamy beds, top tier cleaning and 247 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 to everyone in the chat. Thank you to everyone who's listened to the show. We will see you tomorrow live with Satya Nadella, the CEO of Microsoft.

Cannot wait. Cannot wait.

Have a great rest of your day.

See you then.

Goodbye.

Cheers folks.