Productivity, Innovation, and the New American Golden Age with Joe Lonsdale

>> Jon Hartley: Welcome everyone, especially those who are new here. Really great to see so many new faces. I'm really, really excited, for those who don't know me again, my name is Jon Hartley. I'm host of the Capitalism and Freedom, the 21st Century podcast. It's an official podcast of the Hoover Institution Economic Policy Working Group.

For those who are not familiar with our distinguished guest today, Joe Lonsdale. Joe is one of the world's leading innovators in venture capitalists. And to say the least, it's really unbelievable all the sorts of things that Joe has started. I mean, he's the co founder of Palantir, founder of 8VC, a venture capital firm.

He's also the founder of the University of Austin and is trying to really disrupt education. He's also the founder and chair of the Cicero Institute, which is a think tank that's focused on state policy, and it's based in Austin. Joe also has his own podcast, which is called the American Optimist Podcast as well, and definitely recommend that you guys check that out.

Welcome, Joe. How are you?

>> Joe Lonsdale: Thank you, Jon, for having me. Thanks, Francisco. It's nice to see all of you, I'm great. I'm here in Miami for a day and glad to come by.

>> Jon Hartley: Awesome, well, we're so excited that you decided to stop by and talk to us here.

I want to start with the 2024 election and the Trump transition. I think it's definitely something that's on a lot of people's minds right now. I mean, what are you most excited about when hearing about all these various nominees and Elon Vivek starting Doge, and how are you helping, and how are you planning to get involved in the next four years?

>> Joe Lonsdale: Well, it's definitely a very exciting time. It's a huge weight off my shoulders for our civilization. I was very, very worried where things were heading. And this is a much, even though, even though it ended the way it did, is election was much too close, given the implications for our civilization, it was very scary for me.

And you know, I'd say I'm probably most excited about the fact that, that we have an opportunity with like this vibe shift in our civilization to push back on a lot of the illiberalism. To push back on a lot of the authoritarianism, to clean up a lot of this, like massive corruption with hundreds of billions of dollars going to NGOs that are run by activists that are not accountable, and that are pushing the wrong ideas that push our civilization the wrong way.

And I think I was saying outside, I was exchanging notes with Mark Rowan earlier, who, you know, someone I met, someone who runs School of Finance, obviously, but someone who's very involved, in advising Trump on these things. And I think he's right, I think if we make the right cuts and when you model it out, I think you'd have enough growth that you actually can get the deficit back to below zero and pay it off.

Especially that plus Doge. And the thing I guess I'm most excited about, obviously active things going on, of course, is Doge. I'm friends with Elon and Vivek for a long time. And this is awesome, this is stuff that I've been posting about, and writing about and obsessed about.

Just like this broken thing that's just been this cancer, it's been growing out of control and everyone's been ignoring. And I feel like a crazy person. Like, this is this cancer, we have to get rid of it. And everyone's like, well, yeah, about, we're busy with our everyday lives and we don't want to talk about it.

And finally all these smart people are like, yeah, we have to get rid of it. And this is like, awesome, let's do it, there's like, we can talk a lot more about that if you want. There's just all this stuff to fix in government. It's just, just incentives are misaligned, things are unaccountable, things are being stolen, things are, are, are being wasted, things being managed terribly.

It's just like, yes, let's finally fix this and it helps everyone.

>> Jon Hartley: No, it's amazing. Yeah, we'll see, I guess, you know, there's still a lot of uncertainty, I guess, with how, how Doge is going to work, and I think Elon, Vivek are still figuring it out. I mean, do you think that the best way to go at it is, I guess the bureaucracy that really sort of fights against change, and you can't really fire public employees has been the case since the 70s, is that?

>> Joe Lonsdale: Yeah, I mean, that's one really important thing. Solicit, so it's both about cutting wasteful spend, it's about cutting unnecessary and wasteful bureaucracy, and it's about cutting unnecessary regulations. And all, all of those are related and all those are very important. So I think one of the issues that I wish more people would talk about when they talk about that is that there's really two things that happened in the late 70s.

First of all, it was actually it Wasn't even Jimmy Carter himself. We don't remember this, but Jimmy Carter was actually famous in Georgia for consolidating a bunch of bureaucracy. I don't think he was a great president, but he was famous for consolidating bureaucrac. And he ran on wanting to cut red tape from the left and saying, listen, I agree with the right.

This is bad, his Congress was completely captured by the unions. The Democratic Party was completely captured by the unions in the late 70s, and they're the ones that passed in 7980, these, like, crazy protections that make it impossible to fire anyone. So that's one of the problems that happened to our government, and that was really bad.

And then the other thing that happened in the late 70s was the activist courts. And so, the activist courts basically said, so let's stop for a second. So you go back to 1883, you had the Pendleton Act. And so what happened in the US Is that we had a very small government.

Civil war happens, we start growing the government massively create all these new departments. Just it just if you graph it out, the government grows a lot from that point on. There's a lot of federal power created maybe for good purposes, but, you know, it's something that I don't love, obviously, having lots of extra federal power.

And by the early 1880s, both sides, everyone agreed this is a mess. People are coming in, they're hiring 10,000 friends, our government's running, run by cronies is totally a broken thing. So there's a Pendleton act, which put in some basic protections, which, by the way, are fine. I'm fine with some protections for civil service that aren't crazy.

And it put in tests. So from 1883 for about 100 years, there was like a really hard test if you want to run something in government. So, when we went to the moon, when we fought World War I and World War II, when we did some things that frankly, were pretty competent in our government that we should be proud of in our civilization.

You know, in the mid to early 20th century, the people running those things had to pass hard tests to be running those government departments. Come to the late 1970s, these courts run by crazy activist people on the left basically said, these tests are racists. And then, they said, bo, but the tests actually predict performance, that's not racist.

They said, well, no, just by definition of the fact that different races have different outcomes on these tests, they're racist. You must get rid of them. And so we completely Got rid of the tests. And so, so, so not only can you not fire people who. Which makes them a lot lazier, by the way, in any business, in any organization, if you can't hold them accountable easily, you also can't give tests to make sure you're hiring the smartest people.

So, our government, you know, and by the way, was pretty smart, I think in this, in the 60s, 70s. I didn't love big government, but it was at least like not completely dumb. It's just gotten stupider and stupider and stupider for 30 years. And then, you got the WOKE stuff come in the mid 2010s, and you prioritize identity politics, and then it accelerates its stupidness, you know, for the next 10 years.

So just, so now it's just like completely retarded, it's a broken bureaucracy.

>> Jon Hartley: Well, shifting to generative AI, we'll get back to some of that later.

>> Joe Lonsdale: We're allowed to say the word retarded now, by the way.

>> Jon Hartley: We'll shift back, so generative AI, you know, obviously a hot topic right now in Silicon Valley. How in your mind should VCs be best positioned?

>> Joe Lonsdale: So, I think that, so when you're looking at AI, to see shifting topics a bit here. I was going to talk more about Doge, but that's probably, she got scared of my passion there. We look at AI, there's really, I think, five ways people are investing in AI at a high level.

There's the bottom level, which is the chip companies, there's the next one, which is the data centers. Pretty much all my friends, family, offices who made Money the last 20 years in tech have been putting lots of money in data centers. It's become a very popular trade, which is great because you guys are funding stuff for me to build on top of. 

So I appreciate that, keep funding data centers. I don't know if it's going to work with this much money, but that's great because it's cheap for me. And then the level three is the model companies, which is like OpenAI, of course, famously, which besides this for profit company now too, and XAI and Anthropic and everything.

And then level four is software infrastructure. This is how you deploy AI. And that's obviously there's a ton of, I think San Francisco is probably the center of that. There's a ton of very, very powerful software infrastructure companies we're invested for example, in Ollama, which like over a million engineers use to deploy their llama models.

There's a lot of good stuff there that makes sense. Palantir does a lot of stuff level four. And then there's level five, which is actually this like applying the AI to services itself. Like actually the one doing the health care billing, doing the logistics billing, doing the new legal company, you know, doing the new back end of the insurance or trust or estate or whatever it is.

And so I'm obviously invested in stuff like I'm proud to be invested in stuff that like, if Elon starts a company in level 3X, that's great, I want to be involved. Matt, obviously impalance here and lots of level four things, but most of what I'm building and doing is level five right now.

And the way to think about level five is there's probably about $5 trillion right now paid out in wages in the services economy in the US and about 2 of that 5 trillion, about 40% is in areas of the service economy that we've already shown we could double the productivity.

This is very important. This is not like maybe when models get better. This is today based on things we're seeing. We've already shown these companies can work at least twice as well. And this is things like, I have a health care billing company that's doing a bunch of stuff that we're big investors in and it's already got, I think last quarter was like 56% margins and like most healthcare billing companies are between 15 and I think 25% margins, right?

So it's already more than twice or three times as productive of all of our companies thanks to not just sprinkling a little bit of AI, but thanks to building a lot of stuff and applying AI properly. And so this is a huge fact because if, even if AI gets no better from here, just by scaling these things, you should pull like a trillion dollars of spend out of the economy.

And I think it gets better enough there's probably a trillion and a half to 2 trillion. And you're gonna pull out the economy in the next 10 years. That's a massive, massive productivity boost, and you're gonna make a lot of money. We're going to make America a lot richer.

>> Jon Hartley: So I think a lot of the big generative AI winners so far have been really the large firms, the OpenAI's Anthropics, Nvidia, Google and so forth. The ones that can lay down the fixed costs to actually build the large.

>> Joe Lonsdale: Well, they bet they've been the level one, two, three, four firms mostly. And then level five is like just started to scale still. That's right.

>> Jon Hartley: Really. And I mean, do you think that it's possible for AI upstarts to, to really sort of latch on and without getting sort of outcompeted by one of the big guys? Like, I feel like I remember there's one startup that, you know, created this video, you know, generative AI model and then, you know, raised $3 million.

And then the next day OpenAI builds essentially the same product and releases it.

>> Joe Lonsdale: Totally, totally. I think that's very, very hard. So there's answer this in two ways. First of all, let's just think about the last 15 years of tech to put it in context. So after Palantir started Add a Par, which is a leading wealth management technology company.

It's a giant SaaS company. We sold OpenGov this year, which we built. And invest in a bunch of MY friends built a lot of SaaS companies. and the SaaS companies that my friends built in, like the early 2010s had a very high success rate. And I could say it's because I'm some kind of genius or something, but what was actually happening with these is, is the cloud was newly possible.

There's new things you do in big data. And our competition was stuff mostly from the 1980s, 1990s. So you can kind of be like, like we have access to the best tech cultures in the world. That's my background. So it's kind of like being a bully. It's like we're taking the very best tech guys in the world.

We're going and beating up on these things from poor things from 10 or 20 years ago. And by the way, it's good for the economy to do this. It's not like a morally incorrect thing, but you can kind of think of it, yeah, you're being a bully versus these These always doing things.

Now if you wanna start a SaaS company, if anyone in here wants to start something and compete against something in one of the major areas of SaaS in cloud, you're mostly probably competing against the stuff my friends and I and others built in the early 2010s. And that's a lot harder because there's not going to be as big of a gap there.

And by the way, all this AI stuff, my, our companies are going to use it too. So these 10, 15 year old SaaS companies, they're also using AI. Like good luck trying to start a new AI SaaS company against them. They also have 300 million of revenue and a thousand people and they're already able to customers.

So that's really hard. So that's a dumb thing to do. That's really hard. So the question is, what is there to do? So you go back to those five levels of AI, a new chip company, that sounds really hard. It's probably possible for some, a couple EE geniuses to figure something out new but that's really tough.

It's very expensive to do these things. The data center is fine, everyone's gonna invest in that. I'd like you to do that for me. Like level three is like, like the only person I would have ever backed with a new one of those was Elon. That's great cuz Elon is Elon. So of course they're gonna back Xai even though it was a ridiculous price when he started. But whatever, it's Elon and then he has the top talent, he's probably gonna figure it out. But yeah, I would not suggest trying to do that. So the question is level four, Level five.

Level four is really hard. And the problem of level four is that there will be some things that win, there will be some things that get bought. But it's like the smartest people in the world are all doing it. And you're right, what you just said. Level three is trying to eat level four.

All these level three companies, they just want to own level four themselves. And so it's really hard. Like how do you do something that OpenAI or X or anthropic or somebody's not just gonna eat, I'm just gonna do themselves. It is possible, but it's tough. And you better be really the best in the world and you better be really running fast.

And it's not even clear to me yet what has to escape velocity and what doesn't. So that's a very Fair question. Level five, I think is by far the best risk reward place to build because you have a lot of other things we're doing there. And going back to the model I just gave you of those SaaS companies that were beating out from 20 years ago.

When I'm starting a Level 5 company right now, my competition is these guys from the 1970s, 1980s, 1990s. We're literally competing against Logistics Biller in Baton Rouge for a thing we just won, which is in there. Sure, they're great people, they've been doing this for 50 years, but they just don't have the same ability to build and apply tech the way we do and to service the customer the way we do.

And so if you can build a great operational culture, you can compete and win there. And I do think there's moats and barriers to entry at having great operational cultures that build and apply AI. So I think level five is a great place to build. I think level four for a few geniuses is going to be great.

But man, that's hard. I wouldn't want to try to do that.

>> Jon Hartley: Definitely not something I could do, that's for sure. So outside of AI, what other sorts of startups are you most excited about right now? Is crypto dead? Anything in fintech? Interesting other places that, that you're looking at? Obviously, Jenna-

>> Joe Lonsdale: The thing I've done the most in these areas are built a bunch of SaaS companies. The other thing we've done the most probably and uniquely is defense, right? So there's no, I've been lucky to start six different unicorns. Three of them are in defense right now and, and I'm working on a couple more there.

And the reason this is really exciting is, is a Palantir was like the worst thing to build in the sense that it just. The government was not fair, was not letting a new competition. We had to be so persistent. We had to keep trying for years and years and years.

Even when we won, they would like use the thing that cost 10 times as much sometimes. It's a very frustrating company. But I think the path that was paved by Palantir, by SpaceX and now by companies like Anduril, which was like started eight years ago and has grown really quickly to create these new primes. 

Has created an awareness with the generals, with the admirals, with the people in the House and Senate Armed Services Committee that the old primes, cuz all the old primes consolidated after the Cold War ended in the 1990s and they became completely dominant, but they became very bureaucratic and slow, and they lost a lot of talent.

And they're not as dynamic. And there's still great things some of the primes do. But over overall, there you go. That no problem. Francisco's trying to get me here. Overall, there's just a lot of need for new possibilities. And war for warfare has completely changed. We can talk more about that if you want.

And so. So I'm building a lot there right now.

>> Jon Hartley: Awesome, all right. We're going to talk a little bit about defense tech in in just a minute here, but want to first talk about. We're in Miami. I want to first talk a little bit about San Francisco versus Austin versus Miami.

Joe Joe, you were one of the first-

>> Joe Lonsdale: Dangerous conversation there someone who lives in Texas.

>> Jon Hartley: We'll see if we can maybe convert you here.

>> Joe Lonsdale: The people here may be better looking actually I'm not sure if I'm supposed to say that.

>> Jon Hartley: That's high praise, very high praise.

You were one of the first people to leave San Francisco for Austin. And some other people like Peter Thiel and others from Founders Fund have resettled here in Miami. I mean we're about, I think five years into that experiment. I mean do you see that something that's continuing or is a permanent move or do you see there being a bit of a reversion back to SF amidst this AI boom?

>> Joe Lonsdale: So I originally moved 60% of my firm to Austin and I think now it's about half of my firm in Austin. So we did have to move a few people to be back participating in the. I have to admit there are some things happening in San Francisco that are unique to the entire nation and that are based there.

That said, there are also some things happening in Austin, in Texas and other states like that that are unique in the sense that like you know, the Tesla manufacturing plant that Omi built with Elon took 18 months. The biggest plant in the world. We wouldn't have got, they would have got that permitted in like five years or 10 years in California, who knows.

And we're going to build 500 ships next year with one of my defense companies in Austin. El Segundo has some great talent. I just don't think we could have scaled nearly as quickly and again had the state kind of work and help us move obstacles out of the way for dumb rules in order to help us scale. 

And so there's some things that have to happen in Austin and have to happen in states that are kind of more aggressive about helping us with these things. Miami seems to be attracting a lot of amazing talent. I was saying earlier, I used to come here like every year or two.

I think now I'm here four or five times a year. There's amazing people here to see. There's a lot of the best investors. I think one difference, you know. So I think there's lots of things we can talk about where Florida is run very, very well. I think you guys have like the best legislature because everyone terms out after eight years.

And so you get the Speakers of the House and leaders of those senate here that are really aggressive their last two years they get put in and they just do amazing work. So I love your guys government here. One of the differences is there's probably not as many engineers or engineering culture.

So Austin has a top engineering school, has a history of, for 70 years of like lots of big companies in tech world being built there for tough engineering. And so, so there is more. It is easier I think. I think having it makes more sense that Elon's there and people like that are there because we're building lots of things with engineers.

I think if I was want to be around like top investment minds just investing, I think this would be a really compelling place and Absolutely.

>> Jon Hartley: Well, if there's anyone from the University of Miami here or from other schools, I know that there's a lot of great efforts going on there as well to really invest.

>> Joe Lonsdale: It's amazing how these network effects work where like even though Austin's had great stuff for a keeps getting better. It's just so hard to compete with the depth of Silicon Valley's network and engineering. And then similarly Austin hat does have 67 years on Miami. So I think Miami's going to build it up.

But it takes a generation sometimes to get that depth of engineering. But you definitely see a move from Chicago, like you said, like I said, from California, from New York, both to Austin and Miami and you got just so many great people. It's a really fun time to live in Austin, as I think it is here too from what I'm seeing is you just get like so many cool people constantly coming to town, constantly joining.

I was working out and my friend has a billion dollar consumer company, built an awesome headquarters with an awesome gym right near my house. And like just like the guy who created Solana just happened to be there like this week. He just moved to town. I mean, there's just all sorts of amazing people coming through Austin, it's really cool.

>> Jon Hartley: That's awesome. Well, you're also the founder and chair of the Cicero Institute. And it's one of the most powerful groups in the country when it comes to creating and passing legislation in states.

>> Joe Lonsdale: That's how I know Florida is so great.

>> Jon Hartley: Yeah, absolutely. And you've got teams I think in about 20 states including Florida.

What are some of the big recent successes there in your mind? I know school choice and universe licensing recognition are sort of two big ones, but what are the big wins in your mind?

>> Joe Lonsdale: Well, school choice is big, I'm a big fan of it. We tend to work on orthogonal things that no one else is doing.

So our, our framework is we're fighting special interests, we're fighting things that are broken, and we're trying to fix stuff that's broken by aligning incentives. And so I'll give you a few examples really quick. I know I'm obsessed with this policy stuff. I know most people maybe want to talk about other things, but just a few quick examples.

Vocational schools is something I think really matters. We used to have a lot more better vocational schools by percentage of people in the 60s and 70s. And instead we started sending them to get studies degrees and whatever other degrees at universities that basically lead them to become protesters because they can't get jobs.

And it doesn't make any sense at all to me. We should actually train them and skills to get them good jobs. And so that's what vocational schools do. The problem is people who run vocational schools if they're not held accountable, they also start to get ridiculous, they also start to get broken.

And so if you ask an average politician about this, like, yeah, I want to fund vocational schools and it stops there. But if the vocational school is run by an idiot, what do you do? And so here's what you do. If you have 20 vocational schools in a state, or 27 high end ones like Texas does, what you do is you say we're going to fund the vocational schools, proportion to the average salary is coming out of them.

And you measure, you give them a number which is their average salary over three years and then you give them more money if they're doing well and less if they're doing worse. And you tie their entire budget to that. And what this does is it means the school's incentive is to say, okay, what skills do we teach to get people good jobs, what businesses we partner with?

And guess what, having done this in multiple states, you could double the average salary coming out of these places. Some of them will do worse and get less, some will do better and get more. It just completely changes the whole culture of it. And you take that idea, by the way, and applies to lots of things, let's do another one is prisons.

Prisons are very controversial right now. If you're on the left, you're worried they're trying to make money with for profit prisons. And that's the root of the evil. It turns out only 10% of prisons are for profit and they have just as bad results as a nonprofit prison.

So it actually turns out they're all actually bad. I think if you're on the, if you're on the right, you're worried people are not being put in prison enough, which I think is very fair. And on the left, maybe there's more concerns. I do think some people have been punished for a very long time and are very unlikely to commit a crime again.

There's no reason to keep punishing them. I think both sides have some points here that are legitimate. But here's the thing no one's talking about is we're not holding the people running the prisons accountable to anything. These prisons have terrible cultures. They have no useful goals. It's a total mess.

Let's say we get together with ten of your smartest friends on the left and the right, what can we all agree on? We don't want people coming out of prison to go back in. We want lower recidivism, and we don't want people to come out and be unemployed.

We want people to actually come out with skills to get jobs. Those are two obvious things. So what we should be doing is we should be measuring these things. You should be rewarding the people running it, doing it well, reward the programs that are working and defunding the things that are not working.

And as best I could tell, the best politicians ever get to is just like coming up with their own one favorite pet idea for the prison and giving it to them. Most of them, unfortunately, aren't smart enough to think of the system level. Let's create a system that competes and that measures things and iterates.

So that's the way markets work. And so it turns out if you apply this system, by the way, to probation, to parole, to prisons, you can massively reduce recidivism, massively increase employment, like fix these things. And so that's the kind of stuff that we're working on. And so we're trying to do this in a bunch of different areas and doing a lot of different things with markets and healthcare and fixing education.

But just trying to put common sense and incentives and accountability. And by the way, the moderate left tends to agree with me, the right tends to agree with me, the far left hates me. And the far left is captured by the government unions. And they do not want any accountability, any measurement, because that is totally against how the far left works.

And so. So what we need here is we need people, both on the right, but also the moderate left, to say, listen, we're going to stand up to these people. We're going to demand there is going to be accountability in our society because our government's a huge part of our society now.

We have to hold it accountable to results that we all want. And that takes courage. It especially takes courage on the moderate left. It's mostly not there in our society right now, but I'm hoping with this vibe shift, we're going to get more of them to come along.

>> Jon Hartley: So Chalonsdale disrupted prison warden could maybe be on the horizon.

>> Joe Lonsdale: Okay, you joke. It's a big deal, man. Making these systems work. It impacts hundreds of thousands of lives. And it's an ethical, it's like something we Is our duty. We should be doing these things like just allowing like we all know these places are full of like prison guards who hate the prisoners, prisoners who ate the prison guards.

Everyone's miserable, it's a total mess. It's a broken culture. And like leadership is about not allowing those broken cultures, about going in and fixing them. And so everywhere in our society there should be no room for broken cultures that aren't getting results.

>> Jon Hartley: Absolutely. Yeah, very important issue and thank you for your leadership in that space.

I want to shift to defense tech Palantir. Obviously the world's become I think a more dangerous place in the past few years. Russia invading Ukraine, Iran, Hezbollah, Hamas, China threatening Taiwan. I mean the Assad regime has fallen in Syria. It's kind of unclear what's next. Joe, I mean you're one of the co founders of Palantir.

It's really the leading firm in the defense tech space. And you also have a lot of other very, very serious commitments in the defense tech space to, to other, other firms as well. I mean can you explain the audience what Palantir does and how and how it's a force for good in the, in the world.

>> Joe Lonsdale: So we started Palantir 21 years ago. I think a lot of you were mentioning you loved Alex Karp's clip at the Reagan Library this weekend. He's very proud to be his co founder. He's an amazing guy. Someone mentioned his hair keeps getting crazier as he gets richer.

So quit he's a character. He's a PhD too. He was he not only is a PhD, he was the star student of the top philosopher in the world. So he's a. He's definitely, he's definitely was the only non technical person who was full time on the co founding team.

>> Jon Hartley: Like a Habermas student at one point.

>> Joe Lonsdale: Yeah, exactly. He was. Alex is a genius. He's taught me a lot about people and institutions in the world. It turns out you need kind of these philosopher builder types for these things to work. And he had, he had a big influence on me on that.

You can't just have the best tech. You also have to understand the world and be interested and curious about how the world works. Would which Alex is Palantir took some of the very, really some. One of the very top tech cultures in Silicon Valley we created it was ranked number one for quite a while and applied it to areas of our society that don't have functional tech cultures or.

And didn't have functional cultures for a very long time and solved some very important problems, helped save the government billions of dollars, helped eliminate several thousand terrorists, helped block dozens of attacks, and then help do that for about 40 of our allied countries. And it turns out the hard, technical problems we solved by creating these ontologies and bringing the data and the workflows together in certain ways to build on top of actually are very useful for AI.

So not only did we do something I think that was very important for the world and built like a, you know, when I was there, like a unicorn, I think in some ways we got a little bit lucky because we have the best tech culture. We had this thing we built that.

It turns out it's very useful for AI. So we're kind of in the right place at the right time with that and then seize the advantage and built on it. And so they're now going around and teaching hundreds of the top corporations how to do things in AI.

I think one of the things that you show up and these people have all been spending lots of time, money on this last two years. And so the IT guys are sometimes terrified of Palantir because you come in and say, give us one week, and in one week we're going to do something that's much more useful to you and impressive than the last 18 months.

You just spent time on it. And we consistently do that with the pilots. And then you win it and then you grow. That's why Palantir is growing really fast. So it's in a really good position right now. But I think that the moral kind of background or the mission for Palantir was after 9, 11.

Let's make sure that does not happen again. And you know it's not you can't only credit Palantir with Apple. I think we definitely played a critical role in making that happen.

>> Jon Hartley: That's amazing. Yeah. And I know, I just. There's always these crazy libertarians, like, hate the TSA and stuff like that.

>> Joe Lonsdale: I hate the tsa. I'm also a crazy libertarian, though Kelt gone.

>> Jon Hartley: You want to keep the TSA though, maybe.

>> Joe Lonsdale: No, no. It's like security theater. What the hell are they doing? This is a big waste of everyone's time. Give the airport back the control and then hold them accountable. I don't know. Keep going. What do you want to say?

>> Jon Hartley: Well, well, here instead of tsa, I'd be less concerned maybe, but it's like such a waste of everyone's time and money and they don't.

>> Joe Lonsdale: Whenever we test them, they always fail.

>> Jon Hartley: Do they?

>> Joe Lonsdale: Yes.

It's like really easy to think things past the tsa. You should not try this yourself. But it is, sorry, you're gonna ask a question.

>> Jon Hartley: We won't give anyone any ideas here, but moving on. University of Austin or you know what, any other just, you're such an expert on defense tech. Do you want to talk about the other defense.

>> Joe Lonsdale: Yeah, I think, just listen what they do listen. So, yeah, we mentioned this earlier. Warfare is fundamentally changed. If you want to fight a war today or in five years, it just does not look anything like what the wars of the past look like.

You know, it's very sadly In World War I, you have this kind of maligned aristocratic class that had been like kind of beaten by the progressives for a generation. It was finally failing it. Really excited that it got to prove itself because the whole purpose of the aristocracy in, in the Great Britain and in the past was, was to be the warrior class and to defend the civilization.

And so these like young men got on their horses and they took their swords and they. And they charged the front and they were all mowed down. It was very sad because warfare just totally, completely changed. It was a huge mess and there's a lot of things like that today that are, they're analogous to that.

You. I mean, I have to be careful what I say because I hang out with all the guys who run these things. They'll get pissed off at me. But there's a lot of stuff we're doing now that is equivalent to that. And what warfare should look like today is you did mass swarms of autonomous weaponized, weaponized things.

Whether they're drones, whether we're overwater vehicles, which we're building with Saronic. We're gonna build 500 ships in Austin next year to build these very sophisticated ships that can coordinate with the fleet and attack together and separately. And you know, they're weaponized different ways. Whether it's underwater stuff like the dive stuff that the underworld is building, where you're building 200 from a year out of Rhode island right now, hopefully going to accelerate that.

Whether it's Overland, which is, you know, very best self driving in the country, winning all the awards. Rather than sending 30 tanks to Ukraine, which 21 are destroyed, you could send for the same cost 30,000 of these things that are intelligently swarming and weaponized. There's all that stuff.

And then there's things like Epirus, which is a company actually proud to have Helped start. We kind of anticipated all the drone stuff seven years ago, right after Anduril started, and way back that we backed, we backed, we started Epirus. And it's the best. It's like a force field.

It's the best directed EMP in the world. So you can shoot microwave radiation in a cone and you could turn off like a hundred drones at once, or you could turn off a very powerful large Iranian drone, which I don't think these are Iranian drones over New Jersey, but we could turn those off too, if someone wanted. 

And, you know, so there's things, there's things like that you could do. They're really critical to the future of warfare, basically. Like, it's,kind of like Star Trek. You do need to be able, if you're fighting swarms of these things on land, in the air, in the ocean, you also need to be able to turn them off and fry them, you know, in a distance.

And so there's all sorts of these new things. They're just critical to how war is going to be fought. And it's critical to get our manufacturing base up. So this is I am very pro liberty. I'm very against most industrial policy. But we do need industrial policy from an economic, from a defense perspective, otherwise we're gonna lose to China if we ever have to fight them.

We have to be able to produce more of this stuff. So this is all really critical. I'm very against the adventures of wasting trillions of dollars in Afghanistan. And I'm very aligned with the kind of moderate right, kind of populist right line that we basically don't want to be wasting money overseas.

At the same time, you know, you don't want Islamists to be able to run wild or you don't want a country with nukes taken over by bad guys, because that's going to come back and hurt us. There does need to be a very strong deterrence in the world and very strong.

If America stays the strongest, it does keep us safer. And that's something we need to do in an efficient way.

>> Jon Hartley: No doubt. And pager warfare as well.

>> Joe Lonsdale: Pager warfare is more Israel's thing. But they did a great job.

>> Jon Hartley: Yeah, absolutely. Well, it's fascinating. And also want to talk just a little bit about.

And thank you for. I think it certainly should be said, thank you for lending your great mind and bringing great minds to solve these hugely important problems. And certainly, I think there's, I guess, a changing view, certainly on. Industrial policy and, and rethinking trade when it comes to how to deal with these sort of non economic objectives.

I wanna talk a little bit about just, I guess, do you have any thoughts on fintech? You know you started out of par, any fintech thoughts you wanna add?

>> Joe Lonsdale: You know it's, it seems like there's going to be, there's going to be a need for, for better ways of using stable coins to transact for free.

I think you guys probably saw it came out how the SEC, not even the SEC so much as like kind of leftist senators like Sherrod Brown, others kill, killed Libra and killed a lot of other attempts from large companies to use crypto to create kind of new different challenges and with identity and payment online, which really is the obvious long term answer.

I think it's as much as I don't agree with a lot of what goes on in meta, I think it's actually really bad for the world that the, that they kind of, it kind of used mafia like tactics to threaten Facebook not to launch this. It was legal to launch it, but they're like if anyone works with Facebook and launching this we're gonna look into everything you're doing.

It was a very mafia thing they did that came out and so I think that was really wrong. I think it would be really good for this to happen on the consumer side. On the wealth tech side, add a part just crossed $7 trillion this month. I'm very proud of the company, $7 trillion that reported through the platform.

It's a very dominant company, I think about 1,200 firms using it. And that's gonna keep growing. There's a lot of really interesting stuff we're doing with creating new ways that any RA could offer in aligned way really smart alternatives to their investors. And so I do think wealth tech's going to be smarter, more dynamic.

There's just a lot of fun stuff we're building there. I think the estate stuff's going to be handled way better. You shouldn't have to know the smartest lawyer in New York. You should be able to you feel to automate whatever the Jewish guy is telling you from New York to do for everyone.

It's for their taxes instead of having to all pay, you know I should just have to pay $100,000. Everyone should get access to it. So there's, there's things, there's things, there's things like this that are that are coming, which are just like smarter and better with AI that I'm very bullish on.

But, you know, just, just like everywhere else, there's just a lot to build there, a lot to keep fixing.

>> Jon Hartley: There's a lot of good accounts and finance guys in Miami too for that's worht, not just New York. A lot of our New Yorkers are-

>> Joe Lonsdale: He moved to Miami, that's true here.

>> Jon Hartley: Just saying. Just saying. Why don't you talk a little bit about University of Austin. I mean, you're really the chief benefactor, I think, of the University of Austin. The first class just began this autumn and I think just finished their first semester. How is that going so far in your mind?

>> Joe Lonsdale: That's great. We have 90 something students there. In the first undergraduate class, a lot of students turn down very top schools, have very top scores that came in. They wanna go somewhere, I think that in some ways more patriotic. But it's a place where you can pursue truth.

It's a place where you're taught to be, you know, the philosopher builders that we want to see in our world that are going to step up and fix things. I think there's all these classical virtues which are very important, which we don't talk enough about. And my favorite classical virtue is courage, and you're basically taught the opposite of courage.

Now if you go to Yale or Harvard or Stanford, it's basically like, shut up, go along, don't, don't try to speak out. Echo what you're told. If you speak out, if you're virtue signaling the thing we've told you to say. But if you disagree with us, don't speak out because you're just gonna get punished and you're gonna get hurt for it.

And that is like the worst possible message that we could be teaching our youth. You should go to college to have debates and to speak up, but then at the same time also, by the way, have the intellectual humility that you don't for sure know all the exact right answers.

And that you're gonna listen to the other side, you're gonna learn that you're gonna be able to steal man their arguments. Basically, almost no one on the left who comes out of Harvard or Yale knows how to stealman smart people on the right. They literally have been not allowed to listen to that, which is pathetic.

Like John Stuart Mill, one of the great kind of, you know, classical liberals who shaped all of this, that, you know, if you don't know, if you, if you don't know the other argument, then you don't actually know your argument. And it's just a dysfunctional set of affairs that's bad for our civilization.

We live in a liberal civilization that is based on these open debates and it's based on frankly understanding the conversations of the last thousand years, especially the last 300 years with the Enlightenment, which are also not being taught at these schools. And so there's just a lot of stuff to fix.

And, you know, University of Austin, I think our goal is for about half the kids to be stem. I'd say it's about one third STEM right now because we are starting really deeply with the kind of history, economics, philosophy, and great idea stuff. But, but the, the idea is to have both great thinkers come out of there who are leaders in policy and politics and other parts of society, but also to have STEM leaders come out of there, but STEM leaders who are more well rounded and who are going to fight, you know, speak up and build cultures that are healthy.

>> Jon Hartley: That's great. I mean, how would you describe the University of Austin? I guess just like a few sentences. And like, how would it be different from, I guess, say like Hillsdale or, I know like Thomas Aquinas College makes people actually read. Like, listen, I think Hillsdale is an important institution.

>> Joe Lonsdale: I don't want to throw any shade at it. It's different in the sense that Hillsdale is explicitly a conservative institution. I don't know what cancel culture goes on there against the left or not. But I, but I, but I, I do know University of Austin does have people who, there's people on the moderate left, there's people on the right, there's all sorts of views.

I think University of Austin has scores from students coming in that are similar to the other top 30 universities. So our goal is explicitly to compete with Harvard and Stanford and Yale and those, and build like the next great university that's based on more of these traditional values.

And I guess the other thing is different is we have a talent network with over 100 people in my community who have started and built the unicorns and built the top tech companies who are engaging with our students, are helping us design the new programs and are figuring it out.

So I think having, I think having this philosophical foundation combined with the innovation world is also pretty unique.

>> Jon Hartley: Yeah. Well, it's very impressive what you're doing. I feel like starting school is almost something sort of unthinkable, even just a few years ago.

>> Joe Lonsdale: It's a very crazy entrepreneurial endeavor.

They had to do thousands of pages of regulation, all these things are set up to stop you from building new universities. There's always moats in everything I build, these are some of the worst moats I've seen, we had to be very stubborn. Fortunately, people like Barry Weiss, Neil Ferguson, two of my main co-founders in Pano, who's been our president, they worked very, very hard and pushed this through.

And it's amazing getting to work with those people.

>> Jon Hartley: Absolutely, and finding, I guess, regional accreditation monopolies.

>> Joe Lonsdale: It's definitely a big moat. Well, that's something hopefully this administration will also fix for everyone, which would be great, we'll see.

>> Jon Hartley: Absolutely, and you've really assembled a very impressive board there.

I wanna just pivot a little bit to Wokeness and DEI? I mean, you've been pretty outspoken against Wokeness and I mean, do you see, obviously, I feel like the election seems to be America's kind of spoken. It's the first time a Republican president's won the popular vote in maybe 20 years or so.

I mean, do you see the election as being sort of a resounding rejection of Wokeness and DEI?

>> Joe Lonsdale: I mean, of course, everyone agrees it went too far, and it is great, and we won by only, again, a very small amount. I mean, we won, which is great and it's a rejection.

And I think a lot of moderates said this is enough. A lot of people I know who are moderates, who support the other side actually came to me and said, actually I'm kind of glad we lost because we do need to get rid of this nonsense. And so I do think a lot of people in America have had enough of this.

Now, that doesn't mean that it's going to go away. I think this stuff. Listen, these neo Marxist philosophies have marched through the institutions. They've taken over most of our universities. There's entire departments that are top universities, whether you can't even hire a moderate person, a moderate Democrat, you have to only hire a Marxist.

I mean, these places are completely broken in sociology, anthropology, history. You can't have a history person who believes, not believes in the British Empire, which would be great, but even believes in power of markets. I mean, it's just crazy how broken these things are. And so. So yes, America's said it's gone too far, but does that mean it's going away?

No, we actually have to go and like fight the fights now. So here's what drives me crazy. There's something I'm thinking about with Cicero. Every red state in the country still has state funded institutions where they're paying tax dollars, where there's literally departments that everyone in that department is a neo Marxist of some form who hates markets, hates Israel, hates America.

Like these are red states where the voters don't believe in this, and they tell, it's academic freedom. It's actually a freedom to take my money in this freaking state, take your money and defund these Marxists. And that is how powerful they are in our society, that we're still doing that.

And by the way, on K to 12, if you're a teacher in pretty much every red state you have to go get a master's degree from woke you in order to be brainwashed to hate markets, to hate America, hate Israel, in order to get paid more than come back and indoctrinate the kids.

And school choice is great. Don't get me wrong, I love school choice. It's important you guys started here in Florida. First things to Governor Bush and others, and it's just, it's amazing. And we're gonna try to do it in Texas next year. But school choice is not enough because no matter what, in 10 years there's still going to be a bunch of kids in these schools.

And these red states need to stop telling these teachers to go get indoctrinated in order to get paid more at a workplace. This is crazy. It is crazy, and it's like, it's a lack of courage. They say, well, Joe, people think it's controversial. It's just really hard to fix these things.

I have lots of other things I'm doing. I'm like, no, if you were a governor or, or legislator and you are not fixing this, you are not doing your job. So we have a lot of work to do.

>> Jon Hartley: Yeah, no, absolutely. And I'm curious, like in terms of corporate America, which I feel like came to just recently, but I think corporate America certainly has become a hotbed of wokeness and DEI as well.

I mean, do you see a shift there? I mean there's been some reports of some companies.

>> Joe Lonsdale: No, of course. Everyone's now just shifting slightly, and they're on the new side, and they're quietly firing a lot of the teams that were obviously totally insane people, but they couldn't admit they were insane people until about five minutes ago.

And there's a big shift, and it'll come back, and they'll do the next crazy thing because there's no courage in any of these companies, and it's fine. It's a good thing for our society that the people who are the NPCs and the people that don't have any courage are now going along with getting rid of this nonsense.

But, but yeah, it's just the lack of backbone in all of these people and all of these, in all these CEOs just, it's disgusting, and it's nice. They're coming around.

>> Jon Hartley: And I'm curious, what do you think about the tech world? I mean, I think one of the big surprises ahead of this election was just how many people from the tech world.

I think people like Chamath and others, people who were pretty serious card carrying, toe the line Dems even a few years ago, and I mean even 10 years ago, I mean Elon Musk was I think a card-carrying Democrat and others. I mean, do you think that shift is, is a permanent one?

>> Joe Lonsdale: Yeah, I think, I think this is a big shift for, for quite a while because I think you have a part of the Democratic Party that's completely insane. I don't know if it's 20 of it or 40% of it or what. It's probably 25, 30 of it.

That is completely insane. And I think the moderate left, like I said, lacks the courage to fight that and push back on it. And as long as that insane broken thing is there and is powerful and is running any institutions in our society, you're gonna have a lot of smart entrepreneurs who are gonna be pushing back, you're gonna be fighting it.

And listen, the tech world, the other thing that happened, I think if we're gonna try to diagnose this, first of all these people went insane and you gotta stop them from breaking things. But secondly, the tech world has intersected with government brokenness in a much bigger way in the last decade.

So a lot of the stuff we were building in tech was in completely non government areas that, where the regulation didn't really affect us that much 20 years ago. But if you're building in healthcare, if you're building in defense, if you're building in areas that intersect with manufacturing or education or etc., all of a sudden you're exposed to these like totally crazy corrupt regulations and moats and things set up, and you're like, wait a second, this is broken.

This is hurting our civilization, guys, let's fix this. And you go. And a lot of them like would go and say, okay, this is obviously wrong, let's fix it. And then, they'd be really surprised when it just didn't get fixed, and it was just stayed corrupt and stayed broken.

And you look at all these areas of our society where the costs have gone up, right? You can do this chart where there's this stuff in blue lines and the red lines, and you see the parts there, where inflation has gone up. And these are the parts the government is just clearly breaking and clearly allowing itself to be captured to break.

I mean, health care is probably the worst, but there's a lot of other ones. And this is finally something that a lot of the tech world now is exposed to, is, wait a second, well, let's fix this. And then they realize it's actually their side is the Democrats that are the ones who are not letting them fix it.

And that's been a huge red pill for a lot of my friends.

>> Jon Hartley: And absolutely. Last question, Doge, if you had one recommendation for them that they would, you know, implement with 100% certainty, what would that recommendation be?

>> Joe Lonsdale: So many things I already mentioned, so I'm not gonna mention, the cutting NGOs and fraud and everything else.

The one thing I would say is that when you look at how Elon Musk has his rules, like first you go in, and you like get rid of the dumb requirements, and you cut stuff that's not needed. And only then, once can you optimize. And so I think the optimized thing that's gonna come next to calling it day one is all this really cool firing and cutting and just like stuff get out of the way.

I'm so excited. But day two, if I had one thing for the optimize, is we need to have a, a transparent process driven by tech for every regulation to have to fight for its life every X years. And it needs to be something that, because it's driven by tech and transparent, everyone can see it.

Because basically 17 states have regulatory review processes and they don't happen because the lawyers are busy and no one sees. But if you make it transparent, you force regulation to fight for its life and you make it really hard to stay around. I better be doing really good things that would stop this cancerous growth of just infinite regulations over time that are suffocating the growth of our civilization.

So if we do that right, that could change everything.

>> Jon Hartley: Well, Joe, Joe Lonsdale, thank you so much for joining us. This has been really amazing.

>> Joe Lonsdale: Thanks, Jon.

>> Jon Hartley: All right, if you like to line up over here, you can ask some questions as long as we have some time.

>> Daniel: Fantastic talk. I'd like to hear how you would. I have two problems to solve. How would you tell Taiwan to defend itself? And how would you take out the gangs of Chicago?

>> Joe Lonsdale: Daniel, thank you. I always forget we're being recorded, hoping to say anything inappropriate. You know, listen, I think with Taiwan, I think the Hellscape thing is basically right in some form.

Taiwan is a very hard island to attack. I think there were plans for how it was going to be invaded because I think Japan had taken it in World War II. And then we had to map out what we're going to do with China. And it was just like horrible thing that was gonna cost, 5,000,000 lives and just tons of ships, tons of boats.

And that was even back then. And I think it is quite well defended now, more than people realize. So I think it's more likely that they use other methods to keep it, co-opt it. But if they were going to attack it, slash defend it, you probably want tens of thousands of autonomous weaponized vehicles above and below and around the water.

And you want to be ready to destroy any key ports of entry if they do get a hold of them, to stop them from capturing and bringing them in. So as much as that's kind of scary, you got to buy bomb your own, have bombs in your own areas, in your own ports if necessary, and then.

And then have the Hellscape. And I think that makes it just not worth taking it, makes it so they have to find other ways to co opt it, which they might do in terms of gains in Chicago. These things are always hard to do. I strongly support the US Constitution and our rights, which makes these things very complicated.

But listen, basically you need a really strong DA. Any really strong judges that actually wanna be really hard on crime. And then, you need probably some new form of like the Illinois Rangers. I don't know what they call them in Illinois. They call them the Texas Rangers that basically are trained probably by special forces to do honey pots and to do honeypot operations.

And just whatever kind of crime is going on, like make it a competition and figure out how to. How to entrap them and legally, of course, and get them. But it has to be. It has to be an effort with great leadership and great talent, and it has to be something a state wants to do.

And right now they don't have that leadership at all.

>> Daniel: There's no tech.

>> Joe Lonsdale: Well, sure. If we're doing. If we're doing all this like once, once. I mean, once you get warrants, you could, you could track all these people. You could listen to everything going on. If we're allowed to turn Palantir on the people who were there, once we know that they're guilty, and then use that to find others and everything, sure.

There's all sorts of things we could do with that. We do that with Mexico as well. By the way. If we want to really get rid of the cartels that are bothering everyone is probably use whatever we used on terrorists in the Middle East. We should probably point to the cartels in Mexico.

Same thing. We have to be very careful doing that in our own homeland. I'm very pro civil liberties, and so we just have to figure out what's allowed with that. And at some point, if it becomes a big enough problem and Illinois decides to do it in the right way, that's great. But until then, until we have that leadership, I think the best solution is to move to Miami or Austin.

>> Jon Hartley: All right, next question.

>> Speaker 4: Hey, Joe, how are you? I have a quick question. We recently. Let's see if I can pick this up. I'll just hold it like this.

We recently had an event with the Economic club about the next generation. And so there's like a transformative shift. Trillions and trillions of dollars coming to our generation soon. You're seeing baby boomers wanting to get out of their typical mom and pop shops selling them out. How do you see this unfolding and how would you say we can best prepare ourselves for something like this?

>> Joe Lonsdale: Yeah, this is a huge trend in our society. I agree. There's trillions of dollars of small businesses, where people are going to retire and pass it on. I think probably the right way to do it, and I've talked a lot about this with a few other friends as well.

Is we need an distributed method of getting advisors and training operators and of partnering with operators to buy these and back them. Tech's gonna play a big role in that. It's gonna play a big role in transforming these businesses and making them more efficient. But you still need operators on the ground willing to run them.

I don't think private equity in a traditional sense is always the right way. It doesn't always make sense to go into small towns and just roll up everything into part of something, of 30 things. But I actually do think you need like distributed training and talents and finding smaller entrepreneurs, and a movement for more people to be kind of small and entrepreneurs. And there's a lot to build there.

>> Jon Hartley: All right.

>> Speaker 5: Hello. Thank you very much for this awesome conversation. Have a quick comment before the question. I'm from Spain and I want to go back to TSA, where a socialist country in many aspects, much more socialist than the United States.

And we figured out a way to privatize the TSA. And believe me, Madrid works much better than MIA. So I think there's much progress to be done there. The question, though, has to do with AI and politics. Right now. Most AI I've seen in politics has to do with elections.

In your expertise, have you seen any AI being used for lobbying and, or strategy and predictive predicting how meters will act?

>> Joe Lonsdale: Sure. And thank you for that TSA comment, actually, not only because I was right against Jon. But secondly, because this is what's great about the United States is we have 50 states where you try experiments and this is why the federal government shouldn't do things.

It's also what's great about the fact that we don't have a global government. So we should, as United States constantly be learning. Even if something's more socialist, they could be doing something smarter. So I think, I think competitive governance is very important thing that we keep learning from.

So I love you guys in Florida because you guys do a lot of great things here. We all, we all copy, you know, in terms of AI and elections, I learned a hell of a lot about politics in elections. I don't even really want to know in this last one because I got more involved in, like, I'll tell you, there's things like APIs that are available within 24 hours where you can see if someone's voted or not.

You can't see who they voted for, but you see if they voted or not, which then changes how you follow up with them to, like, remind them to vote again. There's rules in different states for if you request an absentee ballot at one election, they just keep sending it to you forever because that was good for whoever the party was in charge there.

There's all sorts of crazy new rules, crazy new things. And it is a giant, big data problem, and a lot of people are trying to use data in AI to map it out to, to predict, to see where to put their resources. Listen, I think Elon did a very good job of this and was very critical to this when he put money and time into this last election with a lot of his smartest people.

It's something the left was winning at by far before, and now I think it's a little more neutralized in terms of the exact tricks. I can't say too much more, but yes, there's lots of stuff there that's key.

>> Jon Hartley: All right, next question.

>> Speaker 6: I stood in line first so I could ask the last question. What do we got time for like, two more?

>> Jon Hartley: We've got a couple people behind you.

>> Speaker 6: As long as you still cut me in. Not a hard stop at 7.

>> Speaker 7: Hi, Joe. I have a question about autonomous agents. A lot of people are saying 2025 is going to be like the year of the autonomous agents.

Microsoft, Salesforce, Google, they're all building autonomous agents. I just thought what your opinion on it was, if it's gonna be the year of the autonomous agent.

>> Joe Lonsdale: Yeah, well, listen, I'm obsessed, like I said, with the AI services at level 5, I was talking about, of directly applying AI to solve these things.

And a lot of these problems are things we do with agents at different types. I'm not sure a lot of the agents are quite good enough to do everything on their own. But that is how you do these things is you create different agents and you stitch them together and you practice and you iterate and you use them. And so, yes, I definitely think there's gonna be a lot more sophisticated types of agents designed to do different tasks and to complement people and replace people in different ways. And so, that's a great area to work in. All right, next question.

>> Speaker 8: Hi. I have actually a couple of questions. The first one.

>> Jon Hartley: One. One would be great. Sorry.

>> Speaker 8: Okay, well, I want to try to formulate it in one, then.

>> Jon Hartley: Okay.

>> Speaker 8: So when you have kids like me, I have a 4 year old. I'm trying to do something about giving them the tools to know what is good and what is bad before they get a computer to code at the age of 5 or 6.

Your Cicero Institute is working on policies. Is there something going on in any of the nine states where you are working to, to, to help our kids to have a better future? And, and also, I mean it's related also to the level three that you were mentioning. We all have access to large language models, but don't you think that there is an opportunity in the niche and in the small language models and not just for someone like Elon.

>> Joe Lonsdale: So on the education front, there's a lot of cool new companies like Angel Studios has made a ton of content this pro liberty. It's Christian liberty bias. I think it's very interesting company Tuttle Twins was partnering with them, which I was a tiny supporter of, which is this really fun kind of libertarian cartoon.

I don't know if you've seen as they've done a great job. I actually enjoy watching the episodes and my kids enjoy watching the episodes. I have like a four, six and seven half year old who watch it with me. So, I think there's a lot of great content being created.

I think there's some people trying to build various things about walled gardens for kids and there's like a lot of stuff there that makes a lot of sense. It is a really hard problem. I personally don't trust letting young children on a lot of these types of social media.

Or a lot of these types of like certain types of more modern games that use AI to kind of harness dopamine mechanisms in ways that they didn't when we were kids. I think these are a form of drug, I think and I'm very pro liberty, but I think these are a form of drug where it's like.

It's like, it's like it is doing something to your brain and how, in terms of how it interacts with you and addicts you and harnesses you and finds out how to figure out how to get your attention when they're really sophisticated. And I think that's fine for an adult to opt into.

I probably too much time addicted to X or something like that, but I don't think we should, at least for me, I don't want to use that on my kids because I'm not sure I trust them. In terms of, in terms of the language models, I think you could train large language models to act like small ones in different ways. 

So, I'm not sure if you need small ones, but there's lots of ways that you can train them and create context and specify. But I'll have to learn more from you about that later. Thanks.

>> Speaker 8: The education. I was actually referring to AI guidelines. We're actually drafting now the AI guidelines for Miami Dade County with Miami Dade County Digital Commission.

>> Joe Lonsdale: My suggestion is don't draft those. That's, that's really bad. Like you don't want to create any rules at all. And if you, if you're creating rules for your school, that's fine, but don't create rules for what people do with their kids. You should, we should, you should not create top down rules.

I think that's really bad, and we should make Desantis stop it if you do that.

>> Speaker 8: It's happening. So.

>> Joe Lonsdale: Okay, well, Desantis is going to stop it and I'm going to stop you, but thank you very much. Don't do that. Don't do that. That's, that's, that's authoritarian and it's not your job and it's not your business and this is a free state and they shouldn't do that.

And anyone, they try to do it, DeSantis should step in and stop them.

>> Jon Hartley: Okay, next question. Joe Lonsdale, titled Twins Watcher. That I didn't expect.

>> Joe Lonsdale: It's great tonight.

>> Jon Hartley: Great. Next question.

>> Speaker 9: Thank you for your time, Joe. Shifting to the banking sector in a world that's very relationship driven and client facing and you know, building the relationships is, tends to be the backbone of business development.

Do you see or knowing where AI is at and where it could be, do you see AI either augmenting those client facing roles or possibly replacing them down the road?

>> Joe Lonsdale: I don't think it's going to replace them for a long time. I think people like meeting other people and trusting other people and seeing other people, it's definitely augmenting them in new ways.

It's definitely completely changing, outbound sales, is definitely maybe doing some first screening calls and things like that. There's definitely a huge role for like augmenting with AI. But I think what's happening is right now the AI is playing that role for like first emails and for first chats and stuff.

But I think as that gets more and more popular, it's going to get too annoying for people because they're going to get like infinite inbounds and then you're just not going to want to interact with something if it's not like a person you know in your network or trust or have somehow met somewhere. 

So, I think it's like we're in an interregnum right now where this stuff all works better because it's so new and because it echoes a person. But as soon as we all set up ourselves again to protect ourselves from this like increasing deluge of AI, we're all going to face, I think you're going to need to find ways for people in relationships to come back in.

And so, so I, I, I'm, I guess I'm very long people in relationships still mattering for most things. If it's a very simple transaction, maybe the AIs figure out themselves. But for bigger decisions, I, I think, I think this AI stuff that's getting like growing really fast in certain areas there is not going to work long term because it's just too spammy and too weird and people, people want other people is my view.

Maybe I'm old fashioned, but that's my view.

>> Speaker 9: Completely agree.

>> Joe Lonsdale: Great, thanks.

>> Jon Hartley: Next question. We're going to do one question each. You're going to be the last person. We're going to get you out of here on time. And ideally, quick with the ends with the question mark would be great.

Thank you.

>> Speaker 7: So, thanks for Ollama. If I took the entire corpus of your digital data, these recordings, every recording public and private, all your messages, emails, and I threw $10 million at compute and I made a model of you and then I benchmarked both of you, do you think you could still outperform your model?

And bonus question, what would you ask yourself?

>> Joe Lonsdale: I think there were certain tasks at which the model would outperform me. It'd probably be a lot more eloquent in knowing, having all the context of things I was thinking about and saying them well. So, there's if I want to just sound impressive at quoting things or saying things I've said before.

It would outperform me if I wanted to write a new piece. I just wrote a piece on this AI services wave I've been talking to you guys about. It's on my mind. You can go on our site. We published it a month or so ago. And like, I don't think the AI could have written that piece, but now that I've written it, maybe it does.

Maybe in some ways it's more eloquent about it. And that's where we are right now. Because it has the ability to do all these things and have all these things in its mind at once. But, but, but for new creativity. We're not there yet.

>> Jon Hartley: All right, next question.

Would you mind using the microphone? Thank you. We just need you to speak into it just because we're recording it, and it feeds into the video too.

>> Speaker 10: Sure. So my question is, you have interesting political views and based on that, do you think that the government should go towards the direction of AGI building artificial general intelligence, or should it be solely a private sector thing?

And if government has that, so how do you envision this?

>> Joe Lonsdale: So right now, in the US, as I explained earlier, the government is run mostly by idiots. And so that'd be very hard. There are some smart people in the government, but like SpaceX for example, there's, there's like no chance they could have ever built even anything even close.

Palantir. No chance anything even close the way it works right now. And so, I think asking if government should do it itself, I think you don't understand where government, what it is today. Maybe you have a model of government that comes from the 1950s, but that's not where we are.

We'd have to, we'd be living in a completely different civilization if the government actually had the capability. And then at that point I'd be able to answer differently. Maybe if somehow there's some extraordinary cultural merit, somehow was in our government that made it deserve to do these things, I'd be very open to learning about that civilization.

But we live in a civilization where it's like, it's not like a little bit away or a lot away, it's just like infinitely away from doing that. And so I think that's a joke, unfortunately, and with respect, it's just not how government works in America. And it probably should not try to do these things because it will end up just spending lots and lots of money on cronyism and waste and silly stuff.

>> Jon Hartley: All right, second last question here.

>> Speaker 11: Thank you for being here. I'm the dean of the business school at Florida Atlantic University. We have 9,300 students in the business school alone.

>> Joe Lonsdale: Cool.

>> Speaker 11: I've hired some really great people. You hired one that I was trying to get you hired at Austin.

What advice would you give me in the practical world? I can't start over. How do I handle those 9,300 students and the 150 or so full time faculty that I've got? What do I do to improve that the best possible way?

>> Joe Lonsdale: That's cool. Well, you know, thank you for managing something that important.

I would say you should probably have more exercises where you force people to engage with debates and controversial ideas and different perspectives and to learn how to steal me on the other side and learn how to be part of conversations and thought pieces around this stuff that they're not exposed to enough about how the world actually works and about how people like Elon and others view the world and what the great debates are of our time.

And I think a lot of these schools, I think one of the highest ends for professors is not to be controversial and not to engage in these things. And I think that's exactly the opposite. I think they should actually be forced to expose these things. And I think having these debates about uncomfortable topics is actually makes people grow and be a lot stronger.

So, it's something I think about.

>> Jon Hartley: All right, our last question.

>> Speaker 12: I'll go with a question first. Let's say if I want to take my manual heavy business to the next level, leveraging AI, what are the steps do I take? Like, do I go get computer science at MIT or like, is there like a breakdown of the steps you should take?

And I mean the staffing industry, I run a very small boutique domestic staffing firm and as you like, very similar to what you said, there's not many smart people in the staffing world and people are still from the 70s, like using handwritten letters. So that's pretty much where.

>> Joe Lonsdale: The human touch has a lot of, has a lot of good things about it. So, there's definitely, like I was saying earlier though, there's wisdom where people are going to really want those letters and notes and human touch. So, I don't think those are going away fully.

If you're not technical yourself, it's probably not obviously good to start over and try to go become technical. I think you want to find other people who are technical and listen in any industrial wave like this, there's going to be some people who are trying to build new companies entirely, and then there's going to be other companies trying to arm you and trying to give you things.

And there's a lot of companies out there right now trying to give you software, trying to arm you, trying to allow you. So I think experimenting with the most advanced new AI platforms out there, I think finding a smart technical person, helping them help you experiment, having them do tests, is very worthwhile.

And you may actually find there's ways to make it more efficient. So I don't think you have to build something newly yourself. I think you can use what's there, and you can still probably do some cool things with it. Thanks.

>> Jon Hartley: Joe, thank you so much for joining us.

Let's give Joe another round of applause here.