Anduril Founder Trae Stephens on Our Modern Defense System

Yeah. I work in tech. I'm both a partner at Founders Fund, which is a venture capital firm, as well as the co-founder of a defense technology company called Anduril. But maybe more centrally than that, I'm, you know, a midwesterner, who I've been married for 18 years. I have two small kids. Grew up in a lower middle class household and have kind of been on a ride for the last, now, 42 years in my life. Could never have anticipated that I would have ended up where I've ended up today.

I was in New York that day. And I actually walked out at the end of the day and saw Building 7 come to the ground. And I think it fundamentally changed me as well. But I was already an adult with a job, so it didn't shift my path in terms of my profession. But as I understand, that was indelible for you.

And I think on 9/11, there was this kind of moment where I realized all these things that we thought were guarantees in the American experiment don't feel super guaranteed anymore. And I decided at that time that I wanted to serve the country and go into national security. I wasn't sure if that meant going into a military role or going into a political role. But I ended up deciding that I wanted to go the intelligence community route. And that's kind of what set me on my path to Washington, DC.

Well, I was mostly interested in finding a university that was a good path to get into the intelligence community. Georgetown is the best of the schools for that specific thing.

I'm not in the CIA. I went into a counterterrorism office at one of the intelligence agencies. It sounds cool. It sounds much cooler than it actually is.

You know, it's in the government that makes it not exciting. It's not what I expected. And it's highly bureaucratic. So, you know, I grew up watching James Bond, watching The Bourne Identity. And you don't actually get James Bond when you show up, you get like Mr. Beans, right? Like, I had Windows 98 and a CRT monitor in 2006. I mean, no one threw me the keys to an Aston Martin. I didn't have Jarvis from Iron Man to talk to. I was just running what I call coffee break searches, where you open up a bunch of tabs in a very outdated version of Internet Explorer, and you run searches, and then you walk away and you go get coffee, and then when you come back 15 minutes later, hopefully, you have little downloaded CSV files like Excel spreadsheets, essentially. So I think I just expected that it was going to be a lot more advanced than it was, and the people are highly patriotic. They care about the mission. And that was really inspiring. But bureaucracy is brutal.

You know, people matter a lot, like individuals. Individual people matter. They set culture. And I think, you know, there are ebbs and flows of how this goes. I'm hopeful that, you know, the people that are working in our national security apparatus feel supported by their leadership to take risks and do things that I certainly didn't feel empowered to do.

I think there's like all sorts of micro things that you can do. But at the end of the day, like culture is set at the top. And as someone who's now in a leadership position, in multiple different places, I don't want to believe that's true because I want to believe that it's like diffused throughout the organization. But people really do look to like, how do our leaders behave? What characteristics are they bringing to bear? If the leadership is incredibly conservative and incentivizes risk-averse behaviors, then that's what the whole organization is going to do. And so I think what I would have liked to have seen when I worked in the intelligence community was empowerment and promoting people on merit rather than tenure. I would have liked to have seen them say, I don't know if this is going to work, but I believe that we need to try, right? These things like entrepreneurial spirit.

That's right. Yeah. It's like one interesting way to think about this, that's just aircraft specific, is that more time has passed between when the SR-71, the fastest airplane to ever fly, the SR-71, was initially flown to today, than from the Wright Flyer to the SR-71. So we have this tendency to think of history as like it's been so long. It's actually like we made a lot of progress in not a lot of time. And then it wasn't our ability to build things that stagnated. It was bureaucracy that created that stagnation. And one of the former undersecretaries of the army, this guy named Norm Augustine, who later was the CEO of Lockheed Martin, he said in 2054, the entire United States defense budget will buy one airplane. And obviously that's like an extreme thing to say. But I think directionally what he was saying is like, look what we're doing. Like we've created this tremendous bureaucratic bloat that makes it really hard to get anything done. And we're not going to make that SR-71 trajectory as quickly as we did if we continue down this path.

I think that's politically very, very difficult to pull off. And in 2026. Yeah, I do think that some kind of civil service where you're banding together and you're thinking about the country as a whole, especially because things are so divisive. I think if people could engage in that way and contribute while also understanding that they're part of a greater thing, I think it would be great.

No, I think that's like, ultimately that's what you're trying to do is you're trying to give that exposure to, you know, look, we're all getting so tribal and so charged. If people had exposure to the way things worked, I think they would be able to take a step back and say, this isn't really about politics, it's about how ineffective our bureaucracies are. And we should be building these things to serve the American people. And not to serve our own tribal political interests, because that actually doesn't solve any of the problems. It just makes it worse.

Yeah. I mean, I couldn't agree with you more. It's like the, you know, I notice with my own husband who's the best person ever in the world. He's so progressive, like, he has such a sweet heart and he wants to, like, make sure everybody's looked after. And I think in this climate, you know, sometimes I'm like, can you just listen to this? It is very triggering for people. It's become so binary, I think. And I am trying to in my journey through being an American right now trying to weave together lots of different points of view and also to get out of that place of righteousness and anger and fear. And I mean, I'm pretty centrist and my husband thinks I'm a Republican. But I'm not a Republican. I mean, I don't feel anything right now, to be totally honest with you. I feel like I'm completely an independent. But I do think that, you know, we were just in Nashville over the weekend and we went to see bluegrass, and there was this amazing girl playing on stage, a young woman. And, you know, she was talking about it. She obviously had completely different points of view than my husband. You know, as I'm sitting there like, okay, you know, I guess, you know, and I just thought, no, but this is so beautiful to see somebody who clearly is such a good person coming from such a different place, like, isn't that more valuable to understand?

Ultimately, yeah. We don't really gain anything from shutting people out and not hearing what they have to say. You know, I think this is like the greatest risk to a lot of what's happening in technology today. If you think about the centralization of AI, it's just self-reaffirming. Yeah. It's like when you go and you ask questions to a large language model like ChatGPT or Claude or whatever, it basically is just going to tell you whatever you want to hear. I switch between them based on strengths of what I'm specifically trying to do, but fundamentally, they're all just going to basically tell you what you want to hear, right?

Oh, well, that's what we describe. Yeah. So for your husband to do that, I think most people are just, like, pretty happy with being told, you know what? You're right. That seems good. And there goes the friction. And friction is where I think characters build, 100%. And I think that goes especially for things like relationships. Yeah. Where, you know, for millennia people were paired off intimately with others in local communities. It's like people would meet at school or at church or, you know, at the Elks Club or whatever it is. And they would be matched based on we have a long standing friend relationship or I think they're funny. I think they're charming. You know, we have mutual friends. Today, that's not really how it works, right? Like the online dating scene has created this sort of parallelization around optionality. It's like they're meant to date in really high volumes. But not to actually connect in any way. And so if you look at the filters, like I think it's like 10% of men get 90% of the right swipes, the positive matches, and the bottom 50% of men get no matches at all. Well, what does that do for a society where you've created this kind of environment that's super self-selected and like a mass optionality framework, you get loneliness, despair, and, you know, young men, unemployed or underemployed in their parents' basement, you know, never matching romantically with another person, going to LLMs, getting self-validation, having them tell them all the things they want to hear, becoming addicted to porn, gambling. And you can see these like little tech companies that are emerging around these exact ideas. And that's the powder keg for any civilization. I mean, assassination attempts against presidents. Who are these people? They're young men under 40 that are unmarried, that don't have kids. So I mean, this is we've created this as a society.

Well, I think this mandatory civil service is actually kind of a good idea. It's like you guarantee employment. Which is really important in this era of like a reformation of what jobs look like with AI. I think being able to guarantee that, I think getting people back into local communities as a focus is really important. Reskilling, retraining people. The trades, I think will be a big part of reskilling over the next 20 years. Like we need people to be electricians, to be plumbers, to be, you know, all of those things, welders. And this is like the funniest thing to me about recent government policy is that, under the Obama administration, they had this huge push around reskilling the workforce. And what did they do? Code academies. It's like literally the first thing that gets replaced by a farmer. Unbelievable. It was a good idea at the time. But man, they couldn't have chosen a worse profession to reskill people into.

Yeah. You know, these things end up being like, super idiosyncratic, based on what's happening at a time of maturity, like even just a few years difference in age for us. Right? It's like you live through the heart of Reagan rearmament. Yeah. Like the very tail end of the Cold War. I was six when the Berlin Wall fell. Right. And so, like, by the time I have active memories, you know, we're in, like, George H.W. Bush era, the like globalization push. From H.W. into Clinton. So very, very different time but not actually separated by very many years. I think, you know, what's happening today is there's like this sort of speed running of geopolitics that has been happening over the last 6 or 7 years, where you had this transition out of counterinsurgency, non-state actor conflict. This is, you know, the war on terror, essentially, where we're not fighting governments, really. We're really fighting terrorist organizations that are decentralized. It's highly fragmented. They're doing insurgencies inside of urban environments. And transitioning from that into a great power conflict where we're thinking about how do we deal with the threat from China, Russia, Iran, North Korea, very, very different approach you have to take to build protections from great powers rather than counter-insurgency. And then along those lines, we had, you know, Ukraine, we had October 7th, we have the conflict with Iran. And, you know, the threat to Taiwan, you know, Xi Jinping saying we want to retake Taiwan, reunify by the end of 2027. So, it's a very different geopolitical environment than the one that I was working in, in the early 2000s, in counterterrorism era when we started Anduril the company, the defense tech company. I started in 2017. It was super unpopular. Like, I think people were roughly just like, I don't know, we're like sort of coming out of this whole, like, war in the Middle East thing. It seems like everything is basically fine. There were employees at Google who were, like, protesting Google working on a Department of Defense contract, and they ended up pulling out of that. So it was like really an unpopular moment to be thinking about this. Just a few years later, it's like Ukraine, flag and bio, like everyone has sort of changed over to this idea of, wow, the threat of authoritarianism globally. And what that does to free democratic peoples is terrifying. And the whole momentum shifted almost overnight into like a posture of we have to be prepared to draw hard lines around what we're willing to allow to happen.

Yeah. You know, I have been thinking about this space, obviously my whole career. And one of the realizations that I was coming to was, we've relied on nuclear deterrence. And I'll explain what all this means. Against great powers for, you know, like, almost a century now, like 75 years, where you basically say, we don't believe that you're going to attack us, right? Because we have nuclear weapons and we will retaliate. And then you get into this idea of mutually assured destruction that if we fire a nuclear weapon at a nuclear power, they're going to fire nuclear weapons at us and war games. Yeah. And before you know it, everything is kind of toast. And that, like, basically worked. But you see a lot of this like sort of fringe conflict where, the belief is as long as it stays below a certain threshold, no one's going to launch a nuclear attack. And that's like kind of the line that we're drawing, but we don't want any of these conflicts to happen. I mean, look at how many lives have been lost in Ukraine. Like, I think it's like over a million casualties now on the Russian side alone. Yeah, on the Russian side alone. So I think it's really important to figure out how to use conventional deterrence rather than just nuclear deterrence. And what I mean by conventional deterrence is, can you convince your adversary that the cost, both monetarily and in human lives, will be too high to even think about going into conflict? And this is what Reagan did in the closing years of the Cold War, as he said, I'm going to build up and we're going to get to the point where the Russians are going to have to, the Soviets are going to have to look across the ocean and say, do we want to continue down this path? Right? Are we going to win? And ultimately they decided, no, we're not going to win. We can't keep up. And this sounds sort of uncomfortable. You know, it's like if you think about like the spectrum of a quad chart where it's like, you know, feels good is good, feels bad is good, right. Feels good is bad, feels bad is bad. This definitely is in the feels bad is good quadrant. It's like I would prefer to not have to do this. It would be great if we could just say diplomacy works every time. We can convince people without any projection of power, that they should do things the way we want them to do it, but it's just not how the world works. And, we found this in the early days of World War II, where Chamberlain in the United Kingdom was all about appeasement. Appeasement is not deterrence. So you can't prevent people from doing bad things by really believing deep down inside that they shouldn't do bad things. And so the idea that's optimistic, it's very optimistic. It's just not realistic. Right. And so the Anduril story from the beginning was, can we create a new era of conventional deterrence where we can build capabilities that are so inexpensive and so high capability that our adversaries will say, I don't want to get into this exchange, right? Like we don't have any advantages anymore, even though theoretically you profit from a conflict, sort of. Although by the time you're in a conflict, all of the things that are going to be used in that conflict are in inventory, right? Like the way Anduril is approaching this is we're not trying to build technology to solve problems today. We're trying to build inventories of capabilities that prevent the problems of tomorrow. And so I don't view, first off, I don't think like wartime profiteering is ethical really, in any way, regardless. But I think the important thing is we need to build up a capability advantage over the next 100 years, regardless of what's happening right now.

It's yeah, it's super important. We don't want conflict to happen. I have spent my entire career working in defense. I've never once met a general or an admiral that says, you know what I want to do today? I want to write letters to the parents of children who died in combat. Nobody wants to do that. This belief that now we're, like, hard charging, like throwing people, like, as meat bags into combat zones. It's just not true. Nobody wants to do that. And I think this is really kind of the motivation for building Anduril is let's make it really improbable that we'll do that.