The Man Betting on Institutional Reform | Joe Lonsdale and 8VC

When I started Palantir 20 years ago, we were working so hard and we were recruiting all the best people, and man, it is like an order of magnitude harder now to get the best people. I was told by every VC in the valley they should be doing social media and what am I doing and government's insane to do. I think we all have the areas where we're intellectually more disciplined and we have the areas where we're very true to ourselves if you're an entrepreneur. And sometimes the passion is the only reason why you're able to be so obsessed and so eager and to make the decisions quickly and to just put the energy in to actually make it work.

Thank you, Adita. Happy to be on.

Yeah, you know, I grew up in Silicon Valley. I had friends teach me to code as a little kid, which is normal now, but that was, I guess, unique for the time. And was always way ahead in tech. And I knew a bunch of these smart people from like I was a state chess champion a couple times. I was like mental math champion. So, I had all these nerdy friends who were really good at this kind of stuff. And in the AI world, by the way, you see these nerds mattering a lot more now, too. It's very cool seeing how that talent really, really matters. And but I think it mattered back then, too, for tech. And I've always been a little bit of like the warrior archetype. I like running towards really hard problems. I like confronting things no one else wants to confront. I guess I grew up with like a Western civilization mindset. Maybe I'm very typical for someone who was born 200, 300 years ago where you teach all the young boys courage and to fight and you like study all the great heroes and conquerors and that's so maybe a little bit anachronistic but that has always been my framework so I want to run towards important problems. And I was working with Peter Thiel on a bunch of stuff and helping him with a bunch of stuff because we shared a lot of values and had a lot of things in common but I was most interested in fighting the bad guys and after 9/11 that was just such an obvious big thing to work on that I was really drawn towards going after that. You know, a bunch of my smartest friends were helping us with other things and they didn't really like the finance stuff. So, we just started helping them build Palantir with me.

I mean, it definitely came from my father. It definitely came from my parents. It came from books as well. You know, Adita, this is something where I have a really controversial view that'll probably make a lot of people not happy, but

I love it. Well, I think it's really important that we all embrace our tribes and embrace our nationalism and embrace our pride and who we are. So, if you're from Britain and you're a Christian, like the best thing to do is to be proud of Britain, proud of British history, proud of being a Christian and to fight for it. Otherwise, you're going to lose to some other tribe that's proud of itself and it's going to crush you. And so, for me, I happen to be Jewish and I happen to be from, you know, 10th generation or so American. I'm very proud of America and our values and our freedoms. I'm very proud of being Jewish and of Israel. And I have always been brought up to fight for these things and to stand for these things and to stand for Western civilization. And so I think if you don't embrace your tribe, if you embrace modern nihilism, if you embrace the global citizen thing where you're not proud of anything, then some other tribes come and they crush you and then your grandkids are no longer part of your tribe anymore. They're part of some other weird one. And so for me, it's always been like this duty to embrace. And I realize there's negative things about tribes, too, if you're not careful. But overall, it's like a really positive thing for our country for us to have tribes we're proud of that are good tribes. And that's how we see it.

It requires so much and it's so much harder. When I started Palantir 20 years ago, we were working so hard and we're recruiting all the best people and man, it is like an order of magnitude harder now to get the best people. And I'm still fighting like hand-to-hand combat. Like tonight I have guests coming over but then when the kids are in bed I have other guests coming over to work on to try to recruit people to try to partner and it's like you just have to go hard to win and that's why Silicon Valley is so awesome. I totally agree.

Um,

Yeah, it was Adapar and it was kind of OpenGov in short succession which were both correct. And there was really Adapar was more recognizing this big SaaS wave that was coming, seeing the financial crisis, realizing you could solve a lot of problems if you had the platform that everyone represented their wealth on and you could actually connect things and make things more data driven, more legit and that there were going to be these big cloud platforms and that's like a really important one to own. And took longer than I thought but it's been very successful. It has $8 trillion reported on it. It's growing fast. So, but you know, these things take a long time. It's 15 years old now. It probably could go public. It probably won't for a few years, but eventually, you know, they do that. Lots of new AI stuff on it. But then OpenGov, I think, is actually the more interesting story in some ways because I started a nonprofit, one of my first nonprofits called California Common Sense and had like 20 Stanford students and we were trying to figure out why is California spending so much more than other states. Like obviously I came from the bias of not thinking government should waste money but also acknowledging there's maybe useful things to spend money on, right? I'm not a total radical despite what people might think. Let's look at the data, right? And we started for the first time putting data from all these states online doing regressions comparing things and we found some departments were like reasonable, they're about what the regression would expect and some departments were like 50% more, 40% more, 60% more of the whole areas than you expect and those areas tended to have very strong government unions donating a lot of money. And so it was very clear to me there were special interests and it was just funny because my name wasn't even on it for a while. We had all these students publishing all this stuff and there was over a thousand articles in the press attacking them over a period of two years and so people all were hearing about it and it's kind of classic California corruption. And all these cities started saying this is really interesting do this for us too and we'd say to the city oh great send us all of your data we'll do it and the city would say oh our consultant says it'll be $100,000 to send you all the data and we're like what is going on here and we looked into it and like there's 90,000 municipalities in the US obviously only like 2,000 big ones and they didn't have access to their data. They had these consultants. They had these little crafty black and white screens or black and green screens in some cases which I thought was hilarious and just couldn't do it. And so we're like wow there needs to be a SaaS company here too. And so that was kind of the genesis of founding OpenGov to go after that problem as a mission-driven company and fix it. And very hard thing to build but we made that work till we sold it for $1.8 billion a couple years ago, but more importantly, we supported thousands of cities, thousands of government agencies and made the stuff more transparent, more effective.

Back to Common Sense. I love it.

This is 100% where the world's going. I will tell you, I think I mentioned it to you, I have like all this PTSD from trying to work with governments. They won't pay for things that aren't in the budget. It's impossible to put new things in the budget. You have to go after existing processes. It's like one of those things where as soon as you've done something, you don't ever want to do it again. However, it's like such a good thing to do. And yeah, almost died multiple times. Take a lot of resources. It's funny, you know, Facebook and Google kept stealing our best people. We're like, try to build this thing in Silicon Valley. It's a mission. You'll have like amazing people who are mission driven working there for a few years for like three or $400,000 a year. And then Facebook would be like, 'Oh, come here for $750,000.' and their spouse would be like, 'Come on, you have to do it.' I'm like, 'Oh, I'm just losing everyone. This is hard. This is a hard thing to do.' But we thank goodness at least it worked.

Yeah and you know I think we're the bull case is hopefully we're once in a generation in terms of tailwind in terms of AI kind of like providing a buying window for people to really reconsider some of their past software decisions. But one of the things I find interesting to your point about how do you convince a smart team to work on it. I think one of the most interesting byproducts of companies like Palantir and now Anduril and so on is that working for the national good or even like the federal or local good is sexy again, right? It's like

I love it. No, it's definitely the case. I mean, really, Palantir and SpaceX paved the way in the defense complex especially and then Anduril paved the way even more there. And then thanks to that, you know, I'm involved in a few new unicorns I've started in that area and a bunch of things we're working on that would not have been possible without Palantir and SpaceX and Anduril. So, all three of those did extraordinary things. I mean, you probably have heard like Palantir and SpaceX had to sue the government, had to be very persistent over many years. So unfair, but it was great that worked. I'd still say there's just like so many other areas of government that we really should be fixing with tech that no one is doing yet because it's still so hard. But you know these are the types of patriotic missions we should be on.

I think like with Palantir, I was told by every VC in the valley they should be doing social media and what am I doing and government's insane to do. So I think when you're told once that you're crazy and you kind of make it work, then you just it actually it's like a good sign. Like I'd much rather work on something that people are thinking, Joe, this is like not how I would do it. It's not what I'd focus on. Because if I'm doing something everyone else is focusing on, that sounds a lot harder than I have to compete with Mark Zuckerberg and Sam Altman and Anthropic for your engineers if I'm doing exactly what they're doing. I'd much rather do the thing they're not doing. And so I guess it's generally like maybe it has a certain kind of weird personality, but you want to do the things I'm talking to most people, including one point about Adapar, understand why it was valuable, which is in some ways that's a very helpful thing. Maybe it means I'm bad at explaining it, but if I know I can build like a 25, 30 billion dollar company in an area that these people don't even understand, it's probably as good, you know? So I guess that's my approach.

This is what I love even about Palantir still. I mean, I'm working on a dozen other things that I think qualify for this because, you know, I've built companies now. I start a few companies a year, but Palantir even today, it's like so we have these levels of AI. Level zero is energy. Level one is chips. Level two is data centers. Level three is the models. So it's like Llama and xAI and Grok and everyone. Level four is software infrastructure like Lambda for example which we've backed and all these things. Level five is actually doing the services themselves. And so I'm doing all these things in level five. But there's also things and the question is does level three eat level four? Does it just replace the infrastructure itself? Which in some cases it might, right? This is the whole dynamic. And there's things like Palantir, which is just so funny to me because we're doing so much in AI there that's working and it's valuable and it's growing fast. It's the highest valued company by multiple on the S&P 500 in terms of its prospects, which you know, whatever. It's a bit of a meme stock, but it's a great company. Yeah.

And I maybe they want me talking about this, but it's amazing because it's like that is such a good sign that none of them are competing for it, nor even understand how to compete for it. They wouldn't even know where to start. They don't have the conceptual frameworks. There's like many layers of conceptual frameworks there for why this works. And I think this is true for a lot of other areas. Now, it doesn't mean it's going to be the very biggest thing. Like maybe Anthropic and xAI are competing for the biggest thing. That's fair. But there's like lots of things worth hundreds of billions of dollars these guys I would contend don't even understand which is really fun. Yeah.

Well, there's a lot of conceptual layers of abstraction of things they're doing and have done and problems they've solved that people would have to hire big teams, spend years solving. And so, and this is I mean I love this. This is what I'm trying to do in so many areas right now is what are the problems that are really valuable that are like culturally unintuitive for people. Like I'll give you another example which is I'm obsessed with right now which is this company Thorne I'm working on. And at a very simple level, it sounds like other things because I'm trying to make how I do my email more productive and how I do my comments more productive. And there's lots of people trying to work on that problem one-off for people. But the thing that's unique about this is we're doing it for a leader with their team and for processes that exist in their organization. And so an ontology of processes is a very Palantir-esque way of thinking about it where anything that comes in gets categorized ontologically and then kicks off processes and you need reports on those processes and the right people checking it and the people are working with the AI to help the leader take things off their plate. And what I've found from Palantir and many other companies is that engineers are like deeply anti-aristocratic. They're deeply egalitarian and they hate technology that's specifically designed for a leader or for a senior person. They just refuse to think of the world that way. And so therefore, the space is entirely open and there's so many things you could do in AI there right now. So I'm really excited. We have a bunch of great people working on this where I'm just getting baited for my own email. I'm going to show you at some point in the coming months. But it's just really fun, right, to have an area that's just totally unintuitive like no one's going to compete with me because you have to actually struggle with the engineer to even get them to see reality this way. This is an example.