Palantir CTO Shyam Sankar On Refroming DOD

The Hugh Hewitt Show

Shyam Sankar is the Chief Technology Officer of Palantir, and widely tipped to be part of Presiden-telect Trump’s incoming senior leadership at DOD. Sankar joined me this AM to discuss his beliefs about DOD procurement reform, which is summarized at 18Theses.com:

Audio:

12-13hhs-sankar

Transcript:

HH: I’m so pleased to welcome to the program this morning Shyam Sankar. He is the chief technology officer of Palantir. Good morning, Shyam. Welcome to the program. Thank you for getting up early for talking to me. Where are you today?

SS: I happen to be in D.C., so no problem at all. Thank you so much, Hugh, for having me. I’m honored to be here.

HH: Well, I have been pitching the Defense Reformation, which people can find at 18Theses.com, and I read a lot of it in the first hour of the program. So I want to cut right to the chase. We talked about the Last Supper, we talked about Cost-Plus. I want you to explain for the Steelers fans and me, because I’m just as slow as a Steelers fan when we get to this, this line. “The great schism has created a religion in government that is unaware or dismissive of power law outcomes from power law talent. In Silicon Valley, we call them 10x or 100x engineers, meaning they are 10x or 100x as valuable and productive as normal engineers.” What is the great schism? Why doesn’t D.C. want those people?

SS: Well, if we go back to the industrial base that won World War II and won the Cold War, we would realize it looks very different than our Defense industrial base today. You know, when the Berlin Wall was just falling, only 6% of our spend on major weapons systems went to Defense specialists, went to Defense primes. The vast majority of the spend went to what I call as dual-purpose companies. We forget that Chrysler built cars and missiles. Ford built satellites until 1990. General Mills, the cereal company, built torpedoes. You know, we had a very different structure of the U.S. economy. And what that really meant is that everything they were learning commercially, all that R&D was subsidizing the lethality of our U.S. servicemembers. Every car, camera, and cereal box that Americans bought actually enhanced our national security. Today, we’ve gone from 6% spent on Defense specialists to 86%. And so we have really invested in and doubled down on a Defense industrial base that lives on the Galapagos Islands. They’re very far away from all the commercial innovation that’s happening. And that commercial innovation, by the way, is what powers U.S. prosperity. Just look at what the U.S. commercial sector has delivered for the country here. And I think we need to get back to a place where we can leverage that, and the talent that we have in that base to enhance our national security.

HH: Now Shyam, how do we disambiguate a situation that pathway evolution has taken to the six big primes? How do we get out of that in a hurry? My friend, Dr. Jerry Hendrix, whom I’m sure you’ve read some of his stuff, we’ve just got to get back to the pre-1989 competition in the private sector, a variety of primes. We’ve just got to do it. How do we do that?

SS: That’s the most important question. And I think when people look at the Last Supper, the kind of conventional analysis suggests you know, what we really lost was competition in the industrial base. Undoubtedly, that consolidation led to less competition. But I think the high order bit, the real consequence is that consolidation bred conformity. It pushed out the innovative founders. It pushed out the crazy engineers. And you know, this is a country that understands the importance of founders. We call them the founding fathers for a reason. You know, you don’t get SpaceX without Elon. You don’t get the U2 without Kelly Johnson. You don’t get the Apollo program without Gene Kranz. And we have somehow washed over all these efforts as if they’re just faceless institutional efforts. And you, you know, the dichotomy of public versus private is not the right dichotomy. It’s really founder mode versus manager mode. And so I think you can solve this problem very quickly, because $100 billion dollars of private capital has been invested to build companies in the national interest, and you have a plethora of founders who have shown up. You have Palmer Lucky building Anduril. You have the Tseng Brothers building Shield AI. You have Palantir. You know, so like the founders are back. We just need to organize around this problem now.

HH: Now Shyam, there are five dominant elite cultures in the United States. One of them is irrelevant to this. That’s their religious faith culture, and it really doesn’t have much to do with the other four. The other four are Beltway elites, Manhattan elites (Wall Street), Hollywood elites (creativity), and Silicon Valley elites (productivity and innovation). When Donald Trump won the first time, he turned to Wall Street elites, and they got some done, but they also got hijacked and cut off by Beltway elites. Do you think the Silicon Valley elites are better or worse prepared to fend off the Beltway elites’ inevitable attack upon their arrival?

SS: I mean, time will tell, but I’m quite optimistic, because you know, to really execute a vision for the future, you need a definite and optimistic view of the future, that things can be better. You need a group of people that don’t, that aren’t preoccupied with self-loathing and believing that the country is not a great place, and that aren’t kind of cynical about it. They have fundamentally a positive sum mindset. One of the secrets of Silicon Valley that I think even most Americans don’t recognize, we tend to feel like we have this elite tech culture, where did it come from, maybe we imported it from Israel or India. No, we imported it to Silicon Valley for sure, but we imported it from Iowa. It is a culture deeply-rooted in Midwestern sensibilities, a willingness to play positive sum games. And it came to the Valley from Bob Noyce, a great Iowan, who was the co-founder of Intel, the co-investor of the transiter. He coined the term open door policy, and that whole culture of innovation we have is actually descended from Noyce. And I think that it’s very powerful. It’s why America has the best tech culture in the world by a yawning margin.

HH: So when we go through the 18 Theses, and I already read them to the whole audience. They know what it is – Cost-plus contracting makes the nation dumber, slower, and poorer. That’s number two, and I read all 18 of them. National security is economic prosperity is the easiest one to understand. How soon, if Donald Trump looked at this and said do that, how soon could it be done at DOD?

SS: The place you start is that you increase the amount of competition inside the Department. When Admiral Rayborn was building the submarine-launched ballistic missile, he actually had four competing programs running simultaneously. When we were building ICBM’s, we didn’t actually give the Air Force a monopoly. They had to compete and win with Minutemen. Every service threw their hat in the ring, and they were trying to do it. This explains why space is such an innovative area right now, because it is, quite frankly, a food fight between NGA, NRO, Space Force, Space Com, and that is actually productive. We need to get away from the sort of unitary effort, Communist aesthetic that we should have one singular effort, this avoids duplication. We have to embrace the reality, the very American reality that innovation is messy and chaotic, and lean into that rather than circumscribing the whole process in so many rules that you make sure that in an effort to make sure nothing goes wrong, you also have all the rules that make sure nothing can go right.

HH: The most obvious example of this is drone technology, which is being developed not just by a dozen different big companies in the United States, but by hundreds of companies worldwide. And as a result, we’ve got stuff flying around in New Jersey that we have no idea what it is. By the way, what do you think of that story?

SS: I think it’s terrifying. I mean, it shows, I have an expression inside of Palantir, the Orks in the Lord of the Rings sense, they spawn in the seams between teams. And I think a large part of this effort, it’s not a tech problem, really, it’s really we have a division of responsibility between FAA, DOD, and Homeland Security. And we need to reorganize ourselves around the problem, not be so victim to these arbitrary boundaries that we’ve put up there. And I think this is, again, manager mode versus founder mode.

HH: When you say terrifying, what could it be? I want to pick up on this. It’ll be the headline, obviously, so be careful what you say. But when you say terrifying, why?

SS: Because it shows a lack of domain awareness. Like you know, it’s one, we should be able to know everything that’s happening on the homeland airspace, and we should be able to defend against it. Whether you shoot it down or not is a different decision. But you know, we need absolute control, and we need to be able to know with precision where this is coming from and what our options to deal with it are. Play this forward where you go from overflight surveillance of bases to the protection of the homeland, of the stadiums that we have, the, you know, how are you going to protect the people from these assets if you can’t protect the bases?

HH: Well said. Now I want to go to another part of the Theses. Again, 18Theses.com is where you go to read this yourself. The problem with Goldwater-Nichols is it didn’t go far enough. You can’t have a joint department if services have monopolies on their Title X equipping responsibility. That’s profound. It’s interesting. It was foreshadowed. I sat down with a Marine Corps field office who had just come back from Israel, or he was studying the Gaza war, and said they’re so far ahead of us in how they do things and how quickly they learn. The key question is we were very far behind the Japanese and the Germans when World War II broke out. We caught up in a hurry. Do you think we can catch up in a hurry?

SS: I would never bet against American innovation. Like we, you know, at the end of the day, and I mean this in the most positive way, we’re crazy. You know, when push comes to shove, we will throw all the rules out and organize around the problem. The question is how do we do that left of bang here? You know, how can we look at what’s happening in the world and get a head start on doing exactly that? And I’m confident that the things I’m saying, when, if we were in a conflict, people would just intuitively be executing.

HH: Now Shyam, the thing we need the most are submersibles. Submarines and unmanned vehicles under the water. That’s what we need the most. Hypersonic, some people argue, other people argue satellites. But we definitely need submarines. We don’t have enough capacity to build our own, much less what we promised the Australians and the Brits have promised the Australians. What would you recommend Congress do in this CR and the first budget, and in the NDA, which is getting close to pass, to remedy that problem of capacity?

SS: I have a bold vision here. I mean, it’s, there’s no, I think there are no half-measures that are really going to work. It starts with understanding that there’s never been a military sea power that was not also a dominant commercial shipping power. We need to recognize that. So I think the half-measures saying well, we have to prioritize one or the other, how do we just double down on building more and more military capacity? The reality is you can’t get down the cost curve fast enough if you only are building military equipment. This is what you see at the Hyundai, HD Hyundai shipyards in Korea, that it is a whole town in Ulsan that’s built around building both commercial and military ships. We still have a massive asymmetric advantage with our nuclear reactors. I think we should build the first commercial nuclear-powered cargo ship fleet in the world that are American-flagged, American-made, leveraging our military’s greatest technology, driving the lowest cost of shipping in the world, bringing down inflation and the input costs, and use that to bootstrap and rebuild a massive bolus of investment into shipyards.

HH: That’s crazy talk. That’s Rickover talking. But you’re right. I read a bio of Rickover last summer, and I could not believe how much control he exercised. And thank God he was there, or we would be way behind everyone right now, but we’re still way ahead. If the priority is our submersibles, satellites, and cyber, what does a deputy Secretary of Defense need to know before they become a deputy Secretary of Defense, because Pete Hegseth is going to be the outward-facing guy, the public-facing recruitment guy. The DepSec is going to have to do the procurement. What do they need to know?

SS: If we think about the Davidson window, you know, there’s not a lot of big, exquisite systems that you’re going to be able to build and field in the next two, three years. But you can build and field a huge amount of software. And you’re obviously not going to win the war by shooting software at your enemy here, but we, the extant capabilities that we do have, how do we wield them better? How do we have a substantially better OODA loop than our adversary? That is, in fact, a software problem. It’s a problem that John Boyd would recognize. And the other part of this is at the dawn of World War II, we were the best at mass production in the world. Today, unfortunately, our adversary is. So David is going to need a software-defined Slingshot to drive production again. We have to give American workers superpowers in order to compete and produce. You know, we have wargame simulations say that we have roughly eight days of weapons on hand for a China fight. That’s obviously not scaring anyone. So we need closer to 800 days of weapons on hand, and we need to be able to make those weapons in entirely different ways.

HH: Stand by, Shyam Sankar, I’m coming right back with you during the break and afterwards. Gallagher obviously did not brief you on the fine that we assess for everyone who uses an acronym without explaining. OODA loop and Boyd – Observe, Orient, Decide, Act, Boyd is the fighter pilot who came up with it. And he owes $10 dollars to the tip jar. And we’ll talk to him during the break. He’s coming back on the other side. Stay tuned, America.

— – – —

HH: I am back with Shyam Sankar, whose 18Theses.com should be read by every incoming Trump appointee. Shyam, I’ve got to tell you, this is the most expensive interview I’ve ever done, because three weeks ago when we set it up, I sold all my Palantir, because I can’t have a conflict with someone. And you know, the three weeks have been very, very good for Palantir, so I’m thinking to myself, God, this is an expensive interview, but that’s okay.

SS: (laughing)

HH: I love the way you close it. I nailed these theses to the Pentagon Metro. Do you think the DOD wants what you’re selling? Do you think the military wants what you’re selling?

SS: Honestly, when I did this, I wasn’t so sure. But the response inside the building has been resounding. There’s a huge amount of support. I think people really do see these problems. It’s given a framework and some words to attach to the lived experience that our general officers have been fighting through, and the lived experience that staffers in Congress has been dealing with, and provides us a path to kind of chart our way out of it.

HH: Are you going into the administration?

SS: All options are on the table. Look, I’m very happy with Palantir. As you just said, you know, we’ve built the world’s leading AI product here. The demand in the U.S. commercial sector is just unbelievable. I’ve really enjoyed building that over the last two years. I’ve been at Palantir for 19 years, so I’m very happy, but you know, all options are on the table.

HH: So tell me, I hope the President-Elect is listening and brings you in as he does the SecDef thing, but we don’t know, and I have no way of knowing. Talk to me a little bit about whether or not they’re bringing too many people from Silicon Valley, and, because you put Elon, Vivek, you, and a bunch of these other people in a room, are you guys ever going to agree on anything, and gals? Are you ever going to agree on anything?

SS: Well, like my view of this is, we probably won’t agree on everything, but I think we’ll agree on enough to chart the path forward here. And you need a lot of, like, destructive creative energy to get off the kind of sclerotic mess that we’ve gotten into. I think we’ve gotten into a very low energy, high entropy state where you know, things don’t work, but we’re trying to minimize interpersonal conflict. But we actually probably need more interpersonal, you think about Rickover. You brought him up. He was a notoriously difficult person, but where would we be without him? And it took 30, Congress had to protect him for 30 years as the head of the nuclear Navy in order to deliver this capability that we still have an asymmetric advantage on top of. And so I think it really does start with people. As a technologist, maybe that’s a little bit heretical, but you know, the technology is a tool for the commander or the principle’s mind. It’s a software Ironman suit to get more done.

HH: Sure.

SS: I think the kind of conventional view of technology is it makes the median person better, and it does, but it makes the very best person way better. And so it’s really important to get those leadership roles right.

HH: That’s very true in the news media, by the way, so I know the applicability into what I do. I, first article I read this morning when I did show prep was in the Journal. Boeing delays means Trump won’t fly on a new Air Force One. Okay, so they’re not going to deliver new Air Force One’s until 2029 at the earliest. You realize that that was a problem when he took office the first time in 2016? What does that tell you about Boeing and the industry generally?

SS: To me, it tells me about manager mode. Whether it’s the public sector or the private sector, when you have these institutions that are just trying to run based on a playbook, they’re just trying to follow the rules, it’s overly financially engineered, it’s just what the MBA’s tell you, you should be doing, and it’s depriving us of the individual agency and creativity of founder mode, of people who have real insight to drive these things, it’s fostered a legitimization crisis in our, in the world, but also in our country, where we don’t actually believe, it engenders this nihilism. We don’t believe these institutions work. And we’ve got to throw away the playbook and actually build institutions that do work by organizing around the problems.

HH: I’m coming back with Shyam Sankar for one more segment. Don’t go anywhere, America. He’ll be right back.

— – – –

HH: The last part of my interview with Shyam Sankar, who is the chief technology officer of Palantir. Shyam, I want people to go to 18Theses.com and read it for themselves. My big question – there are three lethality innovators in the world, and I mean first-tier. The United States, the People’s Republic of China, and the Israeli Defense Forces and their tech center. Do you think Israel has clearly thrown in with us now, because they danced on the edge of both for a while?

SS: I think it’s unequivocal that they’ve thrown in with us, and I think you know, in crises like these, you take note of who your friends are. Obviously, there’s been probably an unsettling reaction in how divided the West has been in terms of supporting them, but they have also noted who has supported them.

HH: Have you noted in both the Ukraine War and the Israeli War on seven fronts any particular innovation that you sit up and say wow, we need to do that at Palantir or in the United States?

SS: Well, the most important, there’s probably two observations I’d offer you. The first is that the drone obsolescence life cycle in Ukraine is something like two to six weeks. So that tells you that it’s not actually what is your capability today. It doesn’t matter how exquisite it is if you can’t change it. So we need to be measuring and practicing the first derivative of how quickly can we adapt our capabilities to the changing battlespace so that our capabilities are still relevant and still lethal. And that is a big mind shift. We thought, well, we can develop exquisite things, we can put them on the shelf, and they’ll be just as exquisite and just as useful ten years from now. That world is gone. The second major lesson, people love to say isn’t it amazing what the Ukrainians did. Despite not having a navy, they sunk half the Russian Black Sea fleet. And I think they’ve got it backwards. It’s because they didn’t have a navy, they were able to come up with entirely different force concepts that sunk half the Black Sea fleet.

HH: Right.

SS: And this is another strong push for more…

HH: Last specific question. The Congress is going to do a CR, and then they’re going to do a budget resolution. Should they leave a large pile of money in one or the other to the DOD to distribute among the private sector to innovate as the 51 old primes used to do in a competitive, in other words, a sovereign wealth fund for Defense industry? Should they do something like that?

SS: I like the direction. I would say let the combatant commanders decide how to spend that money. You know, that allows us to approximate market forces. I like to quip everyone has given up on Communism including Russia and the Chinese, except for Cuba and the DOD. We need to get away from central planning, the five-year plan, and we need to enable our combatant commanders to express market signal and demand, and allow industry to respond to that.

HH: Who do you read, Shyam? I mean, where do you get your ideas from on Defense and on war fighting? There are historians like Andrew Roberts. There are technologists like Jerry Hendrix. What do you read?

SS: I try to integrate it all. I mean, obviously, there’s a great tradition of the Valley. You know, we’re all students of Peter Thiel. There’s great history on Defense tech that goes back to the Cold War and World War II that I think is very important. And you know, there’s a great quote from Vannevar Bush in 1940. He’s telling FDR the coming war is a technology war. And civilians believe that they can provide incredible capability that the military doesn’t understand. I read that quote out of context, and I thought it was a quote in the present moment. You know, in some sense, there’s nothing new under the sun, and it is our success, the fact that we have been a superpower, that we’ve had no competition for 30 years since the end of the Cold War that allowed us to forget these lessons.

HH: Do you believe the combatant commanders have got the skill sets necessary to spend the money?

SS: I absolutely do. I have seen, the amount of innovation that is coming out of the theaters and the combatant commanders, you know, we already doctrinally believe the commander is the most important person on the battlespace. And we need to really organize around that principle. Look, if we lived in a world where all that mattered was the major platforms that took 30 years to build, the services need to build that. But if we’re living in a world where the capabilities are moving every year to three years, you’ve got to push more of the discretion and more of the speed towards the combatant commanders.

HH: Shyam, thank you for joining me. I hope if you come into the administration, you’ll keep coming back, because people have to preach innovation. Now, I can go back and buy some Palantir again. Get back to work. Make Gallagher work harder, because he’s probably sloughing off watching the Packers, and I appreciate the time. Thank you so much.

End of interview.

The post Palantir CTO Shyam Sankar On Refroming DOD appeared first on The Hugh Hewitt Show.