AI 如何改变战争｜Palantir 首席技术官

My guest today is Shyam Sankar, the CTO of Palantir Technologies. I think you’ll find by listening to Shyam that the business itself, their mission and how they built it, and indeed him in his life, are completely fascinating. We talk about his entire worldview, which I find personally the most interesting: this need to reindustrialize the United States after decades of moving so much of our production, and, as he would argue, therefore our learning of how to innovate and make things, off our own shores. We talk about the history of heretics, people who have, against the odds, designed solutions that have driven so much of our success around the world.

So, Shyam, I think the most fun place for me to begin, because it’s an area that you’re interested in and that I’m fascinated by, is the history of people in American military lore whom you call heretics. You refer to them as heretics: people like Hyman Rickover and Andrew Higgins, who designed 90% of the boats that landed at Normandy; John Boyd, who invented the OODA Loop and whose biography is one of the great biographies of a military figure that I’ve ever read.

Talk to us a little bit about how you would define a heretic in your use of that term and why you’re so interested in that group of people in U.S. history.

Well, I think they’re really founders. They’re founder figures, and they get obsessed with delivering something that makes no sense because they’re fighting, particularly in the military context, the bureaucracy. You’re going to end your career; you pay extreme prices for it. So, there’s almost a pathological obsession with winning, which I really relate to, both personally and because that’s what I see in great founders.

I think in the present moment it’s really important to encourage the hidden heretics who are in the military today to recognize that your heresy matters. Frankly, if you look back at history, the only things that ever worked, the things that helped us win all the wars, were the things that the heretics actually did. Nothing that went through the machine delivered anything.

1. Defining Heretics in US Military History

I think all change comes from these heretics, and they only later become heroes. If you think about Billy Mitchell, who invented the Air Force, he was court-martialed and died penniless and depressed. Only after his death did we invent the Air Force based on his original heresy. I’m just fascinated by what motivates these folks to keep doing this. I relish their rebellion a little bit, and the victory that the rebellion brings about.

If you could have people study any one heretic story to understand this kind of general idea, is it his story? Is it someone else’s? Which one would you pick?

I’d probably pick Rickover. Hyman Rickover was born in a shtetl in Poland and came to the U.S. when he was 6. He had one of these near-miss moments. In those days, when you landed at Ellis Island, you had 10 days for someone to come pick you up. His mother gave someone money to send a telegram to his father, who had already been here, and that guy just pocketed the money.

On the 10th day, there was another guy who arrived at Ellis Island who knew them and bought them one more day. Then he ran out, got the father, and got them. So, it was this near miss where this guy almost didn’t even end up here. He would have been sent back to Poland.

But he was this feisty character. He was, I think, 5'2", very short. He went to the Naval Academy and was so unlikable that, in the Naval Academy yearbook, they had torn out his picture. He was just one of these difficult people.

During the war, he drove a coal ship. He wasn’t some exceptionally high-performing senior military officer commanding a destroyer or carrier or anything like that. But after the war, he went to Oak Ridge, observed the vestiges of the Manhattan Project, and had this idea: we could build nuclear-powered submarines.

He’d had this experience on diesel-powered submarines, which suck. They’re basically surface ships that happen to go underwater for about an hour. This became something he was obsessed with. Talk about chutzpah: Oppenheimer himself thought this idea was going to fail. You have Oppie telling you you’re stupid and this isn’t going to work, and you’re just going to go forward, right?

So, he started on this project, and he built the first nuclear submarine in 7 years, start to finish. The Navy hated him so much that his first office for this project was literally the women’s restroom. They were like, “What can we do to humiliate this guy into quitting?” But he just kept going.

If you actually look at his personal memoirs, you can see that the humiliation got to him, but in a way that he could channel into motivation. It wasn’t like he was somehow immune to the pain of it all, but he could push through it. Every major project you do is going to push you in these sorts of ways. How do you dig deep? How do you grind through the pain?

He created a really unique culture in Naval Reactors, which is alive today. I think part of what’s really cool is that our nuclear submarine force is one of our last remaining asymmetric advantages against the Chinese, and we built it in the ’50s. What a legacy for this guy to have built.

The engineering culture that he created is also fascinating. He was in this unique role of both being a human who would invent the technology and therefore was qualified to think about how to regulate it. Whereas the Russian submariners would spend roughly 6 months at sea and then 6 months in Sochi recovering so their white blood cells could regenerate because the radiation shielding sucked, we’ve had no deaths due to nuclear incidents in our submarine force.

He designed it to a standard: “I’m designing this thing so my son could be in it.” It’s 100 times safer than what we believe the minimum standard to be. That sort of obsession—it’s not a manager’s view of the world. It is a founder’s view of the world.

One of his legacies is that he was an admiral for 30 years, and even as an admiral, he was very difficult. Zumwalt, who was the chief of naval operations for a while, said, “The Navy has 3 enemies: the Soviet Union, the Air Force, and Hyman Rickover.”

I remember reading the book about John Boyd. Three times, he was the best fighter pilot; then he helped develop the F-16; and then he came up with the military strategy concept that was instrumental in some of the U.S. wars in the ’90s. He was a bastard as well—highly disliked, a really bad father, highly disagreeable, but incredibly important in U.S. military history.

You don’t seem to have that kind of personality. You seem, by all accounts, very likable within Palantir and outside of Palantir. Say a bit about your own disagreeableness. Where does it surface?

2. Shyam’s Personal Disagreeableness

It seems like this is a field in which you have to have a very clear view of how things should be, which we’ll talk a lot about in a few minutes, and then this sort of disagreeableness, resilience, and persistence to make it happen. That seems a little bit at odds with your personality, from what I can tell. I’m curious how you think about that.

It surfaces in certain ways. As Karp would say, I’m not everyone’s cup of tea. Where you can very clearly see it is in the early days of Palantir. You’d be dealing with the government, and it’s easy to think about it as a monolith, but you have operators who want to use your software but aren’t in charge of buying it, and you have the IT people or the program office that’s in charge of building what you would be building.

So you're kind of a structural threat to them. The advice is to go along and get along: “Hey, those guys are the people paying you. Why don't you just do what they want?” The problem is, if I did what they want, it wouldn't work, and so that's where the disagreeableness comes in.

It's being ornery about, “No, I'm going to deliver the thing that actually works for the operator. I'm going to piss off literally everyone along the way.” Their knives are out; they're going to try to kill you. We had to sue the Army to compete at some point, and I think in that vein of being truly committed to: Is the product excellent? What else could it be doing?

You could think of the whole OODA loop for deployed engineering as continuously solving, through backpropagation, what the product should be. That's the motor of innovation.

3. Formative Experiences & Worldview

If I think about Palantir's story and success, and your part of that story, your personal worldview seems really important. Having a strongly held worldview seems like a key component of being able to do what you've done over the last 10-plus years.

I would love to hear the formative experiences that contributed to that worldview. I'm curious, at a high level, what your worldview is, but I'm maybe even more interested in how you came to it, because you've got a very unique personal backstory.

I think there's no singular event. I think it's a combination of family history, the environment that I grew up in, and so on. Starting with the family-history part, we fled violence in Nigeria, and we frankly almost died settling in the U.S. That happened pretty young in my father's life, in his early to mid-30s.

It's a pretty jarring experience that gives you a whole new lease on life after that point, where you think about the deep amount of gratitude and perspective you have on the trials and tribulations that may come ahead. But it also grounds you in a profound understanding of the counterfactual: What makes this country so special?

Sometimes it can be unmooring not to have those experiences, because it's easy to take it for granted, and it's easy to become cynical. But that's a really powerful rooting, in part because my father didn't have the classic immigrant story where he came here and became successful, which I think is a great thing if it happens to you, but it's also a facile journey. It's easy to appreciate.

My dad started businesses that went bankrupt. It was very hard, but having that rooting experience created a graciousness and perspective in him. He was always so thankful for the opportunity. I took that away over time, especially as I grew up and could understand and have awareness of what was happening.

The second thing I think is really important is growing up in Orlando in the ’80s and ’90s. It was a profoundly optimistic period. You could say it was probably pretty optimistic everywhere, but in the shadow of the Space Coast, you were bombarded with this idea that technology was going to make the world better, that there was a structural, positive-sum view of everything. It was so motivating, and people had ambition around these things.

I think that really set a frame for me for how I thought about things and how I wanted to spend my life and energy. I think that's really it. It's this commitment to building. Part of what's interesting about the heretics that I'm drawn to is that they're all builders of some sort. The conversion from heresy to heroism is a result of an empirical thing they built.

John Boyd was alive to see the success of his innovations in Gulf War I, where the U.S. destroyed the fourth-largest army in the world. It's hard to think about now because we think about it as almost a foregone conclusion, but leading up to it, there was a lot of uncertainty.

4. What Makes America Exceptional

That worldview is centered in an understanding like, “Look, I think the U.S. is the greatest force of good that exists in the world.” As a country, we understand that there's something special about founders. There's a reason we call them the Founding Fathers. If you think about it empirically, Europe has created 0 companies worth more than 100 billion euros from scratch in the last 50 years—an astonishingly bad track record. We've created all of our trillion-dollar companies from scratch in the last 50 years.

The primacy of people, and being in a culture and environment that allows that to happen, that doesn't subjugate the human, doesn't snuff out human flourishing, and allows you to pursue these things—that may not seem trivial, but the heretics are all a version of that. It's not Pollyannaish, where there's no resistance to these creatives. It's that despite the resistance, you can succeed here, which is really not true, I think, basically anywhere else.

What was your dad like, personally?

The pieces of his personality that always stuck with me were that every spare moment this man had, even though he was so busy and just grinding, he would be on the phone trying to help someone.

I had this weird memory of my dad. He had this sort of engineering autism that I think a lot of us have, but he met someone who had totally disfigured teeth. He had the profound kindness to tell him, “You should get cosmetic surgery.” That guy is so profoundly thankful that no one else had the balls to say that. This actually changed his job prospects. It changed his life.

My dad was one of these people where you could clearly see the positive-sumness in him: “Hey, I'm just going to keep paying it forward.”

If you had to identify the component parts that make America the thing that you think is the greatest country in the world, what are the key components? Obviously, maybe democracy is one, but I don't want to take for granted the things, given your life experience and your unique perspective, that you think most add up to this being a place worth your life's work to extend their interests, protect their interests, and so on.

5. What Does Greatness Mean?

A belief in exceptionalism and greatness. No army that lost its morale ever won the war. Believing in yourself, believing that greatness is possible, is a precondition to being able to express it and realize it. A lot of cultures don't have that.

The other thing, which I know sounds really simple, but is a complete plasticity of thought. Culturally, America is a place where people do change their minds: “Hey, this isn't working. Let's do something different.” Or, “I observe that thing working, even though it goes against the orthodoxy of everything I've learned. We're moving. We're pivoting.”

In Europe, people are still arguing about things that happened 300 years ago. Some of these cultures are so rooted in the past that they can't update their priors. A culture that's capable of learning—you could call it plasticity, the first derivative of learning—but how can you make progress without that?

6. Alex Karp

What does greatness mean to you in general? I'm also curious how you think about it personally. If you are personally in pursuit of greatness, what has that meant? What does that require?

An aspiration that's substantially bigger than oneself.

The focus of American greatness is on American prosperity and the American worker. How can we go do things that are actually enriching our civilization and our people, raising their own aspirations, and inspiring them to go on and do great things?

There's this unstable equilibrium in civilization that pulls you toward nihilism, and greatness is the antidote to that. It's the countervailing force to saying, “No, actually, it's worth investing in our institutions. It's worth going through this painful journey because what we're going to get on the other end is something better, something that we all aspire for, something that you will be proud to pass down to your children.”

Is there an example of greatness that you've witnessed in someone else firsthand that most stands out?

The obvious one to me is Alex Karp. I think his ability to really manage and unlock talent is something like I've never seen before. Those are some of the most profound lessons I've learned about Palantir.

7. How to Unlock Talent

From the very early days, he modeled our entire company as an artist colony. We have more in common with a Hollywood talent agency than we do with a typical software company in terms of our approach to looking at humans. It's this first-principles approach to the individual: How do you grow?

There's a deep resistance to a cargo-cult-like process in every manner or fashion that corrodes all the things that actually create value. I think that's important because if you drive that to the extreme, you actually get something that looks more like the Soviet Union.

On the other end of this: How do you maximize the potential of the people in front of you? You're advocating for their growth and their ability. That also involves a lot: You have to have a structural personality where you're not threatened by other people's successes. You actually relish in other people's successes, and that's the parallel between Alex and my father.

If I were running an organization where my key goal was to unlock talent in the way you just described, how do you and he do that? What's the practice of unlocking talent?

Maybe we could just start with the precept of how we think about the talent to begin with.

Really talented people are highly uneven. They're very good at some things, okay at another set of things, and really bad at some other set of things. Usually, the trick is figuring out where they are in their own journey of understanding what those things are. Particularly for high-achieving people, when they're young, they misattribute what their superpower is.

They think there's this thing they do that requires a fair amount of effort, and therefore they get a dopamine hit when they succeed. It's probably something they're not even bad at; they're just okay at. Superpowers are effortless, and therefore, in some sense, it's almost not even rewarding to exercise your superpower. My analogy for this is: if Superman could fly and see through walls, that wasn't some sort of arduous thing for him to do. It's just something he could do.

That's usually something you learn comparatively. You're like, “Oh, this thing that's lizard-brain effortless, almost thoughtless, for me—other people who are really smart can't seem to do, or I do just way better.” Embracing that is the first step. Then helping them understand, “Hey, all of your contributions to the world are going to come from your superpower. Everything else is basically a waste of time.” So, how do you figure out how to get into that configuration?

There's some releasing of ego around that, as well as understanding, “Okay, this is what I need to focus on.” The other part of this is kryptonite. There are some weaknesses you have that aren't just, “Hey, I'm kind of average,” or, “I'm maybe a standard deviation below average.” It's like you're 6 standard deviations below average here. It's not something you can work on.

8. Identifying Superpowers and Kryptonite

The only strategy Superman had around kryptonite was to avoid it. How can you be okay with that? As the artist, what is your process of accepting that, and how do we help you accept it? How do we help you understand that, actually, at the point of acceptance, you become more valuable? We're going to be able to unleash you further to do things.

Then you have to create an environment that supports them through that, because the discovery of kryptonite usually involves being exposed to it. You're going to make some sort of really bad mistake that probably will have real consequences. Hopefully, you learn from that mistake, and it may take more than 1 time to get there. But you don't want to create a culture that's like, “You fucked this up. I've got to fire you.”

It's like, “I'm so glad you learned that this is not what you do for the company.”

Yeah, exactly. I remember doing something that I really screwed up once that had consequences for the company, and I sheepishly went to Alex and was just completely honest. I didn't realize at the time there was no grand strategy behind this other than my commitment to just always being honest. He was very appreciative. He was also in pain as he internalized what this was going to mean.

But he valued the fact that I wouldn't try to hide it. I took away from that how important that environment was going to be, and I thought, “How do I scale that through the organization?”

What do you think your superpower is?

I think there's actually kind of a narrow sliver. If you thought about it in the Palantir context, it's the intersection of forward-deployed engineering and product. There's an interface there. It's like having one foot on both sides of that interface, owning that almost as a dictator. That's my superpower.

I'm going to come back to forward-deployed engineering, but I want to ask 1 or 2 more questions about the creation of an environment for the discovery of superpower and kryptonite, and doing one's best work and unlocking talent.

I heard a story once that maybe I wasn't supposed to hear. I won't say what the exact story was or who the person was at Palantir, but this person joined Palantir and was very quickly given an extraordinarily difficult project around which this person had no relevant experience. They were given relatively limited resources to solve a problem that was truly global in scale. The impact of this person not solving the problem would have been bad.

They described it as the most formative and important professional—and most stressful professional—experience that they had had. On the one hand, I loved hearing this story because, wow, you can learn so much when there are high stakes and lots of autonomy. But it also struck me as quite scary that someone with this person's experience was given this high-stakes opportunity and responsibility.

I'm sure that happens all the time at Palantir. Some of it seems like letting someone go straight into the deep end is a key part of this. In such a high-stakes business, with militaries and governments and people's lives and things, how do you balance how much leash to give someone like that?

Let me start with an analogy. How did Bruce Banner become the Incredible Hulk? It wasn't progressive overload. It's not like he lifted a little bit more weight every week, right? It was a near-fatal dose of gamma rays: a 50% chance he died and a 50% chance he turned into a big green monster.

9. The Gamma Ray Moment

There's absolutely an element where we're taking people and just irradiating them. Prospectively, you're not sure if they're going to come out the other end, but you have some belief that they have the raw talent and potential. Then you want to create an environment that has enough transparency, flow of information, and exhaust that, if you see this thing is really going off the rails without them having to seek you out, you're able to push in and help them.

A lot of environments don't have that by default, actually. You would have to manage them, have them do reports or check-ins or whatever. You've got to create an environment where it's okay to ask for help. The reason this model is really valuable is that I think the alternative model doesn't actually work, if you judge by the counterfactual.

This structured career ladder with progressions is perfectly designed to make you feel comfortable that you're growing. It gives you some belief that there's a linear path to follow and that you'll learn these things. But actually, it's all fake.

On the other hand, if you're willing to just suspend disbelief and throw yourself off the deep end, your maximal rate of learning will be coincident with your maximum ability to tolerate pain. Now you're in over your head, working on a problem you don't even know. This is the environment in which you're going to learn at the fastest possible rate. You're only limited by your motivation and ability to endure through the discomfort of getting there.

Then you will find that, if you look back retrospectively, you're like, “Holy shit, I grew in all these ways. I learned all these things.” None of which you could have predicted specifically ex ante. If you're just willing to do that serially—keep jumping off the deep end, go after problems you're frankly not qualified to go after, but you know there is some belief you have the raw potential to do—you turn into a superhero. I think that's actually why Palantir is the Founder's Factory.

Did you have your own gamma-ray moment? Was there an original moment like that for you where you discovered your own potential?

The article talks about the deployment at the Co-op. I think that was a seminal one, where I was leading the deployment. It was all on me, and it was never obvious until basically the last second whether it was going to work or not. It allowed me to learn a lot of the lessons along the way, which is really where I internalized that success looks like a lot of pain.

That instinct to figure out how to reduce the pain—how do I do this the next time with less pain?—is exactly the wrong instinct.

10. Palantir's Next 10 Years

If you were giving a commencement speech to a group of people that heard this conversation, how would you urge people to seek that out? If someone is pre-gamma exposure and interested in that sort of test for themselves, how can people expose themselves to that test?

I think this is the conspiracy of the people around you who give you that chance. I don't know why Kevin Hartz hired me as some Stanford grad student coming to talk to him, but the fact that he took the bet and just threw me out there—that was the first time I got to jump off the deep end in many ways.

I really think that, when you're trying to find it, you're assessing the people who are hiring you as much as they're assessing you. Is this an environment that I think is going to bet on me, or is this an environment that wants me to be a veal cow? Are they going to be able to tolerate my eccentricities, or are they going to try to squash that?

How do you know someone's worth betting on? The other side of that equation—new, unproven talent. How do you know?

Yeah. At some point, it becomes lizard-brain. You get enough reps in doing it. But I think early on, the early indicators would be first derivative, proof of some sort of sui generis: did they create anything from 0 to 1? Then there's an expression of agency and initiative.

Here's the simplest way I'd say it: if I give this person an inch, can they turn it into a mile? The worst sort of bet to make is, I gave this person a mile and somehow they turned it into an inch.

It's an incredibly powerful idea, and it seems like the culture at Palantir—I mean, obviously, there are tons of founders who have come out of this with mega-agency. If you think about Palantir's future for the next 10 years, now that you're 19 or 20 years in, what do you hope changes most in the next 10? Obviously, it's now a huge company, one of the bigger companies by market cap in the country and in the world, doing lots of really interesting work with lots of interesting customers. Generally speaking, what are your hopes for it over the next 5 to 10 years?

11. Forward Deployed Engineering

I think, culturally, because all value creation is downstream of the culture, it's about continuing to be an environment where a new person can come in and have huge responsibility, be in charge of people who've been here 10 years or not. That org structure—I call it a quantum org structure. I don't think a flat org structure is optimal. I don't think that actually works.

I think what you want is an org structure that can crystallize into the right structure to solve the problem as it exists today and is able to reform as the problem mutates tomorrow. You can see the pathology of big companies as they go through some sort of big reorg every 5 years. But even if you steelman it, probably most of them are wrong at inception. If you just steelman it and say, “Hey, they just reorganized. This is the right org structure for this company today,” every day the entropy is against them, and every day it's slightly more wrong than it was before.

If you're so ossified and committed to your structure that you have to wait until it's totally broken before you reorganize it again, that's a fundamental weakness. So, that plasticity, your commitment to working backward from the problems—I hope that's how we generate value, and that has to be consistent. If we can protect that and then engender that, promote that, promulgate that culture in future generations, we're going to keep winning.

What critique of Palantir do you take the most seriously?

When you have a structure like this, you don't get to spend nearly as much time with people as you'd want. So, the rate of attrition is very heterogeneous. There's some sort of bimodal distribution for how long people stay at Palantir. The probability that you'll stay for a decade-plus if you've stayed 3 years is incredibly high.

But roughly around the 3-year mark, everyone faces this crucible where they look around and say, “This place is so fucked up and crazy,” and they have to understand, “Oh, but that's the feature of it.” Then they'll stay.

Sign up for it or not, right?

Or they'll say, “This is the bug of it, and I have to get out of here. This is not going to work.” The critique I take most seriously is that, with a little bit of help, a lot more people can make that hump. That's the future of the business. It's those people.

We kind of have you to blame for how many freaking pitches we have to take that involve some forward-deployed something or other in a company these days. You were the one who really pioneered the success of this model at Palantir. Everyone is saying now that every company has to have forward-deployed engineers or forward-deployed so-and-sos to be successful.

Obviously, I think that's not true for every company. It's true in Palantir's case, for sure. Can you tell us the story of what this term means, where it came from, and why it's powerful?

We could start by thinking about my critique of the conventional software-industrial complex, which is that you're sitting, let's say, in Palo Alto and building this piece of software, and your essential feedback loop on “Is this software valuable?” is “Can I sell it?” If someone's willing to pay for it, it must be valuable. That's an okay feedback loop.

There's a stronger feedback loop, which is: Is this software valuable? Did it deliver the outcome? The only way you can assess that is actually in the field with the end user, on the factory floor, in the foxhole.

In our case, it wasn't going to be enough. I thought you had this weird division between the IT buyer and the operator. The validation is going to come from the operator. So, the idea was really almost like building software through backpropagation. How do you go to the person who has the problem?

You're showing up with your best thesis of what you've built in the metaphorical Palo Alto. But then you're not just sending a sales engineer out there whose job is to try to ring up the sale. You're sending someone out there who is obsessed with whether it's going to generate an impact in the world or not and is willing to continuously solve backward. This tech lead will have to understand the full stack. How do we take the specific and learn how to generalize it over time in terms of the product?

One way of thinking about this is that there really are 2 types of engineers. There are people who know how to build the right thing, and there are people who know how to build it the right way. Their dopamine hits come from different things, right? The people who know how to build the right thing are more like MacGyver. Their high in life is, “I solved the problem.”

In their minds, solving the right problem is 80% of the work. On the other end, you have engineers who are more like artists in the traditional sense: “Don't you see how beautiful the thing I built is?” The architectural elegance and the scalability of it—that's what gives them the high. Never mind that it doesn't solve a problem anyone cares about. “Don't you know, it's art on its own. It should be appreciated.”

I think about this like vector math. The forward-deployed engineers skew toward this hacker mentality. The product engineers skew toward this artist mentality. You have an unstable equilibrium where you're trying to align these vectors in a way that you're going to go upwind. If either side dominates, you don't really have a business.

That's why, at that seam between them, you actually need a dictator-like figure to manage and decide. Many of the outcomes—particularly, I think this model is super valuable when you think about your customers existing in some sort of power law in terms of how useful the next marginal thing they're going to be asking for is to the overall development of the product.

If it's a Gaussian, you should just survey all your customers and average it out and figure out what to build. That's a traditional product-development process. If, on the other hand, you think some percentage of your customers are living in the future, yes, only a handful of people have this problem today, but I can assess that this is actually a really valuable thing to solve that no one else has solved.

We should solve it now and effectively have a monopoly on the capability to do that. We'll be 5 years ahead of everyone else if we lean into this problem. You can only surface that in this sort of forward-deployed engineering way.

The other thing about forward-deployed engineers that's worth saying in the Thielian sense is the importance of secrets. Peter talks about the importance of secrets. If you're going to do something really impactful, there has to be something true about the world that you believe that other people disagree with you on.

Warp Speed is born out of one of those secrets. For the better part of 15 years, our software's been used to make things in the world. The one thing you could not tell anyone before COVID was that their manufacturing software doesn't work.

People would point at their ERP system and be like, “Look how great it is. I spent $5 billion, $10 billion doing it.” But as a forward-deployed engineer, when I would go to the final assembly line on the factory floor, everyone's using Excel.

12. Explaining What Palantir Is

So, if you're thinking, “There's a contradiction here that no one is somehow forced to reconcile with,” what they're really saying is, “The software is so ill-fitted for what I'm trying to do, I hit the eject button, pull the data out to Excel, and I'm hand-rolling my own solution.” That should be an obvious sign that something's not working.

What are the limitations of it, in the sense of: If you have X, Y, or Z, you shouldn't try this model, or you need this to try this model? One thought would be that the product needs to be expensive, because it sounds like an expensive process to deploy people on site—highly trained, expensive people, and so on. You need to be selling a very expensive product.

But if you were giving advice about using or not using this concept of forward-deployed engineering in your software business, how would you frame it up to other entrepreneurs?

I would say this: You have to believe that you're going to be able to capture meaningful value for the problems you're solving. The problems you're solving are really big and therefore imply there's a lot of value behind them. If you can solve them in a deep way, you'll be able to capture that.

To get at a lot of the problems we're solving, you're not going to get paid a lot up front. You're actually going to invest a huge amount into the customer to go do that, but you would do that because you believe that, in the end, we'll be able to capture the value—a portion of the value we're creating. Our economics will be downstream of our customers' economics.

The other part is that your ambition extends beyond a linear box. We always struggled when we came into the commercial world, where you had the Gartners of the world and they had, “Well, this is this sort of box. This is my enterprise architecture.”

I've broken it up into all these boxes. What box do you fit in? The answer for us is, “Well, we don't fit in any box. We fit in all the boxes.” It's some sort of heterodox thing. You're solving for all the problems in the seams. You're basically solving for the fact that that architecture diagram was made by some mid-level product marketing manager.

It's never been used or validated in anger. It's never actually solved a problem in the real world, but the cargo cult of the IT buyer just keeps believing they should buy it. I think if you fit into that, doing forward-deployed engineering would be crazy. Why would you do that? But if you're like us and you don't fit into it, you need some way of scraping and clawing to deliver enough value that the institution has to confront it, even though it's heterodox, even though it's heresy to my enterprise architecture, it works.

It probably took until I read Jeremy's profile of you and Palantir that I really understood how Gotham and Foundry worked, and how the product and the business worked. It's interesting how Palantir is somewhat inaccessible from the outside unless you read a fairly detailed description of what is literally going on in the product. I think the early years of Palantir were especially exploratory, trying to figure all this out with all your efforts with Alex and the team.

Could you just explain what Palantir is? It's such a basic, stupid question, but again, until I read a detailed account, I probably couldn't have done so. I'd really be curious how you like to frame it or explain it.

This is actually the hardest question in the world for me. We have built an enterprise operating system where our core thesis is that what makes this stuff really valuable is decisions, not data. It kind of goes to John Boyd's OODA loop. What we're really providing is an operating system that allows you to bring all of your enterprise data together.

It solves for the impedance mismatch: all this data I've collected was collected and stored in a way that makes sense to the underlying transactional system or makes sense to some part of my institution, but doesn't actually represent how the humans of my institution think about the problems that we're facing. So we represent that in what we call the ontology, which is not just a model of the data, it's also a model of the actions. We call that the kinetics. If you're going to make a decision, you might be allocating inventory. How do you do that?

This ontology layer almost becomes an API layer for the business. It makes the business programmable. That changes the rate at which you can compound on top of it. So now you have this abstraction layer that you can use to build applications that manage your value chain from the hand of your supplier to the hand of your customer. You can call it a value chain. You can also call that a decision chain.

Your whole business, almost, if you just squint at it, is just a series of decisions you're making. Every one of those decisions, you want to have an OODA loop. I want to make the best possible decision I can make today. But more importantly, I want the first derivative to be good. I want to learn tomorrow how to make a better decision than the one I had today.

13. The State of the US Military Today

Could you give an example of a customer where you showed up and maybe describe the circumstance of their data or their systems or something? Then what you did—what the process is of establishing and building the ontology, which is, I think, a very good, appropriate, but fancy word for mapping reality onto a system versus jamming reality into a preexisting system.

Yeah.

And then what sort of transformation that enabled at that customer that would have been impossible with the former system? Just to bring it to life with a simple example.

The first thing I want to put down that will make this make more sense is that we want to make software that makes you more different, not more similar. We think about it as alpha, not beta. The second thing is that we are a wildly inductive company. So we would never show up to a customer with some preconceived notion: “This is the aerospace ontology.” I'm going to pick an aerospace example in a second here.

It's rather like understanding all the values and understanding what makes your aerospace company different from all the others. How do I make you more different there? That's the differentiation you're competing on. So when we first came to Airbus, Airbus was in the midst of ramping the A350. The A380 was a wonderful plane, but a structurally unprofitable program. The A350 was this bet that the whole company was effectively riding on in 2015.

The first time you make an airplane, you're always going to have issues. There's a learning curve to the production. One of the most important things to figure out as quickly as possible is: What is an issue that is just a nonconformity? “Hey, I'll give this guy some training and he'll learn how to figure it out.” It's teething pain; it'll go away. And what is a recurring issue that is actually a design defect that I need to go work with my suppliers or upstream?

How quickly you can solve that becomes fundamental to the shape of the ramp and the cost of the ramp. So the first thing we did was just start sitting with users on the final assembly line in Toulouse. These are wrench turners, effectively, and we started understanding: What makes your life suck? What's hard about this? What's the challenge in working on these nonconformities?

Basically, they're going through Excel. They're looking at reports in SAP, trying to figure out: How do I bucket all these issues that are coming in? So we just helped them automate all of that. The ontology at that point was really related to parts, sequencing of work, defect rates—everything related to quality.

Once you had this immense quality asset, though, you were able to pivot it to realize, okay, so now I'm up the learning curve. It kind of naturally leads to the next problem, which is production planning. So quality to production planning to then in-service. Now you have the greatest data asset possible for this tail.

As this plane enters service, how do I help maximize the uptime of this asset for my customers? How am I going to compete against the other aerospace manufacturers? That's the natural evolution that we see with most customers. You start with one problem, but of course they exist in the context of this connected decision chain.

What's the difference between dealing with military customers and government customers, which you did for a long time, and commercial customers?

There are a lot more non-market forces in the military. The military is the only institution in the world that profoundly divides up supply and demand. The beating heart of most companies is that integration, that sales and operations planning process. You have the combatant commands, as we call them, that deal with real-world events. If you're fighting Russia, that's the European Command, EUCOM.

And then you have the services, like the Army, Navy, Air Force, Space Force, et cetera. Their job is to man, train, and equip. That's the supply side. So you have a group that has no responsibility for demand, essentially deciding what to build, how much to build, and how am I going to man, train, and equip this? Then they present these forces to the combatant commands and say, “This is what you have to solve this problem.”

That adds an additional level of complexity. So if the demand side says, “Hey, I really want this. This is the capability we need,” they can't buy it. How do you square this? I think you have all these other non-market forces to deal with.

But I think one of the things that is really advantageous to being both a commercial- and government-oriented company is that the intrinsic rewards on the government side are so high, so motivating. It enables you to get through and solve really hard problems in a way that is kind of like a rising tide for the whole business.

How would you define the state of our military today? I know it's a big question, but it's an important question. You've gotten to see so many sides of it and serve many aspects of it. How would you describe the state of the U.S. military?

We are the best military in the world. Our uniformed service members are better than we deserve.

Every day, actually, I have someone tell me, “Hey, when are you going to stop being shocked at how good this E-4 is at building this app or doing this thing?” The amount of talent is eye-watering. I think we encumber them with a lot of process. We deprive them of a lot of agency, and I certainly have a lot of critiques of the bureaucracy.

But I think the structural challenge we have is something that we need to observe from a distance, beyond the military we have today per se. The whole point of the military is to deter our adversaries from conflict. If you look back over the last 10 years, we haven’t been able to deter a lot of conflict.

We had the annexation of Crimea in 2014, the militarization of the Spratly Islands after they told President Obama, “We will not militarize it”—and then they militarized the Spratly Islands. We’ve had a pogrom in Israel. We’ve had the invasion of Ukraine. We have unprecedented gray-zone, phase-zero operations by the Chinese in the South China Sea.

I would say that, arguably, in the last year we’ve started to restore deterrence. Whatever you think about it, Midnight Hammer was a massively deterring operation. What happened with Maduro is massively deterring.

You have it on multiple levels with Midnight Hammer. It’s obviously the overwhelming strength, the precision of what we’re able to do, and the incredible complexity of the operations. No other military in the world could have done that. But we need to view that as a single strike, and the deterrence capability against a peer adversary is: Can you generate that strike every single day until the enemy submits to your will?

You think about Maduro. It also shows our capability. Literally, the raid happened the day the Chinese were in town. Maduro has all these Chinese and Russian weapons in Venezuela, and they don’t seem to work at all. That sends a very strong message to everyone else who thinks that their security is coming from buying Chinese and Russian weapons.

Now, I think we can trace back—and this is what I really wrote about in the 18th thesis. So, if we start by saying we’ve lost deterrence, let’s be clear-eyed about that. I think in the last year we’ve started to regain it. What happened? Because we had it for a long time. We were the sole superpower in the ’90s, and we spend a fair amount every year on defense. So I think it would be reasonable for the American people to expect that we—

Spending dollars right here.

Yeah. I think there’s something much more profound. People want to talk about narrow things like acquisition reform or the mix of weapons. I think it comes back to people. It comes back to the heretics.

At the dawn of World War II, we were the best at mass production. Effectively, our country invented it. In order to create an industrial base, we took the number two person at GM, who was previously number two at Ford, Bill Knudsen, who was a Danish émigré and really the perfecter of mass production, and we direct-commissioned him as a 3-star general and put him in charge of war mobilization.

Knudsen was a world-class engineer. What did we do? We gave him the authority to go meet other world-class engineers all around America. We had one key advantage in World War II in terms of timing, which is that we had Lend-Lease. There was a period of time when we actually mobilized production while we were not fighting.

We were building things that we gave to our allies, to the Brits in particular and the Soviets. That allowed us to build factories, retool existing factories, and hit a production ramp so that after Pearl Harbor happened, we were basically at full-rate production and were able to really go after these things.

14. How to Re-Industrialize America

Put this in the context of our adversary, the Germans. They were actually better engineers than we were. They just made a very small number of very exquisite things. Now, if you come to the present moment, we sell like the Germans. We make a very small number of truly exquisite—eye-wateringly exquisite—things, and our adversary is the world’s best at mass production. This should give us all a little bit of discomfort.

The second part of this is the nature of the industrial base. Today, we call it the defense industrial base. I think it used to be the American industrial base until the fall of the Berlin Wall. Only 6% of spending on major weapon systems went to defense specialists—that is, companies that only did business with the Defense Department.

Most of that spending, 94%, went to what I call dual-purpose companies. A missile is single-use; there’s no dual-use missile. But the idea that Chrysler would build missiles and minivans meant that all the R&D they were doing on production, materials, equipment, and machine tooling could be leveraged and was subsidizing our national security. We recognized that freedom and prosperity were two sides of the same coin.

We’re so far away from that world today. General Mills, the cereal company, used to build torpedoes and inertial guidance systems. Kodak, the film company, was also a key supplier to CORONA, which was our first spy satellite. This is the industrial base that won World War II, won the early Cold War, and really deterred our adversaries.

It’s only as a consequence of becoming the sole superpower that everything changed. It changed in two ways. One, that 6% number today is 86%. So, 86% of major weapon systems come from defense specialists only.

The other part of it is the nature of the industrial base. We think about it as Northrop Grumman. More accurately, it was Jack Northrop, Leroy Grumman, William Lear, Henry Ford, and Henry Kaiser. You had founder figures. They were their own sort of heretics. They were their own version of what we would recognize in the Valley today.

That’s where this talent was: in the industrial companies of America. We were betting on these people, not just their institutions, to deliver this sort of capability. In 1993, there was a very famous dinner at the Pentagon called the Last Supper.

We had won the Cold War. I think in a democracy, it’s very reasonable for the people to say, “We want to spend less on defense. So what exactly is our threat?” We slashed the defense budget. At this dinner, the Secretary of Defense told a subset—I think it was 15 of the 51 primes—that not all of them were going to survive.

The budget was getting cut. They were given permission to consolidate. Maybe some of them should exit the defense business altogether. Some of them might go out of business, and the government wasn’t going to save them. This set off a merger frenzy that led to consolidation from 51 down to 5.

Now, most people look at this and take away the wrong conclusion, which is, “Oh, this is when we lost competition. When we had competition, it was much more dynamic.” I think the much more profound consequence was that this was the moment of profound financialization and conformity in the industrial base.

You lost the crazy people. They went to tech. They went to other parts of our economy where there was enough positive-sum energy and growth, and it wasn’t just about dividends, buyback ratios, and financial engineering. It was about real engineering.

Why should we celebrate a Hyman Rickover or a Kelly Johnson? Kelly Johnson was the founder of Skunk Works at Lockheed. In his lifetime, he built 41 airplanes. That is insane, many of which we still fly today. He built the U-2 in 13 months. The timelines we’re talking about are just incredible.

You can’t look at that and think anything other than, “This is an artist.” This guy—what he was able to do was divine. So much of what I think we’ve gotten wrong is that those things were messy, chaotic, hard, and frictionful.

Over time, we’ve tried to install lots of process to make these things less messy and less chaotic. As a consequence, though, they end up not working. We need a little more crazy back to get these things to work.

How do we undo this? I’m reminded of the Churchill quote—I won’t get it exactly right—something like, “The United States always does the right thing after trying everything else first.”

15. Perspective on China as an Adversary

A lot of what’s happened is that we’ve shipped the capability to produce physical things off our shores because we optimized for a variety of things. It’s much cheaper, more scalable, and easier, and we’ve become primarily a country with all the market cap that rests on technology and IP and extreme talent—the small number of exquisite things, exquisite people, and exquisite researchers.

We’ve done an amazing job there, but we don’t make much here anymore. A lot of it happens in China and overseas. How can we reverse that? It seems so imbalanced and so daunting as to seem impossible. But of course, I know you think it’s critical. What is even the first step in being able to do this?

What you said is exactly right. It is daunting but critical. We have no choice but to lean into that and start. The best time to start would have been yesterday; the next best time is today.

Let me also make the case for why it’s critical, because I think we’re screwed if we don’t do this. The biggest lie that we bought from globalization is this concept of, “We will do the innovation, and they will do the production.”

What you have to recognize is that innovation is itself a consequence of productivity. To put it in terms that I think our audience would understand most directly, what motivated Google to do the research behind the “Attention Is All You Need” paper in 2017?

If you ask them, they’ll say, “Well, the team was working on an incremental 3% improvement to Google Translate.” You cannot think of something more banal leading to something more revolutionary than that.

If you don't have the input stimulus on improving production, you're actually not going to be able to capture the innovation. And that's literally what we're seeing play out in the marketplace today, which is like, “Okay, they went from making a battery to making the whole car.” WuXi went from being a cheap pair of hands for pipetting contract research in the pharma industry to actually now 50% of all clinical trials being drugs that are created in China. So, we have seeded the innovation.

It's not that it was taken away from us. We just seeded it because we had a completely incorrect precept. So now we have to reattack that. Now, we shouldn't do that. We should do this with some asymmetric advantage, and I think we have a couple of them.

I don't think it's true that we're not great at building things in this country. Look at Elon and the progeny of Elon. We do know how to manufacture things here, and we're going to have to think about manufacturing things in new and novel ways. In addition, it doesn't all have to be new. Ford and General Motors—these are world-class manufacturers, too. We have that capability.

But we also have new ways of thinking about this, with much deeper, vertically integrated production; thinking about the production process itself as something that's more iterative and software-like, that you're going to keep upgrading; and co-locating R&D with production itself. So, I think these sorts of tradecrafts are the things that we can lean on.

Probably the last piece I'd say is AI is David's slingshot here. If we can give the American worker superpowers, if you can make the American worker 50 times more productive than a worker anywhere else, you're going to change the efficient frontier of what can be made here. And I think, as a consequence, we should be reindustrialization maximalists.

The free traders out there like to say, “Maybe we can friend-shore,” or whatever. I think that stuff can let us off the hook too easily, and it fails to solve the problem. The other example I'll give you from Patrick McGee's book Apple in China—that's a great book, right?—is that people love to say something like, “Well, we don't have the machine-tools engineers in America. You could hardly get 2 people in a room to talk about it. You could fill a whole stadium in China.”

But then you look: Apple has spent the equivalent, on an inflation-adjusted basis, in the last 5 years, of 2.5 Marshall Plans building talent and capacity in China. How about we try to spend 1 Marshall Plan here? I think the problem is we're not even trying, in part because we're daunted, in part because parts of our economy don't recognize that this isn't going to keep working. We're going to lose everything, not just something.

It's why I feel the need for an urgent call to action to say, “We are in an undeclared state of emergency.” And it's not just manufacturing the physical things we think about. I even think about this in terms of drugs. 80% of our generic drugs come from China in one way or another, whether it's the APIs or the finished products themselves. That is a huge problem.

Today, your child has an ear infection. That's pretty trivial. You think, “I'll have some generic antibiotics and it'll go away.” If you have great-power competition with China and you're testing the will of the American people to fight, when they have to really contemplate, “Hey, my 5-year-old might die of an ear infection,” we need to take these things much more seriously and further down the supply chain.

I'm glad we're starting with rare earths, but there you could say we had to get punched in the face to mobilize around it. Maybe one of the galling things we should all think about is we've been staring at this rare-earths problem for the better part of 15 years, and we more or less have solved it for roughly 2 billion bucks. It's a rounding error relative to everything that we're doing, and now we're on a trajectory to do it.

So, I think part of it is just starting. I would love to bring everything back. I think making a marginal improvement of even 10% dramatically reduces the adversary's leverage.

16. How to Get More Heretics in Government

Can you tell I was just reviewing my notes ahead of our conversation today from Graham Allison's book on China, Destined for War: Can America and China Escape Thucydides's Trap? Can you give your perspective on how you think about China as an adversary, as a country? I highly recommend the book Apple in China. It really gives you this deep sense of how tightly integrated our arguably most important company is with manufacturing there. It's a very strange circumstance.

I have to admit, I've been to China. It's another place filled with human beings and interesting things, amazing, interesting culture, and all sorts of positives. It's a hard thing to wrestle with, right, that they're an adversary in the traditional sense. But I think there are components of that, and the nuance is important. I'd love to just hear you describe how you think about China as a political entity, as an adversary, as a country. It just feels like an especially important thing to understand your view on.

When we won World War II, we spent our own money to rebuild Germany and Japan. Microelectronics moved to Southeast Asia, as Apple in China and Chip War document, because we helped move it there. Yes, we got cheaper goods in return, but the economic development of these countries—their flourishing, in order to create global stability—was part of the calculus. The same thing happened with admitting China to the World Trade Organization.

The problem is, they cheated the whole way through. If you go back to a pre-Deng Xiaoping era, they called us the big enemy. I think part of it is our own naivety, our own belief that capitalism and human flourishing are a natural consequence of prosperity. So, okay, they're saying these things, but as they get richer, they're going to change their perspective on these things, and we're all going to find some way of having a positive-sum view of the world. That's essentially what didn't happen.

The challenge for us with the CCP is that it's not enough for China to be prosperous; America must fail. You can just see this through simple acts. If you look at agriculture, it is totally their prerogative. It's a business decision whether you want to buy American soybeans or not, or you want to buy Brazilian soybeans. I do not begrudge them one bit.

It's a totally different thing when you're trying to smuggle in agricultural fungi so we can't grow soybeans. Now, if you start going down the laundry list of what's actually happening, that's the shape of what's happening. They're a formidable adversary in the sense that our conception of war is kinetic. It's like the lone, heroic cowboy against the odds, winning through heroism.

Their conception of war is deception. It's the general who tricked the adversary into getting into the valley and then flooded it. You want to win without firing a shot. A lot of the things that are happening are below what we would consider the threshold of conflict, but it absolutely is conflict. It is in their system: systems destruction warfare. It's waging war through all the means, because they're going to wage it where they have symmetric and asymmetric strength.

Our American Calvinist sensibility is structurally positive-sum: we're going to turn the other cheek time and time again, over and over again. There will be some line, there will be some point at which we just react. When we react, the best quote I have from an admiral I like is, “It's no longer John Boyd's OODA loop—observe, orient, decide, act. It's the American OODA loop: observe, overreact, destroy, apologize.”

What we want to do is prevent this from happening by having enough deterrence that we're not provoked into a position where we even need to react.

You've mentioned the idea of asymmetries a few times, with nuclear subs being an example of one of our last huge asymmetries versus others. What are the most important asymmetries in your mind for the U.S. and for China? Where do they have an asymmetric advantage against us? Where do we have it against them?

This is a great question. Their principal asymmetric advantage is long-term planning. They actually can't pivot on a dime. So, let's go back to Gulf War I, which we're talking about with John Boyd. They saw the devastating violence we were able to do in basically 4 days to the fourth-largest army in the world, and they said, “Okay, we have to systematically map all of the critical dependencies of America's military technology and invest against it.”

They have been doing that since '91, and so the frog is boiled in some sense. It's a little bit of an easier game because they're playing a home game, but that's the map they've been on. Then it's been a steady journey, and if you look back, they broadcast their intent the whole time. There were never any surprises.

Our asymmetric advantage is that we're crazy. The AI phenomenon was not in their long-range plan. It wasn't in our long-range plan, but when it happened, our whole economy pivoted on it on a dime, right? Even the Chinese models are a result of distillation.

Our ability to be unpredictable, because we ourselves do not know what we're going to do, is our key strength. That comes back to the heretics. Where does that unpredictability come from? Because it's not our bureaucracy. Our bureaucracy is predictable, right? It's not the steady-Eddie managed decline; it's the crazies.

How do we get more heretics in this part of the American story? There’s obviously a massive profit and, I would argue, status motivator in the private sector. All these companies that are popping up with amazing founders—the potential rewards are extremely obvious.

The heretics that you described at the start of the conversation, that’s definitely not true. You mentioned dying penniless or whatever. How do we create the right set of incentives for more of our most talented people to go be heretics in this part of the American story right now?

There are 2 incentives. 1 is just recognition that we need the heretics. I think Hyman Rickover enjoyed his celebrity for a big period of time, and just being known as the guy who did this is itself a sort of reward.

Then the second part of this is definitely financial. One of the arguments I’ve been making is that the cynical view of the defense industrial complex, in particular, is something like, “These are greedy warmongers.” Well, they picked the wrong industry, because on the greedy side, you should be in tech or some other business. These are crappy businesses. They have, like, 9% operating margins or something. They’re valued at less than 2 times revenue.

So we actually need to make our primes more valuable, and that would help us continue to attract the heretics. How do you make the primes more valuable? The facile critique of the primes is somehow they’re the problem. The system’s the problem. The primes are just downstream of the consequences of the incentives the government gives.

I think the worst incentive is cost-plus contracting. Your upside is capped, you’re not supposed to take any risk, and the government basically pays for your R&D, so you have no skin in the game. You’re still able to make a small number of exquisite things, but that system is anti-heresy. You can only do the conventional thing that the government agrees you should do.

You think about the Higgins boat. Andrew Higgins built that boat on his own. There was no spec. That’s why the Navy kept rejecting it. It wasn’t the Navy’s idea. But then 92% of all boats in World War II were Higgins boats. The heresy is the boat that won the war. How do you create an economic incentive where that’s the case?

I think companies like Anduril are pioneering that on the hardware side, where it’s like, “Hey, I’m building a product. These are my specs. These are my opinions. I’m absorbing the risk. You should buy it whether you like it or not.”

Part of the reason for profound optimism right now is that I would not have started all this advocacy if I thought I was just screaming into the wind. Having done this for 20 years, for the first solid 10 years we were alone in the darkness. It was like SpaceX on the launch side and us.

17. Managing Rapid Pivots & Momentum

If you look at the present moment, there are now hundreds of founders who are building in the national interest with a U.S. capital stack—one of our other advantages—with over $100 billion invested in the projects they’re doing that are actually changing how the department thinks about the future. This is the window and the moment for this stuff to happen.

Increasingly, in conversations that I have in the Pentagon, I hear some of these new entrants talked about as Plan A. It’s not Plan B. It’s not, “Hey, we’ll see. Maybe this will work in the future.” It’s like, “No, no. This is the plan, and we’ll have to have backups if that doesn’t work.” That is unrecognizable from the world I started in in 2006.

If you were the president for a day tomorrow, what would be your executive orders? What would be the things that you think could carry the most freight to create some of the change that you want to see?

I have narrow ideas. This is a narrow idea, but I would try to blow it up. You want to bring pharmaceutical production back to the U.S.? If you try to do that for generics, the business case will never work. It’s too hard. It’s too much of a subsidy.

What you really want is to redevelop a skilled workforce that understands how to make drugs again. You want the production capacity so that, in a time of crisis, you could go from, say, making branded drugs to generics.

Why don’t we manipulate patent length and say, “Look, drug companies, if you make your worldwide supply of your branded drugs here in the U.S., I will extend your patent length such that you recover the extra capex investment to go do that”? That’s a simple business case that every pharma company would underwrite in a heartbeat, and it provides the catalyst to generate that capacity.

I think there are a lot of things in the American economy that look like that. If you go back to the story I was saying between prosperity and freedom, Chrysler built minivans and missiles, and Ford built satellites until 1990. General Mills, Kodak—you just go down the list. The entire economy looked like that.

I think it’s a consequence of being really the sole superpower, and the Pentagon—the old Pentagon, in particular—being the last bastion of communism except for Cuba. Every country gave up on communism, including Russia and China, except for Cuba and the Department of Defense.

That made it both a bad business and then seemingly no threat, so you could just be a prosperity company. You’d be a single-purpose company that just cared about being a company. I’m not sure that’s actually a stable or healthy society or civilization. We have to recognize that we have obligations across freedom and prosperity.

You mentioned that the notion that we are crazy and can pivot and attack opportunity is an amazing part of U.S. culture. What do you think is the other side of that coin? What are the negative aspects of U.S. culture in your view?

I’m going to ask the same question zoomed in on Palantir specifically: how to cultivate the culture that you want. Palantir seems to have an incredibly distinctive culture. I’m going to ask about Shyam bombs and things like this inside the company. But first, at the U.S. level, what are the things that you wish were different here about our culture?

Some problems require sustained effort, right? If you can pivot really quickly, maybe you pivot to solving the problem, but that means you can pivot away from solving the problem and get distracted and lose sight of what you’re trying to accomplish here. So that’s like a feature-bug. It’s not an unequivocal feature.

But if you’re in a competition, you want to play to your strengths, and that usually involves things that the adversary is unable to do. It helps that the nature of populating the U.S. was that all the heretics from the old world came here, right? So we have an unfair advantage on that.

In terms of maintaining a culture, the critical lesson I’ve learned over and over again is that the entropy of the universe is against you. If you don’t invest in your culture every single day, you will lose it. There’s nothing about culture that is self-sustaining. That’s as true for America, the country, as it is for your company.

The best cultures are really almost cult-like. Your culture is highly differentiated. It should actually be hard for someone to inorganically insert into your culture; otherwise, it’s not a very differentiated culture whatsoever. That requires flag-bearers. It requires people who are true believers.

So it starts with the recruiting. Not every human has a disposition that they want to be part of a cult, essentially. Some of them are just structurally mercenaries, and they’re evaluating everything in life really through the lens of, “What’s my next job?”

You want to try to build your company with people who really think of the present job as the calling. That engenders the primacy of winning. They’re always thinking about how to win today because that’s the whole game. If you win today, you have the right to win tomorrow.

The other side of that is getting lost in, “I wish the world could be this. I wish the company could be this way or that way,” which is unmoored. It’s an aesthetic of how you wish the world to work rather than both how the world does work and what is required to win today.

18. Where Will AI Value Accrue?

In a company like Palantir, what do you personally wrestle with the most? The tradeoffs—I’m always fascinated by tradeoffs. It just seems like so much of what happens hinges on a couple of key choices. You can’t do everything for everyone. What do you personally wrestle with the most?

The most important decisions from my seat are really what projects are worth outsized investment that we think are going to define the future of what our customers need and, therefore, the future of the business over the next 5 years.

How do you get the right talent on those projects? How do you discover what those projects are? I like to think about it this way: my backlog is something I reprioritize dynamically relative to reality.

Here’s my roadmap. Here are all the things I want. It’s an approximation at a point in time of what I think are the most valuable things to build. The second some new data comes in, I’m going to rip that up and reprioritize it.

Having a culture that lets you do that is really painful. People don’t like it. Their world is shaken up. It generates a lot of chaos, but it’s the right thing to do.

Managing that so that your culture enables you to do it, and having the conviction that the cost of doing it is worth it because the return is so outsized—that’s the whole business.

I'm extremely sympathetic to that point of view of changing plans very quickly. I've also experienced the whiplash that it creates in a business. How would you coach someone trying this who believes in it but understands it's really hard on a business if you're constantly changing your mind and pivoting your resources around? What have you learned about the ways to talk about that with your team, implement it, and communicate around it? I'm curious how you do it.

There are always degrees of these things, but to the first-order issue, I'd say: Is it hard because you want to exist in a business that isn't painful? There's a great Greg LeMond quote. He's a world-champion cyclist, and someone was asking him about a championship performance. He almost looked incredulous and was like, “You know, it doesn't get easier, right? You just go faster.”

What do you think a championship performance feels like? It feels shitty. It's really painful. It's really hard. Your ability to win is basically: Can you survive more pain than your competitors can?

There's a weird phenomenon where, if you're doing something world-class, it's really painful. Also, if you're totally failing, it's really painful. So then it's about having some discrimination to know which of these 2 worlds you're in. The worst world to be in is where you have no signal. You're just plotting along, but managed decline is really what's happening.

So then, is the pain we're experiencing generating results or not? It's about orienting people to that. The other part I'd say is you can't move the whole company at once. Even a company as small as ours is only 4,500 people.

Every change is itself a mini-insurgency inside the company. I think about it like you're campaigning, right? Who are the people who already believe what you're saying? How do you add incremental weight behind what they're doing? How do you help them succeed more? Are you trying to generate momentum? Are you trying to deal with the drag of bringing everyone along?

My experience has been that momentum is the thing that aligns the whole company. You should be waking up every day in the midst of these pivots just working on momentum. People will catch up.

What are your favorite ways to do that—to create momentum? Is it mainly a communication question? Palantir gets so much attention now from public markets and everywhere. It's one of these exciting places, so there's more scrutiny on the company, its choices, and what it does. You don't get to operate in obscurity like you did in the first batch of time. What are your favorite methods for that internally or tactically?

Celebrating the heretics, celebrating the successes, creating venues where everyone has some ability to revolt. At every employee's first AMA with me, I end it by telling them to tell me to fuck off to my face. It's very uncomfortable for them, actually. They're almost not sure if I'm entrapping them into something that's going to get them in trouble.

How can you say you have a flat culture if you can't tell the boss to fuck off? It's a small gesture, but it's trying to set the conditions to understand: “Hey, we are a company that's committed to taking the very best idea regardless of where it comes from,” and then putting all of our wood behind those arrows.

Most of those ideas are usually some variant of what we're currently doing. Otherwise, it would already be the current best idea. So we've institutionalized—although you've got to be careful how much you institutionalize this—at a minimum 2 times a year, we have these weeks of revolt where people are allowed to go build whatever they want.

19. Reasserting the Legitimacy of Institutions

I almost want people to have a chip on their shoulder. The whole point is to prove that we've been doing it wrong, inspire us with the thing that you've built, and then, if you have a culture that is profoundly truth-seeking, the truth is so charismatic that once I've seen it, I can't unsee it. This is obviously the right answer. Let's go.

It's not as hard as it may sound to move and pivot on these things.

The AI moment is so remarkable. We all have to deal with this new thing at the same time. It's probably like any crisis: There's both opportunity and danger bundled into the coming of AI for Palantir, too. How do you receive this new technology and make sure that you get better because of it and don't get disrupted because of it?

You were a disruptor with this enterprise ontology concept, killing the old, rigid database systems. You had an innovation. Now, if I wanted to create a Palantir competitor, it would be much easier just by virtue of the technology that's available.

The only way you can do that is by disrupting yourself. To the point of institutionalizing rebellion and holding that up, it's the perpetual motor of self-disruption. I think about it as the most valuable thing is not to defend a moat or something, but to go charge after the new opportunity.

One of the things I think is unique about AI is how much it plays to the beginner's mind. Obviously, the first adopters of AI for us were the youngest hires. To the point of generating momentum, I didn't really bother with the people who were cynical or dismissive of it. I let the young people generate overwhelming productivity that almost, in some sense, humiliated or inspired other people to get on board.

There's a similar phenomenon. There was a weird, almost out-of-body experience for someone who's been there 20 years, or someone who'd been at Palantir 3 years, to say, “You know what? The best people doing this are really the young people”—by which he meant people who'd been there 3 months.

That recognition of, “Okay, so here we have someone who's not a gatekeeper. He's actually trying to find the insurgents and empower them and place them in positions where they can go forth and do better, and we can all learn together by doing that.” It's actually been really fun, given the rate of transformation in AI and what it's really made possible.

Everyone's trying to answer this question: “Okay, we have AI. Where is the value going to get captured in the whole stack or chain, or however you want to think about it?” What's your answer to that question?

20. To Do or To Be?

We think that the value is going to accrue at the chips layer and at the ontology layer. If you squint at the stack—I'll skip over some parts of this—you have chips, model providers, AI infrastructure, which is where we think the ontology sits, and then AI applications.

What you observe out there is that the model companies have to run up the stack, away from just providing a pure model, because the model is commoditized. It's really everything they build around the model that's a solution they're going to try to build value off of.

21. Reflecting on Fatherhood

When you look at the AI pure-play applications, they're running down. As they start to scale their businesses and have to deal with more customers and more of their customers' data, they're reinventing the AI infrastructure, essentially AIP, from first principles, recognizing why they need it.

Now, of course, through some divine intervention, we've built this platform over the last 20 years. You can say, arguably, the meat of it has been since 2015, with Foundry leading to AIP. That gives us a huge head start on these things. Even if these models help you code faster, there's the code and there's the secrets that are actually behind why we built the code.

So where are we going to be in 2 years? Where are we going to be in 2 years and after that? I think we feel like we're in a very good position there, and that's expressed by the way we measure this: How quickly can I deliver value for my customers? How quickly can I deliver enterprise autonomy or fundamentally change my customers' businesses?

Not only has that number been awesome, but I'd say 2 years ago we could do that inside of 8 weeks. Now, it feels like a week, so that rate is compounding dramatically.

22. Kindest Thing

One thing I'm really fascinated by you is that, for so long, you were sort of an “if you know, you know” guy, very much behind the scenes. If you asked around, you were the person—or one of the key people—making this business work: the forward-deployed thing, lots of examples.

It seems like you've chosen to be an “everybody knows” guy. What's behind that choice? Why make that shift?

I'd say I'm a bit reluctant about the shift, but I can't look my kids in the eye and not say that I'm ringing the bell that we're in an undeclared state of emergency. Really, the first major public thing I put out there was the 18 Theses, which is to say: We're in an emergency. We've lost deterrence.

I've been doing this for roughly 19 years at that point. Here's my diagnosis of what's wrong. The shot clock is ticking. We've got to coalesce around this problem.

Now, on a note of profound optimism, I've seen more change in the Department of War in the last year than in the prior 19 years. I think, to the point of Americans doing the right thing after having exhausted all the other options, we are definitely in the place where we are starting to do the right things. The first derivative and even the second derivative are positive—we're going faster and faster here.

How would you sum up the 18 Theses? I know we've talked about lots of component parts of it already. Is there anything from that—it was your first big public thing—that we haven't talked about that you think is important?

But if I were to synthesize it, it's that we've lost deterrence.

The reason this is not going to go the way World War II did is that we don't have the industrial base we had in World War II. We have to recognize what our unique American strengths are. The person is the program. Don't forget that Edward Hall built the Minuteman, and Bernard Schriever was the Air Force general who oversaw our ICBMs. We call it the Apollo program, but it was actually Gene Kranz's program, and the primacy of those people and getting the right people in those roles is determinative to success.

It would be like thinking you could found a company by recipe, as opposed to understanding that the company's future is determined by the founders. That's almost like the congenital limiter on what is actually possible. Of all cultures, we should understand that. Downstream of that, we should recognize that there is no process out of this. The reason we're screwed right now is we prayed at the altar of process: we thought if we followed this requirements process, or if we just let the bureaucracy work, it would lead to a good result.

The only thing that leads to is decline. If you look at all the great military innovations through history, Churchill built the tank as the head of the Royal Navy because the British Army wasn't smart enough to realize that horses were not going to win the next war. So he called it a land ship because he could only build ships. It was a land ship. But that's not a one-off. Literally every major military innovation, including the Sidewinder missile, was a rogue project that some guy built with lighters and a new sort of seeker that totally humiliated the Air Force's missile in Vietnam and became what we now think of as the standard.

When Bill Perry became the Under Secretary of Defense for Research and Engineering, he was staring down this thing. He said, "Hey, I can either try to fix the system and lose, or I can take the 2 things I really care about and make sure they work: GPS and stealth." At every crossroads, it's the person who just wants to engage with the system that loses, and the person who's willing to put it all on the line to make it work. It was kind of a call to action: if you're a latent heretic in the department, if you're a latent heretic in defense tech, this is the moment to express your heresy.

Is there anything about your worldview, Palantir, or anything else that you think would be surprising to an outsider that we haven't talked about?

First of all, we don't really think of ourselves as a defense company. We're a software company. I think the fact that half of our business is in government, and 2/3 of that half is in defense, is actually more normal to the historical pattern we've talked about here than to the present. But the part that I think a lot of outsiders misunderstand is that they somehow think we're either in the business of collecting data or doing surveillance. That is probably the part that's least well understood about what we're doing.

We're engaging with the problems as they exist. You could say, "What is the problem that exists?" The core problem that we're set up to solve is the legitimacy of our institutions. Why do doors fall off planes? Why does it seem like our basic government services don't work? If you don't address those problems—if you don't build that state capacity, private sector or public sector, and help these institutions get better—it breeds nihilism. It breeds this sense that we should just burn it all to the ground. That doesn't actually work. It's just strictly a worse world. But that sort of energy is what you're fighting against.

Our diagnosis of this problem is that, if you thought about these institutions, let's give the C-suite, government or private sector, the benefit of the doubt. They have the steering wheel, and they're diligently trying to steer the steering wheel. What they don't understand is that it's a Jungle Cruise prop from Disneyland. It's not connected to anything. Then you have the people down below who are looking up and saying, "How could my leaders be so dumb?" That is where that nihilism starts.

What is the point of this operating system? Why are we trying to make these decisions better? We're trying to create something that's more responsive for the people who are leading these institutions and the people who do the work. So much of the 20th century has been this wild managerial revolution that has concentrated power away from both these ends to an amorphous middle blob. That's where all the disconnection happens in the steering wheel. Reasserting that, so that our institutions work for the people they're designed to serve and our companies are actually the best in the world, functioning and competitive—that's how you underwrite freedom and prosperity.

This idea that you have all this data, that you're doing surveillance through this panopticon in America or something—does that weigh on you? Does it weigh on you that you are cast as an easy enemy or villain in certain people's stories?

It used to weigh on me. Part of it is, okay, you've been doing it for 20 years. How long can this stuff keep weighing on you? But I think it would be fair to say it weighs on every new employee when they come in. There's some journey of dealing with this.

But that's also the lot of the man in the arena. We can end with John Boyd: to do or to be. Are you trying to be someone, or are you trying to do something? We're trying to do something.

That was one of my favorite lines from that book. Expand on that one a bit more.

John Boyd's advice to other officers and other uniformed service members was always, "Look, you can either be somebody or you can do something. But you can't really have both." If you're going to be somebody, you've got to play this game. It's all kind of theater, but you're going to get promoted, get the accolades, and get the rewards of doing that, and you're going to feel good about yourself.

Or you can actually do something, and it's going to feel really shitty, and no one's going to appreciate it. But you're going to have the intrinsic reward of knowing that you did something.

You mentioned your dad earlier in the conversation. There's this visual story where I think he's held up at gunpoint or something early in his life and went through a lot. I'm curious for you to reflect on the nature of fatherhood in this whole story. It seems like such a key part of your story from the outside—your relationship with your dad—and you're a dad yourself. Would you be willing to reflect on that a little bit?

It would be hard to overstate the impact that Dad and his example had on me. He was such a selfless man who just gave everything to his family, and could weather any humiliation in service of his children and his wife.

When you're a kid, you learn this stuff subconsciously. You're like any normal kid: this thing Dad did is annoying, or you don't fully appreciate it. But the culmination of all these events over time profoundly shapes who you are, and I hope I can set an example that inspires my kids to do the same sort of thing. I think a lot of fatherhood is being a man who inspires your family to be better.

When I do these, my closing question is the same for everybody: what is the kindest thing that anyone's ever done for you?

The greatest thing, which is covered in the story, is that Ms. Ethel Danhoff was the head of admissions for this private school in Orlando. My dad took me for a tour of the school, having learned that it was the best school. That was the other fixation my dad had: education was how you were going to achieve some prosperity. It was like, "I'm going to give my son every opportunity I possibly can."

It turned out we'd missed the admission window by a mile, and everything was kind of done. I was actually very reluctant to go on this trip and see the school. I was very happy with the school I was at, but when I saw the science facilities there, I thought, "I need to go here. I need to learn this stuff. This is inspiring."

Now I was all spun up and excited, and of course I'm learning, "Hey, you totally screwed this up, and you'll have to wait a year." And Ms. Danhoff—I had no data. It wasn't like I had standardized test scores or something. She said, "Give this kid the admission test. Just do it."

Maybe she had a modicum of data after that test, but we were a family that needed financial aid. So it's like, here it is: let's admit a kid that we're going to have to subsidize to come here. That opened the entire future. If you look later in life, maybe people have cynical reasons to help you. You're further along; they think there might be something in it. But there's nothing in it for her. It's just a profound act of kindness that fundamentally changed my life.

Great place to close, Shyam. Thanks for your time.

Thank you.