The Exchanges

Q How different was it in the Facebook days? Because you had sort of the same dynamic.

A I think, first of all, it's kind of funny because by historical standards, then it was an incredibly meteoric rise. Facebook's valuation grew unbelievably slowly compared to the labs, right? And There were obviously generations. Like I would argue I was Peter, you were kind of generation one and a half. Maybe I was generation two at Facebook. There's now been generation three and generation four, but there's not 20 generations, right? There was like a set of people who worked for four years, did great work, and then passed the torch. And that's happened like a few times. And actually the core leadership team at Facebook has been unbelievably stable now for like a decade, right? Like Which is a very different dynamic.

AI assessment note: “Facebook's valuation grew unbelievably slowly compared to the labs”

Q Apple agent. I've got all access to everything on your phone. Don't worry. It will never be in the cloud. It will never be exploited by a third party. What do you want me to do for you now? And you just like go through and give it connectors and access as you need certain things. It saves your login as you need it. Why wouldn't they just win this?

A Well, I think Apple might win it, but I think the interesting thing is I don't think you want to be the one who takes on that liability. If anything, I think the goal is to like radiate out risk to other people and be like, well, we're not like to keep yourself safe and be the one that's aligned. Cause like the reality is you're like, Peter, you're saying that you think the ban is going to go up. The reality is the hacks are only going to continue to grow and scare people more and more and more, right? Not less and less and less. Everyone's going to have them, and so I just wonder if the real thing is going to be, if you really want to win in, like, the new economy, in this AI thing, you have to somehow figure out how to hoard the data and the trust and build the trust, but then when things go badly, always have a third party to point at to blame.

AI assessment note: “I don't think you want to be the one who takes on that liability.”

Q Is data moat or have we lost that too?

A Can be, can be if you can only uniquely create it. So I think this is, like, this is my new precision, right, on this, which is, like, these people, like, I was talking to some entrepreneurs, actually two today, who I felt like I was listening to, like, a 2021 pitch, where they're like, we did the user research and talked to the buyers, And the problem the buyers have is blah, blah, blah, blah, blah. So we were building software to address the buyer need. And I'm like, what universe are you living in? Right? Like, if you believe that the tendency is to like, all this shit is like commodity software that can be cranked out, which is the direction. I'm not saying we're there yet, but we're in that direction. You can't build a business doing a thing a buyer thinks they want right now, just because it's like, and it's just software. It has to be. That you're doing something where you write software, right? And you basically give it away, right? Or it's like basically free. You, you don't bank on the software, but you use it to build some sort of durable compounding data asset or trust asset or marketplace asset where like they have to keep using you because you're fundamentally structurally better at something that's defensible, right? Where you have a compounding advantage. But this idea, like anyone who, who believes That they can build a business because something is hard right …

AI assessment note: “Can be, can be if you can only uniquely create it.”

Q On that front, Sam, I was going to ask you, you had a post about You've had several posts, but one this week about, I'm going to get it wrong, but when intelligence goes to zero, what happens?

A Yeah, yeah, yeah, yeah. And blinded intelligence is bad business. It's like, everyone's like, all these computer scientists running around gleefully talking about how computers will commoditize human intelligence, and so all the humans will have no jobs, and they're kind of like, sad about it, but they're also kind of proud of their power. But what they fail to realize is that The, you know, the original narrative was someone would get to AGI and then that would be the end of the game is not how this is playing out. Everyone's gonna get to the same shit. It's all commodity. It's all getting a price to the cost of energy. And it's all worthless because if things are abundant, then they have no value. So I'm just like, guys, if you really, like, if you really believe that we're moving towards a world of abundant machine matrix multiplication intelligence, y'all in the pretty bad business. Being hard is not a moat anymore. Software is not a moat anymore. I mean, I've been, I'm, like, having fun with this because I've been saying for a while.

AI assessment note: “It's all commodity. It's all getting a price to the cost of energy.”

Q in that. And so, you know, is there any signal in being like an alliance of sorts? Like, do you think the university presidents are all talking about how should we position ourselves in this together? Is that, is that a, would that be a good thing, bad thing? Is there any way that You know, there's just, like, a quicker way for the dominoes to fall into place here.

A I don't think so. I think these guys, unfortunately, don't really coordinate enough, right, in terms of how they think about or respond to these things. They think about their campuses and communities differently, and they're very hamstrung, is my honest sense, in a lot of cases. I mean, there are exceptions. Like, University of Florida has done a really, I mean, there are places that have done a good job with this, and there are places like Chicago, which have really clear principles, right? Because they have really clear principles, and they articulate them really well, The president of Chicago's letter was fabulous, being like, this is the process allowed, this is what's not, you do this, you get arrested, you do this, it's fine. And you're like, great, thank you for the clarity, right? But I think the problem is, is that a lot of these, you know, complicated organizations without enough clear direction, they're just kind of drifting. And the students, you know, they have nothing, they think they have nothing at stake, although I'll never hire any of them.

AI assessment note: “I don't think so. I think these guys, unfortunately, don't really coordinate enough”

Q Sam, does Bitcoin hit a hundred K this month?

A It might. I mean, it's, that's within random walk at this point. I mean, the only, well, the, the only thing that I've heard contrary to that is that Look, I think there will be a lot of cell pressure into like just around a hundred K just because a lot of people have been up and down and say that's a good number and I'm out. And so I don't, I don't know. I mean, I, you know, like many of us on this call, I'm like a long term crypto holder and will continue to be and have no intentions of selling or anything. But, uh, you know, I think the hundred K the hundred K thing is kind of an emotional thing more than a practical thing.

AI assessment note: “It might. I mean, it's, that's within random walk at this point.”

Q what does it sound like when a tool bit is going to break and how do you, what is the path that you take? I mean, those are, there's a bits of data that I think, you know, in some ways, Sam, I think you and I saw this at Facebook. It's like, what did Facebook do? It just digitize the information that was in our head. Who's your friend?

A Yeah. A hundred percent. I mean, I used to, we used to talk about, I mean, like talk about the composer and Facebook era. It was a data pump, right? It was an efficient, free data pump to bring data online that couldn't exist anywhere else, and the reality is, if you go way back in the Facebook history, the real story was on day zero, internet's scary, so no one wants to post on it. You're never gonna put a real photo on it. Facebook created the trust and security for people to take a bunch of real-world data, photos, relationships, messages, et cetera, and bring it online, right, for the first time, and created an enormous amount of value Doing that, right? And so now we're in this weird multiverse where it's like, the internet has all been scraped to hell, right? Like good luck New York Times with their lawsuit, right? But like that has been like copied a thousand times and distilled a thousand times. You have personal private digital information, what's in Gmail, what's in da da da da, where people have some say in theory, as opposed to it as all being like dumpable effectively. And so that's where we were talking about the trust, things like that. There's then, like, the, the whole class of companies that are out there trying to, in some ways, build new data sets, the handshakes, the scale AIs, where you just pay a ton of money and you turn financial capital in some form in…

AI assessment note: “It was an efficient, free data pump to bring data online that couldn't exist anywhere else”

Q Are you still using just one? I have, like, So many agents, because I'm trying all the different agent Harnesses and platforms and like, because I want to compare them as we're moving along in this race, right? But are you just using one, Sam? And is that where you think most people are going to net out here?

A I, it's an interesting question. I mean, I, I basically, I do, I guess it depends if you define what an agent is. Like I use Gemini for things that I actually care about security in, right? So like for health and finance stuff, I just use Gemini because honestly I trust them. Like, and I think there's like value to that. It's like, I'm not even sure they're not like, I have no opinion about them being fundamentally better. I just like, I generally trust. It's like, it's kind of like my dad used to say, you can go on Disney World rides because they're a public company, right? They're not going to screw you because it's too expensive. And so I'm like, okay, like, it's like, that is my Gemini thing. And like, I use like Grok for image stuff because candidly, it's just way better. It's like not even close. Like Grok is the only one that's cared about image stuff. So like their image stuff is just way better. And then, I mean, I use tons of services and agents, but they're all for me wrapped through like a central like setup I have built around Right, so like I have, like, Claude controls everything else, and it controls a huge number of services, but like, it's kind of the tip, and I have multiple instances of it running, and I email, I mean, like, it's like, but it's really hard to even define what an agent is. It's like, I have an infrastructure that has a huge amount of access t…

AI assessment note: “I use tons of services and agents, but they're all for me wrapped through”

Q that Got, you know, a silver spoon and screwed it up. But I feel like a lot of them will also not graduate with college debt or debt for whatever education ends up looking like in the future and have a little startup capital to take a risk in their lives and all the other stuff. So we're not going to solve it for everyone, but isn't that net better?

A I'm very pro the Trump account. I think it's good and smart. I'm pro it. I I'm not convinced that the picture of how it compounds Is so certain, especially as we're talking about resets and all this other stuff that, you know, we're trying to figure out right now, but like, I'm certainly pro it. I just, except with the exception of when people start lobbying for regulatory capture by giving a 500 bucks that are stocked to someone, right? Like that seems insane to me, and we got to figure out how to like not have that happen. I want to pivot one or two more times before I lose you guys, and I really appreciate you, you joining for this fun conversation. OpenAI Exec Exodus. Brad Lightcap out. Kevin Weill, who we used to work with, out. Peter, you've been out for, like, years, so, like, you were just, you're a forebearer on this. Like, why and, like, what, what's going on?

AI assessment note: “I'm very pro the Trump account. I think it's good and smart.”

Q Yeah. But I think like the macro points to like, there are just shakeups happening at all these like frontier labs. I think a lot of Google employees might see this as a sign that maybe they should leave. Is it unstable? You know, I think anytime that just one, not only two leaders like this leave, It creates some instability.

A The world has changed a lot. Like, you just think, like, these AI things and the types of orgs required to run them, the leaders required to run them, has just changed a lot. Like, you know, a lot of these guys have been doing this a long time. They're a great research leader. They're not gonna, like, be the operational leaders you need to do. By the way, half of them are so fucking rich. They don't want to, like, lead some big team. That sucks. Like, they want to work on, like, frontier cool stuff, right? Which is, like, health or things like that. So this is, like, we're in the cycle where Especially at, like, big companies. The right people aren't the right leaders at different phases, and I think it's good and healthy to, like, recognize that and then reshuffle the decks when you can. Like, the worst thing you want is to have some guy who's, like, the wrong leader for this phase. They're not in it because they're too rich and they want to work on health. They're, like, an amazing research leader, but they don't care about driving some random metric up. Like, you want to cycle people out In healthy orgs. So I actually think turnover and these types of changes are good. Now, the interesting thing when you're at a really big company and you kind of keep people around cause you can afford to. And so it is sometimes comp inefficient. You don't fire anyone. You're just like letti…

AI assessment note: “So I actually think turnover and these types of changes are good.”

Q they all kind of went way up and then they kind of crashed out and nobody remembers any of them in terms of like, it was just simple photo upload. And then Instagram came out with like, okay, we're going to have the common unit and we're going to have the feed and a network structure and all the rest of it. And so that's the thing that stuck, right?

A Elad, how do you think about that? The thing that strikes me is so different about this AI wave though, versus mobile and the fuck around experimentation when you have something early, new and weird is just, The scale of capital that's being played with right now, where the amount of money, I mean, it's one of those things where like, yeah, like you have a bunch of small teams of two people fucking around with some APIs and maybe they build an app and maybe they don't, it's super cheap and dah, dah, dah, dah. This is an era where like, you're talking about billions and billions and billions of fuck around dollars, right? Being thrown around. And how do you feel about that as an investor, right? And as a capitalist, when you think about, yes, I mean, I agree. In enterprise value from some of this fuck around stuff, but the cost is going to be a trillion dollars of capital sunk to create that three hundred billion dollars.

AI assessment note: “The thing that strikes me is so different about this AI wave though, versus mobile”

Q And you're suggesting we're inviting China in right now?

A Well, we invited Kimmy in to clean up our own mess on OpenAI. OpenAI will invite an anthropic. Like, all these things will be like, oh, shit, we screwed it up. Let's open the doors, and we need their help. And then once the army's in, they say, well, I think we're gonna stay here now for security, of course, because we need to make sure everything's safe for you, right? Like, that's how these things will play out, right? And so, it's gonna be really interesting. I mean, there's, I was in, there was another conversation going on in Silicon Valley this week that I got a debrief of, because clearly we're not there. And it was really interesting. A bunch of, like, political people and AI people And the tenor of it was basically, which is unsurprising to any of us, is, like, every country is into, like, sovereign infrastructure and sovereign models for obvious reasons. Like, that's how you, like, leave your doors open or close your doors, right? And what's gonna happen from conquest perspective, digital conquest, will be, like, a bunch of these countries will try that. It will go badly. Shit will start breaking. They'll be like, oh, shit, we need, like, the Americans. We need the, we need someone to come save us. And then whoever gets to come save you now runs your country.

AI assessment note: “Well, we invited Kimmy in to clean up our own mess on OpenAI.”