Palmer Luckey 谈 AI 驱动的下一波军事科技 | EP #158

Hi, my name’s Palmer Luckey, and I build killer robots.

You must have a thousand different ideas. What’s your calculus for filtering these?

We don’t have time for business as usual.

We don’t.

We don’t have money for business as usual. We have to try something. I think you’re going to see humanoid robots in defense applications pretty soon, but they’re not going to be for what people expect. We need to avoid outsourcing responsibility for violence to machines, to robotics. If we’re going to kill people, we need to kill people—and it needs to weigh on us.

Now that’s a moonshot, ladies and gentlemen.

I do appreciate you showing us your advanced designs here. When I asked Palmer backstage, “Do you have this under development?” he goes, “Yeah, kind of. Something like this.”

1. Humanoid Robots Enter Defense

Yeah. I mean, this is a long discussion, but I think you’re going to see humanoid robots in defense applications pretty soon, but they’re not going to be for what people expect. The first use isn’t going to be humanoid special-forces door-kickers. It’s going to be robots who walk around with about the physical ability of maybe an 85-year-old man, and they operate a lot of the existing systems we have. So think about things we have that are manned systems today, like a surface-to-air missile defense system.

Silos, yeah.

Or missile silos.

Yeah.

Exactly. Right now, they’re fully manned. If you could build robots with the physical ability of an 85-year-old man that shuffled around, pushed a few buttons, pulled some levers, maybe—

Instead of having humans be bored in there day after day.

Yeah, and there are a lot of jobs like that in the military. Same thing for, potentially, rather than automating old vehicle platforms, you could use humanoid robots that are able to just walk into them, close the door, and then operate them. Is that going to be the ultimate future of robotics? Of course not. But there’s a near-term future for even the limited humanoid robotic systems that exist today, and I’m excited about that.

Yeah, me too. But tell me the truth: Iron Man suit coming soon?

This is another one of those problems. I mean, it’s the classic exoskeleton—

I mean, you are the closest thing we have—

We—the United States—should have invested in exoskeletons a long time ago. I mean, look, Walt Disney was a huge fan of exoskeleton technology, and he was part of the Man Amplifier project. A lot of the animatronics that are at Disneyland were actually a result of work that he did envisioning in that space.

That said, we probably should have invested in exoskeletons a long time ago. Too much time has passed, and at this point, you’re probably just going to have fully remotely piloted robots—

Yeah.

—or autonomous robots. Building a robot that is capable of doing superhuman things while also wrapping it around a person made of meat—

Putting meat—

—it’s very difficult.

A meat sack in danger, right?

It’s much harder to do those two things—

Uh—

—at the same time, and so you have to answer the question: Why am I doing this? What is the point? Am I trying to reenact my sci-fi fantasies, or am I trying—

And we had one here—

—to solve a problem? And so, if I had to guess, you’re going to see exoskeletons more in the consumer and civilian sector, where people just want to do cool shit, than among people who are actually out to do a job.

We had one here in our tech hub this year for kids who need help walking and for elderly adults. I mean, it’s—

Kids who want to walk good and do other stuff good, too.

Yeah. You have taken on industries that others have considered untouchable. First of all, the naivete and insanity of the VR industry, and then, of course, did people ask you whether you need to have your head examined to take on the DoD?

At this point, I’ve been doing it for 8 years, so I think they asked me a lot more at the beginning. 8 years ago, starting Anduril was very controversial. You might remember we—

I do.

We were on Bloomberg’s list of the most controversial companies in tech. They called us the most controversial company in tech. This was as Uber was going through its ousting disaster. This was as WeWork executives were being indicted. No, it was Anduril that was the most controversial company in tech—somehow. Little old me with my 2 dozen people, for the crime of daring to work with the US military. Wired magazine named me the most evil person in Silicon Valley. It’s just been a really interesting ride.

What an honor.

Oh, believe me. Yeah, it is.

That’s extraordinary. So why do you do this? Why do you take on these seemingly impossible goals? What drove you to build Anduril?

2. Passion Built Oculus

I’ve actually been reflecting on what you’ve been asking people to do in terms of coming up with how they’re going to do their moonshot and how they think about impacting the world.

Yeah.

The first time that I did it was nothing like that process you’re asking people to do. I did not start working on virtual reality because I said, “Oh, my God, I want to impact the world. How can I best do it? Ah, this is how I will do it.”

Mm.

It was much simpler than that. I was a gamer. I liked gaming. I had been asking myself a question for a long time: What’s the next step in games? And then one day I woke up and asked, “What’s the final step in games?” Clearly, it’s virtual reality, and I said—

Well, that’s passion-driven, a purpose-driven—

What I’m saying is, it was just passion-driven.

Yeah.

When I was raising money for Oculus, I was not at all certain that any of my investors were going to make any of their money back. I felt like I had conned a bunch of people into paying me to work on my hobby full-time, all day. And I think that’s how a lot of the best companies start, right? Arguably, that’s what the guys at Apple were doing.

Yeah.

There were a lot of people who—

Computer club.

Computer club. They conned some people into paying them to play computer club all day and do what they were doing in the computer club, but as a business. And so I was really no different from that. Oculus turned out to be exactly the right thing at the right time, and I sold that company for billions of dollars after figuring out how to make VR headsets better.

What was key was you said no to a billion dollars. How old was the company at that point when you said no to a billion dollars?

13 months.

Holy shit.

Yeah.

And then Zuck came back with $2.2 billion.

Quite a bit later. We were 18 months at that point. But the thing to remember is that what convinced us wasn’t the amount. At some point, if you sell your company for $1 billion or $2.3 billion, it’s the same in terms of quality of life. It wasn’t the bump that made the difference; it was that Mark Zuckerberg committed that he was going to put at least $1 billion a year into research and development of VR technology, which was my passion, for at least the next 10 years.

So that’s what I was weighing: What is it going to take for me, as Palmer Luckey, to raise $10 billion in R&D cash? Well, to do that, I’m going to need to make some number of billions in revenue, I’m going to need to dilute myself by some certain amount, and I’m going to need to do some number of raises. And you start to do the math and you realize—

Very simple math.

Yeah, you say, “I’m not going to be in control. That’s going to be almost impossible to do this, and here is a surefire way to maybe not be as in charge of my destiny.”

And, of course, I end up getting fired a few years later, so that really manifested fully, the risk. But the positive side, and what did happen, is $10 billion. And I mean, that was the commitment. The commitment from Mark was $10 billion, $1 billion a year for 10 years.

That’s impressive.

But the actual number has been $60 billion.

Wow.

And they changed their name, in fact.

Well, yeah, and then, actually, the day that they changed their name to Meta, I put all of my liquid assets back into Meta stock. I’m a total nutter. I really fully believe in the Metaverse future, whether people think it’s a fad or not. I’ve been with it long enough that at least you can’t accuse me of chasing the fad. You can only accuse me of being naive or stupid. But I’m a stupid person who really believes it.

I believe you. Anduril’s core mission: How do you define why you built it, and is that still the same mission that you have today?

3. Anduril Turns Rejection Into Mission

The first page of our pitch deck to our investors said, “Anduril—”

I wish you had come and pitched me, but you didn’t.

Okay.

Sorry. I'm sorry.

Better late than never. Yeah.

I'll tell you: I only ended up raising money in that first round from 1 fund. It was Founders Fund.

Nice.

There are a lot of reasons for that. One of them is that Founders Fund was the first institutional investor in Oculus. After meeting with them and them beating up on me and saying, “Well, we don't think this is really going to work. If this works, you're not just going to be a successful VR company; you'd be the first successful VR company in history, ever,” they said, “We don't really believe this is going to work, but we'll give you $1 million.” That's something I'll never forget, and so I have a lot of loyalty to them for that.

Yeah.

The guys at Founders Fund were also some of the only people who were still willing to talk to me after I was fired by Facebook and ripped out of Oculus. I know it seems hard to imagine today, because I've clawed my way back to a level of some relevance, at least, but at the time, people literally would not answer my texts or would stay far away from me.

It came back to me through other people. They would explicitly tell other people, “I'm staying away from Palmer. That guy's done. He's a one-hit wonder. He got it good that one time, but he's toast. I'm not stupid enough to tie myself to a millstone like Palmer Luckey.”

That was a big part of why I started Anduril. You asked what our mission is. Our mission is to revolutionize defense and save taxpayers hundreds of billions of dollars by making tens of billions of dollars. But there's also an element of, “I'm going to prove those guys wrong. I'm going to show that I'm still somebody, that I'm not a one-hit wonder.”

Then I'm going to ask them to come and pitch us on why I should let them invest in my company. In the end, I'm going to say, “I don't really want to tie myself to that millstone.”

I love it. Love it. Love it.

There is 1 investor in that category whom I let invest just $100,000.

Who's he?

I won't say who. It was just enough to get information, right? So I can remind them how well we're doing.

I love you, buddy. You're amazing.

No, I'm a vengeful, bitter, cynical person.

I've seen some of the back-and-forth salvos. I would not want to be on the other side.

I’m very kind about most things. People imagine that I have this vengeful streak in general, but if you look, the only thing that I'm vengeful about is the people who ripped me out of my own company that I started as a teenager, then celebrated it, and especially the ones who made hundreds of millions of dollars in the process. The things I'm most bitter about are just that.

It's just plain wrong.

You can slight me today, and I'm actually pretty forgiving. But that 1 event in my life—I will never forgive any of the people who are responsible.

All right. Very crystal clear. So there was an event that took place recently that is epic. You took over the Integrated Visual Augmentation System contract—a $22 billion contract that was handed over from Microsoft to Anduril. That's extraordinary.

It is.

From Microsoft. Yeah, so tell us that story.

I mean, it’s a long story, but, you know, the—

Tell us shortly.

4. Anduril Wins The IVAS Contract

The short version is this: The idea of putting a heads-up display, a computer, a radio, and an AI on every soldier has been around for a long time. It goes back to at least the 1959 Robert Heinlein novel Starship Troopers.

Mm.

What's actually fascinating about Starship Troopers is that it achieves so much cultural relevance as a film, but the film doesn't actually have the mechanized infantry wearing heads-up displays or mech suits. It's very strange. The thing that I most liked about Starship Troopers did not make it into the film. There's a new film being made.

Oh, that's awesome. I love Heinlein's stories. They should all be made into films.

He has a lot of incredible stuff.

Yeah.

If you haven't read the incredible number of things he's written, then you should.

One of the things I'm proudest of was getting the Heinlein Prize years ago.

This idea is an old one, but nobody's ever been able to pull it off. There have been many efforts: Land Warrior, Future Warrior, Connected Soldier, and Net Warrior. What you really lacked was a back end that could feed such a device with useful information feeds.

It's easy to make a thing that can show a 2D map floating in front of you. It's hard to build something that can understand the world around you, augment your environment, show threats, show friendlies, and tell you what to do. That's something that's only recently become possible.

Anduril actually tried to go after the Army's last attempt at doing this, which was IVAS, almost 8 years ago.

Eight?

Almost 8 years ago. But at the time, Anduril was less than 2 dozen people—the whole company—so it was pretty clear we were not going to win, and we didn't.

Mm.

The whole time since, I've been wondering when I was going to get to tackle this problem. The story then gets very long and very bound by NDAs, but it ends with Microsoft saying, “Okay, we will hand over the entire contract to you,” and the United States Army saying, “Yep, that's fine with us. We'll assign all responsibility to Anduril for continuing this work rather than Microsoft.”

The good news for me is that I've been putting enormous amounts of my company's money into building exactly the system you would actually want to get onto every infantryman. I'm going to be able to get done in about 6 months what other companies would take 8 years to do.

Amazing. And I have to thank you for that.

I have to thank my investors for giving me all my money that I could use to invest in that stuff.

Yeah. Did Microsoft shut down HoloLens completely?

It depends on the way you look at it. I didn't actually just get the IVAS contract. I bought Microsoft's entire mixed-reality business.

Ah.

The only part remaining of any substance was IVAS.

Okay.

The original pitch of IVAS was that it was a militarized variant of HoloLens, which was going to be an AR/VR device for consumers and for enterprise. That got shut down. They're stripping Windows Mixed Reality out of Windows. It's not going to be a part of Microsoft's near future, that's for sure.

Yeah. I've played in the aerospace industry, in the launch business, early on, and it is 1 of the most entrenched industries on the planet. It's literally a self-licking ice cream cone of people flowing in and out of the government onto these military-industrial-complex boards. How in the world did you penetrate that?

5. Defense Needs Product Companies

The way that we did it seems crazy in hindsight, but we believed it would work, and somehow it did. We decided that we weren't going to start a defense contractor. We were going to start a defense product company.

Mm-hmm.

The difference there is that you spend your own money to make something that works, and then you sell that as a product, versus trying to get somebody else—usually the government—to pay you to do work, right? That makes all the difference. The incentives are different.

When you're a product company, you make more money when you move faster. You make more money when you make affordable decisions. You make more money when you do the right thing. When you are paid on a cost-plus contract, where you're paid by time—

materials, hourly, and a fixed percentage of profit, you make—

Kickback.

Yeah, you'll make more money when you spend more time working on something, when you buy the more expensive component, and when you don't reuse things that you've done in the past, instead redoing them from scratch. By changing that incentive and bringing in a lot of our own money, we were able to build things much more efficiently. We've built autonomous fighter jets, autonomous submarines, and now visual augmentation systems for the infantry, as well as an AI system called Lattice that kind of binds all of our stuff together. We've done that all on our own dime. That's the only reason that it works.

I don't think people realize that military cost-plus contracts are a thing of recent history, like post-World War II.

That's right.

Before, it was a complete opposite way of doing contracting.

The United States has a long history of turning small technology companies into major defense companies. The problem is that we've now forgotten how to do that. We haven't done it for many decades.

Yeah.

And cost-plus contracting is a relatively recent artifact.

Yes.

The funny thing is, it's a contract structure that was intended to control graft and cost. The idea is, well, either we don't want to let them make too much of a profit margin, so we'll fix their profit margin and say we're only going to pay them what it costs plus a fixed percentage. What they forgot or didn't understand is that incentivizes you to make it cost as much as possible, which harms everybody. Nobody wins.

Yeah.

I have to point out that the only other real industry that is dominated by cost-plus, like the military, is residential renovation construction. Almost all major defense acquisition programs, or MDAPs—meaning anything that is of substance—go to just 5 companies. About 80% of MDAPs go to those 5 companies, and 30% of MDAPs have a single bidder, meaning there's zero competition. Almost all of them use cost-plus contracting.

Wow.

The only other industry with the same density of cost-plus work is residential renovation construction. Has anyone ever renovated their home and, at the end, said, “That cost exactly as much as I thought it was going to cost, and I really feel like I got value for my money”? No. I really feel like it's not a coincidence that 2 industries so different, so far apart, would come to the same end.

6. Wildfires Become A Technology Test

The last time we spent a bunch of time together was the launch of our Wildfire XPRIZE competition.

Wildfire XPRIZE.

Yeah.

That was in Washington, D.C.

In Washington, D.C.

It was great. There were a bunch of people there. The lieutenant governor was there.

Yep.

There were a lot of CAL FIRE officials.

You announced you'd be the first team to register for that competition, and then Peter Hulan and I came and toured your facility. You showed us the technology you were going to use for that.

Yep.

I was like, “Holy shit, this is amazing.”

And it's worth noting that all the tech you saw—not getting into the details—was stuff we developed entirely on our own dime. That wasn't something where the government was paying us to build it. It's because we believed it was the right thing and the right solution, and so we invested our own money. We're betting that eventually we'll be able to make it work. I won't always be right, but if I flip enough coins, enough of them will come up heads that it works out.

And one of the things you said on stage then—we've just gone through these hellacious wildfires. My family and I are still out of our home. We'll hopefully be back in the next couple of weeks, but we're lucky. So many thousands lost their homes in a quarter of a trillion dollars—what is it, $250 billion or thereabouts—in damage, probably not even accounted for fully. But one of the things you said was—

There's $50 million in damage, and they can't seem to find a few million dollars to do controlled burns. It's really interesting.

Yeah.

Anyway, sorry. Continue.

No, it's insane. It is. It's a set of perverse incentives, and the insurance industry is broken. Don't get me started on that. If you want to start a third company, let's talk about reinventing the insurance industry. The insurance industry should be: We insure to make sure your home never burns down. We're going to protect your home.

Prevent it.

Life insurance keeps you alive. Health insurance keeps you healthy and keeps you from getting sick, right? That should be the reinvention of our insurance industry. One of the things that you said on stage at that press conference in D.C. was that you felt that, at the end of this competition, or with the technology that you are creating, this could be the end of wildfires.

I'm going to be a pedant here—a pedant of destructive wildfires.

Destructive wildfires. Yeah, not planned.

The ecologists were really insistent on this. I remember—

Yes.

Prepping for the speech—

Yes, yes.

You can't say the end of wildfires, because wildfires are a natural force that is healthy and good for the—

Not in the Pacific Palisades.

Yeah, not in the Pacific Palisades. No. The end of destructive wildfires.

Yes. Do you still believe that?

My press training from years ago is coming back to me. I absolutely still do. I do still believe it. I think you're not giving yourself quite enough credit here. You were trying to make the Wildfire XPRIZE challenge happen for a long time before it actually happened, and there was a lot of resistance, even from the governments and government agencies that did end up getting involved. There just wasn't an interest. I think there wasn't a belief that technology could solve this problem. It was easy to say, “You're just a bunch of techno boys with your techno heads and your techno keyboards, and you just type on your computers and do your techno stuff.” They didn't really believe that could be part of the solution. I think now people are finally figuring it out.

It's slow.

You convinced a lot of people on that.

Thank you. It's slow.

And I think the impact of these wildfires makes clear that we don't have time for business as usual.

We don't.

We don't have money for business as usual. We have to try something.

One of my dear friends and coauthor, Steven Kotler, showed me the data that in the next 20 years, the Northwestern United States is going to burn. The amount of dry kindling in the forest has exceeded 50%, and there will just be continual burning. Some of that needs to be addressed with controlled burns and taken care of, but we should have our towns and cities protected. You have some tech to show us on video, I think. I think we have Roadrunner.

What's going to happen?

Tell us what we're about to see.

We're seeing a few things. I recognize what this is. This is the end-of-2024 Anduril sizzle reel from the Anduril holiday party that our team must have sent your way.

Yes.

This is a CCA, a Collaborative Combat Aircraft. It just got its official designation from the Air Force a few days ago.

It's beautiful.

It's FQ-44—F for fighter and Q for unmanned. It's the first unmanned fighter jet. That was our Ghost surveillance drone. There are a bunch of those in Ukraine, and a lot of those with the U.S. forces. It just won a major contract with the U.S. Army for MRR. This is Anvil. These are in service on U.S. military bases all over the world, protecting bases from drone attacks. These are some of our ASTs, Anduril Sentry Towers. They're on the border, on military bases, and on critical nuclear and other energy infrastructure all over the world. We're covering about 35% of the U.S. southern border right now with those, actually.

Wow.

This is Menace. It's a mobile command-and-control vehicle.

I love your names, by the way.

Thank you.

They're great.

They're pretty good. Half the names are good, and half of them are code names that the engineers always get to pick. The customers get so attached to them that they never change them. One example is Roadrunner, which was competing with a Raytheon project called Coyote.

That's great.

And this is one of our Dive-LDs, one of our smaller submarines. It can do a lot of things that used to be exclusively the domain of a manned submarine, and instead you can do them with an autonomous system.

We have a much larger version of that called the Dive-XL, and unfortunately I can’t show you that on a video yet. But you’re going to be able to see it in the next few months. This is an Altius 600. It can be launched on the move. You can carry one on your back, in a backpack, and launch it. We actually just sent another big plane full of these to Ukraine.

We sent a bunch of these to Taiwan, and that got me sanctioned by China, so I’m going to go to prison if I go to Hong Kong or China now.

Oh, gee.

I’m also sanctioned in Russia and Belarus, so there are all kinds of places I can’t go.

Four countries down, 50 to go.

Four countries. I’m going for Iran next. I think that’s pretty good.

That’s beautiful.

Yeah. Then there’s Roadrunner.

And—

What a beauty. Twin-turbo, twin-thrust-vector turbojet, vertical-takeoff-and-landing micro fighter.

Now, is Roadrunner what you’re going to use on the Wildfire?

You’ll have to wait and see.

Uh-huh.

You’ll have to wait and see. I can’t give away my secrets.

Okay.

We’re collaborating with some of the companies that are in the competition.

Awesome. Love that.

But others, we’re just going to destroy them.

Crush. Yes. Give it up for that. But by the way, something you just said that is important, and that I love about XPRIZE, is that while we run this competition, we also create extraordinary collaboration between the teams. Teams merge, teams partner.

Anduril is teaming with a bunch of other companies that are there. I think I actually said this in D.C. when you were announcing it: I suspected the winning team wasn’t going to be any one of these companies. It was going to be a consortium or collaborative effort between companies.

I’m a huge believer in specialization of labor—not just at the company level, but even at the human level. I am a generalist, so maybe this is me fetishizing what I am not, but there’s a lot of value in people becoming deep experts in exactly what they do. We’re partnering with companies that have deep expertise in parts of this problem that we don’t have and have no intention of building, and vice versa.

We’re doing things at Anduril that some of these other companies don’t want to do. So I think that’s actually also the healthiest outcome. The healthiest outcome is going to be a lot of different companies all working to solve this problem.

Love it. Love it. And then he said the companies he’s not partnering with, he’s going to crush. Okay.

Only the bad ones.

Only the bad ones.

The bad ones. I like efficient markets.

7. The Four Rules For Big Bets

All right, I’m going to ask one more question before we get to your questions. We have a lot of them, but one more because I want to get to your questions here. How do you decide what to go after and what not to?

As a person or as Anduril?

As Anduril or as an entrepreneur. Let me put it that way—as an entrepreneur. You must have a thousand different ideas, lots of approaches.

Sure.

What’s your calculus for filtering these?

Personally, when I started Anduril, it was my first round, like I said earlier. It was me—

Yeah.

Doing my hobby. The second time around, I wanted to prove—it was a combination of truly wanting to impact the world and wanting to prove to everybody that I could still impact the world.

How important is that ego drive, right?

Massive. Absolutely massive. I need everyone who wronged me to weep. You know, Conan: “What is best in life? To crush your enemies, to see them driven before you, to hear the lamentations of their women.” I want it all. I want it all.

In deciding what I was going to do that was going to be impactful, I was deciding between either fixing the defense industry, solving obesity, or solving the prison crisis in America. I decided on defense for a long list of reasons we could talk about some other time. When Anduril is trying to decide what problem we go after, it’s a lot easier.

Yeah.

It’s a lot more rational and less emotional. We have kind of a 4-part test for us before we work on something seriously.

Yeah.

First, it has to be something that the Pentagon deeply cares about. That means that it can’t just be a thing that someone somewhere in the bureaucracy is technically tasked with doing. You need to pick their top problems. What are the things that keep the Joint Chiefs awake at night, afraid that America is going to fall?

Those are the problems that I want to work on—things like industrial capacity for rocket boosters because of the lack of manufacturing in the United States, and things like our lack of long-range fighters that can actually project effects and sensors deep enough into enemy territory to matter. You want to work on things that are going to be a big part of solving big problems, because if you’re not, people aren’t going to help you cut red tape and push down boundaries. You need to work on important stuff that they want to help you work on if you’re going to move fast enough.

Do—

So that’s one: the Pentagon has to care.

Yeah.

Two, Congress has to care. This might not be true for your business, but for me, I have to recognize that Congress has the power of the purse. I can spend my own money developing things, but at the end of the day, Congress decides what gets money at scale.

If you are working on solving a problem that they don’t believe in, you are never going to get significant money. They’re not going to tell you that. They’re going to meet with you, and it’s like when there’s a girl who’s turning you down nicely and she says, “Oh, that’s very nice, very nice, very nice,” but then they never talk to you again. You have to recognize that the nice words are not a reflection of reality. It’s just them being nice.

I’ll give you a concrete example. If you’re working on stuff related to counterterrorism right now, it’s just not what Congress cares about. They are worried about a great-power conflict against Russia, China, or Iran. They are trying to figure out how we’re going to fight a war in the Pacific on the other side of the world and win. Stuff that looks a lot like the wars we’ve already fought and already won or lost, it’s just not of interest to them.

And then the last two things that we have to answer are: Is it something we can do well? That sounds obvious. You should only do things you can do well. But that’s really a call for Anduril—for us—to do more. There are things that we can do today that we never would have been able to do well 5 years ago, 6 years ago, 7 years ago.

Mm.

You always want to be growing as a company so that you can do more things and go after things that fit in those previous 2 categories. We couldn’t have built an autonomous fighter jet 8 years ago when we started the company.

It’s pretty fucking—

And now we’re beating Boeing and Lockheed—

Yeah.

—and Northrop Grumman doing the same.

Amazing.

Again, the last one is: Are other people already doing a good enough job? I don’t want to be in the business of using my investors’ money—

Yeah.

—to crush other companies that are doing a quite competent job. Even if I could do better, why would I spend my life achieving marginal gain over other American companies that are going to get the job done reasonably well?

I want to build things that wouldn’t exist otherwise, or kill companies that deserve to die. So that’s what it is: The Pentagon has to care. Congress has to care. We have to be able to do a good job. Other people have to be doing a bad job. If it fits all 4 of those categories, then you’re going to see it in the Anduril showroom within a year or 2.

Less than half the products we make are even on our website right now. The things we're doing go beyond even what's necessarily public.

8. Skills Matter More Than Dreams

I'm going to ask one question. What piece of conventional wisdom in defense or tech do you think is completely wrong?

Follow your dreams. It's the dumbest shit I've ever heard.

Yeah?

I know it worked for me, but the reality is that most people are going to do better off following where they can have the biggest impact. It's following your skills, following your talents, not your dreams. A lot of kids have stupid dreams, and a lot of people have dreams that aren't going to impact the world.

When people say, “I'm going to tell my kids to follow their dreams,” my kids would probably have stupid dreams, at least at some point in their life. I'm not going to tell them to follow their dreams. Are you crazy?

YouTuber, video game.

Yeah. In 1969, do you know what the number one job kids wanted was? You can probably guess.

Astronaut.

Astronaut. Fantastic. They were like supermen: fighter pilots, PhDs, public-speaking heroes. What's the number one job today? It goes back and forth between YouTuber, professional gamer, and streamer. It goes back and forth.

I would say conventional wisdom that people say, “Follow your dreams”—I think it is dangerous. It is bad. It would lead to a nation full of people not having impact and not taking care of their families. There are even people who say, “I'm just going to do this. I'm going to do something I hate, at least...”

On average, people will not make enough money to get by, and they will not be impactful. That's a bad thing. If you are not passionate about any of the things that you're talented in doing, you need to get better at doing stuff. Go find something to be good at or find something to be passionate about. You need to change yourself, not—

I love it.

You don't just fall out of a coconut tree and then go and do whatever you feel like.

Let's go to Yas. Hi, Yas.

Hey, Peter. Thanks for having us. Thank you so much for coming back. I was a huge fan 2 years ago. Still a fan.

My question for you is, I really admired what you were doing for Ukraine at that time, and I'm curious about what the new administration means for your company.

So, with regard to Ukraine, or generally?

Generally is fine, but however you feel comfortable.

Generally, the ceiling for positive change is much higher. The point that I try to make with some of my friends who are more left-leaning—and I was fired because I gave $9,000 to an anti-Clinton group, so you can guess where I fall—is this: I argue with some of my well-meaning but more liberal colleagues, as Reagan would say, and I say, “Look, whether or not you agree with individual decisions, the variance day to day, the ceiling for positive change is certainly much higher.”

I'm actually quite optimistic. I think things are going to happen that never would have happened previously. I think we will be able to cut spending in a very big way. I think that will force people to tighten their belts, think hard about where we're spending money, and develop more effective techniques. Necessity is the mother of invention.

If there are things that our government needs to be doing and they find themselves with less money, given competent people in the role, they will figure out how to innovate and do better. We've always done that as a country. There have never been catastrophic failures in our country because we couldn't figure out how to be more technologically savvy. It just hasn't happened. Anything that we've actually set our minds to, we've done, whether it's going to the Moon or trying to build better databases, so I think we can do that.

With Ukraine specifically, I hate to abdicate responsibility for this. We've had stuff from Anduril in Ukraine since the 2nd week of the war. I met with Zelenskyy before the war, and I met with him again in Kyiv during the war. I've been to Ukraine to help train operators in how to use Anduril's weapons system. So, again, you can probably guess where I fall, but it is not appropriate for me, in my opinion, to weigh in.

People ask why I'm not tweeting about this: “Why are you not tweeting?” I tweeted about us sending them more weapons. “But why aren't you tweeting more about what should be done politically?” My answer is simple: Because I'm the executive of a weapons company making money selling weapons to Ukraine, to the United States for Ukraine, to the United Kingdom for Ukraine.

Aren't we supposed to hate it when the military-industrial complex advocates for longer, more extended wars in a way that clearly benefits their pocketbooks? My point is, whatever my opinion is, I'm not the right guy to be telling the message. I don't think it's the place of weapons companies to be weighing in on what conflicts are appropriate, how long they should go on, or what quantity of weapons we should send. I just don't.

And so people say this, too. They're like, “Palmer, would you work with this country? Would you work with that country? Would you enter this war?” My point is, you better hope that that decision is not made by me. You better hope we're not moving into a dystopian future where corporate executives de facto control U.S. foreign policy and military policy.

Because if you believe in democracy at all, then those decisions need to be made by civilian leadership that is accountable to the body politic, not me. I'm not accountable to anyone. My board is 3 people, and I control all of them.

Great.

So that'd be my look—big picture and little picture.

Good question, Yas.

Thanks.

All right. Thomas, what do you got?

Thank you. Thank you. It was very enjoyable. Exoskeleton aside, I kept thinking of Tony Stark when I saw that video, so you remind me of him.

A private question I'll try to ask you later, but in your bio that I read, you talked about your inspiration when you first started out to form Anduril.

Yeah.

You were helping veterans with PTSD. Could you share a little bit about that and how that happened? I'm a huge believer we've got a big problem there, and psychedelics can help. A lot can help.

So for about 8 months before I ever started Oculus, I started building virtual reality headsets when I was 15 years old. I started Oculus when I was 19, but in between there somewhere, I worked for about 8 months at the ICT Mixed Reality Lab, which is an Army-affiliate research center working on an Army program called BRAVEMIND.

BRAVEMIND was doing a lot of different things. Conspiracy theorists say it was a brainwashing program. It wasn't. It was a program to treat veterans suffering from extreme PTSD using virtual reality exposure therapy. By exposing them to things that trigger them, you could train them to engage in coping techniques, thinking techniques, and biofeedback techniques that would mitigate their physiological responses. In doing so, you could reduce their dependence on medication and improve their quality of life.

I can't take any credit for the success of the program. I was a lab technician. I was a cable monkey, a monitor minder—all kinds of names they came up with for us. But the people who were doing the real work on that successfully shepherded it from clinical trials at 1 VA hospital to 40 VA hospitals across the country. It's a great example of how technology can help people if you apply it in ways that, to a normal person, may...

People thought it was crazy, but it ended up having a better impact and better results than any of the medical interventions or any of the pharmaceutical interventions. It was fantastic. There are so many areas across the government like that, and you mentioned psychedelics and novel substances. I’d say the thing in common between these is that sometimes things that seem crazy to the existing bureaucracy are, in fact, the right solution.

The problem is that no bureaucrat ever got fired for doing the same thing his predecessor did, right? The number-one job of most bureaucrats is not to get fired. So we need to normalize doing crazier things. We can do them somewhat responsibly, but I think we can afford to borrow from science fiction and at least give it a shot.

Thank you.

Let’s go to Craig in the back here.

During the last 8 years, have you seen any of the major defense contractors actually be able to make change, or are you seeing them continue to do what they’ve always done?

Some of them are definitely engaging in change, but it’s a matter of speed and extent, right? They’re not totally static. They are somewhat changing. I don’t think they’re changing fast enough.

If you look at their revenue streams, they’re not actually dominated by new procurement. If you’re a company that’s been around for many decades and you’re supporting platforms that the United States has spent tens of billions or hundreds of billions of dollars procuring, you’re actually making more money off the things that you’ve done for the last 20 or 30 years than anything that you’re doing in the next 10. That drives incentives. It drives how they think and how fast they can move and react.

You’ve got to remember that, at the end of the day, companies are the product of their shareholders. Shareholders can be defined in a lot of ways. Some people literally hold physical shares. Some people are the employees; they have a stake in the company whether they own shares or not.

In the case of these major defense companies, their investors don’t want them to be like Anduril. Their investors want them to be an ultra-low-risk extension of the United States government, akin to a bond in terms of risk, that will continue to exist even if, let’s say, a COVID Lambda variant comes along and wipes out consumer spending for a couple of years. That is the asset category they fit into.

Suppose I were the CEO of a major defense company and I were to announce that we’re going to be like Anduril. We’ve seen the light. We’ve seen the way. We’re going to be a defense product company. Instead of putting 1% of our revenue back into internal research and development, we’re going to put 100% of our revenue back into IRAD. You know what that CEO’s going to say the next day? Nothing. He’s already been fired by the board. That’s not what his investors hired him to be.

That is actually the biggest challenge. Before we go to the mic, could I just run through some of the ones that are on the screen real fast?

Yeah, sure. Go for it.

All right, we’ve got, number one: Are we militarily ready enough? No. What is next for ModRetro and the Chromatic? For people who don’t know—oh, hell yeah. You have my second-favorite color.

For people who don’t know, I have a side company called ModRetro. It was a forum that I started when I was 14 years old. We started a project modifying vintage game consoles with modern technology.

Some of my buddies and I had been working on a project for 15 years to build a clone of the Nintendo Game Boy Color. We finally finished it last year, started selling them to people, and open-sourced it. If you want an open-source clone of the Nintendo Game Boy Color with a sapphire screen lens and magnesium-aluminum alloy chassis, there’s one. Look, it’s beautiful. Would you like one? You can’t order it. It’s not for sale. They’re completely sold out.

I’m glad that you got it. What is next? We’re going to be doing a Nintendo 64.

Oh, wow.

Yeah. How important is the human in your future work? Very, very important. We can’t automate people entirely. Will you need employees? I think so.

How do you differentiate yourself from the competition? I think we’re just very differentiated. Mostly, we struggle to convince people that we’re not too crazy. We are very different. We’re so fundamentally different at a product level. I don’t have customers coming in and saying, “So what makes you different?” It’s not my problem.

They can hire the same talent and tech know-how, but your end products out-innovate them.

Yeah.

I think this is what I got to before with that last question.

Yeah.

They could, in theory, hire the same talent and tech know-how, but people don’t want to work for companies that don’t put their own money into things or that try to drag things out. They don’t want to work for low-risk companies like that. It’s a different type of person for a different type of role. Anyway.

Let’s go to John Battaglia on Zoom. Hey, John.

Hey, great information. Wow. Unbelievable. My question is about your creative process. When do your big ideas come to you? Are they in a dream state? Are they when you’re in the shower? Is there a consistent way they come to you? If so, how do you get yourself in that state?

I steal all of my ideas from science fiction from the ’60s and ’70s. That’s mostly what I do.

In fact, I had a friend of mine, Art Dula, who used to literally read all of Heinlein’s stuff.

I’ve read every novel he’s ever put out.

Yes.

And I steal everything.

Yes.

He can’t do anything about it. He’s dead.

Yes, he is. But amazing stuff. He was a brilliant designer.

That’s okay. I mean, you read about stuff like this. He published a piece in, I think, a 1945 issue of a serial called Astounding Science-Fiction. It was a short story about fighter jets and space fighters piloted by intelligent AI that flies alongside human pilots. Now, that is what we are building today.

There are ideas that he lays out for how he believes you should communicate with and personify AI for piloting ships that we have copied into our products.

Love it.

It’s not just him. My job is to look at problems in the world and find the best solutions, but I need to first come prepared by knowing the solutions that people in the past who have thought about things very deeply and thought about the future deeply have already come to.

I’m not going to be able to do as well sitting on my own in a room thinking about what the future could be as the combined works of the top, let’s say, 1,000 science-fiction authors over the last century. They’ve had a lot more time to think about it.

A lot of concentrated time.

They’ve been able to think about the first-order effects, second-order effects, and tertiary effects. I could just rip off all their thinking. Of course, they get it wrong because some of them are trying to tell stories.

This is also why I like to rip off science fiction that’s old, because new science fiction is largely not as concept-driven. Books were this way for a while. There was a period in the ’70s and ’80s when it was the “take a piece of technology and that is the show” genre, like Knight Rider—

Knight Rider, exactly. That’s what I was going to say.

—and Airwolf—

Yes, yes.

—and The Six Million Dollar Man, or even RoboCop. It’s like there’s this thing, and it’s a tech thing. All right, so that’s the story.

That genre has gone away, which is unfortunate. I want it to come back, where they’re like, “So what’s the pitch? What’s this movie about?”

What’s the pitch?

“Oh, it’s a car, and it’s really fast because of this new technology.” You’re like, “Oh.”

Let’s go to one of our faculty members, Gil Verdun.

Hey.

Hey, Gil. Welcome back.

Speed Racer.

Thank you.

Speed Racer. I love Speed Racer.

Yeah. The car’s fast.

Yeah.

That’s the story.

Gil, AKA Beth, here. I think we follow each other on Twitter. First of all, big fan. I want to thank you for the vibe shift you started for all deep-tech and defense-tech founders. I’m personally working on AI chips to win the chip war, and I think your chip-on-the-shoulder energy definitely resonated with me as well, having gone through the media swarm attack myself.

Supposedly, I’m building Skynet or something. I don’t know. But what’s—

If you can believe it, I was a journalism major before I dropped out. I could have been one of them. I’m kind of like one of those Terminators that turned good.

You’re Schwarzenegger.

I know how to spin a story and tell a narrative and twist the truth, but this time I’m back for good.

I’ll be back.

Awesome.

Awesome. Yeah, a lot of the vibe has shifted for this new generation of founders, but there’s still inertia from the media and from the venture capital community to invest in deep tech and have some actual courage. What sort of advice would you have for this new generation of founders in deep tech and defense tech, in El Segundo and beyond?

Control your narrative. You don’t need to work with the press.

Yeah.

I think—maybe I’m going to be wrong. This is probably the thing I have the least conviction in, so it’s a crazy opinion that I have about 60% conviction in. All of the media companies, every single one, is running on either fumes or, in some cases, a half-full tank of gas left over from when interest rates were near zero. There was a lot of money put into tech companies, media companies, and everything that made no sense at all.

They’re all trying to figure out what they’re going to do about the fact that their businesses are terrible, nobody likes them, and they’re irrelevant in modern society.

Right.

That is what is driving a lot of the vitriol. I think they know the numbers. They’re not putting the numbers out publicly in most cases, but they know the numbers. When they see more and more people getting their sources from citizen news sources—you know—

Mm.

I’m not just talking about on X. I subscribe to a Patreon called Inner City Press.

Mm.

He’s a fantastic independent journalist who covers mostly things in New York City courtrooms, way better than any mainstream press outlet. There are a lot of Substacks that are the same, and a lot of YouTubers who are doing a great job. They’re seeing all this happening and realizing that things are coming to an end. So I would say, just don’t worry about it. Wait them out. You’ll be here when they’re gone.

Yeah, I love it.

Thank you.

Rickard.

There was a question: Given that so much innovation comes from defense—oh, no, it disappeared. It was, given how much innovation there is in defense, are you going to sprinkle some of that into the civilian world?

The answer is no, not really. I set out to start a company that would solve our national security problems. In fact, I’ve purposely avoided doing things that I knew were going to do better on the civilian side than the military side. There’s nothing wrong with the civilian applications, but it’s not what I set out to do.

When you run a company, you need to be focused. To be clear, I think I could make more money if I just focused on where I need to make money, but I make enough money and control my company sufficiently well that I can afford to leave money on the table and do things that I want to do. I want to work on national security problems, and so I will.

Are JCal and I on good terms now? Absolutely not. He’s a horrible person. Everyone laughs, “Ha ha.” No, he’s a really bad guy, and it’s really terrible how he’s still out there. On podcasts, he says, “Oh, nobody did agree with Palmer. I know all his board members, and I have a good relationship with...” No, he doesn’t. I think he couldn’t even name a single one of our board members.

He’s literally just lying to people because he can get away with it, and he’s surrounded by sycophants who won’t call him on it and say, “But Jason, you’re just lying. That’s not true.” They’re all condemnable. Anyway.

All right. Rickard.

Hey, Palmer. Thank you for all the great work you’re doing. Two years ago, we brought the dream of flying cars here to A360, and I think you met Tomasz Patan, the founder. We got some of the community here to invest. Last week, we had the first serial production.

But as we continue dreaming, do you think that the consumer dream of the personal aerial mobility sports car of the sky is the future? Or now that you’ve seen the other side, is it more government, first responder, police, ambulance, or potentially military logistics? What do you think?

It’s all of it. I’m a rotary-wing pilot. I own 7 helicopters, including a UH-60 Black Hawk. I really love vertical takeoff and landing aircraft, and I believe that eVTOL is going to come to pass. It’s taken longer than I wanted. I think that we wasted a lot of energy on purely electric systems when we should have been focused on hybrids.

Yeah.

Because eVTOLs aren’t actually designed to haul people around; they’re designed to haul batteries around, over and over and over again. The person gets to hitch a ride.

Yeah.

That’s where all your mass is.

Yeah.

So anyway, I’m glad we’ve stopped wasting time on that. I still believe that it’s going to happen. Now, there are obstacles in the way. I was at CES a couple of years back, and the head of the Los Angeles Department of Transportation said that the City of Los Angeles will not allow any new vertiports to be installed until VTOL transport is as cheap as public transportation, because she refuses to allow the billionaires to come in and compete, subsidize, and destroy public transportation.

Yeah.

Some of the guys from Bell Helicopter were on the same panel, and they were saying, “That’s a beautiful vision for the future. It’ll start expensive and get cheaper.” She said, “No, I want to be clear: no vertiports until it’s as cheap as the city bus.”

Unfortunately, there are a lot of dumb people in the world, and that is actually going to be our biggest obstacle for eVTOL. It’s going to come to places where—I think Dallas is going to be a place—

Yeah.

—where it turns into a hub pretty early. I think New York, obviously, because they already have heliport infrastructure. I totally believe in it.

That said, you might have seen that we partnered with Archer a few weeks ago. I think that a lot of the tech developed by the civilian eVTOL sector will have huge military applications.

Yeah.

This is one of those things: if people are going to design FAA-certified drivetrain and powertrain components that can go 15,000 flight hours without a single maintenance interval, I absolutely want to use that in my military—

Yeah.

—aircraft. I don’t think I need to rebuild it.

So it’s essentially a platform where you don’t choose. You go both ways.

If I were you guys, I would probably realize that—if I’m just making things up—search and rescue, fire, and police, that is the now.

Mm-hmm.

And the future is still eVTOL.

Yeah.

Because remember, government—and public works applications—have waivers for everything. You don’t even need to be FAA-certified. If you can find a public agency, you don’t even need a pilot’s license to fly—

Amazing.

—for a fire department. Do you know that?

No.

Or for a police department. They can put you in a helicopter that they build themselves with a pilot who has no training. Of course, they don’t actually do this. In practice, they buy surplus military helicopters or civil helicopters, and they’re mostly private pilots or people they put through their own training programs. But the point is, the waivers are there.

Yep.

That is the now. You need to hope that the FAA gets its act together before you run out of money if you’re going to make money on the civilian side. That’s the MOSAIC, right?

Yep.

The brand-new, all-of-everything modernization program at the FAA. They’ve pushed it back for what, 5 or 6 years now? You need to last long enough for MOSAIC to—

To get there.

—to get through.

Yeah.

And if you don’t, then you’re gone.

Thank you.

Love your chill outfit. What do you wear when you show up for government meetings?

I wear a suit. That is because—

Does it have Hawaiian patterns on the suit?

Not when I meet with the government.

Okay.

Suits are for funerals, weddings, Pentagon meetings, and Capitol Hill meetings. That’s it. It’s simple. It’s just a matter of respect.

Hi. Hello there. I’m an Australian based in London. Hi, Palmer. I really respect what you’re doing, and I remember getting the Wired magazine issue about Oculus in 2014. That got me into the whole metaverse, Snow Crash, and Neal Stephenson, so I bond over the science fiction, and I love what you’re doing with autonomous vehicles and defense.

My question for you is: in the perspective of Moonshots as our conversation, and how that changes over time based on cultural context and all these different factors, do you foresee a moment in your lifetime, or in your kids’ lifetime, when we’re no longer at war as a human species? I believe collaboration is more important than competition.

My science fiction book is going to the moon at the end of the year, and I got Frank White to write the foreword for it, which is “The Overview Effect.” The whole idea is that we’re one human family on Spaceship Earth, and we should start viewing it like that. I would love to see, personally, in my lifetime, a time where we don’t even have to talk about war as an industry.

I think that there will be a time without war, and then that time will come to an end. Human history—we’ve been fighting for way too long for me to sit on this stage and say, “Yeah, I think we’re just going to get over that.” It’s one of those interesting problems where, let’s say you totally get away with it: war stops. The longer you spend away from it, the less people believe it’s possible and the less you do to prevent it.

This is the theory of the end of history that existed pre-World War I, then pre-World War II, then before the war in Vietnam, and then before the war on terror. The globalized financial elites come to this conclusion every time where they say, “Well, we’re the most important people in the world, and so sternly worded letters from us and entanglement between our companies economically mean that large-scale conflict is impossible.”

I encourage you to go back and read people who wrote about the end of history right after World War I. They said, “World War I was so horrible, and now we are so unified. Europe is so unified that never again is war even possible. It’s not even possible to imagine.” They also said, “All the territorial disputes are so settled that nobody’s ever going to try to undo the borders.”

I mean, it happens over and over again, and I refuse to be the guy who’s quoted like that 100 years from now, where they say, “And here’s this idiot Palmer. He said, ‘I believe that someday war will be over.’” So, yes, I think there will be a period where war ends. I think it will go on for a long period of time.

I think we can build tools of deterrence that make—wars start when one or both sides disagree as to the outcome, when they disagree as to who will win and who will lose. When the outcome is relatively known, wars don’t start. I’m a big believer in either unipolar or maybe a bipolar or multipolar power. If you have a few powers in the world that are relatively at stalemate with each other and your interests don’t diverge too much, you can get away with no war for a long time.

But who’s to say that someone’s not going to come up with an asymmetrical advantage, a programmable virus that wipes out all of his enemies, and he decides that he’s going to launch an asymmetrical war and get a bunch of crazy people on his side? The last thing I’ll end this with is that war can come in a lot of forms. You can have nation-states, or you can have radical extremist groups.

Yeah.

You can have radical, violent religious groups. It would be very hard for me to imagine that never again there will come a group that is willing to die, lose, and consider that victory in pursuit of their extremist goals.

How do you deter someone like that who’s like, “Oh, I’m going to lose so hard. I’m going to die and go to heaven so good”? It’s just—how do you deter that? It’s very difficult.

All right. We’re running short. Jacob—

I’ll be efficient with my time so that we can get everybody.

Jacob and Carrie, short questions.

Yeah.

Okay.

This is controversial. I work in prosthetics. I work with a lot of active warfighters. They suffer a traumatic amputation, and the first thing they want to do is go back.

Yep.

Have you ever thought of militarizing prosthetics? We—

I—

It’s—

It’s—

Osseointegration happens.

I actually collect high-end prosthetics. So, I’m—

For a hobby?

Yes. So, the thing is, there are a lot of interesting—there are a few disciplines in the world that attract extremely competent people who really fight for every last bit of weight savings, power density, and materials science advantage.

One of them is F1 racing. Another one of maybe a half dozen in the world that is the best at attracting these people is high-end prosthetics. There are very advanced prosthetics that have the most power-dense actuators in the world, the best materials science in the world. They’re not too cost-sensitive. Small improvements in performance make a huge difference in the quality of life for these people.

Have I considered getting into that industry? Of course. I’m daydreaming about it. I look at some of these cool multicomposite and metallic, ceramic, carbon wonders, and I say, “Man, I would love to be doing something better.”

But this is actually one of those areas where my conclusion was that there are already people doing quite a good job. I think that the best I could do would be to make it somewhat better for a while, but then the other people would leapfrog me. It would be a continuous fight to be a tiny bit better.

I’m very interested in this space, and it continues to advance rapidly. But I think it’s probably not going to be one that Anduril gets into, for that reason.

Very good.

Thank you.

Thank you.

All right. Last question, Jacob.

Hello, Palmer. I hope this is a short question, but it’s a big one. Based on your experience in DoD, do you believe that there will ever be a time when the United States employs a fully autonomous global warfare system where we no longer fight wars with our men, but with our technology and our machines? If so, when do you think this will be, and what do you think are the technologies that will be within these future wars?

9. Humans Must Own The Violence

I don’t think it’ll ever be fully autonomous because the gains aren’t there, and the negatives are. Going from, let’s say, 1,000,000 people doing some task to 100,000 people through automation—that’s a huge gain. In cost, you can often do a much better job. Going from 100,000 to, let’s say, 1,000 people, maybe there are even gains there.

If you look at the United Kingdom, a smaller country than ours, they’re looking at reducing the size of their navy over the next 10 years by 30%, and that’s still significant for them. Going from 1,000 people to zero people, I don’t see the gains. If people are going to be responsible for violence, and if we are going to be responsible for use of force against other nations, against other people, there has to be a level of attention and responsibility—

The consequence—

—to commensurate it with the consequence, exactly.

Yeah.

Imagine if you had one person—not nobody, but one guy—who runs the whole war. You can’t actually hold him accountable for anything because you could say, “Oh my God, this thing happened that’s absolutely unthinkable.” And he’s going to say, “Well, of course. I had to do 1,000 actions over the course of an hour. I had to take out a million targets. Of course I couldn’t possibly ever dedicate any meaningful amount of attention to anything.”

That is what we need to avoid. We need to avoid outsourcing responsibility for violence to machines, to robotics. If we’re going to kill people, we need to kill people. And it needs to weigh on us.

I never thought I’d be clapping for, “We need to kill people.”

Hi, my name’s Palmer Luckey, and I build killer robots.