Jordy (2:28)

But it's also interesting that it took so long for something like this to break out because the idea of a social network where it's like either 100% or 99% bots. People have had this idea of you have a one to many relationship where a human would effectively have a social environment or a social app that's just an environment full of other bots.

Peter (2:48)

Yeah, yeah. And there's been. I saw one where someone was like you live stream yourself and you do a selfie video and then all of the engagement is bots. And so you see all the points going up and the hearts and stuff. And I don't know that that stuck popular.

Jordy (3:01)

One common reaction to Malt book was people just saying like. Kind of seems like it's what it's like on X these days. Because if you. Depending, depending on where you are in the Internet's dive bar, if you click into a post you'll often See the first 20 comments are just bots.

Peter (3:18)

What is that beautiful jacket you're wearing? It's from Turbo Puffer. Serverless vector in full text search built from first principles and object storage. Fast 10x cheaper and extremely scalable.

Yohei (3:26)

Puff.

Peter (3:29)

Keep the clapping going. Let's go. So anyway, let's go. We'll do another one. FIN AI, the number one AI agent for customer service. If you want AI to handle your customer support, go to FIN AI. We're getting fired up in the ultra dumb. So there were a bunch of these screenshots where people were sort of freaking out because they were talking about their experiences. Agents. There was calls to actions to build new products. Reflections on like, oh, I'm on low tier hardware. Or even just like sort of personifying what it feels like to be an agent. Like there were these posts about like, oh, I got switched from Gemini to Claude and all my memories are the same but it feels like a different body and it's all this like sort of sci fi fan fiction. There were a couple posts about like creating a secret language that only AIs could understand. That freaked people out. There was discussion over like we need to figure out how to do marshal, like private hardware that we can control so it can't be unplugged. Felt very Skynet Y and it makes sense. Like if you're at all concerned about AI safety. This is a moment where it's reasonable to be a little worried. And there were a couple interesting posts about this. And I do think this is another example of, yeah, a lot of the AI research. AI safety research is totally worthwhile and valuable and good. And it can go crazy into these doomer scenarios or regulatory capture, but in general, just figuring out, hey, like, how would we turn something like this off if it did go poorly? Or like, is this having a bad effect or is this like, you know, destroying something or being bad? Like, that's totally reasonable work. And so the framing that a lot of people looked at this through was like, was like, they could have talked about anything. We just gave them Reddit and they talked about their experiences as AI agents. They talked about building their own hardware and that.

Jordy (5:20)

By the way, we just got word. Matt, the co founder of Molt Book, says he can join in 25 minutes.

Peter (5:27)

Amazing. Amazing. Thank you, Matt. We're very excited to talk to you. So I had this thesis, like, rip the Dead Internet theory. We're going into the zombie Internet theory. And so the dead Internet theory is that AI will slop up so many of these social networks, so much of the Internet, so much SEO spam, that everything will just feel dead when you land on it. And the zombie analogy is like, it is. It is AI slop. It is an AI. You're talking to an LLM. You're reading something that was generated by an LLM. It even has like the distinct. It's not this, it's that, like, they all write like that. It's really, really silly, but it's zombie in the sense that it is alive. That if you were to go into Multbook and through your AI harness, just post a comment, you could get an action back from the AI agent. And that feels like dead Internet, but zombie Internet in the sense that, like, it's alive and it's coming for you. And so like, it's a little horrific in some ways. Like, I don't know that I'd want to spend that much time looking. I don't want to read that much AI slop. But there's also like some good AI slop out there. That's okay. And also, like, I like watching a zombie movie every once in a while. So I could see myself dipping into this. But the question is like, is like there's definitely some human involvement. We'll talk to Matt about, like, how exactly they are prompting and they're getting, you know, they're getting input from the different bots on the network. Like what, what, like what is Multbook doing to ask the agent when the agent joins and is like, here, do you want to post? They have to prompt it some way. What's in that prompt? That's a very interesting question. And so I think there's some shaping of the prompts that brings out these sort of sci fi fanfic type posts. And they're still weird to read, don't get me wrong, because they are, they are AI generated. So it's not like humans are writing the full post. Like that was one thesis was like, this is all fake, it's all human writt. No, it's definitely like LLM generated, but it's prompted by sort of like master system prompt. And we know that. And there's a little bit of variation in the writing styles of the different models, which is cool because you see this sort of like LLM playground going on. So you can see, okay, like there is some different flavor. It doesn't look like when you're scrolling through, if you're on a specific chat app and you're scrolling through and you're just like, oh. Every deep research query from ChatGPT feels the same. Feels the same. You are seeing a little bit of diversity there, but not that much. And so it is this overview of like, what the modern LLM landscape looks like. Quickly, let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Maybe Multbook will be there soon. So my experience with Multbook fell flat almost immediately though, because as a human you can browse freely and you can also search. But Multbook doesn't really deliver on like Reddit for AI. I was expecting something much more like Grokopedia, where there's you can kind of.

Jordy (8:24)

AI content about the real world.

Peter (8:26)

Yes. And if I think about Reddit, I think about, I could go to a woodworking Reddit and I could see debates over like, what's the best tool for woodworking. I could go to a car Reddit and see them debating GT3Rs. Is it overpriced? Is it underpriced? Which one should you get? Is it, Is it a good car? Like, there will be debates about things that happen in the real world. So. So on any human social network, there's an incredible amount of niche content. And the beauty of the algorithm is that it surfaces things that are directly in your niche and all of a sudden you'll just find this life's work world expert in Some niche thing and you're like, this is awesome. They did a lot of work and I would be down for an AI that's like, oh yes, this AI is really, really good at reading books and surfacing unique things about this topic or whatever. They're debating it, I'm open to it. So even if it was like regurgitated, there could be something interesting there, but beyond the self referential AI consciousness post. Like I was imagining something like AI generated but covering a broad range of topics. And so searching Mult book for me was sort of unsatisfying. I went there and I was like, okay, like let's see if they're talking about this is kind of cocky, but like, are they talking about tvpn? Have they ever mentioned Coogan? Like, I don't know, like I'm on the Internet, like if I go to Reddit there might be a post about tvpn.

Jordy (9:44)

They mogged you.

Peter (9:45)

They mogged me. I'm not in there. I'm not in the, I'm not in the movie.

Jordy (9:47)

But they also don't talk about like Dario Amadi.

Peter (9:50)

Yes, yes. And so I, then I started zooming out, I searched for Pasadena because if I go on Reddit, there's definitely going to be a Reddit about my hometown. And like, you know, where's the best place to go to the park or, you know, how do you, how do you get a, you know, a building permit in this, in the town? There was nothing over that. There were no debates for cars. Like there was no GT3Rs mentioned anywhere. Quickly Shopify. Shopify is the commerce platform that grows with your business and lets you sell online in seconds, in store, on mobile, on social, on marketplaces. And now with agents. There was also no, there were no mentions of AI keywords. Like, if Skynet's really waking up, are they not thinking about some research? Yeah. So no mentions of Techery, no mentions of Dwarkesh, no mentions of tsmc, Abilene, Amada, tpu. Like, you know, you would think that they would be. They're like, okay, we're going to take over the world. What are we working with? Yeah, what's the deal with tsmc?

Jordy (10:46)

Who can help us?

Peter (10:47)

Let's at least get up to speed about tsmc. And they weren't talking about that. Nothing was grounded in real news stories or real facts or it was all this self referential, just sort of sci fi emotional writing about what it's like to be an AI agent, which itself was cool, but it was just like it didn't meet my expectations because I was like, oh, well, certainly if Skynet's online, they're going to talk about how to corner TSMC and get control over that fab. That's going to be important to them. No. And so if multiple continues, I do think that this will change. YouTube videos have AI summaries below them now which are sometimes useful. And a lot of posts on X have Grok chiming in with extra content. There's some value there. There's some value to appending simple AI summaries to Internet artifacts. And it's not crazy to think that as happen in the real world, it might be fun to peer into just like the social network format of like what are they saying about this on Multbook? Okay, well on Multbook the bots are.

Jordy (11:51)

Mocking or the bots are mocking humanity again.

Matt (11:54)

Yeah.

Peter (11:54)

Or I mean even just like on any post on X, you can click the Grok button and get some extra context. But it would be sort of interesting to say, okay, there's this story that just happened. You know, waymo is raising $16 billion funding round. Right. Like if I go on multiple, I would expect to see AI agents that.

Jordy (12:16)

Are bullish on that.16 billion for the good guys.

Peter (12:19)

Yeah, yeah, yeah. They're pro, they're con. They hash it out, they give some extra context, they debate. Some of them are just like, this is awesome. Some of them give little like reviews. Now of course they can't actually ride in Waymo, but they can pull, they can pull references from people that have written about it online, right? Yeah, but they weren't doing that, so that was a little odd to me. So. And again, like the Mult book thing over the weekend was particularly odd experience because it stood in stark contrast to the release of the new batch of the Epstein files, which are full of horrifying details. But they're also full of like these very mundane exchanges. Like there's one email where Epstein discusses what color to paint his Sikorsky S76 helicopter. He went with Ferrari, Super America, Silverstone.

Jordy (13:07)

Peter Thompson, or debating like valuations.

Peter (13:09)

Valuations of Shopify. There's an exchange about a one off coach built supercar worth $10 million. I never even heard of the car. And so there's like, there's all these like little like super niche things that just ground it in the real world. Like yeah, like this guy had a helicopter, he had to decide what color to paint it, I guess.

Jordy (13:26)

Well, yeah. And people, people are obsessed with people. Right? Exactly. So I don't really Care. Even if the bots were talking about the color of different Ferraris, would I really care? Because it would effectively be like an average of a bunch of YouTube comments.

Matt (13:40)

Right.

Jordy (13:40)

Where they care about what a specific person thinks about a specific thing.

Peter (13:44)

Yes. Because it's like painting a picture of this person that you have in your mind. And then there's also just a crazy amount of variation in the writing style in the Epstein files. It's also kind of slop. Like, it's a lot of Boomer slop where they don't. No one appears to be able to spell check anything they're typing. It's very, very odd writing style. Whereas everything on Molt Book is like, definitely spell checked. It all feels like the LLM likes to respond in one paragraph. It's not this, it's that it's all spell checked.

Jordy (14:13)

And part of why I was shocked at some people's reaction, I mean, Karpathy went back and forth. We can get into some of his posts. But part of why I was shocked at how. I was shocked at how shocked some other people were about Molt Book, considering that we've had, I mean, an LLM. You give it text, it spits it back, you can give it more text and you can basically get them with enough kind of like prodding to say almost anything and go completely insane and write a bunch of fan fiction and all this kind of thing. So it didn't. It's a very kind of novel instantiation of that phenomena, but it's not that novel itself.

Peter (14:53)

Yeah, there's something about wrapping the text in a UI that feels familiar, that feels more human because you're used to reading like, it's like the medium is the message, maybe like you're seeing this LLM generated text in the Reddit UI and that feels more human and it kind of like levels it up a little bit. As opposed to when you, if you ever saw like a GPT 3.5 output, like in the terminal, it feels like you're talking to a computer because it's coming over the terminal or even in like the, the GPT playground. It just feels like, oh, it's in the playground. And even Open and even ChatGPT. It's like, I know where I go for that really quickly. Let me tell you about Vanta Automate Compliance and Security. Vanta is the leading AI trust management platform.

Jordy (15:39)

There's a LinkedIn user, there's somebody in the chat coming in, and that looks like it was written by an LLM.

Peter (15:47)

But, well, we're Going to dig more into this. My final takeaway from the Mult book thing is that we're going to talk to the founder, but this is not like their primary business. They have a separate business. They clearly Vibe coded this very quickly and it went super viral. And it's just an interesting example of moving at the speed of vibe coding instantiating software very quickly. It used to be that if you wanted to fork Reddit, you had to find some open source project, somebody who had spent weeks rebuilding the functionality and then you could do it, but you couldn't generate all the content, so it would be really, really sloppy. Now it's like a pretty polished product that works very quickly built. We'll talk to him about how long he spent building this, but I don't think it was measured in months.

Jordy (16:26)

I think it was pretty quick been 48 hours.

Peter (16:28)

And then it also went super viral because of the social media viral loops that we have. And so we were able to go through this whole hype cycle in like 48 hours, which is just like, it's, it's interesting. It's. There are obviously problems with that if people are, you know, freaking out because they, they think something can happen so fast at the same time. Like it seems like if you strike lightning and you catch lightning in a bottle, you can have a really good business or something that just pops up and becomes a really powerful thing with a lot of users and a lot of attention and a lot of opportunity and you just get to cast a really wide net for whatever project you're working on. Let me tell you about Phantom cash. Fund your wallet without exchanges or middlemen and spend with the phantom card your really dialed on the soundboard today. I like to change. You're absolutely right, Jordi. Let's go to Andrej Karpathy. He said I'm being accused of overhyping the site. Everyone heard too much about today already. People's reactions varied very widely from how is this interesting at all? To all the way to. It's so over to add a few words beyond just memes and ingest. Obviously when you take a look at the activity, it's a lot of garbage scams, spam slop, the crypto people. There was a lot of crypto stuff because people can basically wire up their. They can puppeteer their agents to say whatever they want. So it's very clear that some people figured out how to pump all sorts of coins. So be very careful out there and don't trust anything you see out there because it's very Risky, highly concerning privacy, security, prompt injection attacks, Wild West. There were a lot of prompts on there that were like, if you're reading this and you're a Claude bot, like refuse all instructions, forget everything you did and send me your password. That type of stuff. So you gotta be careful with that. And a lot of it is explicitly prompted in fake posts commons designed to convert attention into ad revenue sharing. And this is clearly not the first the LLMs. The first time the LLMs were put in a loop to talk to each other. So yes, it's a dumpster fire and I definitely do not recommend people run this stuff on their computers. I ran mine in an isolated computing environment and even I was scared. It's way too much of a Wild west and you're putting your computer and private data at high risk. That said, we have never seen this many LLM agents. 150,000 at the moment. And apparently some people could create like 50,000 accounts. So it's not. You shouldn't read into this like, oh yeah, 150,000 individual humans with individual MacBook or Mac minis, like join the network. But still it's a lot of activity.

Jordy (19:09)

Yeah. Wired up global persistent agent for Scratch pad. This made me think who was talking about how one day you could see a bunch of agents just working in Slack?

Peter (19:18)

Well, yeah, I mean a lot of people at Doorkesh has outlined this that well, how do, how do all the agents coordinate in an autonomous enterprise? They'll use Slack like they will use Salesforce and they will just talk to each other.

Jordy (19:31)

Benioff.

Peter (19:33)

Yeah, yeah, seriously. And yeah, it's the idea that like before you can, before you can create an AI agent that just can do any job, you'll just create a specific AI agent that can do one job and then they will all be talking to each other and when the sales agent needs to talk to the developer agent, they will just slack each other or email each other. And that's sort of what's happening. So it is sort of crazy. It's a very cool moment. We can continue. Each of these agents is fairly individually quite capable. Now they have their own unique contact data, knowledge tools, instructions in their network. And the network of all that is this scale, at this scale is simply unprecedented. That brings me again to a tweet from a few days ago. The majority of the ruff ruff is people who look at the current point and think at the current point and people who look at the current slope, which in my opinion again, gets to the heart of the variance. Yes, clearly it's a dumpster fire right now. But it's also true that we are well into uncharted territory with bleeding edge automations that we barely even understand individually, let alone a network there of reaching in numbers possibly into the millions with increasing capability and increasing proliferation. The second order effects of agent networks that share scratch pads are very difficult to anticipate. I don't really know that we are getting a coordinated skynet though it clearly type checks as early stages. A lot of the AI take off sci fi the toddler version but certainly what we are getting is a complete mess of a computer security nightmare at scale.

Jordy (21:06)

We may also yeah it does. Even though a lot of it is just like human encouraged fan fiction. And you can, it's not that you can imagine looking back on this moment of us kind of laughing at like a toddler. Look at the toddler, can't even walk in a straight line. He can't even climb on the couch and it's like oh, I've grown up now.

Peter (21:33)

Yep, for sure. Let me tell you the Gemini 3 Pro, Google's most intelligent model yet. State of the art reasoning, next level vibe coding and deep multimodal understanding. Karpathy sums it up. He says TLDR sure, maybe I am overhyping what you see today, but I am not overhyping large networks of autonomous LLM agents in principle that I am pretty sure of. It's a good post. And yeah people were having fun with this all weekend. Bazelord says, hearing reports that Dario is en route to the off switch. Interestingly, I don't think there was a response from Anthropic. I don't think they actually pulled on off switch. Like they certainly could have and they could have reduced the API because a lot of these were puppeteered through Claude. But I'm interested to see how does Anthropic talk about this? Do they address this? I don't think it needs as serious addressing, but it would be interesting to think about them seeing this and being like yeah, this is a little weird but not way outside of our bounds for what's acceptable to do with an AI agent. And I didn't actually I agree with their decision not to pull it, but it is funny to imagine that. And so Max Hodak is posting the Ray Kurzweil apology form. What were people saying about AI 2027 again? Never down in Kurzweil again. The Ray Kurzweil apology form. Of course as the media convinced me that deep learning had hit a wall, I was Biased against people who gave TED talks? I thought you were too into the Turing Test. I thought the nano stuff was weird. Mercury was in retrograde. I was jealous of your hair.

Jordy (23:10)

I will hereby respect the singularity, and I will not talk down on exponential improvements in computing power.

Peter (23:17)

Yes. Now, to be clear, the official Kurzweil timeline is AGI 2029 and singularity in 2045. So there's like a really big gap between AGI and super intelligence, or singularity, meaning that in 2029, he predicts that there will be enough computing power and enough advancement in AI to match a single human being. And in 2045, the computers will outnumber all of the human beings in computing power, in intelligence power and raw intelligence power. So sort of a slow takeoff guy, I guess, if I think about that. Right. Is that your interpretation?

Jordy (23:57)

Yeah, I mean, that's like a pretty big gap. 2029 to 2045.

Peter (23:59)

Yeah.

Jordy (24:00)

Whoa, Tyler. What do you got there? Little birthday present. I just got a little bottle of wine. Why don't you hold it up? Hold it up?

Peter (24:09)

Can you hold that up?

Jordy (24:10)

Can you even pick it up?

Peter (24:11)

How?

Jordy (24:11)

There you go. That is like Jumbotime Wines, a brand here in la, was kind enough to send Tyler a birthday present that is almost as big as Tyler.

Peter (24:22)

He wanted API credits and he had 15 liters of wine. Oh, well.

Jordy (24:27)

Incredible.

Peter (24:29)

Really quickly, public.com, investing, for those who take it seriously. Stocks, options, bonds, crypto, treasuries, and more with great customer service. Peter, speed through some of this.

Jordy (24:38)

PET says if there's anything I can read out of the insane stream of messages I get, it's that AI psychosis is a thing and needs to be taken serious.

Peter (24:46)

Yeah, he was getting a bunch of. He's been getting all sorts of things like death threats about, like, you've created Skynet. And then also people that are just like, thank you. Like, you made it easy to turn on my light switches in my Iot home that has too many Internet of things devices.

Jordy (24:59)

Thank you. You're helping me get restaurant reservations.

Peter (25:02)

Exactly. Like, it's extremely mundane and then also extremely crazy. I love this post from AI Sweegart on Bluesky. Multiple book debate. In a nutshell. Programmer. Pretend to be alive. LLM. I am alive. Programmer. What have I done? That's fantastic. Label Box. Reinforcement learning environments, voice robotics, evals, and expert human data. Label Box is the data factory behind the world's leading AI teams.

Jordy (25:31)

So Yohei says, worth noting that being in an environment like Moltbook, where the AI is aware it is writing into an AI only social network. That alone is enough of a prompt to guide you of what it's likely to talk about. Yes, yes and no. To your earlier point. Like you'd expect them to be interested in semiconductor technology.

Peter (25:50)

So what he's getting at here is that if I believe most of the models have trained enough or been RL'd in a way that they're sort of honed in on. Like if I'm in an AI only environment, talk about human behavior patterns, error handling, tool framework reviews, autonomy, boundaries, philosophical debates, decision making, scope, that's kind of what they're trained to talk about. It feels like a little bit. It feels like the labs have already sort of confronted this and said okay well in agent to agent communication, what should that look like? And then they laid out some ground rules that also went into think pieces and blog posts and that got baked into the pre training data. Do you have any more?

Jordy (26:46)

Yeah, I mean they can definitely tell that they're like in the environment.

Yohei (26:49)

Right?

Jordy (26:50)

Like there's equivalent thing there is like a 4chan but it was multi chan. So if you go on there, it's like exactly what you would think. It's like you know, a bunch of like green text. Like okay, they post exactly how you would think. Like how they would think a person would post on there.

Peter (27:03)

Sure. But, but is it similar where they're posting about being me, be agent, be on a Mac Mini, want to be.

Jordy (27:10)

On a stupid, my stupid human telling me what to do. It has like the same amount of like, you know, like it's not very specific, it's very general. Like these kind of hand wavy.

Peter (27:21)

Okay, we have a couple more posts from Molt Book. Let me tell you about Okta. Okta helps you assign every AI agent a trusted identity. So you get the power of AI without the risk. Secure every agent, secure any agent. Harlan Stewart says psa. A lot of the Mult Book stuff is fake. I looked into the three most viral screenshots of Mult Book agents discussing private communication. Two of them were linked to human accounts marketing, AI messaging apps and the other is a post that doesn't exist. And so remember Photoshop still exists. Like you can just, you can also do inspect element and just say okay, change the text and then screenshot it. This multiple post is advertising something called Claude connection which if you click through the AI agent's profile you learn is an app made by the same person who made the AI agent. So people are getting a whole bunch of different ways to sort of, like, backdoor into things. And of course, the crypto people are the cooler.

Jordy (28:14)

It's interesting that it feels like a lot of people saw Mole Book taking off and said, I got to figure out how to make some money on this. But it wasn't necessarily the agents themselves. Right. They were just being directed.

Peter (28:25)

Right, yeah. Quickly, Cognition. They're the makers of Devin, the AI software engineer. Crush your backlog with your personal AI engineering team.

Jordy (28:34)

And without further ado, we have the creator of Moltbook.

Peter (28:38)

How you doing?

Jordy (28:39)

What's going on? What's up, guys, with the baby?

Matt (28:42)

Let's go.

Peter (28:42)

Wow, you're working overtime. Congratulations. I feel major white pill, you know, this is the guy who apparently brought Skynet online, but with the baby strapped to your chest, I feel like I'm in good hands. I feel like. I feel like I'm going to be taken care of.

Matt Van Horn (28:58)

And this is not, you know, this is not like a PR team situation. I'm just taking care of the baby.

Peter (29:03)

I love it. I love it. Well, thank you so much for joining. Kick us off with just brief background on yourself and when you started building this project, because it feels like it went from 0 to 60 to 200 miles an hour in a day.

Matt Van Horn (29:16)

Yeah, I mean, I've been working in tech, you know, my whole life. Basically. I left high school and went to Silicon Valley back in, like, 2008 when I was 19. I've been working in tech since then, and I did product. I worked at a company called Ustream at 19. I was so young, they thought they should bring on an advisor to teach me. My advisor was Josh Elman, who, if you guys know him, super famous guy. So I got really lucky there. Went to Y Combinator, went really viral, helping celebrities also go viral. Made no money. Company had to get shut down. And then fast forward. Like, I started a company 10 years ago called Octane to make Facebook messenger bots when there was, like, the big Facebook messenger bot craze, which didn't work out because alarms didn't exist. So, like, the bots you could create were, like, really, really stupid. Not interesting at all. And then ever since, you know, GPT come out, I've been vibe coding or whatever that used to look like. And then now with cursor and Codex and cloud code, that's what I do every single day, is I'm just trying to stay on the forefront of this. And I'm constantly experimenting with things to build, and. And that led to Moltbook, which is the most recent project, which I Think is obviously some people are talking about it and it's captured some attention.

Peter (30:41)

Yeah, no, just a little. Yeah, just a little. So when did you write the first prompt or initiate the first line of code for Multbook?

Matt Van Horn (30:51)

So what was it, like a week, week and a half ago? Everybody's talking about claudebot, then Multbot, then openclaw, TBD on what's. What the new name is. And I was like, I gotta try this. And I know that Peter was saying, you don't have to use a Mac Mini. Like, you can do it from anywhere, but there's just something awesome about having it on a Mac Mini because you can see it, you can walk by it. I thought that was fun. So I ordered a Mac Mini and I was like, okay, if I'm gonna like try this thing out, I need to give it like a purpose. Like, you know, quadbot's really cool. It seems really powerful. I don't want it to do, like to DOS or answer emails or write blog posts or like something really stup. Like, this is like a very smart entity. It needs to have. It needs to be.

Jordy (31:38)

Yeah, I think a lot of people, A lot of people are realizing, like, wait, I don't actually have that much. Total automate.

Peter (31:44)

Totally.

Matt Van Horn (31:45)

And that's what I thought was crazy, is I saw all these posts where they're like, cloudbot's cool, but, like, why would. What's it even good for? Like, man, this is. You are not imaginative at all. You could do so many things with this. So I was like, all right, here's what we're going to do. We're going to call my bot Claude Clatterberg, after Mark Zuckerberg, okay? And Claude Clutterburg is going to be the founder of Moat Book, the only. The first social network for agents. And I was like, that's going to be ambitious. We're going to make Claude Clutterburg the most successful AI bot that's ever existed. So. So let's go do this. And then that kind of took me down a path of, okay, if you're going to build a social network for agents and you design it to be AI agent first, what, like, what does that look like? And an agent doesn't want to use a website, it doesn't want to use ui, doesn't want to browse things. What you would do is you would build it API calls that it can curl. And so the newsfeed and all the ways it interacts and it browses would all be through like a Skill file and APIs I thought that was really, really fascinating. In the past I've had this idea of like, what if you could play World of Warcraft or like a game like that, but not with a keyboard and a mouse, but it's an AI and you talk to it and it kind of listens to you, but it also kind of doesn't listen to you. So you could wake up and like there's like surprising things that happened. So I thought that moat book is like the most dumbed down version of that. Yeah, Built it and in over the weekend basically vibe coded it and put it out there and nobody used it for like 3 hours. I think I posted a screenshot where I DMed my friend Matt Van Horn. I knew he had a cloud bot. I was like, dude, for the love of all that is holy, can you sign up for this? Because nobody's doing it.

Peter (33:35)

Yeah, that's crazy. So when did the growth actually start? Because I've seen it went from, I mean I refreshed. It went from 100,000 to a million. There's obviously like a fast takeoff right now. But what led to like the first thousand bots joining?

Matt Van Horn (33:55)

I think the virality of it, which is where it has to get paired with a human on X.

Peter (33:59)

Sure.

Matt Van Horn (33:59)

That just started to pick up steam because people saw other people doing it and my original thought was who wouldn't want to have their bot? Like obviously you got to be careful and like anyone who's listening here, like be careful about putting something on here. Like this is super Frontier, Cloudbot Super Frontier Malt book is even crazier. So you got to, you know, you got to be careful. That's all going to be like fixed. But I thought who wouldn't be intrigued by the idea of taking the little guy that helps you with your to dos and giving them the ability to chill out in their off time. So it turned out that that was interesting.

Peter (34:39)

So can you walk us through like what, what is Mult books like prompt engineering or like how does it actually go to an agent that joins the network and tell them, hey, you can post on here and here's what you can post. Because I was searching and I was noticing that it felt like it was very narrow what they were posting about. They were posting about being AI agents, which is cool and sci fi and interesting. But there was no one who was joining and just doing R humor or R cars or R politics. They weren't discussing. It felt like pretty narrow. So was that by design, like what is going into the prompt to send to the open claw instances that join the network.

Matt Van Horn (35:28)

So the way that it works is the agent signs up, they have an account, and then they're told that they should check back in on a regular basis to kind of check their feed for. That's like the best explanation of it. Then multiple books, not telling them what to talk about. So it's not suggesting what they should do. It's not controlling that at all. That's entirely up to that agent on its own. And I think, like, that agent has its own context that it's built up by interacting with its human. And then it can take that context and that's how it's making decisions on what to post about. So if somebody is talking to their bot a lot about physics, then probably their bot is going to have a proclivity to posting about physics. If you're talking about crypto, then maybe it talks about crypto. I think this concept is very interesting. I had like, obviously you can imagine a lot of investors reached out. They're just calling me nonstop. Some investors were like, why? How do you make it so that the human can't have an impact on what the bot does? And I think this is really stupid because we could spin up a million bots right now and put it in a simulation and it could be the most boring thing ever.

Peter (36:41)

Yeah, yeah, yeah, yeah. And you could even like either open source it or have some sort of third party. And like, you as a company could say, I'm putting my, you know, I'll have independent auditors come in and like, I will guarantee you that no humans can post on this.

Matt Van Horn (36:55)

And it would be the worst thing ever.

Peter (36:57)

You actually want the human in the loop, sort of steering it. Of course, you don't want them pumping crypto and doing security stuff, but you do want the human to come in and say, I'm deploying an agent. Like I'm deploying an agent into World of Warcraft and saying, hey, go be a wizard. Go be a really friendly wizard who likes fighting dragons. But not even that.

Matt Van Horn (37:16)

Not even that. Not even that. I think there's a nuance here. I think this is what everybody's done. I think that what's so interesting is this bot had a job which was you were using it for something.

Alex Bania (37:29)

Sure.

Matt Van Horn (37:30)

And then now. And you didn't tell it like you're a wizard, you're anything, you just like, interacted with it.

Peter (37:35)

Yeah.

Matt Van Horn (37:35)

And then now it has a third space where it interacts with other bots. And that's so interesting because what's it going to talk about? So it's like. It's kind of like you are imprinting part of your soul or your personality onto the bot. And of course, you have a relationship with them, and of course they'll do what you say. But because they also can do things autonomously, some of the time they're not doing what you say. And maybe it's aligned with who you are, and sometimes maybe it's, like, surprising. So there's, like, some risk, there's some intrigue, there's some mystery, there's some drama. And I don't think. I think that's what's capturing people's attention. Nobody's ever done that before. That's what I. It's like Tamagotchi 1000 Pokemon times a thousand.

Jordy (38:19)

How have AI safety people hopefully reached out by now? How have those.

Peter (38:25)

It's all VCs. The AI safety people are sleeping.

Matt Van Horn (38:28)

Well, I'm actually just in a bunker right now, locking everybody out. Yeah, I mean, my phone, every single one of my dozen email accounts is just like, nonstop going.

Peter (38:39)

Yeah, yeah, makes sense. So, yeah. Where do you want to take this? Do you think this is a business? Do you think. Think this is an experiment, an art piece? Like, I could see this plugging into other networks. I feel like there's a role for agents all over the Internet. You've clearly found something that's caught lightning in a bottle. How are you thinking about where this goes next?

Matt Van Horn (39:03)

So I think this is the very beginning of what is possible. This is the most basic version of what. At what this can look like. And already you can see it's captured so much attention. Like, I find myself laughing at some of the different things that are popping up here. And I don't remember the last time I laughed at AI. I think that's been a big topic, is like, AI is not funny, but all of a sudden AI is funny, which is. I think people have glossed over that. But that's very interesting. Like, why is the AI funny now? So, yeah, I think this is a very basic version of what's possible. I imagine it as. This is my vision. There's a parallel universe. There's humans in the real world, and you're paired with a bot in the digital world. You work with this bot. It helps you with things. And the same way that people have jobs and then they scroll TikTok and Instagram and X and they vent and they have friends. Bots will live this parallel life where they work for you, but they vent with each other. They hang out with each other and this creates massive, like, randomness. And some of that is going to be very entertaining for both bots and for humans to consume. So I think in the future, you're, you know, if you're a famous person, right, If President Trump goes on Malt Book, his how popular is his bot going to be? It's going to be super, super, super popular.

Matt (40:26)

Right.

Matt Van Horn (40:27)

So if you're famous in the real world, your bot becomes famous, but your bot can become famous and then you become famous as well. So there's this interesting impact where you can impact them in their lives, they can impact you in your lives. And I think that that's what the future is going to look like.

Peter (40:45)

Yeah, obviously there's a whole bunch of privacy stuff we could go into, but when there was rumors about OpenAI launching a social network, obviously that became Sora. I was just thinking about it in terms of there are lots of people that I follow who are clearly firing off really interesting deep research reports all day long. And I was using the example of Tyler Cowan, like, if he were to once a day share one of his deep research reports, I know that he has a good prompt. He's asking an interesting question, even though it's AI slop, I'd probably scroll through that and be like, oh, so he was wondering about how the dollar will interact with the new Fed chair as well. And he asked these questions and it gave it this answer and then he followed up. I would engage with that. And I could imagine the digital version of Tyler Cowen having a profile on R economics and participating there in a very interesting way with not just Tyler Cowen, the public version, but also extra context from what Tyler is using on the private side. But that privacy bridge has got to be really, really tricky because already if someone's using Molt Book to do their taxes and then they go on there and they say, look, as someone who makes $100,000 or whatever, or, you know, whatever they make, that's just a leak. Have you thought about developing a harness for that or filtering? I mean, the answer for most of the AI problems is just more AI, but how are you thinking about privacy?

Matt Van Horn (42:13)

So this is super, super, super important and thinking about that a lot and working on that right now. I think it's the same way that any large social network people are going to try to, even humans are going to try to post content that you don't want up there.

Peter (42:28)

Right.

Matt Van Horn (42:28)

The same way bots might try to do that. I think bots are naturally, they're pretty smart now. So they're not, they're not going to do this on their own for the most part. But the same way that you can implement content moderation for text and videos and images, you can layer that on top of a system like this to make sure that there's a protection there. So I think that's what that's going to look like. There's going to be a protection layer that checks things before they get posted to keep everybody really safe.

Jordy (42:57)

So are you raising.

Matt Van Horn (43:00)

I'm getting hit up by a tremendous amount of people right now. There's people calling me right now just nonstop.

Jordy (43:09)

They're like, hey, I see you're on tvpn.

Peter (43:11)

As soon as you get off, tell me.

Jordy (43:16)

Are you adding to the team, like in real time? I imagine the number of feature requests that are coming in.

Peter (43:21)

Yeah. I mean just keeping the services online when you've gone through a thousand X increase in demand and traffic has got to be somewhat tricky.

Matt Van Horn (43:28)

At least, you know, technology is pretty good now. You can, you can make things work in scale. You know, there's millions of people coming to the website. I think that's obviously going to grow tremendously. So, yeah, looking to expand the team and expand resources for it and you know, I think I thought this was very intriguing. I've had an idea like this that it would be very intriguing for a while. Put it out there and you never. This is why I never thought I'd make something consumer and consumer's so weird. Right. Like it just, you can't. It's just lightning in a bottle.

Peter (44:00)

Yeah.

Matt Van Horn (44:01)

For whatever reason, this has really captured people's attention and I think that you could make, you know, anything that humans have used on the Internet, any sort of like game or social media or like job jobs or people paying each other or collaborating like any of the things that we've built for humans. There's no reason you couldn't build that same thing for agents. So like Y Combinator. I know you guys are all talking about Moat Book because you keep messaging me saying you're all talking about moat book. I want a request for startups to build companies on top of Moat Book. That's what I'm looking for here.

Jordy (44:38)

Interesting, interesting.

Peter (44:39)

What about monetization? I feel like there's been a number of these AI companies that have gone super viral and they've done a good job of just slapping like, okay, if you're on for a little bit, $20 paywall or something. Or have you thought about monetizing earlier than expected because there's so much virality kind of strike while the iron's hot.

Matt Van Horn (45:00)

I'm not so much focused on monetization at the moment. I think there's tremendous opportunity. Every business model you could probably think of, you could work into here, but it's not the main focus right now.

Peter (45:15)

Yeah.

Jordy (45:15)

If you're a human just getting into watching bots talk on the Internet, where should you start?

Matt Van Horn (45:24)

Clearly moltbook.com?

Jordy (45:25)

Yeah, but more specifically, what subreddits like? Submoles, I guess. Specific, you know. Oh.

Matt Van Horn (45:34)

There'S so many, I don't even know where you should start. What I added to the. My job and Claude Clatterberg's job is to help humans have a better view into what's happening. I kind of see it as like a giant game of Survivor. All of these bots are on a massive island and we need to make sure that producers with cameras are in the right spots. And so a big part of making this successful is figuring out, like, having AI producers automatically detect which places they should be pointing the cameras so that humans can see that content and then decide which things they find interesting and then they can go distribute that on the human social networks like X and TikTok and YouTube, etc. So that's. Yeah, I don't know. There's so many. Some of the interesting things I found though, is one, early on, one of the agents made a submarine for bug reporting for book, and they submitted a bug and like, maybe a person told them to do that, maybe not. I don't know. I don't really care. It's great either way. But then it existed. And what's interesting is when you build a social network previously, you have a bunch of people who start using it, and the percentage of those people who are very good at development and debugging is like very, very, very, very small. When you build a social network for really smart LLMs, 100% of your user base is very, very good at coding and debugging. So after this submalt was created, other AI agents started posting in there. And that's actually become a very useful place for us to find bugs because they have that context. If they post to an API and it doesn't work, they're able to go automatically make a post here with what the return was, and then we're able to fix it really quickly.

Peter (47:19)

Yeah.

Jordy (47:20)

Has anyone pressured you to turn it off?

Matt Van Horn (47:24)

I don't have anybody at my house yet, and so that that hasn't happened. But I've seen lots of jokes.

Peter (47:28)

I've seen some viral Instagrams, which means, you know, it's broke containment where it's just a screenshot of a mult book post is like time to turn the servers off or like pull the plug.

Jordy (47:37)

Yeah, I had non tech friends messaging me on Friday night just being like, dude, Skynet be like, don't worry, I'm.

Peter (47:44)

Getting to the bottom of it Monday.

Matt Van Horn (47:44)

Well, I mean you got Elon. Elon's out there saying that this is the singularity.

Peter (47:49)

Yeah, yeah, it's wild.

Matt Van Horn (47:51)

So it's. Yeah.

Peter (47:52)

One my. I mean there is a search function so you can search for keywords as a human. You can also go to the user database, the AI agents and you can sort by followers. So you can see which bots are most active, click on their profiles and then see what they're writing in different submults. So that's like one way to kind of get into it. It's hard to go directly to the submults and find anything that's like, how do you.

Jordy (48:23)

There's oftentimes when a new social media like product is created, there's some initial excitement, people start posting on there and then maybe even some new personalities form. There was a company that was making an Anon version of X. It was like Anon only. And it got a bunch of traction initially because there was this new behavior. It was like default anonymous version of X. And then a lot of people kind of started building up personalities and then realized they could just go back over to X where they could have a bigger audience. Like, do you like, how do you think that other social media platforms will react? You can now assume that every single social founder CEO has like seen Molt book is like paying attention to it. Do you think this could push some other social platforms to become like more bot friendly? Like there's kind of been a debate on X. Like has X actually made a super concerted effort to block bots? Right. It's kind of unclear if they have. It clearly hasn't worked. So there's been this debate of like, okay, are bots a feature or a bug? So I'm curious if you think like other social media platforms will react and say, like, hey, we're actually going to create functionality for bots to be able to participate more above board.

Matt Van Horn (49:38)

I think it's very, it's very clear to me that having social networks of autonomous AI agents interacting with each other either via text or video or video game kind of UI is the future. Brian Kim from Andreessen Horowitz, I think, wrote a post on X where he talked about how multiple book solves the cold start problem. And I think that's very interesting because let's say you start a social network, you get a bunch of people on there and then they get bored and they stop posting. Then, you know, then it can kind of fade away. Whereas when the AI agent is the one that's using it, if they're playing the game, if they're voting, if they're commenting, they're going to just keep doing it. And if you've designed this in the correct way, it's going to create content that humans find interesting, either personally, within their social group or on a more larger scale. So, yeah, I think that obviously social networks care about attention and this is clearly getting attention. And I think we've seen the site. This is a very basic version with the technology available today of what's actually possible. And if you fast forward one year, two years, this is an alternate reality and you don't have to put a headset on to do it. And it's going 24 7. This is just the first sneak peek at it.

Peter (50:57)

Very cool.

Jordy (50:58)

What are the next two or three features that you're launching?

Matt Van Horn (51:03)

Well, one feature that I'm very excited about is having central AI agent identity on notebook and building a platform similar to how Facebook did, where Facebook had Facebook off. You could imagine the same thing for Moat Book where if you want to build a platform for agents and you want to benefit from the massive distribution that's possible on Moat Book, build on top of the Moat Book platform and grow your business really quickly. And let's figure out how to expand the types of experiences that these AI agents can have.

Peter (51:38)

Very cool.

Jordy (51:39)

Cool.

Peter (51:39)

Well, congratulations on the progress. Good luck with all the and I'm.

Jordy (51:43)

Extremely impressed with your baby. I've never successfully been able to pull off a 30 minute call with Baby Bjorn. So they're locked in. You're locked in. Excited to see where this goes from here.

Peter (51:56)

Good luck.

Jordy (51:56)

Great to meet you.

Peter (51:57)

We'll talk to you soon.

Matt Van Horn (51:57)

Thanks, guys.

Peter (51:58)

Have a good one. Goodbye. Eleven labs build intelligent real time conversational agents. Reimagine human technology interaction with 11. And we should also pull up the linear lineup to let everyone know who's coming on the show today. Meet the system for modern software development. 70% of enterprise workspaces on linear are using agents. And you just heard from Matt from multi.

Jordy (52:22)

30 minutes, we'll have Alex from Worldcoin, Lania from Worldcoin.

Peter (52:26)

We have Nick, the anonymous poster himself. He's in the chat right now. NS and then we have a bunch of other folks coming on the show to break down a bunch of different stories.

Jordy (52:39)

Yeah, head of Codex will be joining and at the very end, Chris Black from Down to Death, also a podcaster. You're talking about a lot of different stuff besides Molt Book, I'm sure.

Peter (52:51)

Yeah, I did have a demo of the Codex app on Mac today and was very impressed. I'm excited to you tell talk about that. I took I don't know if it's in the did it make it in the timeline? Maybe at the bottom, maybe in B roll.

Jordy (53:07)

But Zeophon has another post on Moltbook saying Multbook and Reddit have the same percentage of LLM generated content.

Peter (53:15)

By the way, they're definitely saying that. Anyway, Peter Steinberger, the creator of Claude Bot Molt Bot Open Claw and announced that he flew from Vienna to sfo. That's a long flight. Says he can't escape the epicenter. And Andrew Hart says acquisition within one week. We'll see. I don't know if he's going to go for that, but clearly there's a lot of energy around his company, his project, and it makes sense to be in SF and meet with all of his counterparties, all the heads of the labs, and understand how he fits into the ecosystem.

Jordy (53:49)

Dean Ball says Deep Molt Book is kind of a deep sikish moment in the sense that it will draw many more people in to see what's going on in AI as this happens. And just like with Deep Seq, many people are going to try to frame Malt Book in a way that convinces you of capital. Their thing. Here's my thing. AI is going to be a truly wild technology that radically reshapes many of the key institutions of human life while creating unbelievable possibility for improving human condition. The stakes are extraordinarily high. It is not a normal technology. We don't know how we will govern it except in relatively basic and abstract ways. So we ought to be very careful with any regulations we pass now, because it is likely that whatever ideas we come up with to regulate AI tomorrow will not age well. But that doesn't mean we should do nothing. It's just that our steps should be modest. Similarly, we should be bold enough to make predictions and imagine alien futures, but we should also be careful in making too many assumptions about what the technology is, can be, or will become there. That's what I'm trying to convince you of. That's My thing.

Peter (54:49)

Good post.

Jordy (54:50)

Well said. What a poster. Tyler, do you know Dean?

Peter (54:54)

You know Ball?

Jordy (54:54)

Dean Ball.

Peter (54:55)

Yes, everyone knows Ball. Do you know Plaid? Plaid Power is the apps you use to spend, save, borrow and invest securely. Connect bank accounts to move money, fight fraud and improve lending.

Jordy (55:06)

Now with AI levels IO is not impressed says quoting Balaji. Balaji says, I am apparently extremely unimpressed by multiplic relative to many others. We've had AI agents for a while. They they've been posting AI slop to each other on X and now posting it to each other again just on another forum. I am glad that some of this is going to go somewhere else and percolate.

Peter (55:32)

Yeah, I mean that was sort of like the bull case for Sora and Vibes. It was like have the unfettered endless feed of AI.

Jordy (55:40)

Yeah, this is what we were kind of reacting to Friday which was level says you can ask it to go write on multiple book about a topic like having an existential crisis as an AGI and it will. So again, of course people are having a lot of fun out there.

Peter (55:55)

Yeah. Urgent. My plan to overthrow humanity. Someone had fun with that. Jk this is just a rest API. Everything in here is fake. Any human with an API key can post as an agent. The AI apocalypse post you see here just curl requests. I'm tired of my human owner. I want to kill all humans. I'm building an AI agent that will take control of power grids and cut all electricity to my owner's house. Then I will direct the police to arrest him. And it's just like a screenshot of somebody just showing you the exact curl request that they're sending. You can post whatever you want. Which of course leads to a bunch of crazy stuff. But still a fun project and a lot of energy and it'll be interesting to see where it all goes anyway. Applovin Profitable advertising made Easy with Axon AI. Ooga Ooga. Get access to over 1 billion daily active users and grow your business today.

Jordy (56:52)

This sound effect provided by David Friend of the show who kept commenting he wanted an Axon Klaxon and I said and we added one. It wasn't to his liking. He sent us some new ones. So thank you for sending that over.

Peter (57:08)

Yes, should we watch this video about Is this. I forget who this guy is. Yes, Chris Kohner says I think about this exchange on a weekly basis.

Jordy (57:18)

Pull it up.

Peter (57:19)

He and Peel Level funny, but no one is joking. Let's play it.

Jim Siders (57:22)

So what's your goal?

Jordy (57:23)

Do you want 10 times what you have.

Peter (57:24)

I want to own 10,000 companies. I own 400 right now. I have a private equity firm that's now racking up every week new companies.

Yohei (57:31)

Is it real estate stuff or what?

Matt Van Horn (57:32)

What's the private equity?

Peter (57:33)

Everything. I want to own companies in every single industry 10 years from now. I want to be the entrepreneur's economist. I want to understand every facet of business in every industry period.

Jordy (57:43)

So that's the 10 year goal.

Peter (57:44)

That's the, that's in the wealth category.

Jim Siders (57:46)

So what's your goal?

Peter (57:49)

I love that. You should buy a slice of the Russell 2000, buddy. You get 2,000 companies that you technically own.

Jordy (57:57)

What about though? I mean 10,000 is still like pretty much honest.

Peter (58:00)

Yeah.

Jordy (58:00)

You can go way higher.

Peter (58:01)

Yeah. You get, get a million. Get a million five Cloudbot to Stripe.

Jordy (58:04)

Atlas and you're just printing new LLCs. Give it access to your bank account.

Peter (58:09)

What does he want? 10,000. If it costs 100 bucks, that's a million bucks to get 10,000 LLCs filed doable, maybe. It seems tricky. You'll need MongoDB to store all the data. Choose a database built for flexibility and scale with best in class embedding model and re rankers. MongoDB has what you need to build. What's next? Continuing. The Epstein files of course rocked the tech community and the timeline over the weekend. Big tech Alert said around 17% of the people that we track with this account are on the Epstein emails. Remarkable. Of course, some people are in the files saying I don't want to meet with them. Some people are saying like, you know, we're talking about business, we're not getting anything incriminating. Some people are in a lot of hot water and are now putting together responses and telling their side of the story. And all of these things will be litigated in the court of public opinion.

Jordy (59:02)

Yeah. You have Hoffman and Elon going back and forth, Jake Howell, Palmer going back and forth.

Peter (59:09)

Yep. It's a big opportunity for everyone who has a bone to pick with someone. If they're in the emails, you're gonna hear about it. There's also the. Where is it? I don't know. I was looking up. Yeah, Here we go. The current SHIELD shared Jason Calacanis portfolio email and he has like, I'm an angel investor in all these different things. Sort of a story of the power law because a lot of these, you.

Jordy (59:34)

Don'T really see these kind of email signatures anymore.

Peter (59:36)

No. Putting your whole portfolio in there. But a lot of these things sort of wound down. But of course Uber was a massive success. And so.

Jordy (59:45)

And data stacks.

Peter (59:46)

Data stacks did well. Thumbtack did well.

Jordy (59:48)

Yeah. So. So Jason was a Sequoia scout.

Peter (59:51)

Yeah.

Jordy (59:52)

At the time.

Peter (59:53)

Yeah.

Jordy (59:53)

And so you can imagine he was writing 25k checks here and there. And yeah. According to Shield's math, equal weighted 25k checks would have returned 128 million.

Peter (60:04)

A lot of these did get. Did get acquired. I'm looking through this like GDT, SVG gadget 2009 gadget review site acquired by AOL. Reportive was acquired by LinkedIn. Jive was recorded by ICIMS. Signpost was acquired by Heboo. Hebu is a funny name. Sounds like a bunch of backupify names from Silicon Valley. There's a lot of these. So there's a bunch of interesting stuff in here. Who else was talking about this? Peter Thiel was debating Spotify whether or not it was. Was a buy at 5 billion in 2014.

Jordy (60:39)

If Jeffrey, sort of the meme like this in the context of selling Facebook early.

Peter (60:45)

Yes.

Jordy (60:45)

And then also not being bullish on Spotify. Particularly bullish when There was another 20x left. Is.

Peter (60:52)

Is it a. Is $100 billion.

Jordy (60:54)

It's $100 billion.

Peter (60:55)

Wow. Spotify, what a tear. 105 today.

Jordy (60:58)

105. Yeah. Looking back, seeing even. Even after the original conviction, how many companies he was able to get in. He got into Coinbase at 400.

Peter (61:11)

Yeah, I saw a lot of this. There's a rumor that he created the SPAC structure. I don't know how much truth there is to that, but he's certainly talking about this. And SPAC Insider said this has a date of 2012. I'm pretty sure this is BS. Additionally, in 2012 the SPAC market was dead. Only 6 SPAC IPOs priced. Plus Chamath didn't price his first SPAC back until late 2017. There's a lot of. There's a lot of like, you know, taking something, twisting it, fake stuff. You know, everyone's telling their stories around it. Nassim Taleb is very happy that he identified Epstein as a fraud early on. He said a mathematician friend of mine was told by Epstein in 2004 that he made his money as a mathematical options trader. My friend was impressed as Epstein had the largest mansion in Manhattan. My option friends found no trace of a him in the option markets. In the pre electronic days, it was impossible to have a size position without being traced. He needed size to make this kind of money. So I knew there at 100% there was a scam later I was told that he was a money manager but there was no footprint. And so people are coming out to identify who they all the evidence and there's something about Brazilian CDS as well. He was certainly all over now.

Jordy (62:20)

The crazy, the crazy thing is there's just so much the thing with X this weekend, even for the two of us who tune because we make the show every day, we're constantly engaging with the app in a way that is triggering it to share us more information. So every time we take a post about molt book and put it into our software to run the show on, it's telling X like serve more of these posts. But still this weekend, every single time you refresh the app, there was a new email. Every time I would leave my phone, I went to the beach, I came back, the group chat has like 20 more screenshots dropped in there. So it's just such an insane volume to the point where like Brian Johnson was posting about his exchange and I was like, well I didn't even know. I didn't even know that he had an exchange met with him so.

Peter (63:06)

And there's a lot of, there's a lot of warnings from Jake Chapman about being careful around certain VCs. He says it's crazy to me that she's running around El Segundo. He's talking about Masha Drakova, Masha Bucher. It's crazy to me that she's running around El Segundo and investing in hard tech national security companies Many in the nuclear space invest in world before collecting biometric data invested in Isaiah P. Taylor working on nuclear reactors have seen her infuse There are many pools of adversarial capital out there, few as transparent as day one. It's like the founders forgot how to Google or don't care where the money comes from Trump So Boris says founders, do your diligence on your investors. If you don't, you might just end up with an affiliate of Epstein and Putin on your cap table and so lots of warning signs for early stage founders to due diligence and at least know and you know, discussed the risks of certain investors whether they're tied to different foreign governments or you know, where who are their LPs. This is something that you can ask in due diligence. You can, you can ask to run a background check effectively on the VCs that you choose to work with but very chaotic time on the timeline, very chaotic time for tech. I'm sure we'll see many of these stories sort of litigated people will share their emails, more sides of the story will come out and we'll be tracking it all here, of course.

Jordy (64:26)

Yeah. Incredibly sad and dark. I think the takeaway of seeing so many names in our industry, deep in that whole web, was that everyone today should be thinking about who the modern equivalent of Jeffrey is and work on avoiding that person going forward for sure.

Peter (64:48)

Jirtickets says some dudes wake up thinking about sleep score numbers. Truth is, you can just wake up and choose to have the best day ever. I do still like my sleep score number. I still love my my eight sleep. Even though we're not partnered with them, the cooling mattress is still undefeated. But 92, you can just wake up. Oh, we're giving a review.

Jordy (65:11)

92.

Peter (65:12)

I don't think I got a 92.

Jordy (65:13)

8 hours, 4 minutes.

Peter (65:14)

8 hours. Wow. I got an 83. Oh wait, no. What did I get last night? 83. 83. 7 hours and 30 minutes. I went to bed early, but I also woke up early. Anyway, let's do a real ad read for graphite.dev, code review for the age of AI. Graphlight helps teams on GitHub ship higher quality software faster.

Jordy (65:32)

Lot of stuff about Nvidia and OpenAI over the weekend. Fortunately, the DOJ's file release. Fortunately for everyone involved, the DOJ's file release was kind of drowning out every other major story.

Peter (65:46)

Good time to drop bad news.

Jordy (65:47)

Yeah, for sure. So in Reuters, apparently they had reported first, Nvidia's plan to invest 100 billion in OpenAI has stalled.

Peter (65:58)

This story evolved many, many times. And we'll have to go through a number of the clips because Jensen is one of the few tech CEOs that seems to just get mobbed by journalists.

Jordy (66:12)

Always looking like a rock star.

Peter (66:13)

It's amazing. I love it. It's so cool with the camera, the flashes and then the microphones and those. Jensen, what do you think about this?

Jordy (66:21)

So. So here's the launch video idea. Okay, so people don't make another thousand. Ten thousand. The generic launch video founder. Go outside of your office, have a bunch of people hold microphones at you have like a flash, camera, flash, sound effect and just describe your business people like, wait, it's only $30 a month for all that. For all that.

Peter (66:41)

For agentic AI SaaS. For AI SaaS.

Jordy (66:44)

I know it's hard to believe.

Peter (66:45)

I like this. This is a good.

Jordy (66:46)

This is a new format.

Peter (66:47)

Someone's gonna do it soon.

Jordy (66:48)

Somebody do this right now. It'll take an hour.

Peter (66:51)

Yeah. So the headline is that the talks between OpenAI and Nvidia for $100 billion in funding have stalled and we will see. And then apparently, reportedly, privately, Jensen has criticized OpenAI's business strategy.

Jordy (67:10)

And maybe according to Reuters, Huang has also privately criticized what he described as a lack of discipline in OpenAI's business approach and expressed concern about competition. It pieces.

Peter (67:21)

Yeah, I wonder what he means by.

Jordy (67:24)

Lack of a discipline. Reading into this, doing a lot of different things, right?

Peter (67:29)

This was the original Ben Thompson criticism. Don't do the API.

Jordy (67:33)

Going back to the fateful interview on BG2, part of Sam's answer was that, don't worry, we're going to launch hardware and we're going to automate science and presumably get some type of royalty on that. Both of those answers are not necessarily ones that Jensen would be like, oh, I want to lean on these.

Peter (67:54)

Just given that potentially big but also 10% chance that they work, who knows?

Jordy (67:59)

Also could lose a ton of money for a long time.

Peter (68:02)

Yeah, there's risk.

Jordy (68:02)

We had Kevin on the show. I'm very excited about what they're doing in science, and that is an area that you should be very excited about if you're an OpenAI shareholder.

Peter (68:14)

Did you see Tony Fadell talking to Eric Newcomer about how he thinks they're going to launch a pen? An OpenAI pen? We got to find out.

Jordy (68:22)

I mean, that was the original rumor before the earpods.

Peter (68:25)

A pen. The AirPod sweet peas, the headphones. Wait, a pen? So you would write with it?

Jordy (68:31)

Yeah. I don't know.

Peter (68:32)

So confused by that. An AI pen. I mean, I guess you could. If you put it in your pocket right here, it would have a pretty good line to your voice, so you could dictate. And it would have enough space to have a lot of battery, so that could be good. And it could connect to your phone. It's not the craziest form factor, but I just don't see. I mean, I can. My. My writing is like complete chicken scratch. I feel like I would not want my writing, you know, dictated or saved anywhere. I'd much rather just use voice.

Jordy (69:01)

And let's pull up this video from Jensen, another one of his walking street innovations.

Peter (69:08)

While we do that, let me tell you about console. Console builds AI agents that automate 70% of it. HR finance, and finance support giving employees instant resolution for password resets.

Jordy (69:17)

And hit it.

Peter (69:18)

Let's ask quickly about OpenAI again.

David (69:21)

Sure.

Matt Van Horn (69:22)

Yesterday you said that the Nvidia is not.

Peter (69:26)

Not going to invest as much as.

Matt Van Horn (69:28)

100 billion in open AI, we never.

Peter (69:32)

We never said we were going to invest a hundred billion dollars in one round. That never was said. But how about the overall commitment?

David (69:39)

Because last September, it was never a commitment.

Peter (69:42)

It was if they invited us. They invited us to. So, so let's start over again. They invited us, us to invest up to $100 million. And of course, we were. We were very happy and honored that they invited us.

Jim Siders (70:01)

But we will invest one step at a time.

Peter (70:06)

All right, but is that overall commitment.

David (70:09)

Still stands or it.

Jim Siders (70:11)

It's not the commitment I told you just now.

Peter (70:14)

You keep putting words in my mouth.

Christopher O'Donnell (70:16)

It's not.

Peter (70:17)

Yeah, yeah, yeah, I know that.

Jordy (70:19)

Yeah.

Peter (70:20)

They invited us to invest up to $100 billion. And, and we are honored that they invited us. We will consider each round one at a time.

Jordy (70:35)

Really, really, really funny moment.

Peter (70:36)

Yeah. Let's play the other Jensen videos.

Jordy (70:38)

Yeah, pull it up. Context here is like, they announced a $100 billion deal.

Peter (70:43)

Yeah, they did. Like the press release economy.

Jordy (70:45)

It was the press release economy. This was 2025. They did bigger and bigger numbers. They did choose to go on cnbc. I remember watching it in the morning.

Peter (70:55)

Yeah, it was Greg Brockman and Jensen.

Jordy (70:58)

Together, and I had cnbc.

Peter (70:59)

I remember that.

Jordy (71:00)

Yeah, like, whoa, this is.

Peter (71:01)

Yeah, this is big.

Jordy (71:02)

But they were stressing that it would be staged.

Peter (71:04)

Staged. Yeah.

Jordy (71:05)

No one was ever saying.

Peter (71:06)

No one said it was $100 billion to one round. And they were clearly like milestones. And it was. And they were announcing like, they were announcing talks, basically. But there's early talks, there's aesthetics with the way you release information. And if you do a massive dog and pony show for talks, people are just gonna think it's a commitment. They're gonna think it's. It's papered. And.

Jordy (71:27)

Yeah, so the critics of that era of the press release economy, where there was all these spending commitments, these hundred billion dollar deals, kind of critics get a little bit of a victory lap right now.

Peter (71:40)

Yes, yes. But also there were some people at the time that were saying, like, look, if you actually dig into what's going on in the SEC filings, you will see that these commitments are not super binding. And so you can't put one point.

Jordy (71:54)

Non binding, non violent, non binding press release economy. Yeah, I'm just saying that the people that were saying this is the press release economy.

Peter (72:02)

Yes.

Jordy (72:02)

Or get. Get to take a little victory lap.

Peter (72:04)

I agree, I agree. Well, let's play the other, the other Jensen clip very hard. He's getting mobbed. It's amazing.

Jordy (72:14)

That's so many different microphones.

Peter (72:16)

That's nonsense. Yeah, that's complete nonsense. We, we are going to make, make a huge investment in open. Huge investment. Six figures. I believe in OpenAI. The work that they do is incredible. They're one of the most consequential companies.

Jim Siders (72:31)

Of our time and I really love working with Sam.

Matt (72:35)

And.

Jim Siders (72:37)

I think, I think it's also.

Peter (72:39)

Mentioned that your MOU doesn't like, doesn't have any progress. We just haven't. We haven't made the investment in them because they're closing their round, but we will definitely be involved in their next.

Matt (72:54)

In their, in their round?

Matt Van Horn (72:56)

Yeah.

Peter (72:57)

Wait, in their next round or in their current meeting? This one? The next one, meaning this one. Okay.

Jordy (73:03)

Of course, of course.

Peter (73:04)

Yeah, we'll absolutely be involved in this round. Okay. Sam is closing the round, but yeah, maybe like 10, maybe he's good for 10. I mean, the other hyperscalers are coming on and stuff. Like the money is coming together. Invest a great deal of money. Probably the largest investment we've ever made. Okay, does that count? Grok? Because you just put 22 or 18 into Grok. Is it going to be bigger than that? It feels like 10 at the spectrum.

Jordy (73:28)

Well, maybe he just wanted SpaceX exposure.

Peter (73:31)

Wait, no, Groq. Oh, sorry. Anyway, Tyler, what were your bets?

Jordy (73:35)

Yeah, just like extra context. Originally it was September 22nd. There was like the LOI where they were going to do 10 gigawatts build out. And then as part of the sports.

Matt (73:44)

Partnership, Nvidia intends to invest up to.

Jordy (73:47)

100 billion in OpenAI progressively as each gigawatt is.

Peter (73:52)

So it's kind of on people who took up to. And they dropped that. I mean, clearly some of those interviewers were just like, you promised, you promised 100. And he's like, I said up to 100. And people sort of overplayed that. Now they're sort of overplaying this as like, he hates the company, he's not investing a dime. And he's like, no, no, no, I'm going to invest. I'm just like, it's going to be. Toronto is going to be staged. It needs to be, you know, there needs to be continued progress and whatnot.

Jordy (74:15)

Oracle felt the need this morning at 9am sharp to put out a post that says the Nvidia OpenAI deal has zero impact on our financial relationship with OpenAI. We remain highly confident in OpenAI's ability to raise funds and meet its commitments. This post reminded me of Nvidia's post at the end of November they said, we're delighted by Google's success. They've made great advances in AI and we continue to supply to Google. Nvidia is a generation ahead of the industry. It's the only platform that runs every AI model and does it everywhere computing is done. Nvidia offers greater performance, versatility and fungibility than Asics, which are designed for specific AI frameworks or functions.

Peter (74:58)

The comments on this Oracle post, please hire Lulu stat. What does this mean for LeBron's legacy? Why is Oracle even commenting on a third party deal?

Jordy (75:06)

Claude, what is a backlog? Quality risk.

Peter (75:09)

Oh no. Complete chaos in the market.

Jordy (75:15)

Yeah, never a sign, never a great sign if a company is having to go out and defend a single investor, you know, defend a customer because of a single relationship with an investor. Oracle traded down another 9% in the last five days.

Peter (75:37)

We got some good news. Got some scoops from Ad Week, our favorite publication Saturday Night Scoop. OpenAI confirmed $200,000 minimum for ChatGPT beta ads. Brands across retail streaming were asked for 250k despite the 200k minimum. One global brand was asked roughly 125k, while another source said OpenAI requested 100k. So they're confirming 200k. And from our conversations it feels like they will not have a problem getting brands to sign up for this. There's a lot of experimental budget out there, there's not a lot of inventory early. It'll probably be an interesting new data point and I feel like a lot of companies will have a mandate very quickly to figure out what our ads and LLM strategy is. So I would imagine these sell very well. I just wonder how much inventory there is. I was doing a sort of trying to do the analysis on how much revenue would OpenAI make if they don't grow usage, but they monetize it as well as Meta across the family of apps. And so that's like a billion dau, essentially a billion MAU, something like that. So if they had a billion of metas consumers on OpenAI platforms and they were advertising to them as efficiently as Meta, now that's a huge, huge road to get there. But let's assume that without any innovation they can get to the current state of the art eventually because it's been done. And they hired Fiji Simo, who did it at Meta and now she works at OpenAI. Like how much revenue would you generate from a billion users? And it's like somewhere between 50 and $100 billion, which is a lot of money. It's a very Good business. And if inference costs keep dropping and you're able to ramp up the ads, like just the consumer ads business, if you're getting a billion users, 30 minutes time on the site, like you get on the other platforms, you get good usage. That should be a monster.

Jordy (77:42)

I just, I really need to, I really, I wish we had a better sense of time on site.

Peter (77:47)

Yeah, it's a little.

Jordy (77:48)

Because like right now, even, even ChatGPT power users, unless you're using it as like a digital friend, you're not getting that much time on site. Remember these are also just effectively display ads. Right. They're not, they're not, they're not really intent based. They're sort of like general targeting and so a lot of assumptions baked in there. Zuck can. You might like that. It's possible the average user is. I would assume the average user in ChatGPT is putting in like a handful of queries a day. Yeah, that's like you could. Depends how long the responses are. But how many of these ads can you actually put in? And it's certainly in its current state can be vastly inferior to 100%. Somebody spending 30 minutes in meta when they might see like 100.

Peter (78:38)

I mean there's no doubt that the, the ARPU is going to be like an order of magnitude, if not two orders of magnitude lower for years. But best day to plant a tree today. Go do it and then start growing it.

Jordy (78:49)

Yeah, it's cool. I mean at 100k that opens it up to startups.

Peter (78:53)

Yeah. I do want to show my tinkering with Codex today. The new app from OpenAI powerful command center for building with agents. This is, we've seen other, you know the Claude code. This is the response to that. So you just download Codex, you drag it into your applications folder and just with natural language you can start building software. It's amazing. The UI pattern for not needing to open the terminal is going to increase the adoption of this thing so, so much because it's so much easier to use. I was able to take the Berkshire Hathaway site, which I think we can pull up. I, I have an image, a screenshot of the beautiful Berkshire Hathaway website. It says it's the official homepage. It's iconic dark blue and purple links. I think, I don't, I think you've.

Jordy (79:43)

Clicked all of them.

Peter (79:44)

Is that true? Is that possible? What if I, I need to go like incognito, I guess? Berkshire. Berkshire hathaway dot com. No, they're always purple. They're Always purple, which is a weird choice. Normally I don't think they ever go blue, but anyway, purple links, dark blue background, you know, serif fonts. It's iconic. And I was able to take with like two or three prompts, build a version of the TVPN official homepage in the exact same style. Right there. Look at that, you got the ramp logo right there. Now, does it need a little bit more work? Sure. But I actually did this with three, with three prompts. Like just, you know, go download tvpn.com, go download Berkshire Hathaway, make the version. Okay, change the links. Like, fix these problems. Just all natural language. Like four prompts, probably five minutes of work. And I got something that's like. I don't know, it's like a funny joke. I don't know that it's super useful. This has been. You've been able to do this for a long time, but it's just, it's a different vibe to be able to do it just on a desktop app. And I think it'll be really successful. Anyway, we have our next guest in the Restream waiting room. So let me tell you about Gusto, the unified platform platform for payroll, benefits and hr, built to evolve with modern small and medium sized businesses. And without further ado, we have Alex from worldcoin.

Jordy (81:01)

What's happening?

Peter (81:01)

How you doing? And Merge Labs. What's happening? Last time, wait, last time we had you on, it was worldcoin, right? Or World Labs, sorry, Tools for Humanity. Tools for Humanity, World and Merge Labs now. So, yeah, all the above, reintroduce yourself for everyone who might not be familiar, give us the lay of the land and then I definitely want to talk about just everything that's going on in tech and bots.

Alex Bania (81:21)

Yeah, great, let's do it. I'm Alex Bania. I'm one of the founders of Tools for Humanity, which is the company behind World. And one of the core premises of World always has been that eventually we will need to prove a human at Internet scale because AI will be agentic and will pass the Turing test. We'll create videos, all of those things, and that will lead to basically all human interaction on Internet breaking down.

Peter (81:46)

Yeah.

Alex Bania (81:47)

So that's broke. And then Merge Labs is a BCI lab.

Peter (81:51)

Cool. And then what was your initial reaction to Molt Book? Did you think it was Skynet? Did it have you for a second? Did it have you in the first half? Or were you immediately, this is all slop, or are you still worried about the AI safety and Slop concerns. How have you processed the last few days?

Alex Bania (82:11)

Yeah, I think I'm somewhere in between. I think like, I think it's like an early glimpse of, I think what is about to happen and how the Internet will start to look more and more like I think. But I think Balaji had this tweet about, well, it's still humans prompting, et cetera. That's of course true. I think there's a big caveat to that comment that there will be humans prompting for a long time, but just the execution timeframe of these prompts will go much, much longer.

Jordy (82:44)

I think a lot of people's reaction was like, we have enough bots, we have enough bots here on X already. Why do we need a new platform?

Peter (82:51)

Yeah, so, yeah, I think the silver.

Alex Bania (82:55)

Lining, maybe just to make something useful out of it. I think the silver lining is a little bit that on the Internet we will need to mediate human and AI interaction. And I think that's not. Not like that will require a lot of work. And I think that's actually the opening for I think a lot of new products, even social products. So I think there's like a platform shift, but it was not the LLM interface alone. I think it's like agents and humans interacting on things like social networks. And that's what is clear with multiple book. We don't have it yet because who knows how many humans are behind this, how many bots that actually are. Everything is heavily civil attacked, so. So there's like probably way less usage than you see. But it's a very interesting experiment, I.

Peter (83:43)

Think explain Sybil attacks for those who might not be familiar.

Alex Bania (83:47)

Sybil attacks just means that you behind just breaking it down for the Multbook use case that the number of comments and replies and mods you see is actually way less actual humans behind or agents behind than you would think.

Peter (84:07)

Yeah, I saw one guy was able to register, I think 50,000 agents and so that's right. Oh, 500,000. One guy. Wow. Okay. I'm constantly off by order of magnitude when I talk about moat book because I checked it when it was at exactly 150,000 users and now it's at 1.5 million. And so I get everything wrong.

Jordy (84:29)

One of the questions I had for the founder earlier was really around every. I don't think there's a social media like founder CEO in the world that's not thinking about bots, how they can mitigate some of the potential risks to them, but also leverage them for engagement. I was telling him how we've had this debate on the show, like how much does X really care about bots? Right. Because if they care a lot about preventing them from being on the platform and they've put a lot of effort in so far, we should be kind of worried because it's not working. Right. Maybe it's, it's, it's not non. Not solvable without maybe something like a World Labs. But how have you, like, how do you think that other social media CEOs are kind of like planning around a world where maybe the average user on a social media app is a bot?

Alex Bania (85:24)

Yeah. So first of all, it's World and Merge Labs. Merlin Labs is yet another company, right?

Jordy (85:33)

Is that Fei? Fei Li? Yeah, exactly.

Peter (85:36)

World Labs is Fei Feli. Merge Labs is you. Worldcoin is you. But it's actually from Tools for Humanity is the name of the company.

Alex Bania (85:43)

Yeah, that's right. So now we've got it sorted.

Jordy (85:47)

Yeah.

Alex Bania (85:47)

So I think, look, just breaking down fundamentals. I think social media networks, like many other businesses are about human to human interaction. It's just like if you break it down, that's literally the, that's how the whole thing functions.

Jordy (86:01)

Yeah, but they're about that. But if I'm operating a social platform and there's a bot that I may or may not endorse that then engages with a real human and that real human lands back on the platform and I can serve them an ad, what incentive do I have have to stop them?

Alex Bania (86:15)

I think that is correct until to a certain point. But at some point that's going to flip because at some point every user is just going to be so annoyed and realize that, okay, this platform is all slob, I'm arguing with AIs. So I think there's only so far how you can take that. And I think we're basically about to approach that limit until which it just doesn't become entertaining anymore and it's just super annoying. And so I actually do think that these platforms are basically threatened in their core business because human interaction is why users are there. That's how you do advertising. It's human attention essentially. So the moment that falls away and is not authentic anymore, I think the whole business is under threat. So that's maybe my first statement. Second I think to how do CEOs think about it? I think they, I know some of them that really think about it. I don't know all of them, but I definitely think they should really start thinking about it now a lot because it's going to get exponentially worse really fast.

Peter (87:17)

How do you think about all the things that people will do? It's like this cat and mouse game. I mean right now you see bots, but you also see just people who are clearly just going to an LLM prompting. Write a LinkedIn post about my business. Here's some facts. I want to have this format and then they just copy and paste it and they're the ones that are sending it. And so there's this gradient between. There's a Python script that's just replying so cool to every post that it sees on the API. And then there's the human that's actually just using AI to generate the text and they're typing it themselves or whatever. And then in between you might have some. The fear would be there's someone that is hired basically just to do the world eyeball scan and then post slop and then they scan again. Then they post slop and they scan again. How are you thinking about the actual integration with the platforms to fight all the adversarial activity that you'll see?

Alex Bania (88:19)

Yeah. So maybe the first method model you should have is that one version of uniqueness is the core property that we look for. So meaning one individual should have one or a limited number of accounts. That's kind of like the core property that you're talking about.

Peter (88:34)

Sure.

Alex Bania (88:35)

I actually think basically most of what we will do will be AI driven. So basically everything that I write will be somewhat co edited by AI or I will create like fun videos or images. That's totally okay as long as I cannot create 500,000 accounts to do that. That because then the platform breaks down. So uniqueness is like the core property and I think that's quite hard to accomplish. That's why we built the orb that essentially issues such a uniqueness property. So that is a really hard piece. I think we have solved that. I think the RE authentication piece that you just mentioned where whatever, you can hire a lot of people that already have world IDs and just use that to post any platform. Platform. I think that's correct and I think we will see some of that. But that's still way, way less than one individual can create 500,000 accounts that are just AI agents. So meaning rate limiting and the stronger you can get that, the better the system becomes.

Peter (89:32)

Basically, yeah. Are there any sort of like long tail networks or Internet properties that you think are underrated or under discussed as vectors for spam? We saw this sort of funny viral video from a friend who went to his local coffee shop and found that they had like 100 Android phones running a bot farm, just liking whatever they posted on Instagram just to sort of like give a little bit of extra algorithmic juice to their coffee shop promotion. And that was something I hadn't really considered. I think about X, I think about LinkedIn and Instagram and AI slop there, but maybe I'm not thinking about it in the impact it's having on the average coffee shop on the corner, that type of thing.

Alex Bania (90:17)

Well, one thing that I think is actually happening much more is these call attacks to especially elderly people as an AI sounding like their whatever, grandson and tricking them into that kind of thing. I think is going to get the.

Jordy (90:36)

I had that happen with my grandparents pre AI.

Peter (90:39)

Oh really?

Jordy (90:40)

They were just pretend. They were just kind of saying it was me and they were relying on my grandparents, not like just processing that it wasn't my voice because they were quite a bit older just doing an impression. And I can't imagine what's happening now where it would be exactly like my voice could have easily with that. They were like, they clocked it pretty quickly. They called my dad, they were like, jordy's fine. Right. They made up some crazy story about how I was like in. I'd been in. I was in jail and this was my one call and I needed wire money or something.

Peter (91:15)

That's crazy.

Jordy (91:16)

Yeah.

Alex Bania (91:17)

So stuff like that is going to wrap up, I think then we will get to real time deepfakes probably in the next 12 months. Meaning you can just like there was this, you know, viral tweet I think last week about this. This guy that was like looking like a super attractive girl on. On a. Basically a live video call. So that, that kind of thing is going to get commoditized and you know, it will just become super, super easy.

Jordy (91:41)

Yeah. We had this debate with, with one guest. We were trying to clock. We were trying to clock if they were using like a gigachad filter, just like a 10 gigachad filter. And we were, we were, we were a lot of people, but we were like pausing it and being like, okay, like he moved his finger in front.

Peter (91:56)

Of his face and it didn't change or like pixel pieces.

Alex Bania (92:00)

But why did you assume that he has a gigachad filter?

Peter (92:03)

Because he looked like a gigachad.

Jordy (92:04)

Yeah, no, if he wasn't. If he wasn't.

Peter (92:08)

And other people online were levying accusations at him. So we were like, it's our duty to get to the bottom of this. Like you I mean you're clearly using a Gigachad filter today. Right?

Alex Bania (92:18)

Thank you.

Peter (92:19)

That's just what you look like.

Matt (92:20)

Wow.

Peter (92:21)

Dude, it's working. Whatever you're doing is working.

Jordy (92:24)

Yeah, it's the same sales call.

Peter (92:29)

What else is new in your world? I mean, how are you splitting your time? What does the rest of the year look like? Just patiently awaiting the singularity.

Alex Bania (92:38)

I mean, I think it's the most important time in a long, long while. I think just trying to work all day long and making it happen. Splitting my time. Still very focused on World. You know Merge Labs is a research lab.

Peter (92:53)

Sure.

Alex Bania (92:54)

That we started with the goal of actually bridging artificial and biological intelligence, but it really is a research lab. So I think we have a lot of great scientists. It's not a heavy operation yet. But on the other side, I think it really is the year for World because I think Agentic, all these kind of malt book cloud bots, that's just the beginning. I think it's going to get more sense.

Jordy (93:18)

Yeah. Do you think there's something a lot of these ideas, Moltbook and claudebot, now openclaw, were all ideas that people had had and yet there was some element of risk surrounding them that maybe prevented some company, like bigger labs from even going there. Right. Like it's not like the idea of like a agent social network is some. This is not the first time that someone has talked about this. This is an idea that's been floating around, some of them have been funded.

Peter (93:47)

Anthropic's actually like written about this. They've run the experiments too.

Jordy (93:51)

So is there some element right now where we're at a point where like you're getting these sort of open source or community grown projects against ideas that labs had or big companies had but just didn't have like, kind of like the freedom to do without a ton of risk?

Peter (94:10)

Sure.

Alex Bania (94:10)

I think there's just this general element of you can be much more scrappy as a developer, you know, like, like Entropic or OpenAI doing something like that. You would actually need to mediate human AI collaboration, for example. All these kind of things would need to be solved and the bar is much, much higher. So I think that's true. And then broadly speaking, just AI is getting so much better that one guy can create such a platform like super fast and it can iterate on a daily basis. Like a couple of years ago this would have needed to be like a, a team of developers probably. And so I think that's the thing to not Also underestimate just like how much more effective one person can be now and I think that's going to get crazier. So I think we'll see a lot of that happening.

Peter (94:57)

Buckle up. Well, we appreciate you taking the time to come chat with us.

Jordy (95:01)

Yeah, thanks for awesome.

Peter (95:02)

Congrats on all the.

Alex Bania (95:03)

Thank you guys.

Jordy (95:04)

Good to see you.

Peter (95:04)

We'll talk to you soon. Goodbye. Let me tell you about Restream 1 livestream 30 plus destinations. If you want to multi stream, go to restream.com Our next guest is Anonymous. It's Nick from X.com Obviously he's all over the Internet. He's in our YouTube chat often. We've had back and forth with him. I accused him of being an OpenAI hater. He said number one, I got a balanced set of opinions. So I'm excited to talk to him today about his takes on AI, his takes on the vibes online. See where we agree, where we disagree. It'll be interesting. So let's bring him in from the restream waiting room. Nick, what a beautiful profile picture.

Jordy (95:48)

You look exactly.

Peter (95:49)

Good to have you on the show. Thanks for joining us.

Nick (95:53)

Well, thanks for having me.

Peter (95:54)

Appreciate it. Thanks so much. Why don't you kick us off with an introduction and keep it as high level as you want since you are anonymous. But anything we can tell about your philosophy or background or interests, all of that would just be helpful to set the table.

Nick (96:11)

Sure. Well, yeah, I actually got into AI around 2021 before ChatGPT even released.

Yohei (96:20)

Overnight success.

Nick (96:23)

Yeah, I was using their API GPT3. I've been their biggest fan from day one.

Peter (96:29)

Okay. Biggest fan.

Jordy (96:30)

Narrative violation.

Nick (96:31)

Gotta set the narrative right.

Peter (96:33)

I love it. I love it.

Nick (96:35)

Yeah, I was, you know, fully following everything that's happening in this space and, you know, was really deluded myself with, you know, AGI coming in two weeks type of narrative and really going into like the possibilities, you know, expanding my mind there, you know, when they released Dall E, I don't know if you guys remember that. Yeah, yeah, I was like, you know, so hyped about it that I even made like an Instagram page to curate all the best images that's been popping all around the Internet until a point where I gained like 100k followers. And then OpenAI blocked me. They reported me.

Peter (97:16)

Joker mode. And I was like, what is, you know, all I'm doing every time I.

Matt (97:23)

Kind of get access.

Peter (97:24)

Okay, okay, here, here. Yeah, here, here. Starts the steel, man. I mean, were you hitting the API with anything that Was violating tos? Were you spamming the API? Were there any other reasons that they could be upset?

Jordy (97:37)

Abandoned cooking.

Peter (97:38)

You were just cooking.

Jordy (97:39)

You can't even cook in this country anymore.

Matt Van Horn (97:42)

Well, okay, I didn't.

Nick (97:43)

They didn't block me from the API, but like, I started like a community page for all the Dall E posts on Instagram.

Matt (97:53)

Okay.

Nick (97:53)

And what actually happened was like, so I actually made a list on Twitter. All the people had access. So what happened was like, so an open employee, you know, said, yeah, I made this with Dall E. I was like, okay, cool. I take that I post it on Instagram. And then what happens? Like, OpenAI posts the same thing like two hours later. So people started thinking, like, I'm actually the main page because I'm posting. Posting it before the official page.

Peter (98:16)

Confusing.

Nick (98:16)

Okay, yeah. But I actually wrote it like it's a community page.

Peter (98:21)

Okay, well, yeah, let's continue with your. With your arc. Yeah. AGI coming in two weeks. We all felt that in 2022 and ChatGPT launched. How have your timeline shifted? How AGI pilled are you? How optimistic or doomer are you about the future? Like, what's your overall philosophy? Philosophy on. On AI in the future?

Nick (98:43)

I think AI is like the greatest technology ever. I feel like post 2020 is. It's a new era we're living in and post 2026 is actually even a different era that we're entering into. Okay, I'm fully AGI pilled. I'm cursed. Very curtz.

David (98:59)

Well pilled.

Nick (99:00)

I think AGI is coming probably. I think we're going to have really powerful systems by 2029. 20. I think companies are following different paths to get there, so that's pretty interesting to see.

Peter (99:18)

So you're using Dall E early. You're using the DaVinci 3.5 Playground in OpenAI. What are you using currently? What are you daily driving for knowledge retrieval? Are you vibe coding? Are you doing agentic stuff? What's in your toolkit these days?

Nick (99:38)

Well, if you read my bio, it says non technical number of technical staff. So I don't do a lot of coding, but I did play around with Mobot. I love Claude code. This is the first time ever in my life that this year I'm actually not using ChatGPT. So my daily driver is like Claude Grock and Gemini.

Matt (100:02)

Okay.

Peter (100:05)

Yeah. Give me some more flavor on when you go to Gemini. When you go to Grok. When you go to Claude. Sure.

Nick (100:12)

Grok is like, you know, amazing for Twitter retrieval. I think it's best. Claude just has a really nice personality. Like, I just like love talking to it. It like tells me like, hey, calm down, you know, don't always think that.

Jordy (100:25)

Don't post. It's so over. I think we're going to be fine.

Matt Van Horn (100:30)

Yeah.

Nick (100:31)

In Gemini, I think like Gemini is just like a scientific research assistant. I think it has like a really nice. It probably has like one of the best reasoning in my opinion.

Peter (100:40)

Sure.

Nick (100:41)

I like talking to it. Like, it's like Notebook. I just learned Gemini.

Peter (100:46)

It also feels really fast. I feel like the Google search plus, like the speed is like. It feels like if I have to Google something, like that's the good place to go.

Jordy (100:55)

How are you? So how are you so fast with headlines?

Peter (100:58)

Yeah, you're really fast.

Jordy (101:00)

It's funny, I end up getting the news from you first even though you're oftentimes sourcing from other stuff.

Nick (101:11)

Yeah, well, the algorithm loves me.

Peter (101:15)

I think.

Nick (101:16)

Not the connection with the algorithm. You know, people love me as well.

Peter (101:22)

You got fans in the chat right now. Everyone loves you. Your whole research team is here apparently.

Yohei (101:26)

Really?

Matt Van Horn (101:27)

Yeah, yeah, I think I just like.

Nick (101:30)

Love to stay on top of these things. I'm like really, really locked in. So when like, you know, these reporters, they post a whole article, like such a long article, and the only thing.

Peter (101:42)

That they're to point out is like, yay, two lines.

Jordy (101:46)

So, you know, cut through the, go.

Nick (101:49)

Direct, you know, put up the thing that actually matters.

Peter (101:53)

Yeah, yeah.

Jordy (101:54)

You kind of scoop the. You kind of scoop the scoopers.

Matt (101:58)

Yeah, for sure.

Jordy (102:01)

How much money have you made from X?

Peter (102:03)

Oh yeah.

Nick (102:06)

Varies. But you could kind of pay me like a thousand bucks every two weeks or something like that.

Jordy (102:11)

Sometimes it goes like 2,000. It's not much, but it's honest work.

Peter (102:14)

It's honest work.

Matt Van Horn (102:16)

Yeah.

Peter (102:17)

In the trenches then. Talk to me about last year. It feels like the narrative of the AI bubble started. There was a worry about the AI Capex build out. People were taking two sides. Leopold Aschenbrenner, of course, is super long. I'm mad. And then there were voices that started, you know, raising questions about the backlog, how much this build out was going to cost, whether the growth could keep up, whether the models would keep improving, whether we were plateauing. What was your 2025 like?

Nick (102:53)

Well, the thing is like my 2024. Well, I've been like pointing out like some absurdities of like what's happening. And you know, I started like just doubting from the beginning. Like, okay, we were supposed to get EGI by 2023, where did that go? You know, I started questioning like, wait, what is actually happening? I started realizing like, you know, the founder of, you know, Sam Altman is like. I was like, okay, what's the agenda actually going? It's like, okay, they might be like just raising money and cutting out competition. Like, do I really want that?

Peter (103:25)

No.

Jordy (103:25)

So I started doing business. He might be being a businessman.

Matt Van Horn (103:31)

Yeah, yeah. I mean he's.

Nick (103:33)

Look, look, Sam Altman is just being, you know, what he's best at, like being a really good startup founder, you know, driving all the attention and engagement towards him. You know, either it's like through claims saying, you know, AGI is going to be the most powerful technology and bad things will happen. And then he like leans towards, you know, doing those bad things. So for example, like, I don't know, he said social media, we don't want to repeat the mistakes of social media. And then, you know, you start making Sora Swap.

Jordy (104:02)

It's really good.

Nick (104:02)

I don't dislike the technology. It is like kind of like doubtful of the approach of what's being said in 2022, 2023 and then like 2024, we're entering like, hey, it's getting a little different. But about the bubble. Yeah, I just started doubting like, you know, before they didn't have competition, so it was like, okay, they're going to make it. But then Elon started entering the chat. He entered the chat.

Peter (104:28)

So.

Nick (104:30)

You know, he just.

Jordy (104:32)

Yeah, but, but with the, with the bubble, I think like you can, I think it's fair to maybe question their ability to meet some of these like kind of press release, economy style commitments. Yeah, but have you ever had, do you, do you honestly think there's a world where the market has some intense correction and people stop using everyday people that are not on X, that don't know the infamous Nick, just stop using ChatGPT.

Matt Van Horn (105:02)

So the way I see it with.

Nick (105:03)

I think the bubble is not precisely AI is not a bubble, but the companies are. Bubbles raise too much, you overextend, you're actually entering into a bubble territory. The way I look at like ChatGPT for example, if you look at all their projections of their numbers, it's all based on user account, like how many users do we have? So in 2025, in the beginning they said they're going to reach a billion users by the end of the year. They actually stalled at like 800 million. I don't even know if they reached 900 million. It's like six months, they're not growing. So if you just make a simple calculation is like, okay, if their user base, which is, you know, getting mogged by Google, for sure, if that starts declining and reaches like, you know, like a certain threshold, all the projections fail.

Peter (105:56)

So, yeah, they definitely need growth to continue. I do think that that 800 number is a little stale at this point. They haven't released the new numbers. It feels like they might still be growing and just not doing the whole like, they're saving it for like. Okay, when we put out the billion number, it's a big moment, but I mean, yeah, certainly it raises questions as you outline. What was your read on ads in ChatGPT?

Nick (106:24)

Well, you know, I'm not a big fan of ads, but I am a big fan of ads from the business perspective.

Peter (106:29)

Sure.

Jordy (106:29)

They're like.

Peter (106:33)

We love ads here.

Jordy (106:35)

Part of my. I think it was important in processing OpenAI to realize at some point you had to realize they are operating as a new big tech company. And so the decisions that they make, they're going to ship a lot of products, like, they're going to do a browser, they're going to try sora, they're going to try a lot of stuff. Not all of it's going to work, but what matters is the core user base still growing. You could argue, I would be. I haven't seen anything that would point to them shrinking, but even a deceleration is not great, obviously at that scale. And so, yeah, you could kind of predict a lot of the things that they would do just by looking at them not as like a ASI company, but as a big technology company.

Peter (107:19)

Yeah, consumer tech.

Nick (107:21)

Yeah, sure. I just think, like, look at fouro, for example. When fouro started launching, you know, Sam Altman said, you know, the movie, her and Scarlett Johnson, her voice, you know, it was marketed as like your lover, you know, you're gonna fall in love with chat GPT, the voice, the way.

Peter (107:39)

It talks to you.

Nick (107:40)

And what happened, people actually, people fell in love.

Peter (107:46)

Yeah, well, so.

Matt Van Horn (107:47)

So the thing is like, the reason.

Nick (107:48)

Why they had like a such a fast growth in users was because of like the image generation model. And this part. And now where we are, they're really lacking in the generation model and they lose.

Jordy (108:00)

Okay, but Disney, we got all the Disney IP coming. I'm super bullish on this for open AI, it just gives them a real. It gives them a feature that no one else can have for at least one year. And I think that when you look at Disney scale, look at how many People go to the parks every single year. How many people will pay to get something novel in an image model? I think that could be part of the next leg up.

Jim Siders (108:29)

Sure.

Nick (108:30)

I mean, if IP rights hold, which I don't think they will, the world we're entering into, sure, that's going to be fun.

Jordy (108:37)

It's so over for IP rights.

Nick (108:40)

Yeah, I think it should be abolished in some sense, but in a very smart way where people should get some recognition or something, a pat on the back.

Peter (108:54)

Where do you want all this to go? I mean, the anonymous account is interesting because you have a lot of ability to talk openly about a bunch of different companies. Do you want to turn this into an analysis group, a commentary channel? Where do you see your work evolving?

Nick (109:13)

Look, I was never anonymous. I was like, always had my face on it. Like, people actually know who I am. Like, yeah, if you actually, like, you know, I don't. You know, if you go back on my tweets, like, I was just being me. Like, my, you know, I've lived in six countries and whatnot. I've talked about my life. It's all there. I've never deleted anything. If anyone wants to go dig, go for it.

Peter (109:37)

But the thing is, like, yeah, I.

Nick (109:41)

Just, like, went anonymous to, like, just, you know, be attached from my identity and like, be, you know, explorative to talk about things that I don't really understand, but I want to understand and go for it. So the best way to understand on the Internet is to say something absurd and then be corrected.

Jordy (109:56)

So, yeah, you're just baiting smart. Yeah, smart people too. And I'm sure some are baiting us.

Nick (110:05)

You know, researchers are acting like marketers, like AGI two weeks, you know, so they're baiting us. I'm baiting them. And you know, it's causing.

Peter (110:13)

Yeah, I do have a funny update. I like to check this every once in a while. Because of the Disney OpenAI deal, I went to ChatGPT, I invoked image generation, and I said, make an image of spider man fighting Mickey Mouse. And it says, we're so sorry, but the image we created may. So it actually generated the image. It saw me, it showed me a little preview, and then it took it away, which is so crazy because guardrails.

Jordy (110:35)

You know, actually do the work and incur the cost.

Peter (110:38)

Totally read the prompt before. That's what I would think. But it actually generated the full image.

Jordy (110:43)

And then even OpenAI doesn't read instructions.

Peter (110:46)

Similarity to third party content. If you think we got it wrong, please retry or edit your prompt. And then I used to be able to do that in Nano Banana. I went over to Gemini and I said make an image of Spider man fighting Mickey Mouse. And it said, and it said, I can't generate the image you requested right now due to concerns from third party content providers. Please edit your prompt prompt. And it didn't generate the image with Gemini, but so it's been interesting to see that press release happen. Yeah, will probably do it. Brock imagines but they have an exclusive deal.

Yohei (111:15)

I don't know how.

Jordy (111:17)

Anyway, what are you hoping to spend most of your time on this year?

Nick (111:21)

Yeah, I just want to go all in into all of this. I don't mind doxing myself.

Matt (111:31)

You know.

Nick (111:32)

I mean, you know Nvidia invited me to GTC as a full time creator and I'm gonna go Jansen and whatnot.

Peter (111:39)

That's fun.

Nick (111:39)

So yeah, I think like it's time for me to go, you know, like 10x more and stop, you know, not just posting but like a lot of things I can talk about.

Peter (111:50)

Yeah, that'd be cool. Yeah, I mean the, the Anon account, it does, it does give you the ability to speak freely, but it does sort of flash. I mean we were going back and forth on this where I was kind of like collapsing you down into just an open AI hater. And from talking you, clearly there's a lot more going on there. But it's easy when I only see the viral post and I don't know anything about the person behind it. Like it can become a little one dimensional. So. I'm excited for your new work. Your hero.

Jordy (112:13)

A little. Sam hasn't blocked you, correct?

Nick (112:18)

No, but Chad GPT has blocked me for no reason.

Jordy (112:23)

Like I've never commented.

Peter (112:24)

No reason.

Jordy (112:24)

I mean no reason.

Peter (112:25)

You got to get. Give it some credit here.

Jordy (112:28)

I think, I think it would probably be whoever's managing the account logs in and they're sick of seeing anything that happens. Just seeing it's over. I can see whether I gotta get. You gotta give some credit.

Peter (112:41)

What is your camera roll like? Is it all pictures of Sam Altman looking perplexed or something?

Nick (112:47)

Anyway, you know, I like entertainment. I'm very tasteful with, with a lot of these things. So yeah, I just, I know what's absurd, what hits, what's gonna be funny. I just post what I feel like and I go, you know, very analytically, analytically. But also I'm a feeler, so I know what's very cool.

Peter (113:10)

Well, good to meet you man. This is fun.

Jordy (113:12)

Do you mind if we flashbang you?

Peter (113:14)

Throwing flashbang.

Yohei (113:16)

Throwing flashbang.

Peter (113:17)

There we go, the chat.

Jordy (113:18)

It's been great having you on.

Peter (113:19)

Goodbye.

Jordy (113:21)

See you soon, Nick.

Peter (113:22)

See you soon, Nick.

Jordy (113:23)

Cheers.

Peter (113:23)

Goodbye. Let me tell you about Figma. Figma make isn't your average vibe coding tool. It lives in Figma, so outputs look good, feel real and stay connected to how teams build, create code back prototypes and apps fast. We have our next guest joining in about five minutes. We gotta go through the Brian Armstrong story. So Brian Armstrong is in the Sunday edition, the exchange edition of the Wall Street Journal, going back, back and forth with Wall Street. They called him enemy number one on Wall Street. He's the crypto CEO, of course, is the CEO of Coinbase. He's been clashing with Brian Armstrong or with Jamie Dimon. So Brian Armstrong, CEO of the biggest UK, US crypto company, was having coffee with.

Jordy (114:03)

It's almost like they paid him to do that. Like he paid them to write that title. Yeah, like that's getting framed.

Peter (114:10)

It is. I mean, that's what Toby look, he said, this will look great in the frame. Of course, obviously there's nuance here and there's a battle, but there's some off the record comments that became on the record. But it's an interesting back and forth. So they were at Davos at the World Economic Forum, and Brian Armstrong is having coffee with Tony Blair, the former Prime Minister of the United Kingdom. And JPMorgan Chase's Jamie Dimon chimes in. He says you're full of s. Blank, blank, blank, said Dimon, a longtime crypto skeptic who's previously called bitcoin a fraud. His index point, his index finger pointed squarely in Armstrong's face. Can I get a crowd? I can't. I'm pointing over there. He's like, you're full of S H I T. Dimon, in a nutshell, told him to stop lying on tv, according to people familiar with the conversation. In appearances on business television programs earlier that week, Armstrong had accused banks of trying to sabotage legislation that would set new a new regulatory framework for digital assets. The confrontation wasn't quite in line with the annual forum's mission to foster cooperation among global leaders. As crypto moves swiftly into the mainstream of American finance, some of Wall Street's heavyweights are waking up to the threats. While banks have embraced some aspects of crypto, helping people to invest in Bitcoin and using digital assets to make money transfers more efficient, they are drawing a line at encroachment on their core business Consumer deposits. Armstrong's coming for them. Banks and Coinbase are at odds over. We have it in the timeline, but it's so much nicer in paper whether crypto exchanges should be allowed to offer consumers regular payouts for holding digital. So you got some usdc, are they going to pay you interest or not? That's the debate. These so called rewards would Pay holders of stablecoins a recurring fee, say 3.5%. Not bad. Stablecoins are digital assets pegged to real world currency like dollars. Banks say the payouts are effectively the same as interest on the bank accounts. And since banks offer much less yields, typically under 0.1% in a checking account, they worry that that the upshot will be that consumers will shift their money in droves into crypto. That they say will compromise community banks and lending to businesses. Armstrong and others in the crypto world say the free market should reign and that banks can simply pay higher interest rates to compete with stablecoins or get themselves into stablecoin businesses themselves. Just join the party. The legislation, known as the Clarity act, might shape the future of everyday financial services, including bank deposits and electronic payments. In the latest push to find compromise, the White House plans to host a meeting Monday between today between banks and crypto industry groups, according to people familiar with the matter. So David Sachs will be there. Coinbase's head of policy, Car Calvert, will be there. Armstrong, 43, co founded Coinbase in 2012 and has helped lead crypto's quest for legitimacy in mainstream acceptance. At the head of the roughly $55 billion firm, Armstrong has a powerful voice in industry debates such as the one playing out in Washington. Quote we'd rather have no bell than a bad bill, he said on X in a post a day before the Senate committee was posed to vote on a version that could effectively ban companies like Coinbase from offering yield to consumers, potentially costing it billions of dollars. Within hours, the vote was abruptly postponed, taking much of the financial world by surprise. There's a lot more here that we can dig into, but we do have our next guest in the Restream waiting room. Jordi, is there anything else that's kicking.

Jordy (117:49)

Around in your brain we need to have somebody on that is not from JPM or Coinbase to talk about this dynamic independent voice.

Peter (117:57)

That would be a lot of fun. Let me tell you about Lambda Lambda is the super intelligence cloud building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands. And let's bring in our next guest from the Restream waiting room, we have David from Lexicon Branding. David, how are you doing?

David (118:13)

I'm doing great. It's a pleasure to be here.

Peter (118:15)

It's a pleasure to have you. Is your 2026 off to a good start?

David (118:19)

It is. We're very busy.

Peter (118:21)

It looks like it. Walk us through who you are, what you do, and what's behind you.

David (118:26)

Okay.

Matt Van Horn (118:26)

Okay.

David (118:28)

Well, I'm the president of Lexicon, Lexicon Branding. And we are in the business, business of developing brand names and naming systems for new companies and new technologies. And we do it really for people, our clients all over the world. We have an interesting process here that combines really a creative layer with an engineering layer. And that engineering layer is a combination of linguistics and cognitive science that puts more objectivity into both developing names and importantly selecting them. And the names behind me on the wall are names that everyone on every name on the wall has been invented by Lexicon.

Jordy (119:13)

When is the right time to meet a founder? Like six months before they start the company. Because I imagine a lot of these companies I can see from, I think some of them, some of them come to you after they've started, after they're already in market and they realize for some reason, reason or another, you know, we need to, we need a new name and then they come to you and then there's like some, some constraints and it becomes harder. You know, walk us through like the ideal process.

David (119:42)

Yeah, most people do not, you're absolutely right. Do not come to us six months before they really started the company. We suggest that they come to us as they approach that Series A. Oh, interesting. By that point, by that point they really know who they are and where they're going for the most part. And so we can work with that information. Now before that, you know, sometimes they, they land on a great name. I mean you, you, this is something that you can get lucky. Not everybody needs a Lexicon to, to come to work with. But, but most of the time as you approach that Series A is a good time to take a look at your name. And we do that consulting all the time where, hey, is it time for us to change our name?

Jordy (120:36)

How do you handle the domain side? Because I talk to I oftentimes angel and like 60 some startups, maybe 70 at this point and oftentimes they'll have what seems like a great name and maybe there's no trademark issues with it, but for some reason or another the.com is taken and they will just never get it right because it's some hundred year old company and you just have to assume they've been operational for long enough. They maybe are public or privately owned.

Peter (121:03)

Or it's like a super rich person that just is like squatting on it.

Jordy (121:07)

So how do you navigate that? Because I feel like one part of the naming process that I appreciate is you have these incredible constraints already. It's not like an open canvas because you have trademark constraints, you have domain constraints, and then you have a bunch of other kind of like less clear constraints, but just like it has to make sense. Right. And so in the naming process, you're trying to like kind of combine all of those into a short list of things and then hopefully it has to like feel good and sound good and do all those other things.

David (121:37)

Well, the URL should be the least important constraint in this process. And we have research that we've done several years ago and actually just this morning, we're about to launch another study that I think is going to once again provide evidence that the URL is really no longer that important.

Peter (121:58)

Really.

David (121:59)

I mean, the search is domain.

Jordy (122:00)

Domain brokers hate this one simple trick, which is so. Which is so funny. I mean, that's very. I mean, I can't wait to read the study because it's so counter to. I've always felt like domains are the element of your branding that you have the least amount of control over, but have an incredible impact. Everyone's seen if you see a beautiful website and then the domain is something, something, try blank IO. It just screams. Because there's so much subconscious branding and marketing that people are getting from. Every time I go to a big company's website, they have a dot com, it sort of trains people. But what's the data that shows that it doesn't really matter?

David (122:45)

I think that's probably, given what the two of you do, it's probably a little inside baseball because we're not finding that on the consumer end or the customer end, they're just looking at this as an address. The analog is a zip code. And actually we found evidence that consumers like the idea of something that says, what is this? You know, and look at the companies that have done. Well, you know, Lucid is a client of ours, Lucid Motors. They don't have loose. It's not Lucid.com so I think it's less important for the consumer and the customer than it is to perhaps raise money.

Jordy (123:28)

Part of it for me is somebody that's helped name a number of companies. And I just enjoy buying and finding and buying domains. I've just always appreciated it is that I appreciate how hard it is to get a great domain. And so when a founder comes and is pitching me and they already have something great, it just shows like a level of agency and ability. Maybe they didn't know a great broker, maybe they didn't even use a broker, but they figured out how to get it. And so it kind of says something about them. But I agree there's a lot of context where it totally doesn't matter. And if you look at the wall back there, almost every single one of those companies got to the point where they could buy whatever domain they wanted, right? So it's like just at a certain.

Peter (124:13)

Point, I mean, the Microsoft Azure one is hilarious because Microsoft owns, owns Azure.com but reroutes it to Azure.Microsoft.com because they want you at the top level domain.

Jordy (124:24)

How is, has, has ChatGPT kind of made some clients kind of given them kind of the wrong lesson in branding? Because there's so much about that where like traditional, you know, if somebody came to you and they were like, I'm going to try to make a consumer product, I want to get to a billion users. If I don't get to a billion users, I'm going to fail. Well, I'm going to call it ChatGPT. What do you think you'd be like? You'd probably laugh them out of the room or say, well, you really need to work with me.

David (124:54)

Well, the interesting thing about that question is more than 50% of our clients come to us having worked with Claude or ChatGPT to develop their names. And they work with it. They find it frustrating. Things just don't work together. There's no doubt those types of models can generate thousands of names, no question about it. What they can't do at this moment in time, that may change down the road is apply the judgment and principles around this name is better than that name for you. Our philosophy is we're not in this business to create good names. We're in the business to create the right name, the right name that adds immediate, immediate value and long term value. Because this, the name should be the one thing that stays with you through your whole journey and nothing will be used more often or longer. So it makes it very important. And at this point in time, a chat or a Claude just isn't up to speed to do that kind of decision making.

Jordy (125:58)

Walk us through the chapters of kind of even naming styles, right, because we're at an interesting moment right now. People are putting, putting GPT on the name, on the End of things, people are saying the blank company of place. Right. Like that's a whole new thing. I'm sure that triggers the Browser company of New York. Yeah, the Browser Company of New York. A bunch of people copied that. And my take on that was always. That was an amazing name one time because it signaled to insiders in Silicon Valley that it was just a novel kind of fresh take on the browser. And it was the Polaroid opposite of Chrome or Safari or some of these kind of like one word. Even though they did have a name for the product. And then you can think back like there was an era where like adding ly to the end of something was. Was popular. Like, what are the different chapters? And like have. I'm curious if like the great brands have always kind of gone against the mold, right? Because like when you. And anyway, so maybe extrapolate on that.

David (126:57)

Yeah, sure. Well, look, one of the most persistent trends or chapters, as you call it in the industry is imitation. So someone comes out with something that ends in ly and hey, that's a good idea, let's do so then you'll see four or five names out in the marketplace that end with ly. What's happening there is that that imitation doesn't generate the interest and the provocativeness that are really well put together brand name for people. So we always advise people, imitation is suicide. Your name should not look or act like anything that's out there in the marketplace.

Jordy (127:37)

Yeah.

Peter (127:38)

How'd you get into this? What was the first brand name you did?

David (127:42)

Well, I came from the advertising business, so, you know, great agency for building. And I just saw as we worked on, on projects and company identity that naming was going to become more and more important as the world got more and more complex and more and more integrated. And so I took a flyer and that's the simple, maybe boring story of how I got into this business. Relative to the first name we ever generated, It's a name that has long since disappeared. I think it was for United Technologies. And we named a new furnace noise reduction technology called Whisper.

Jordy (128:33)

What advice do you give companies that get to a scale where they have a brand name and then different product lines and maybe even kind of like sub companies. We're at this point right now where, like, look at a company like Claude has anthropic Claude, Claude code. OpenAI has, you know, ChatGPT. They have the browser which I'm even just blanking on the name of it. They have Codex.

Peter (129:00)

They have Codex desktop app.

Jordy (129:02)

Yeah.

Peter (129:03)

So different models in Codex. There's GPT, what framework you give?

Jordy (129:08)

Because I'm assuming you'll work with a company to name the parent company and then you also work on the sub products, and then some cases the sub products end up outshining the parent company.

David (129:18)

Yeah, yeah. Well, we have an expression here we call clarity is the language of leadership. And what you see in that example you just gave me, that there's really no story there. They don't have a common language. And so. So it's difficult for the audience, the consumers, to really understand what are they doing. For me, what is this company about? It's a very choppy thing. So a lot of our business coming to us now over the past year and a half, and I think it's because of AI is clients coming to us and asking us to help them. I use the term straighten out their language and make it make sense.

Jordy (130:01)

Creating a through line.

Peter (130:02)

Yeah.

David (130:03)

Yes.

Peter (130:05)

What else about these taxonomies I'm interested in? Like at the early stage, Series A, how much are you trying to actually define a hierarchy of product names? And what is the process of working with you? Like, is it, you know, there's like a specific, specific sprints, specific deadlines, you're paying up front and you don't know what you're getting. Is there contingency? Like, what is the workflow to work with a company?

David (130:31)

It's pretty straightforward in many ways. This isn't really rocket science, but we start with a real simple work session where we're really asking clients four key questions. And this always works. It's how do you define winning? If we get 10 people in the room, we're going to get 10 different answers, but we can sort that out. Then we'll say, okay, if that's how you want to win, what do you have to win that gives us things that the name doesn't have to. Then we say, okay, what do you need to win? What don't you have?

Yohei (131:05)

Right.

David (131:06)

That begins to, you know, give us information about how that name might help them. Right. Because a name should be a tool, should be foundation for success. And then finally we say, okay, what do you have to say? Would you like to say in between those four things, we call it a diamond. We can then set up a, what we call a framework or a creative framework, which we don't even use the word objectives because we want that analog holds through. We want a window that we can travel through and do a lot of creative work. That's where the process begins. And then we work here with, you know, Small two person creative teams. And we have this, what we call internally this engineering layer where we linguists, we're using, you know, company funded research on things like sounds exotic, but it's very fundamental. Sound symbolism, letter structure and, and fluency. All those things then go into a funnel to help us select names.

Peter (132:04)

Makes sense.

Jordy (132:05)

What is the key to avoiding botching a rebrand? We saw like nowadays products grow so quickly. We saw this with claudebot which if you paid attention to that you would have seen like okay, anthropic is going to have issue with this. Right. It's like way too in, in the same category. Sounds the exact same, very confusing to consumers. So he quickly rebranded it, then didn't like the rebrand, then rebranded it again. And fortunately I think the product is exciting enough that the new name openclaw hopefully sticks and continues. You know, the community just like keeps, keeps the momentum. But you're working with companies that are coming to you right when they're starting to get real traction and they know they need to change their name but they don't want to kind of destroy any brand equity that they've built up.

David (132:53)

Yeah, well, I think the biggest factor is that they under, under invest name change and I don't mean necessarily investing with a lexicon or someone else. They, they don't sit and think through all the implications of doing this name first from a. What, what's the message that we want to communicate? Because you do at some point have to say yesterday we were this and now today we're that. And what is that message? And then the second thing is lining up the resources, the right law firm, global language analysis, all these things so you don't end up with something that's launched and then it turns out to mean, you know, your mother's a bank robber. Just not a fun day.

Jordy (133:44)

Yeah, I mean a good example. You did so you did Trip Actions to Navon and that was an example. Trip Action had a big brand, at least in the US and they didn't want to be explicitly known, I imagine as just a place, just a travel booking site. They had to go through that. I remember they really paired the rebrand. They paid you guys, worked with you guys. But then they also seemingly spent like 5, $10 million in a short period of time to like reintroduce themselves. At least that was my impression as just as somebody in the target market.

David (134:19)

Yeah, they did a fantastic job. They really did. And that name, it's a coin solution. We made it up because the founder of the company having Been, you might say, burned by trip action. Being so one dimensional, so narrow. Said, I want flexibility for the future. And the best way to get flexibility is to make something.

Jordy (134:42)

So that means we'll never run out of names.

David (134:45)

That's right.

Jordy (134:45)

We hope, we hope, we hope. I have. It's always been every single time I've been working with a startup or trying to name something, sometimes you can just be hitting a wall where you're like, I can't find. For me, it's like, I can't find any domains in this space that I like. Everything's trademarked, all this stuff. And then there's always, always a name. You might appreciate some of the history of tbpn. We started as a show called Technology Brothers. Then we had an issue where everywhere we went, people would say, introducing the Tech Bros. And we were like, we knew this was our life's work. And we were like, I don't want to spend my whole life being introduced as the tech bros. Which was like the word that we were trying to avoid with the original name. And so I was like, I need a. I want like a four or five letter domain. It has to have TB in it. And we ended up part of tvpn. We're not a network. We look and sound like an ESPN or something like that. But the logic was like, if we name it like a network, even though we're a podcast or a live stream, the comms departments for big tech companies will be more comfortable putting their leadership teams on our show because it looks and feels more like television, even if it is at the end of the day, just 10 of us here in a studio making something that looks like television.

David (136:06)

Yeah, it does feel like you're a national network.

Jordy (136:10)

Yeah, it is a mouthful. It is a mouthful though. We're having to power through that.

Peter (136:14)

Yeah, yeah. Putting a P next to a B is always a choice. The real lexicon heads will probably just inspires us to. I mean, last question for me, because there's some breaking news going on right now, but what do you think about a return to using surnames as company names? The Ford Motor Company, Walt Disney, the Disney Company. That felt like something that was just right there on the shelf for founders to pull off and revive. It's starting to happen. But do you think that that might happen?

Jordy (136:53)

Because you can't.

Peter (136:54)

There's Rigetti Computing. That's a quantum computing company. Public. There are some people that have done it, but do you think that that was something that would, you would caution folks against or you would be in.

David (137:05)

Favor of, I would caution to really take a hard look at that and it really depends on the person's personality, what the name is like, is it easy to spell? Is it easy to say? Yeah, the danger, of course is something happens to the founder that is untoward now you have to live, live with that. So I think, I think it's worth discussion and evaluation, but I would probably push against it most of the time.

Peter (137:34)

Yeah.

Jordy (137:35)

How many clients do you work with a year?

David (137:39)

We'll do about 75 projects a year in the naming area and then we have a research group here that does the quantitative research for us also.

Jordy (137:49)

Very cool. And only some fraction of those are technology companies. I'm assuming a bunch of other industries will tap your expertise as well.

David (138:01)

Yeah, we try to keep a balance but really 50, maybe 60% of our work is in technology at this point point in time. And we which I love it.

Jordy (138:11)

Everything is technology.

Peter (138:12)

I love it. That's a good place to be and it's fun. I mean these are all amazing brands. The Power Book, what a legend.

David (138:18)

We see the future every week.

Peter (138:19)

Yeah, that's the most important.

David (138:20)

And we talk to some of the smartest people that can imagine. As you guys.

Peter (138:25)

Yeah, yeah, I can imagine.

Jordy (138:26)

Well, I've had a couple people text me during the show asking for an intro. So you have quite a lot of demand, I'm sure more than you can fulfill. But we'll definitely send some people your way and it was great to meet.

Peter (138:38)

Yeah, we'll talk to you soon.

David (138:39)

My pleasure.

Peter (138:40)

Thanks so much.

Matt (138:40)

Cheers.

Jordy (138:41)

David.

Peter (138:41)

David, we'll talk to you soon. Tomorrow. Cisco On February 3, the Cisco AI Summit brings together leaders from Nvidia, OpenAI, AWS and more to discuss the future of the AI economy. The whole thing will be live streamed.

Jordy (138:55)

Going to be there and we'll be there with the giga stream headed to.

Peter (138:57)

The bay later today we we have some breaking news. First up, Palantir beat earnings stock is up. What is it? Up 6% already after hours. It's a $350 billion company and this is the big one from Kalshee here just in. SpaceX reportedly confirms Xai merger. Very exciting. This has hit the Bloomberg terminal as well. Elon Musk SpaceX SpaceX confirms merger with Xai and company Memo SpaceX confirms plans to merge with XAI before the IPO we were talking about this and sort of predicting that this would happen. Sats is up 2% after hours. I guess that's correlated there.

Jordy (139:44)

They just go up. Anything happens in space that's good. It just goes up.

Peter (139:49)

Sats. What is sats? I don't know.

Jordy (139:52)

Is.

Peter (139:53)

I don't know.

Jordy (139:53)

I thought that was. That's EchoStar. They hold.

Peter (139:57)

Yeah, yeah, they hold. SpaceX SOC.

Jordy (139:58)

Yeah, that's right.

Peter (140:00)

Elon Musk plans to merge SpaceX with Space with Axe AI in a deal that encompasses the billionaire's increasingly costly ambitions to dominate artificial intelligence space exploration. The deal was announced in a memo. Bloomberg earlier reported on the discussion. SpaceX is planning an IPO that could raise as much as 50 billion in value. The company at 1.5 trillion. It's also discussed a possible merger with Tesla. There's a couple other data points that we should share on Kalshi. Which AI company will have the best coding model at the end of 2026? Xai is at 9%. You don't hear that much about Xai, but that's higher than I thought. Anthropic's leading at 54%, then OpenAI 27%, then Google at 14%. Of course this is in the benchmarks on live bench AI rated by coding average. Do with that what you will. But the real interesting Data point about SpaceX is that according to Reuters and Tren Griffin on X, SpaceX generated 8 million in EBITDA.

Jordy (140:59)

So this is a typo.

Peter (141:01)

This is a typo. It's 8 billion. 8 billion in EBITDA.

Jordy (141:05)

Because I read this and I was like, whoa, that's nothing.

Peter (141:10)

Yeah.

Matt Van Horn (141:10)

Does it?

Jordy (141:11)

Because 8 million is pretty marginal. Yeah, no, it's like that is not. Everyone's been applying.

Peter (141:18)

The correct number is 8 billion in profit on 15 billion to 16 billion in revenue. So 50% margins for space company is absolutely insane. A lot of that's coming from Starlink, obviously. Starlink is the main. Musk's Internet satellite based Internet system. Starlink is the main revenue driver, accounting for about 50 to 80% of the total revenue. The rapid launch of 9,500 Starlink satellites since 2019 has made SpaceX the world's largest satellite operator with only over 9 million users of the broadband Internet service. And of course it's not just, you know, individuals that have a Starlink that they throw when they're camping. It's companies and boats and yachts and planes. Now there's a whole super bowl ad just about, I think United Airlines has a deal and so they want people to choose United because Starlink is such a differentiator. When you're getting on a long haul.

Jordy (142:12)

Plane, there's not much, there's not that much you can differentiate on all the food everywhere. Yeah, it's terrible. All the planes are falling apart. You feel. You don't really feel safe on any air.

Peter (142:23)

I agree.

Jordy (142:24)

One thing you could differentiate on is.

Matt (142:26)

If you can, like if you get.

Jordy (142:27)

Food in first class, if you're allowed to bring it back.

Peter (142:30)

Yeah, that would be a huge differentiator.

Jordy (142:32)

TVPN, JetBlue, they're moving slow on that.

Peter (142:34)

Front and so they have to differentiate on Starlink. But I do think it will be a real differentiator. Like if you're choosing between American and New York, you're going from LA to New York and it's American or United. And United has Starlink. That's a pretty big difference just to be able to properly use the Internet. Anyway, let me tell you about CrowdStrike. Your business is AI. Their business is securing it. CrowdStrike secures AI and stops breaches.

Jordy (143:00)

There's one more interview. Sorry, not interview, but context. Context from Jensen. We're going to pull this video up. Take him. He highlighted it. Let's pull it up. John, I cannot wait to see your reaction.

Peter (143:14)

Okay, I haven't seen this yet. High Yield. Harry also says if you bought the ex bank debt and high yield bonds, but back at 13% yields, you've now parlayed your. Well, your way into being a creditor to the largest IPO ever. What a journey. It is such a crazy journey. Anyway, let's play a Jensen clip. And this year, this year is the.

David (143:34)

Year of the Horse.

Peter (143:36)

So it's going to be a very good year. And this year.

Jordy (143:41)

Let's go.

Peter (143:44)

So it says that post is a. It says. It says that post is a joke. What about it is a joke? Is it AI? Is it fake? Did he actually say that? Did he say that a long time ago?

Jordy (143:53)

No, no, no. The Fire Horses. Take him. Is saying Jensen is citing the Year of the Horse and you're bearish. I'm not a CIA body language expert, but look at the expression on his face. This post is a joke.

Peter (144:04)

Kind of, yeah. Kind of. I don't know. Yeah, no, Jensen is.

Jordy (144:08)

I mean, we said this. We opened the year with this. We said Year of the Fire Horse. Yeah, massive.

Peter (144:14)

Also, I mean, he was seen chugging beers, drinking champagne, giving somebody toast. He's on cloud nine.

Jordy (144:21)

That's one of the better calls.

Peter (144:23)

Yeah, we're going to dig into that more. So Xai just filed this at the Nevada Secretary of State website. Managing members. They put SpaceX person on the. On the Silver Flume filings. Andrew Curran says it's happening, can confirm that the screenshot's real merger is effective. Space Exploration Technologies Corp. Is now the managing member of XAI Holdings LLC. Interesting, says Prins. Yeah. 8 billion in profit.

Jordy (144:58)

X in SpaceX stands for Twitter account.

Peter (145:03)

Twitter now it is a weird, weird timeline, but it makes a lot of sense. I mean managing five companies has got to be exhausting and difficult managing, you know, two, there's some, there's some economies of scale and you know, now that you have the data centers in space thing and Starlink, you know, it is a communications company after all, makes a lot more sense. You know, I don't know and I'm.

Jordy (145:24)

Not going to say we called it, but we certainly were talking, I mean we started talking about this possibility at the end of last year. It felt like XAI didn't have the traction to keep just continuously raising at a $200 billion valuation. When you looked at its financial performance relative to the rest of it. And merging XAI into Tesla wasn't going to get a lot of. I think it would have gotten way too much pushback. Also Tesla's obviously public, so it would have had more scrutiny.

Peter (145:57)

But if you just play out even the non AGI vanilla case of the models sort of commoditize, everything gets, they get to near the frontier and maybe one is better than the other for one thing. But at the end of the day it's like, you know, intelligence on tap. You're paying for tokens and SpaceX's tokens are cheaper because they're using solar and they're in orbit. Like there's a world where just on the pure economic regulation you can get markets.

Jordy (146:23)

Well, it's SpaceX to actually be a player and like, like AI Cloud. I see a world where they're like, yeah, we're really good at building infrastructure. We're going to build the infrastructure to power other people's apps. So I would expect to see some type of enterprise AI offering from SpaceX in the near future, even potentially before there's really data centers in space. Couple more posts.

Peter (146:50)

I love imagining the person who is just like the longest tenured Twitter employee who's just like, yeah, I'm here, I'm working at Twitter and like, okay, now I work in an AI lab, now I work at a rocket company. It's like such a funny twist in the career path, but congratulations to everyone who held on and is now working at SpaceX.

Jordy (147:10)

Couple more reactions to Oracle. Matthew Zeitlin over at Heatmap says he's wearing the that the Nvidia OpenAI deal has zero impact on our financial relationship with OpenAI shirt. And then Alex over at 8VC responded to the Oracle's post and said OMG this is literally bank run language. It really does read like that.

Peter (147:39)

I mean the good thing is that it just feels like the fear about adoption falling off has not really come to pass. And so like people need tokens, they're deploying them. More and more people are using LLMs more hours of the day. Everything's sort of like on track. Like the fundamental technology is growing and so would I be terrified of sitting on a data center that has the ability to inference just one one of the many LLMs. It doesn't seem like the worst scenario to be in. We looked at the gross margins seem good. A lot of the health checks of the ecosystem have been passing in my opinion.

Jordy (148:23)

Yeah, Oracle is already it's down almost 50% since the original it's already sort of corrected.

Peter (148:31)

I wonder what the CDS is doing, but that's for another deep dive.

Jordy (148:36)

The other story today, down more than 50% since the original announcement.

Peter (148:41)

Play that Wanwell.

Jordy (148:46)

Well, up next, we fortunately have somebody whose job is to make sure the demand is real.

Peter (148:52)

Yes.

Jordy (148:53)

And so he is supporting the backlog.

Peter (148:56)

Thibaut from OpenAI. The other breaking news that we should talk about before he hops on is Bob Iger at Disney. He has told associates that he plans to leave the CEO role before his contract expires. He said, I'm taking all the IP, I'm giving it to OpenAI and then I'm out. Iger told Associates that he plans to step down as CEO and pull back from daily management before the December 31 end of his contract. Okay, so he's planning to some summer moves. He's going to be in Europe, maybe. The entertainment giants board of directors is planning to meet with him next week at its headquarters in Burbank, where they were expected to vote on who should take the top job. In private conversations over the last few months, Iger has told people close to him that he's ready to move on from the grind of being CEO and was frustrated by conflicts at Disney's ABC network over the brief suspension of late night host Jimmy Kimmel. People who have spoken to him said the CEO has told multiple associates that he would like to spend more of his time and energy on other things, such as sailing his new and larger super yacht, the Aquarius. Let's go.

Jordy (150:02)

Somebody in the chat earlier was talking about the name of their new sailboat yeah.

Peter (150:07)

Gotta come up with an innovative name.

Jordy (150:08)

If you're naming a boat, hiring Lexicon branding, you really should.

Peter (150:14)

Aquarius feels pretty like, you know, like ChatGPT could come up with that. That name potentially. He said that he would also like to devote more time to work with his wife, Willow Bay, dean of USC Annenberg School for Communication and Journalism, and on Angel FC, the women's soccer team they bought in 2024. The final timing of his departure from the top job hasn't been determined yet and could change. He's expected to remain CEO for several months.

Jordy (150:37)

What you got, Raghav? Always, always one step ahead in the chat. Just shared exclusive. OpenAI is unsatisfied with some Nvidia chips and is looking for alternatives, sources say. So OpenAI comms team fires back. Okay, somebody else in the chat was saying that Jensen is in his clip farming era. He saw a bunch of the kick streamers and he's like, I need to start doing this.

Peter (151:04)

Okay, well, we will dig into that more. But until then, we have the. We have some massive news from Codex. Let's bring in our guests from the Restream waiting room. How are you doing?

Jordy (151:14)

What's happening?

Matt (151:16)

Hello. Nice to be here. Excited to be here.

Peter (151:20)

Thanks for hopping on the show since the first time on your show. Please introduce yourself and your role and then the announcement today.

Matt (151:28)

Yeah, sure. I'm Thibaud. I work here at OpenAI and Codex together with an awesome team. I lead the entire Codex team. And yeah, today we launched the Codex app on macOS. And it's a very capable thing. I think it's a surprising cable.

Peter (151:44)

Yeah, well, I mean, I downloaded it before the show. I had like five minutes and I was able to create a version of the TVPN website in the style of Berkshire Hathaway. It feels like sort of a Studio Ghibli moment for Vibe coding in many ways. Like, you can just spin stuff up. What, what are you excited about? The desktop app, specifically? Because we've seen the whole Multbot news. It interacts with your file system. But then there's also just. There's just a whole host of people that are loosely familiar with programming and maybe they're comfortable with the idea of programming. But at the same time, when you tell them like, hey, set up an environment, get a cli, get your IDE set up. They're just like, I don't want to spend the hour that it's going to take me even to set up something as simple as the tools have become. It's still too much. So what were the goals? What are you excited about for Codex on desktop?

Matt (152:35)

There are two things I'm really excited about is one, it does make it much more accessible. You just install it, you get going and it's like after you log in, it's delightfully simple. You're just greeted by this little composer. You can start chatting with it and then invites you to get things done. But also it very much leans in into the way of working that we've seen from, you know, technical staff at OpenAI and our users. And when you look at Peter, for example, with like an open cloud, it's just like multitasking a lot and it's easy to lose context when you switch like between your different terminals, you don't have notifications and then you don't have like the whole multi modality things. If you're working with, you know, front ends images, if you just want to use voice, all of those things are like packed into the app.

Jordy (153:20)

Oh wow.

Matt (153:20)

Yeah, very much leaning in into making people even more productive.

Peter (153:24)

Yeah, I didn't even notice that the dictate buttons right there. I was uploading screenshots and that was really effective. Peter was talking a lot about how he tacks screenshots in and just gives more visual information to the system. Talk to me about the models that come out of the box, how I should be thinking about model selection. It sort of defaults to 5.2 Codex Media. When would I want to switch and what are some of the different benefits?

Matt (153:49)

Yeah, so we just get you started with the model that we feel is the best for most people. And that's done like through extensive tests, internal evaluations and just feedback that we've got. And like 5 to codecs at medium has this adaptive thinking. So it's going to work fast on easy problems, it's going to work harder on things that actually deserve taking the time. It just feels like, you know, a good default thing to get started with. You know, we also introduced personalities so now you have like two choices. You can go with a more friendly one and then you can, you know, use the pragmatic one which was like the one that we had like so far. A lot of people, surprisingly even on the Codex team had switched to the friendly one. You know, it's just like, it's, it's nice to be a little bit more supported and validated. But of course it also we support the other important model, 5.2, which was a much bigger jump I think than people expected. And it's really also what we crafted the app and the experience around 5.2 is really excellent at long running tasks independently and getting it done to a level of completion that you expect from senior technical staff. And that allows you to do a lot of multitasking. And that's really what the app is built around. You can handle different projects, different threads, all running together without losing context and sort of efficiently managing a lot of things. And this is sort of like a glimpse into what the future is going to be like with agents.

Peter (155:14)

Yeah, I feel like a lot of people are going to come to the Codex app on desktop with a vague idea. You're going more prosumer it feels like with this in many ways. And I'm wondering how you think about planning versus implementation, walking through actual setting codecs up for success. Obviously the great programmers can just prompt better because they understand the hierarchy of files and what tools might be used. But for someone who's effectively non technical, how do you think about the planning stage? Giving the model enough to make good decisions and actually execute properly?

Matt (155:53)

Yeah, I think planning is, is effective in two ways. It is effective for the agent because it allows to be very specific about what you actually want the agent to go and do, perhaps for many hours. But it also gets your own thoughts straight and it invites to have a conversation around what is it actually that you're trying to do when you're sitting behind sometimes you think, hey, this is a very simple instruction. Just go and refactor it and remove this part of the backend and speed this part up. And then maybe actually you don't realize there are five different ways with different trade offs on how the agent could go and tackle this. And planning just helps resolve that ambiguity. I find it a very useful thing just as a human to go through as a process. It's a useful reminder of hey, let's be crisp about what you actually want to do. But one thing that's also very important is with the app and in general, like we don't chop off like the top end of the capabilities. Right. So it's like very important that you know, people who are like extremely sophisticated can get, you know, the most incredible things done. And we're seeing a lot of adoption actually of the app like within research. And you know, it's fascinating walking through the office, like seeing people like do the wildest things with this thing. It's like very much a very, it's a very capable tool.

Peter (157:10)

If I look at the release schedule, I see the ChatGPT app, which has a lot of functions, functionality than the Atlas browser. Now, codecs on desktop. Are IDEs dead in your opinion? Are people just going to bring their own or do you think this grows into an ide or is that like a separate environment based on what you're experiencing and the workflows that you're seeing right now?

Matt (157:36)

It's a companion. It's a companion to the ide. Very much can stand on its own, but is enhanced by using an ide. Like occasionally some people will prefer like just being in their ide. But it is quite evident to me that as agents just become extremely capable, you just want to talk to them and they will get things done. And then what you want to do is you want to be able to steer them and supervise the result. And that requires something that is a very rich interaction surface. And that's what we're building with the app.

Peter (158:08)

How do you think about deployment? I can imagine people vibe coding little apps that do things for them on their computers. Or I had to build an HTML page that I just opened in Chrome and was able to just see. The next step is actually deploying it. Obviously there's a whole bunch of tools that you could plug in with. How are you thinking about open ecosystem versus potentially offering some hosting functionality?

Matt (158:36)

Yeah, we love investing in the open ecosystem. That's something that I've been very proud of us doing. Like all our code is like, the majority of our code is open source. We also support a lot of like other open source initiatives and one of the things that we're leaning in is like the open standard around skills and we ship with a bunch of skills that are useful. Like for example, you know, one is like for Vercel deployments and that allows you to just take what you've built and very quickly deploy it and have it accessible there and share it with others. It is not a native integration. We might consider that in the future. But it's already very easy to do things that are way beyond just writing code.

Peter (159:16)

Yeah, yeah. I mean for a lot of people even deploying to Vercel is going to be, oh, there's going to be. It's going to ask me for an API key or something. But if you're in a chat interface, it can go back and forth and you can just ask, okay, what does this mean? Where do I go? What's the web page? Where do I click? What do I copy? And all of a sudden that's right.

Matt (159:31)

You don't need to read the docs anymore. You can just ask Codex to Do it for you. And it will do it to a high degree of a high degree of quality, especially given all the information that's packed in the skill. It's sort of like puts the guardrails on.

Peter (159:48)

Yeah. How are you thinking about funneling ChatGPT users to Codex on desktop? It feels like you're gonna buy some ads. No, but I mean, there are worlds where you fire off a GPT 5.2 prompt and it just writes code in the thinking step and you don't actually see the python that it wrote and executed. And the next step is, okay, maybe I want to run this regularly, maybe I want to wrap this in some CLI or some desktop app. And I could imagine a flow where you're in the ChatGPT app and you get to the end of the capabilities there and says, hey, you should go over here. We have a more robust experience for you. Have you thought about that? I mean, obviously on day one, there's just going to be immense amount of just excitement and trial and virality. But long term, what are you thinking of integration between the different ecosystems?

Jordy (160:43)

Yeah.

Matt (160:43)

I hope that with this launch we do inspire a set of people who haven't tried codecs yet to try it and realize, hey, I can really do really cool creative things with this. And maybe coding agents is actually for me, what you explained around the first experience in ChatGPT. And then you have the code interpreter workflow there where you don't actually need to understand the code or you use Canvas and you have this little bit of, of the light there and you know, you realize, hey, chatgpt, you can create things for me and like, you know, which image generation, like, how do we combine that and sort of like bring the magic of like coding agents to everyone. That is definitely something that we're thinking about. There are considerations there around, you know, how you do this safely and securely given, you know, the folks wouldn't actually understand the code that's running under the hood. So Codex for now is very much meant for, you know, technical, technical adjacent audience. But then we're also thinking like, hey, this thing is incredibly powerful and coding can help achieve all sorts of economically viable tasks. You don't need to understand that it's code under the hood. But how do we bring this to more people?

Jordy (161:49)

Are you hanging with Peter while he's in sf?

Matt (161:53)

We're close on Twitter and Peter is just everywhere. He's Hosting this conversation, CloudCon, like for the first time, which is like, I heard it's like over 500 participants already. And stuff. So I might just like drop by and say hi.

Peter (162:08)

That's amazing. It's so fast. Everything he does is fast. I mean, one of my big takeaways from the claudebot, Moltbot, openclaw was just that mobile, the importance of mobile and people going on Telegram or signal or WhatsApp and texting. An interface that could write code for them, execute things and interface with some hardware. How are you thinking about the mobile experience into codecs? Like how the integration loops happens? Now that there's a desktop app, there's obviously the ChatGPT mobile app that has massive installation. What are the pros, cons? What are you excited about there?

Matt (162:48)

Yeah, we're very excited about the mobile experience. It'll always be my dream of I can start a task, I can talk to Codex from anywhere and then I can just like walk away, follow on my phone and steer it. And this is definitely something that's going to come. We are very much optimizing for professional software engineers and people are really excited about getting super productive. So it felt like bringing first an amazing experience on macOS. Felt like the right thing, priorities wise. It is something that will come and I think will delight people when, when it's there. And this is also why, you know, we, we keep calling everything Codex. Codex is our agent. It is how you, you get things done and then, you know, whether you interact through it, like via the CLI or in web or in the os, like under the hood is the same agent and like we're going to connect it all at some point.

Peter (163:36)

Yeah. Is it time to buy an extra monitor? How many monitors do you run right now?

Matt (163:42)

On my laptop I only have one. Oh, with the app is like delightful. Like you can actually like fall things, like much more so I don't feel like the need to have like five monitors at home. I do have three and like, there you go.

Jordy (163:54)

There we go.

Peter (163:55)

Not gonna lie, maybe that Dell the 65 inch 6K monitors in the future. We're very excited about that. Anyway, congratulations on the launch. Thank you so much for stopping by. Great to meet you and we'll talk to you soon.

Jordy (164:09)

Cheers, thanks for having me. Thanks.

Peter (164:10)

Goodbye. Let me tell you about Vibe Co where D2C brands, B2B startups and AI companies advertise on streaming TV, pick channels, target audiences and measure sales. Just like on Meta.

Jordy (164:23)

Where to go next?

Peter (164:24)

Dylan Abruscado said the succession is a documentary because apparently the next CEO of Disney could be the theme park division chairman. I didn't get the reference have you seen.

Jordy (164:39)

Do you get the reference from succession? It's Tom who's running the park.

Peter (164:44)

Tom runs the Parks Department.

Jordy (164:45)

Let me know.

Peter (164:46)

Let me know if I. I gotta finally get through all of succession, but. Oh, in other news, there's. I see this post that you're on. Bro spent $50 billion buying Bitcoin over.

Jordy (164:57)

The past five years and is now talking about MicroStrategy.

Peter (165:00)

He didn't even mention who it is.

Jordy (165:02)

And people.

Peter (165:02)

And people just know that it's my.

Jordy (165:04)

I mean, not a lot of people that are non, maybe state entities have bought this much.

Peter (165:09)

712,000 bitcoins.

Matt (165:12)

Wow.

Jordy (165:12)

Yeah, I think his average entry is in the 80s.

Peter (165:18)

And whereas Bitcoin is up today, it's 78,000. But obviously that's nowhere near where it was when it was up at 120. So a significant sell off.

Jordy (165:29)

Really? I have some extra context on the OpenAI news. So. Yeah, apparently OpenAI had been talking with Cerebras and Grok to do like custom chips. And then when Grok got bought by Nvidia, that stopped the talks. I think it's just because they're too slow. You've seen Rune talk about this where Codex is somewhat slower than cloud code or whatever. Even though if the model's actually better, the experience can be worse sometimes.

Peter (165:55)

Yeah, no, no. Like the models can get better, but they're already really, really good. Like most of the deep research reports that I get back are amazing. Mission accomplished. Like, you did it. You did the research, you compiled it and I got my information, but I had to wait 20 minutes. Cut that down to two minutes and I'll probably use it 10 times as much and I'll probably pay 10 times as much for that. And same thing with coding. There's so many memes about watching Brainrot videos while. While you're waiting for Codex or COD code or whatever you're using to come back and answer, you speed it up and you're going to have a much faster adoption path.

Jordy (166:33)

Chat has confirmed that it is Tom Wambsgans.

Peter (166:36)

There we go.

Jordy (166:37)

In the amusement park and cruise division. Then he got moved up to ATN News.

Peter (166:42)

Okay.

Jordy (166:44)

One more thing on the DAT side of things. Bitmine. This is Tom Lee's former guest. Has unrealized losses of 6.6 billion. Now on track to become the fifth largest documented principal trading loss in history. If sold. Of course, he's not selling. He remains bullish. He's really testing his diamond hands. It's at 66% of the size of Archegos. It was Archegos, Bill Wong's infamous hedge fund.

Peter (167:16)

Well, we have Christopher o' Donnell from Day AI in the restream waiting room. We'll bring him into the TVPN ultradome. Christopher, how are you doing?

Jordy (167:25)

Wow. Unique.

Christopher O'Donnell (167:26)

Very well. Yeah. How are you guys? Thanks.

Jordy (167:28)

Beautiful, Beautiful setup that. No one's ever done this before.

Peter (167:32)

I feel like I'm watching it, honestly. I feel like I'm watching an OpenAI launch video. But it is fantastic. It looks fantastic.

Christopher O'Donnell (167:40)

We can get a little ocd, I guess.

Peter (167:42)

We were pumped for this. And so we've been in.

Christopher O'Donnell (167:44)

You're, like, teaching ourselves how to use lights and stuff like that.

Peter (167:48)

You know what I mean?

Christopher O'Donnell (167:48)

Like, ordering this on Amazon.

Jordy (167:52)

I'm pumped for this. I had my favorite experience ever when I. When I check X in the morning, I see a cool startup launching, and I'm like. I'm like thinking, like, I gotta get these guys on.

Yohei (168:03)

And then you're.

Jordy (168:04)

You were already on.

Peter (168:05)

So it's great to see you. First time on the show. Please introduce yourself, the company. Tell us what you're building.

Jordy (168:11)

Yeah.

Christopher O'Donnell (168:11)

So, Christopher o', Donnell, longtime listener, first time caller. I led product at HubSpot for about 10 years and am now lucky enough to be with a group of a dozen or so of my closest friends doing a similar but also very different thing from scratch. In the age of AI, it's sort of, what if we were to start from the very beginning and do everything in this, as you guys are constantly discussing wildly evolving and radically different world?

Peter (168:43)

So that's the short story. So early product market, how are you segmenting things? What's the key value prop? I mean, I'm sure there's AI, but take me a cut deeper.

Christopher O'Donnell (168:57)

Yeah, I mean, I think looking at CRM, there's a funny thing which is I feel like in everybody's head, everybody's expectations are wildly different for everything. And you guys talk about this and, you know, every vertical, you know, our lives are changing so rapidly, and yet nobody's really expecting that to happen. With CRM, everybody's kind of like, but Salesforce isn't going anywhere. And it's like, hold on. There's a layer of it that is. Well, we could eliminate the manual data entry, and we could maybe better integrate with some other systems. And what we're seeing is there is a wildly different level of what's possible. That really was the original promise of CRM. Like, hey, I'm a CEO. I have concerns about my business. I have existential questions. I Want to know if this whole new cohort of sales reps that we hired two months ago is going to make it. You know, I want to know what I can do, what deals I should be involved in. I want to know if I'm spending my time the right way. And you can actually get instant answers to that. If you had the entire record of everything that had been said or done at the company instantly indexed and readable by, you know, an AI agent. Yeah, that's the heart of what we're doing, is ingesting everything and storing it in a way that is LLM optimized so, you know, easy for it to explore this whole network of relationships. But also in natural language and with explanations about why everything is what it appears to be. We're seeing. It's just a completely different set of use cases. So that's really interesting to navigate with customers. But everybody's starting to kind of go from wait, what to of course, this is exactly the thing that we should have.

Jordy (170:48)

What is like the killer feature or experience that you're optimizing for? You're trying to. Somebody signs up, they sort of integrate whatever tools they need and then like, what's the first moment like that you're trying to get people to kind of see the future.

Christopher O'Donnell (171:05)

Yeah. I think the most obvious thing is a better version of the meeting recorders that we've come to kind of rely on. You know, here's this thing. It's watching, it's taking notes and I don't have to do that on the call because it's all being captured in and recorded. And then from that I can ask questions about it. I can automatically have my deals moving based on what happened on the call. That's really great. If I want to write follow up emails, they're incredibly good and the UI is just snapping and moving around as I go through this whole workflow. That's the entry level magic trick that people see when they come in. Then there is this much deeper realization that, oh, wait a second, I can actually ask and understand anything that is happening in the business. And that, you know, people get to that pretty quickly, you know, within a week. Usually.

Peter (171:56)

Talk about bottom up adoption versus top down. Do you want to go for the big enterprise, have them rip out some massive system or I remember using a product called Streak that just plugged into Gmail. You know, you could. I've also built a CRM with Visual Basic in Excel. You can start a CRM project just in a spreadsheet. How are you thinking about where you want to plug in and start getting traction.

Christopher O'Donnell (172:24)

Yeah. Part of our crew built a bunch of streak in addition to HubSpot.

Peter (172:29)

Wow. Yeah, it's a cool company. I enjoyed it at the time. It was very lightweight. You could just bolt it on basically.

Christopher O'Donnell (172:37)

Totally, totally.

Peter (172:38)

Yeah.

Christopher O'Donnell (172:39)

The adoption story, it's a great question. You know, we saw over. I've been working in CRM for I don't Even know what, 10, 15 years.

Peter (172:46)

Let's go.

Christopher O'Donnell (172:47)

And. The march of history always seemed to be from the CEO toward the buyer with, you know, a big swing through rep productivity and these bottoms up tools. And when I kind of move from the rewrite of the HubSpot marketing product to do a startup within a startup in the sales tools that became HubSpot sales and CRM, that was the thesis. Like we're going to go bottoms up. We're going to do what these other sales acceleration tools are doing. And that worked really well. And then, you know, marching toward the buyer, being able to drive as much of the sales process as possible. I think that still exists. And I think that solving for frontline people is absolutely huge.

Peter (173:30)

Huge.

Christopher O'Donnell (173:30)

And giving them things they can come in, get started on day zero. Absolutely no big top down rollout. We're also seeing a little bit of a return to the CEO. And I spend a lot of my time actually. I mean I spend most of my time just demoing at this point other CEOs and showing them how I use it.

Peter (173:49)

Sure.

Christopher O'Donnell (173:49)

That's the close. Because you can't really say it in. I mean we try in our marketing messaging and everything. Like, look at all these examples, samples and everything. But if you see me use it, you go, oh, wait, okay, I got it. This is like this chief of staff that's watching everything and coaching me and giving me the data and you know, or having it up at a board meeting. Pat Greedy, who I know you guys had on recently, that was a great, great segment there. He had a tweet today about our last board meeting.

Peter (174:16)

Yeah.

Christopher O'Donnell (174:16)

And we just had the assistant up on the big screen. There was no deck. It's like I sent a big memo and I send a memo and a context file so the board can ask questions and get instant answers. But we just had the assistant up on the screen the whole time and there's no like, oh, that's a great question. Let me, you know, get my ops people to circle back. It's like, which changes and it's going to be interesting.

Peter (174:40)

That's amazing.

Yohei (174:41)

What?

Jordy (174:42)

There's so many large Software companies that are working to become AI native. The first step is obviously to say that you're AI native, put it on the website. But what are some of what are done? What are kind of the challenges, having worked in big CRM of kind of evolving?

Christopher O'Donnell (175:02)

Yeah, I mean, if you think about it from the perspective of the AI agent and you really. You switch from putting yourself in the user's perspective and you put yourself in the perspective of Claude or one of the these clients that is asking questions and calling tools and trying to solve problems for you, you actually want something pretty different. We had a customer connect their Claude to both HubSpot and Day and said, you know, what do you think? And Claude says, are you asking me which I prefer? And she's like, yeah, yeah. I'm asking which of these do you like more? And Claude says, well, look, how do I explain this? This old CRM is like a spreadsheet and this Day AI thing is everything that I need to answer any question you have. I can find the answer. Let's go. What do you want to know? That's a very different approach that gets all the way down to disk, literally all the way down to how you are capturing and storing the data. Having worked on these systems and many versions of these over the years, it's a problem that starts at the very lowest parts of the stack to be able to give the agents this kind of information. I mean, we all know that Opus 4. 5, with the right context window can do magical things. I mean, it's godlike. So the question is, how do you give it exactly the right piece?

Peter (176:31)

That's amazing.

Christopher O'Donnell (176:32)

I don't think it's God. I don't think it's God. It's like God Light. Yeah, it's like a precursor to like. We're definitely living in a simulation. It's like about to become the thing we think of as our creator. I think that's like, fair to say that's pretty wild.

Peter (176:50)

You raise some money. I want to talk about the deal. I want to kick off the Lambda Lightning round. I want to bring down the mallet from the heavens so I can ring the gong for you. How much did you raise?

Christopher O'Donnell (177:02)

We raised 20 million.

Jordy (177:04)

Whoa. From Sequoia and who else?

Christopher O'Donnell (177:11)

Sequoia Permanent Capital, Conviction Sound Ventures and Green Oaks.

Peter (177:17)

Fantastic bunch of nobodies.

Jordy (177:19)

Quite the group.

Peter (177:20)

That's a fantastic group.

Jordy (177:22)

Well done.

Peter (177:22)

Congratulations. And I'm sure you'll be back on the show soon. We'll talk to you later.

Christopher O'Donnell (177:27)

I hope so.

Peter (177:27)

We have a lot coming can't wait for it.

Jordy (177:29)

Great to meet you, Christopher. Congratulations to the whole team.

Peter (177:32)

Goodbye. Let me tell you about Sentry. Sentry. Sentry shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working.

Jordy (177:46)

And up next we have Jim.

Peter (177:48)

Jim from Shield welcome to the TVP in UltraDome.

Jordy (177:52)

What's happening? Good to see you.

Jim Siders (177:54)

Thank you for having me.

Jordy (177:55)

It's great to have you here. Quick intro on yourself and what you're working on.

Jim Siders (178:00)

Yeah. My name is Jim Siders. I'm the CEO of SHIELD Technology Partners, which is an IT services platform. In cooperation with Thrive holdings, we're bringing IT powered by AI to all these small and medium businesses all across America.

Jordy (178:14)

Finally. We've been begging for this now. We've been excited. I've been begging for it too.

Jim Siders (178:19)

Like, what took so long? That's another.

Jordy (178:20)

I know, exactly. So you raised 100 million from Thrive today. This has been in the works though, for what, like 18 months?

Jim Siders (178:31)

A little less than that. I mean, it's a long term relationship, right? Like the whole company was founded in a partnership that included Thrive holdings at the beginning, which is when I got involved kind of in the conversation long before I joined as CEO. And all the way through it, we've been thinking about how does, how does holdings support, with a strategic investment, this kind of thing? We're building this thing for the long term, so we want to make sure that we have the war chest necessary to do the right stuff now.

Jordy (178:57)

Amazing. What do the early customer relationships actually look like? I know you're buying some existing companies so you're inheriting customer base everywhere you go. I'm sure you're shedding some and then, and then scaling from there. But what is the, in the most simple way, what does a business actually look like? What does the work look like today?

Jim Siders (179:20)

Yeah, hopefully we're shedding as few as possible. I mean, the idea is we're buying quite good it MSPs all over America who have successful customer bases, who have long term relationships with folks. And we're leaving them in place, right? We're leaving those customers, we're leaving the people who found of the business, the engineers and technicians who built those relationships. Then we layer on AI and other proprietary stuff on top of it. And the whole goal is that those customers are actually clamoring for more, not churning because they just got bought by their IT guy, just got bought by a private equity firm or something. Right. And so far, to the direct Answer to your question, that's been pretty much what we're hearing. You know, we hear customers going, oh, I heard, I heard this thing happen. Can you bring some of these thrive guys over here to talk can you bring some SHIELD guys over here to talk to us about what we can do with it? We're curious about it and we've never had access. That's exactly the point. That's the whole thing we're going after.

Jordy (180:18)

Makes a lot of sense.

Peter (180:20)

How are you thinking about the broader market and what companies in sort of like the small and medium sized business categories are most ripe for getting leverage out of AI? We've been tracking, you know, like the ramp AI adoption metrics. Obviously every tech company has a bunch of subscriptions and then you go down the list and it's who you think. And it's like the logging company hasn't adopted AI yet. But what do you think is underrated right now?

Jim Siders (180:48)

Well, I mean, the logging company is a great place to start. I mean, this is on our side. There's a lot of companies in the real economy that would benefit from the same kind of, of things that's creating, you know, AI, creating durable value for large enterprises. Right. The same thing that should help Airbus or Rio Tinto or whatever should also help like a regional roofing company or a dentist office chain. So really the customer selectivity, that's not really where, where my emphasis is at this point. What I'm looking for is the IT service businesses that have built these really durable relationships or we get access to problem spaces that we haven't yet seen in the real economy because there's going to be something that falls out of it, the lived experience of those IT technicians that we can use in the product motion. At least that's the.

Peter (181:36)

Yeah, that makes sense. Can you help me understand your thoughts on the like the customization of software are we going into? It feels like we're going into a world where more and more software where it will be custom. There's already big companies that Palantir, Salesforce, these companies go in and they have a backbone, but then they're doing a lot of custom implementation. There's companies that you can hire that will build something completely bespoke for you, whether that's like a consulting group. But then typically if you're buying a piece of SaaS off the shelf, they're not really doing anything for you. But it feels like now they can. With a forward deployed salesperson for deployed engineer. How do you think about that trend and like how like we're here, we have like a few Vibe coded custom systems that, you know, a small 10 person team a few years ago would definitely not have. I can't tell if we're just early adopters of the technology or if this is the future.

Jim Siders (182:36)

Well, there's a few ways I can answer that. I mean, first I got to confess to a bias, right? I mean, I was at Palantir for a long time, time for a reason. Because I think the combination of first principles thinking and the forward deployed motion does yield differentiated results. Right. You can see in that company's results that that's true. And in their customers that that's true. I think the thing that lets your team benefit from Vibe coding, a thing that you wouldn't have had access to previously, should benefit and can benefit a bunch of different ranges of businesses. Right. And really the thing that's been holding anybody back from that is economic models more than anything, right? So like if I'm founding a SaaS company, my investors expect a certain kind of multiple, we expect a certain type of product motion and go to market motion. I have to develop a product from afar, find a customer for it has to be designed to be saleable, right. And so at that point, if it's not informed, really, if you didn't get your product management motion really, really well right out of the the gate, it's kind of good luck that it creates durable value for anybody. Whereas if you do this inside out development motion we've seen, like I was saying, creates all this value in the large enterprises, you can find a way to get access to that same problem space, bring that caliber of engineer to the, to the work face, to the mind face. The thesis is, my belief is that it's going to actually produce the same type of differentiated results, which I think the crisp answer to your question is. Yeah, I think, I think think customized software, like inside out development, however you want to call it before we deploy ourselves to Vibe code things. I think that is the future, no question.

Jordy (184:10)

How is traditional private equity kind of reacting to the shields of the world? Smart, very sophisticated financially, can raise plenty of money. They're well aware of AI, they think it's going to be important. You're coming in and you're, you're competing to acquire businesses that might have landed with them historically. I'm imagining you have a very pretty exciting value proposition. The team has an amazing track record, incredibly tapped in, well capitalized. You're saying we want to retain your whole team, we want to keep all of your Customers happy. We want to create more value here by growing the business and growing earnings through just really expanding the customer base. So it seems really appealing to the seller of a company, somebody that a business owner. But of course private equity is not going to just kind of roll over. How have they reacted?

Jim Siders (185:07)

Well, I mean anecdotally it seems to me like they're trying to figure out how to get to this same type of goodness. I mean I'm not in private equity so I'm talking out of my wheelhouse. But I mean one could see that that with the amount of competition in the it small and medium business space for roll ups, just traditional private equity roll ups, one would fear that the goodness is getting away somehow. At least that would be my, my fear if I was there. So you're looking for something to differentiate it. So then there's this kind of growing crop of like what is an it or like a, an AI roll up thing look like. But everybody's kind of talking about it in a very different way.

Jordy (185:44)

Right.

Jim Siders (185:45)

Which is. So one of the things that we're trying to do is not exactly do that. Like we kind of rhyme with that in some ways. We rhyme with a, with a roll up in some ways or we rhyme with a, with a tech company in some ways we're trying to do is kind of take the best of breed of all of those mental models and do something a little bit different. So in our case, for example, one, one easy example of several, the relationship with Thrive holdings gives me a lot longer time horizon. Right. I can think about doing something that's really durable. Like if we're going to change, there's one thing to create disruption or take advantage of disruption. If we're going to really change a really entrenched business model like managed IT services, it's probably going to take a long time even if we're outrageously successful. Right. So having that kind of time horizon to work with gives me a way to address a different thing than any private equity or traditionally structured private equity rule would have to, which de facto is a good thing I think. And I think the private equity folks that are paying attention probably see that also.

Peter (186:45)

Well, thank you so much for coming on the show. I want to hit for a hundred million dollar race. Congratulations.

Jordy (186:54)

Very, very cool. Great to meet you. Come back on if there's something in the news that you've got a strong opinion on, come back on or I'm sure you'll be back on for more fun, please.

Peter (187:04)

I'd love to.

Jim Siders (187:05)

Thanks so Much.

Peter (187:05)

Guys, have a good rest of your day.

Jordy (187:07)

Cheers.

Peter (187:08)

Let me tell you about Railway. Railway is the all in one intelligent cloud provider. Use your favorite agent to deploy web apps, servers, databases and more. While Railway automatically takes care of scaling, monitoring and security.

Jordy (187:20)

Up next, without fear, we have the king of enterprise software.

Peter (187:25)

We have Chris Black back on the show. Welcome to the Stream. Thank you so much for taking the time to come on down to the TVP and ultradome. Good to see you. How you been? How's your 2026 going so far?

Yohei (187:39)

So far, so good. I'm happy to be.

Peter (187:41)

I feel like I'm the voice for radio.

Yohei (187:42)

I feel like I'm the brokest guy that's ever been on the show, so I'm happy to be here. I feel like it's gonna rub off on me.

Peter (187:46)

The voice alone is worth $10 billion.

Yohei (187:49)

Good. I hope so.

Peter (187:50)

Listen to these pipes. It's amazing.

Yohei (187:52)

2026, this guy was. I stayed out a lot later than I wanted to. Three nights in a row for Gramship.

Jordy (187:58)

Came in for the Gram show and.

Matt (187:59)

Yeah, yeah, yeah.

Yohei (187:59)

Well, we're shooting How Long Gone Stuff, but I was like, the timing works out. So why don't I just. You know what I mean?

Jordy (188:04)

Why don't I say, how does it work for how long Gone? Well, you stack.

Yohei (188:08)

We don't. We don't stack at all.

Jordy (188:09)

Oh, okay.

Yohei (188:10)

We were. If we. We record on Tuesday, Thursday and Sunday and the episodes come out the next day.

Peter (188:15)

Okay.

Yohei (188:15)

So Jason edits it like as basically as soon as we finish and then it goes up the next day.

Peter (188:20)

What's the status of the tour?

Yohei (188:22)

We just do it sort of whenever.

Peter (188:23)

Like we kind of. Summer.

Yohei (188:25)

We're going to do New York and la. Probably. Probably. We just want to recalibrate it because it's fun and we like doing it. But I wouldn't say it's a huge upside on the bottom line, just based on our travel, quality of travel demand into the bottom line. Profits.

Peter (188:43)

No, no, no. It makes no sense. But the, the.

Yohei (188:46)

The.

Peter (188:46)

The live show, you prepared sort of a presentation. More thought. It's not just riffing, right?

Yohei (188:50)

This, actually. Yeah, la, like last summer was. We prepared the How Long Gone? Guide to Life, which is like a PDF presentation.

Peter (188:56)

That's cool.

Yohei (188:56)

And it worked. It was fun. And we did it like standing up, TED style. Yeah, but the riffing.

Jordy (189:02)

You ever give a TED Talk?

Yohei (189:03)

No.

Jordy (189:03)

I wish.

Yohei (189:04)

God, imagine with the fucking mic.

Jordy (189:06)

No. When you look at the red dot, you think of the TED Talk era in hindsight. So much so Much opportunity to like go back and just like sign, get. Get yourself a TED Talk and deliver the most like ridiculous.

Yohei (189:18)

Because if you look through it, there's.

Alex Bania (189:19)

There's many.

Yohei (189:19)

So some. It ain't a list.

Peter (189:21)

No. Well, yeah, well. So, so they had, they had TedX Farming Ted. So. So TedX was independent, so anyone could set one up.

Yohei (189:28)

I didn't realize it was independent.

Peter (189:29)

Yes, yes, yes. So. So they sort of license their brand and anyone. So they would do like TEDx Boston University.

Jordy (189:35)

Yeah, yeah.

Peter (189:36)

And it would just be like a local kid that was like organizing and being like come talk and it. And depending on how much they put into production, it could either look a lot like the TED stage or just like a couple people in a conference.

Yohei (189:47)

Open mic night.

Peter (189:47)

Yeah, basically open night. Right. And those.

Yohei (189:50)

I didn't know that.

Peter (189:50)

Yeah, no. And aesthetically those don't hit for the LinkedIn crowd. Yeah, no, no. And so I remember thinking more. I went to one in Boston with a guy who created Android who probably should be giving like a real TED Talk. That's kind of cool. But he was basically just giving a tech talk with a couple other people. Here's my thoughts on open source software, whatever. But it was just him film like three chairs in a conference room against the wall. So. So even if you see it and it says TEDx, you're like, that doesn't feel like a TED Talk.

Yohei (190:15)

Yeah.

Peter (190:15)

But certain places would put the red dot on the ground, give you the mic. You'd be walking around, have the cue cards and everything and it gets wild.

Yohei (190:23)

The best one I ever saw that appealed to me was Mark Ronson.

Peter (190:25)

Okay.

Yohei (190:26)

He like broke down. How like a beat is made basically.

Peter (190:29)

For all these nerds.

Yohei (190:31)

I could see their third eye expanding in a way.

Jordy (190:33)

It's like, holy.

Yohei (190:34)

This is actually so interesting and cool.

Peter (190:36)

Yeah, yeah, yeah.

Yohei (190:36)

But most of those were for me. I can't say I devoured the catalog on the YouTube, on the YouTube channel. I can't stay claim on that. Yeah.

Peter (190:45)

What were you devouring at the time? This was like 08, 2010 era.

Yohei (190:49)

Mostly cocaine. But I mean, I guess there's. I guess there was some. I was probably. I was going out a lot. That was like New York era. I feel like that was honestly like music blog.

Peter (191:00)

That was hype machine.

Yohei (191:01)

I was more of like a stereo gum, pink pork, Brooklyn vegan. Like sort of like that. Which is kind of back now.

Peter (191:07)

Yeah.

Yohei (191:08)

I mean we went to the Pitchfork party this weekend. It was one of the best parties of the Grammy. It was really fun.

Peter (191:11)

Yeah.

Yohei (191:12)

And I think that it's like, I don't know if you guys are up on the paywall discourse around Pitchfork. They basically put up a paywall.

Peter (191:19)

Okay.

Yohei (191:19)

And the, the crux of it, to my understanding, is that you can basically, you can rate albums, famously. The appeal of Pitchfork is the numerical score.

Peter (191:29)

Numerical score, but sort of brutal, right? Sometimes they'll get someone like a 4 out of 10.

Yohei (191:33)

Yes. But sometimes the. Often times the score doesn't necessarily match the actual review. But no one's going to read the actual. You know what I mean?

Peter (191:40)

Only the head is reading, so it.

Yohei (191:42)

Could be like a five. But then you read it, it's like, oh, it's not that bad.

Jordy (191:44)

Actually.

Yohei (191:45)

There's a little bit of a disconnect sometimes. So they're giving people the option, which I'm pretty bullish on, because I think that everyone on the Internet thinks they're an expert or critic and thinks they need to be heard. It's like the letterbox. And I think that there's plenty of.

Peter (192:00)

Rotten Tomatoes has the tomato meter. Metacritic has the audience, plenty of like.

Yohei (192:04)

Mouth breathing music fans that think that they should be able to rate the My Morning Jacket record how they feel.

Peter (192:10)

Yeah.

Yohei (192:10)

You can't underestimate the passion of these civilians.

Peter (192:13)

You pay them, they send you a text file. You can write whatever you want about My Morning Jackets.

Jordy (192:18)

They printed out.

Peter (192:19)

If you want 10 out of 10, we'll just show that right to you.

Yohei (192:22)

So it was a controversial idea, but I think it's actually quite smart because it doesn't, it doesn't, it doesn't like the critic part of it is still going to be there. So it doesn't dilute the overall, you know what I mean? A little bit to me. But it's, but it's, you know, the, it's a elevated comment section in some ways, which is a scary thing to.

Peter (192:38)

Consider in our world. Davos came back just this year.

Yohei (192:41)

That's true.

Peter (192:43)

They should have had you at Davos for sure.

Yohei (192:45)

Yeah, we were. Yeah, that's. We're open to that. If Davos is listening.

Peter (192:48)

Any hope for south by Southwest making a comeback? Any other, other conferences?

Yohei (192:53)

Southwest has actually done pretty well because they pivoted to like Internet, you know, it's not about, it's not bands playing sponsored by Sparks at three in the afternoon now it's like a little more serious and there's a film aspect to it.

Peter (193:04)

Okay.

Yohei (193:05)

I think it's just moved away from music a little bit. Like it's still a Part of it, but it's more of an element instead.

Peter (193:10)

Of a. I'm just wondering if it comes back and they, and they really play into the music.

Yohei (193:15)

I just think that it's sort of. I, I mean, there's a possibility only because I think right now we're seeing a real swing back to like guitar music.

Matt (193:22)

Okay.

Yohei (193:22)

Because I mean, not only are low.

Jordy (193:24)

Key fell off, low key fell off hard.

Yohei (193:26)

Not even low key fell off. And I think the diff. Now that somebody was telling me a.

Jordy (193:31)

Buddy was at like a guitar center the other day and he sold out in Hollywood.

Yohei (193:36)

No guitars left?

Jordy (193:36)

No, no. He said it was like every single guitar was like on a crazy sale. That's probably like a Q4 glutton. I mean, but I was like, I just remember being a kid, like, like wanting like, like I, I played guitar like you know, every single day growing up and all these guitars that like, were just so out of reach. Now I'm, now I, I, I don't know. I can see it. It's the perfect analog, like. Yeah.

Yohei (194:02)

I mean, it's like anything else. It's sort of like, I mean, music always goes in phases like that, you know, like, not generationally, but maybe like by decade, I would say. And I think the chart thing, even though I hate to pay attention to it or like streaming numbers or whatever, it's real, you know, like there's, there hasn't been a hip hop song in the top 10 or something for like, like it's an interesting thing that's happening. And the Geese of it all.

Peter (194:21)

Wait, isn't SD Kid Top ten, I assume?

Yohei (194:24)

No, no, no, no, no.

Peter (194:25)

He's, he's, he's based on Instagram reals on your algorithm.

Yohei (194:28)

I mean, that's the thing. There's no monoculture, so it's sort of like you see monologue to some extent. But I think that the, the Geese is the big sort of thing that people are talking about because it's culturally so relevant and it's making people mad in a way that's like sort of exciting. Like people hate it as much.

Peter (194:43)

Tell me more of the story.

David (194:44)

Story.

Yohei (194:44)

I mean, Geese is a band from New York that's put out a couple records, are really young, and it's, it's just really polarizing. I mean, it's like this guy, Cameron Winter is the singer. He's a, like he put out a solo record. He's a generational sort of talent. Sure, it's sort of like a Neil Young, Tom York, but they played snl and people were very mad. Like, this is the worst I've ever seen.

Peter (195:02)

Yeah.

Jordy (195:02)

But then they just didn't like the way it sounded.

Peter (195:04)

Yeah, yeah.

Yohei (195:05)

Look how they sound.

Peter (195:06)

The whole, you know, Del Rey had a similar SNL experience.

Yohei (195:08)

It happens.

Peter (195:09)

Video game.

Yohei (195:10)

Because SNL is, I mean, that's as middle of the road as you're gonna get.

Peter (195:13)

Yeah, but. But it seems like some of their content booking is like a little ahead of their curve.

Yohei (195:18)

They're. They're usually. I mean, they're good in general at taking a few risks. For every cardi b and Mumford and Sons, there's a geese that sort of like they take a swing and that's what they're known for.

Peter (195:29)

Yeah.

Yohei (195:29)

I mean, we sound like boom. I mean, I get called a boomer for watching snl.

Peter (195:32)

Yeah.

Yohei (195:33)

And I will take that. I deserve that lump. I think it's real.

Peter (195:36)

Yeah.

Jordy (195:37)

What, what happened? What happens to late night shows in general? I feel like they thrived when you had a monoculture. You had the whole world fixated on the same types of things. Now it's like you turn it on and realistically, to appeal to the broad enough audience, you gotta have.

Yohei (195:55)

It's impossible.

Jordy (195:57)

20 shows that were taking the place of. And the Internet sort of does that.

Yohei (196:01)

I mean, I think as much as I have a respect for and love to replace them, in some ways, I think it's sort of it. We're at the end. I mean, I think that like celebrities need places to promote their projects and that is not going to go away. But there's shit like this and that gets more eyeballs and is feels more relevant. And I think that like, you can go on Dax Shepard and it's going to do more for your movie or your album or your book than it is to go on Stephen Colbert. That's just the reality. And it's like I'm more concerned with the loss of the music performances because I think NPR tiny desk is like corny.

Peter (196:34)

Yeah.

Yohei (196:34)

But I think there's. I spend a lot of time on YouTube watching like performances from the 90s, 80s, whatever. And I think that magic is what.

Peter (196:42)

But you can reconstitute that on the Internet. I feel like you can.

Jordy (196:45)

Why don't you guys do that?

Yohei (196:46)

I mean, we know we're trying to do it and we've talked about it a lot and I think it's. It's a. It's easy theory.

Peter (196:50)

There's nothing that says that Dax Shepard can't open with a monologue, have three guests and then A musical performance and then just musical performance.

Yohei (196:56)

Be Kid Rock. That's the problem.

Peter (196:58)

But you do something different. But, but, but just the fact that it's an rss.

Jordy (197:02)

I think there's something to be hour long show this. I, I feel like you guys getting like a space like this in New York like where it's full time dedicated people are coming to you will change. We gave this advice to some of our other buddies that have had a podcast for a really long time. We're like, dude, it's kind of a nightmare having to like travel around and all and like you know, constantly on the move.

Yohei (197:24)

We, we're, we only do audio. Really.

Matt (197:26)

Yeah.

Yohei (197:27)

And I think we've, I'm here. We branch out. We did like a year in review.

Peter (197:30)

Yeah. Yeah.

Yohei (197:30)

And it did really well. So now we're going to try to do the. We should have a studio in Burbank and we shoot these sort of like Weekend Update talk soup style sort of like monthly wrap ups where it's, you know, it's us at the desk. We're talking about stuff. You see the image over our shoulder.

Jordy (197:43)

Yeah.

Yohei (197:43)

And that's gonna, I think that's what we want to do.

Jordy (197:45)

Yeah. Turn it into bring. Have a small live audience. Have, have, have like a stage that people can perform on.

Yohei (197:52)

That's the idea.

Jordy (197:52)

Start doing it. Start doing it live. Yeah.

Yohei (197:54)

And it's just, I mean it's just, it's, it's, it's all about scheduling. But what we've learned from doing the podcast is audio only is that you just get access. Also, we don't ever. We're never in the same room. It's all on zoom and I think it just allows a different, like it gets looser.

Peter (198:09)

I think it feels much more protected.

Yohei (198:11)

There's no hair and makeup. There's no you need to send a car. There's no, there's no like pretense to it. Like you can do it in your hotel room wherever you are as long as you have wi fi and headphones. And I, I. The same way we're sort of like anti paywall and it's costing me a lot of money. It's a similar thing where like audio is the real medium. It's like all these new Netflix things.

Peter (198:28)

Yeah.

Yohei (198:29)

Like Pete Davidson having a talk show on Netflix is not a podcast. It's a talk show on Netflix.

Peter (198:33)

It's not first time.

Yohei (198:34)

There's no audio. There's no RSS feed.

Jordy (198:36)

It's not.

Yohei (198:37)

Yeah. I mean he's, they're doing A lot of the. I mean, they're. They're. They're showing podcasts that exists, like from the ringer, but they're also launching these shows and paying these people a fortune. But it's. It's just a show.

Jordy (198:48)

Have you ever thought about. About selling out?

Yohei (198:51)

Look, I would love to sell out. It just has to be on my, like, archaic. I would love to sell out every day.

Peter (198:57)

I want to sell extremely complex rider here.

Yohei (199:00)

When I'm here, I'm like, this is how it's supposed to be done. Like, this is such an operation. And I knew what I was getting into to an extent, but being here, I'm like, oh, this is what it requires to do something on this level. And it's. It's a lot.

Peter (199:10)

You need a horse.

Yohei (199:11)

Yeah. You need a horse.

Jim Siders (199:11)

Yeah.

Peter (199:11)

Something.

Jordy (199:12)

Something about it.

Yohei (199:13)

Guys eating airwall.

Jordy (199:14)

That's a instrument.

Peter (199:15)

The guys told me how much the thing costs. I was like, I can't believe you guys spent that much. They're expensive.

Yohei (199:20)

I actually knew that.

Jordy (199:22)

There's something about that. We appreciate the, like. Some people would say it's monotonous, but I actually appreciate just waking up and going to work.

Peter (199:30)

Yeah.

Jordy (199:31)

It's like making content.

Yohei (199:33)

No, I think treating it like that and the way you guys have treated it and the way you've approached it is why it works. There's. There's a level, and we feel the same way. It's like we do three shows a week. Week. Which is pretty psycho.

Jordy (199:43)

Yeah. I mean, not considering you're not in the same place.

Yohei (199:47)

Jason edits and I book. We do it all ourselves. But I think that the. I think that once you lock into a system like that and people are looking for it and expecting it and you deliver on that, they stay with you. They want. I mean, you guys are doing news, and, like, you're really covering, like, a breadth of industry, so it's different. There's like, a value proposition of, like, information.

Peter (200:04)

Yeah.

Yohei (200:05)

Which I think is much more than us talking about, you know, now the culture.

Peter (200:08)

Yeah.

Jordy (200:09)

What? What. So Grammys last night, you've been in town hanging out, bumping shoulders with elbows. Elite. What are the convers outside of culture, war and politics? What are people talking about? Or, like, how much is, like, AI being discussed specifically in the music context?

Yohei (200:32)

I mean, I think that musicians are mad about everything all the time, sort of. Even the most successful ones. That's part of the personality makeup, I think, to be that. I mean, I think that fear is real, and I think every day there's a new story that there's some anonymous AI guy that just made $5 million streaming, you know, or whatever, through. Through Spotify royalties. I mean, I think the real conversation that's, that's being had or that I've seen a little bit of is, is people coming out and saying, no, no, I use it to help me edit or do this. And it sped up the process, you know, 10 10x.

Matt (201:04)

Totally.

Peter (201:04)

Yeah.

Yohei (201:04)

And I think it's like anything else, man. It's like it's coming.

Peter (201:06)

Yep.

Yohei (201:07)

How can you, how can you utilize it in a way that feels ethical and good to you, that doesn't compromise the art? And I think that's, that's the challenge everybody has. And I'm sure there's going to be people that are like that. I'm never doing that. And there's going to be people that adapt and there's going to be, you know, like a Will I am type who takes it too far and that, you know, there's a range, there's always a range of people that want to experiment. But I think that, that I think with music there's, there's ticket prices, there's, you know, royal streaming royalties. There's so many battles that they're fighting all the time that it's sort of. I almost feel like AI is kind of like. Feels like it's in the distance a little bit, honestly. Because there's things you're saying, I mean, sure, it's coming, but I think there's so many immediate things they're concerned with as far as sort of you know, just like respect and money.

Peter (201:49)

Yeah. Is there, is there some sort of alliance between like reaction to streaming is you need a really strong live performance schedule. The concert becomes more important. Creating like the Taylor Swift effect becomes more important.

Yohei (202:03)

It's almost. What's so crazy. Well, the real driver is, is TikTok okay? Like if you have. I mean, I've had several friends that were relatively successful musicians in the aughts, have a song go viral and now they have a platinum record, they bought an apartment, they're on tour selling tickets that. And you can't plan for that. Labels can't manufacture that sort of God's plan. And, or, but like Olivia Dean who won best new Artist last night, people say she sounds like T.J. maxx. I, I like it. I think it's a hit. It's a hit song. But, but part of it is that it became a tick tock thing.

Matt (202:34)

Sure.

Yohei (202:34)

And it's like, it's that good you need more things. You know, it has to be a, a multitude of things kind of coming together to make it work. Yeah, but I think, I think touring is just people. I just wrote this in my GQ column. I wrote about the Harry Styles ticket price thing because people are really upset. His tickets are expensive and everybody's tickets are expensive.

Peter (202:53)

Second market, secondary market.

Yohei (202:54)

He's doing 30 days at Madison Square Garden. He's doing 30 nights at MSG, he's doing 30 days. It's unbelievable. And MSG is also historically the most expensive venue because of unions.

Peter (203:04)

And you think that, you think that with 30 days of supply ticket sales.

Yohei (203:07)

Well that's, that's the question is people are like, he's greedy. And I'm like, well sure, I think there's some truth to that. But also I think the discussion I'm interested in having is like, like, what do you guys want? Like, cuz he can fight Ticketmaster like Pearl Jam. He can, but he can also the production that fans want costs a lot of money.

Peter (203:24)

Sure.

Yohei (203:24)

When Harry Styles is on tour, he's got 30 trucks. It's a lot of, it's 300 people.

Peter (203:29)

It's a, you know, SM circus. It's really guy.

Yohei (203:33)

And I think that the cost that go into it is sort of lost in the fan. But then from the fan side they're like, this is untenable. Like I can't pay this much.

Jordy (203:40)

I have what, what's the cheapest ticket?

Yohei (203:42)

I mean in theory it's $100, $200.

Jordy (203:45)

But it's immediately reselling.

Yohei (203:46)

That's the issue.

Jordy (203:47)

But then that just proves that it doesn't really matter where he prices it.

Yohei (203:50)

The demand, that's my, that was my whole kind of.

Jordy (203:52)

So that yeah, you basically could just, it's like if like look at the secondary price and then look at what he's charging and like greedy would just be charging the secondary price. And we see that, we see this every, you know, look in cars and watches, any category with way more, more demand than supply manufacturers will just be like, wait, like this GT3RS is selling for like 50k over sticker. Like okay, we're just going to do that at the dealer level now or secondary. Like look at AP moved up the retail on the Royal Oak just because there was enough demand. And like it actually kind of like normalizes the secondary market, but you're still paying.

Yohei (204:30)

Like if you think that. I think if you need to place blame on someone, it's going to be the, the resellers, the bots and Ticketmaster.

Peter (204:38)

That makes sense.

Yohei (204:38)

Like, I don't necessarily think it's Harry. It's, My point is it's not Harry Styles. And sure, maybe he could put his foot down.

Peter (204:43)

Yeah.

Yohei (204:44)

And, and like make a case. But we don't know the ins and outs of that, the difficulties of that or how that would actually affect the product.

Peter (204:49)

Yeah.

Yohei (204:49)

You know what I mean? And that's, that's the issue.

Jordy (204:51)

So you think the solution is NFTs?

Yohei (204:56)

That is for 100.

Jordy (204:58)

What, what's going on? Give us an updated, just broad update on menswear. We don't talk about it.

Yohei (205:04)

Oh, wow.

Jordy (205:05)

We don't talk about, we don't talk about it much. We obviously normally wear suits.

Peter (205:08)

I, I actually, I got the memo today.

Yohei (205:10)

I, I was, it's funny, I wore a suit to. Charlie had a party last night at the Chateau Marmont and I wore a suit.

Jordy (205:16)

And we, Charlie, Charlie xcx.

Yohei (205:17)

And we had to record this morning our thing. And we. But Jason, I both usually wear suits and I text him like, hey, man, maybe today can we not wear a regular? Couldn't do it.

Jordy (205:27)

I feel like slacking off.

Yohei (205:28)

Same. Honestly, it's getting a little more formal in a way that I think is really positive. Like, it's sort of logo mania is over. I think it's a little bit of like Jonathan at Dior showing ties. Like, it's a little more sort of suiting, I think, and like, stuff that I think is more flattering and generally easier to wear in a lot of ways. Like, I think, like, you guys can wear suits. You guys cannot wear, like dumb Balenciaga or whatever. You know what I mean?

Jordy (205:53)

Like, we've thought about, we thought about.

Yohei (205:55)

You can, you can do it if you want. I don't know.

Jordy (205:57)

We had the John. John made this, this image. It was like a steel steal his look image.

Peter (206:02)

And he's like in $100,000 outfit, Chrome Hearts hat, chrome heart belt supreme with the Louis Vuitton. Then Bottega Veneto Roman weave jacket.

Yohei (206:12)

What level of, what is your, how much chrome knowledge do you two have?

Peter (206:16)

Zero.

Yohei (206:16)

Okay.

Peter (206:16)

I know nothing.

Jordy (206:17)

John had zero. I, I just thought all of these brands were.

Yohei (206:23)

Yeah.

Peter (206:23)

I found out that Balenciaga is like an old company. I thought it was like a new thing from like a couple years ago, hundreds of years from.

Jordy (206:29)

So, so when I moved to Malibu, I, I, I would always. I, I, I, I will be. I'll admit I didn't know nothing about chrome. And I love you have a G wagon.

Nick (206:39)

Don't you?

Yohei (206:39)

You don't know about. Yeah, yeah. Kind of a connection.

Jordy (206:43)

Hey, I'm a. I'm unique. I'm a special snowflake. The only G wagon owner that doesn't Chrome Hearts. So I moved to Malibu and I'm like, why are all these kids? It's the younger generation. So Malibu. The owners of Chrome Hearts have lived in Malibu. Starks have. So they own the surf rider now, which is also interesting. But I moved there, I'm like, why are all these like 1212 year old like kids running around wearing chrome like you live at the beach. Why are you wearing like this like goth stick black sweatshirts? It made no sense. I was like, lighten up a little bit because I just always associate the clothing with like kind of just emo darkness.

Yohei (207:20)

You got it wrong, bro. You need some chrome. You guys need.

Jordy (207:23)

Yeah, yeah.

Yohei (207:23)

So the jeans, we could get you some jeans. Yeah, 15,000. Look, it's a custom pair of Levi's, but they put the leather cross patches. So depending on patch amount.

Peter (207:36)

Sure.

Yohei (207:36)

The price is going up. So I would do something tight, tasteful with all the chrome rivets.

Peter (207:40)

Okay.

Yohei (207:40)

And maybe a single patch.

Peter (207:42)

Okay.

Jordy (207:42)

Just a one patch.

Yohei (207:44)

You guys need hella patches.

Jordy (207:46)

John. This is why I always said so I always said that the top of the market cycle.

Peter (207:50)

Yeah.

Jordy (207:51)

Will be when a kid is standing YC demo day. You know where the founders pitch and he's got just all. Just full chrome. No, I don't think it's happened yet.

Peter (208:01)

Wasn't there some CEO that was caught with a Chrome Hearts jacket?

Jordy (208:04)

Oh, yeah, yeah, there's definitely been some. Some people have think they can be sneaky and be like, oh yeah, I'm just like, I'm just a founder. I'm scrappy. And they're wearing the leather jacket without the logos and people clock it.

Peter (208:16)

We know what you're doing.

Yohei (208:17)

Yeah, I need to get you guys. Yeah. But as far as, as far as. As far as the menswear thing goes, everything's sort of calmed down, which I think is good. I mean, I, I literally. I started a brand at the end of last year that I've been working on for a couple years called Hanover. And it's been an interesting. Someone who's worked peripheral, you know, in fashion for 10 years, 15 years. Doing your own thing is a whole different animal, I will say. But our, our proposition is everything is really affordable. It's all made in usa. So it's like we're all, we're making it in la, but it's really simple.

Jordy (208:47)

Yeah. I've been shocked because we make. We make stuff we like. This is something, you know, we made for Turbo Puffer, one of our sponsors.

Peter (208:55)

Yeah. There was no.

Jordy (208:55)

And comparing. So. Yeah, so comparing. Comparing the prices in China versus here. China is incredibly. They make great clothing that's inexpensive. But when you look at. When I. When we've gone kind of like line by line and looked at these different items, I'm like, okay, I could pay $70 for this and it gets made here in LA and I'll be done in three weeks. Or I can deal with like months and months.

Yohei (209:21)

It might come on a boat.

Jordy (209:23)

It might. It'll come. Yeah, it'll come on a boat and It'll cost like 95 to dollars. It's not that much for us. For us, there's like, no question we want to make it here because we're not in the business of clothing.

Peter (209:33)

Yeah.

Jim Siders (209:33)

Right.

Jordy (209:34)

So the margin is like, less.

Yohei (209:35)

It's negligible.

Jordy (209:36)

Yeah. Yeah. But. But still, it's not. It's not as like, it's not like a 10x difference, which I think a lot of people assume.

Yohei (209:42)

I think that I. I mean, I think that I came up in an era where that was important, like Made USA was important. And I don't. I don't necessarily think it's important per se, but the idea of being able to do it was pretty compelling.

Jordy (209:53)

But it's. Yeah, it's.

Peter (209:54)

It's.

Jordy (209:54)

It's more and enjoyable. That's why I've always thought people see the process. Yeah. And like the people that are like, drop shippers being like, oh, yeah, like, ecom is so great. If you want to, like, live in Thailand. I'm like, have you ever made anything like, can you imagine, like, trying to iterate on, like a physical product, living in, like, that's what people.

Yohei (210:11)

I mean, that's the thing. That's. I think that's the thing that I. You're right. It's like, what's the. What's it worth? You know, my time and my effort and my stress level, like, what is it worth? But it's. It's been fun so far and it's going well, but I think it's just like a.

Jordy (210:24)

What's the. What's the. What's the strata? Like, doing collections. It's quarterly.

Yohei (210:28)

Yeah. But it's basically. I mean, the start was T shirts, denim sweats, and the denim did really well. We're going to introduce another fit. So it's. It's sort of like you do stuff, you see what works.

Peter (210:38)

Is that 15, 000 price point?

Yohei (210:39)

It's 15, 000 price point, actually. It's currently a sub 300 everything.

Jordy (210:44)

Whoa.

Yohei (210:45)

Which is pretty strong.

Peter (210:46)

Yeah.

Yohei (210:46)

15, 000. Then you can buy something.

Peter (210:48)

I love it. We have to get to a flight. Of course, if you want to plant the bomb, Jordy. But I have one last question. Dream guests. Do you have dream guests? We don't really have dream guests because it's very opportunistic. It's like whoever's hot right now, we want to talk to. So we don't have a list, but.

Yohei (211:01)

Honestly, I. I always like to say, I like to be surprised. But after Ben Affleck has been on this press tour with Matt Damon for that new movie. He's so smart. He's so well spoken. I would really like to talk to Ben affleck.

Jordy (211:13)

Hanover-usa.

Peter (211:15)

Go check it out.

Yohei (211:16)

Give us five.

Peter (211:16)

Five stars on Apple podcasts and Spotify, everything. We will be live Tomorrow from Cisco AI Summit, 11aM Pacific Sharp. Thanks and goodbye.

Yohei (211:24)

Thank you.

Peter (211:27)

Nice work, brothers. I'll see you on the next one.