Why the Co-Founder of LinkedIn Is Betting on NFTs Again

When I bought my first Bitcoin in 2014, and by the way, I haven't sold any since then. I'm a believer in all crypto. It was like actually in fact, this is part of a design feature that this is how DNS should work. This is how identity should be working generally when you get to the internet.

Thank you.

I don't think so, and if it was, it was in the like prehistoric stone age 2015, etc.

So, there's probably an addition. I'm still focused on app layer. I do think there was a number of areas that have been experimented with in the last 10 years that now as we've gotten this new level of stability, it's almost like internet one and internet two. Like Webvan went out of business, yeah, but actually now we have Instacart and all the rest. So, I think there's a lot of things that will be coming back, and that's part of the app layer focus that we were talking about. But the other thing, and that's part of buying a CryptoPunk, is I started thinking a lot. I've been spending a huge amount of time on AI. You know, kind of co-founded a couple of AI companies including one trying to cure cancer, and I've been thinking about what AI and agents mean for everything. And one of the things I went, oh, wait a minute. This is going to be another important thing within crypto. And not like for example, agentic commerce, yes. You know, kind of payments rails, yes, etc. You know, kind of the enablement of payments within agents, yes. But actually in fact what happens is when you begin to think we're going to have more agents than people, what does the identity layer look like? What does the kind of the notion of, hey, when your agent's talking to my agent and we book this talk here, you know, what actually in fact like, you know, is it a trustable, you know, transaction? Do you have identity, etc. And that got me back into thinking about NFTs. And that's part of why I ended up buying...

Yes.

Well, so precisely one of the things that I think is interesting is if you say that the entire world of work changes because of agents and the fact that you might have agents that are like your not just your talent agent, but also and scheduling agent, but also doing various kind of predictive productive work, but we still have this kind of super agency of humans doing that, well, what does that whole identity ecosystem look like? And that identity ecosystem, while there'll be important identity ecosystems within the companies, like it's going to be kind of free-range on the internet. And how does that work? And crypto's the obvious answer. Now, you know, is it Ethereum, Solana, etc. Who knows? And maybe all in terms of all this stuff. But that started getting me thinking about it. Then of course, you know, part of the thing that was very clever about CryptoPunks NFTs was like profile pictures, online identity. And then, you know, I met the CryptoPunks founders. They're great, you know, and then I was like, okay, I should buy a CryptoPunk. I should start, you know, getting because I always think you should get your hands into it when you're trying to learn, right? And so it was like one of the ways to try to learn.

Well, I think so. And like for example, like an obvious one is this obsession flash concern with deep fakes. So, you go, okay, is that really a video of Sam or not? And is this really me given that I've also made Reid AI, and I put Reid AI to speak at various conferences and so forth. And I think that part of that provenance, part of that, okay, this is, you know, certified, and how do you have, you know, part of the reason why, for example, crypto networks is certification authority being done in a very deep way. I mean, when I bought my first Bitcoin in 2014, and by the way, I haven't sold any since then. I'm a believer in all crypto. It was like actually in fact, this is part of a design feature that this is how DNS should work. This is how identity should be working generally when you get to the internet. So, that part of it, I think, becomes all the more important in an AI age, whether it's information, whether it's questions around deep fakes, whether it's question around this is really Reid versus Reid AI, this is actually information this conversation at this conference actually happened. Right? In terms of how it's going. And all of that, crypto is the technological solution. Right? And it's the natural kind of and it's part of the reason why like the deeper identity and trust questions. Yes, of course, identity trust really critical for financial. But it's critical for many things in addition to financial.

Well, it's not against every social business model. I mean, as you know, LinkedIn, etc. And part of it is I think that it's part of the reason why one of the things that I had the earliest kind of discussions, you know, when we started, you know, kind of web two, was why actual real identity matters. Because we want to know. And by the way, there's legitimate claims for some areas where you want to have pseudonyms. But your actual identity then creates more responsibility, more reliability when you're functioning within a Western society, even though we have, you know, some problems. You know, in Western society and also generally, but like it's one of the things that I've always been a strong believer. And that's part of the reason why I think when you look at the critiques that, you know, kind of like whether it's, you know, a lot of misinformation and a bunch of other things that social networks get, LinkedIn gets a lot less of them because of that. So, I do think that it's important to bring back kind of questions around where do we need to have robust senses of identity and that those robust senses of identity then lead to trustable information. And you know, for example, you know, in markets, like if someone were to say, you know, suddenly like, 'Oh, X has happened.' Well, you actually really need to know if X has happened. Now, and I think part of the question around information is not only do we need to say is this a source of trustable information, but like for example, let's take like one thing that's frequently quoted in media like polls. Well, are these polls genuine? Are they run well? Are the people answering them? I mean, how do you know like for example, it's even as like well, not may not just be AI bot farms, but how do you know it's not like a bunch of people being paid for it? Like this was one of the funniest things that we saw on X, which is for like a small number of hours, they started posting the actual locations. And then a whole bunch of people who were tweeting MAGA influencer stuff were based in Nigeria and were doing it for economic reasons. Then of course, they took that down. Right? So, like that kind of identity really, really matters.

Well, I think it would be wise on lots of levels because as you know from our conversations, I actually think that the key question, you may have challenges to what's going on in crypto, but you should try to shape it to what is good, not just go 'Eh.' That was one of the problems with like earlier, you know, administrations and approaches is what's good about Genies, about clarity for doing this. But if the industry goes, 'Oh, we're overly reacting against Gensler, etc.' and then being kind of as it were anti-Democratic Party on this, the problem is is that the pendulum swings, right? And it's good to be bipartisan from a viewpoint of what we care about is the ecosystem. We care about how it plays a good role in society, and if you overly locate within one partisan group, like it was like, 'Oh, we got to make sure that, you know, all the MAGA Republicans are really on our side.' Well, that works for now, but then when it turns around, it's like, 'Well, that's what you were doing.' versus what you were trying to do was build a really important part of our financial ecosystem. That's what actually, in fact, we should be focused on. So, I think it's good to be, you know, especially looking at what's likely to happen in the midterms, to be much more bipartisan anyway. And I think if you look at a history of decades of American business, they try to do that because they go, 'Look, we understand this thing changes around, and stability through it, not being specifically partisan on one issue, really matters.'

Yes.

So, you know, it's a little bit like, for example, people have been now predicting for a bit that there will be no software engineer. And actually in fact that's not true, but what is true is what's going to happen is like for example, like the rote thing of learning how to use Rust, Perl, you know, C++, Objective-C, etc. All of that stuff will be easily available. Now, it's still useful to say I'm an orchestrator, like I have multiple agents running. So in giving over cognitive function, it isn't that when I have this like for example, you know, working and writing and building an app for me and everything else, it isn't that I've given over cognitive function, it's I've taken on a different role and the notion of now, oh, my unique differentiation is I know how the rules of this language work and I can compose it, now it's I'm a thinker and a strategist in directing it. And I think that's the right way to think about it. And of course, you know, one of the things that can happen is, you know, if people haven't done this, they should go do this. Go use the frontier models for something that's really seriously matters to you. And like for example, for me as an investor, I find it's really really helpful in doing competitive analysis, due diligence, etc. But if I tell it, well, should I make the investment or not, it gives me what I think of as the bad business school answer, which is a really smart business professor who can sound really good and who would lose money as a private placement tech investor because that doesn't actually understand how the game works. And so that's the reason why I'm not worried about, oh, we're giving over our cognitive facilities. Now, people can still sometimes do that. Like you can go to Google and say, you know, should I take Hydroxychloroquine as a vaccine against, you know, COVID and then do it, you're like, okay, that's a mistake, don't do that, but engage.

Yes. Well, and so, for example, one of the things that I've been using for does when does AI really hit the workforce is like customer service. Because any place where we're following a script, where it's following a script, the machine will do it a lot better. Now, it's more than that in terms of like coding and knowing how to use things. But, the real key question is how do you get that amplification? Like the real business sustaining business advantage is not going to be a cost advantage. Namely, oh, you know, you have 10 engineers and now I can do three. It's how do I use it for go to market, product development, iteration for customers, depth of tech stack. That's one of the things that people get wrong in the SaaS apocalypse. It's like, oh, sure. Now, it's much easier to build these kind of SaaS softwares. But, that means that's the new competitive minimum bar in terms of what you're doing. And so, I actually think that what we've seen so far and every company said, I'm doing layoffs because of AI, maybe other than Meta, right? Is not out of productivity, but is just out of reshifting like, oh, we've over-hired because of the pandemic. We need to change. We're going to call it AI for a position of strength, but we're not actually in fact that's actually not happening. And the sustaining one will be how do I use AI to advantage?

So in crypto, I have, you know, like for example, part of when we were talking, you know, I've actually been thinking that things in a lot and that was part of the reason I gave the apps answer when we talked 6 months ago was I've been thinking a lot about the rebirth of a set of different things that I think were tried, just like things were tried in Web 1 and now you have Web 2, that will have a new version. NFTs is one. I think, you know, DAOs in other areas, I think it's going to be another. And there's this stack of stuff that I've been doing there. Then, as part of my general focus on AI, whether it's, you know, co-founding Manifold AI, procuring cancer, other things, I've been thinking about AI as it touches everything. So, and that includes like funny fun things. Like, I don't know if you saw that I released a Pi Day music video of me singing. Right? And by the way, I'm the kind of person you would pay not to sing, but my AI does it actually in fact reasonably well. And so, that kind of like where all the areas where there's enduring value because, you know, part of being a venture investor is you look to invest in things that you would essentially kind of hold forever. That means you have a 5 to 10-year plus, you're not timing a market, you're not doing anything as part of that. And I think that part of that has gotten me back into we're going to need like what are the things when you have agents everywhere? What are the things you're going to need to have? One of them is identity. That's one of the reasons why for, you know, NFTs and crypto and I was partially thinking about that. I think that also the notion of how AI is going to change somewhat of what the nature of a firm is in terms of information flows. Well, crypto may play into that, too, in terms of information, identity, etc. And so, those are all areas that I'm looking at.

Is there such a thing as an exit price?