Shaun Maguire on why foundation model companies are the new operating systems, and his unpopular take that only ~10 defense tech companies will win big

bring in Sean Maguire from Sequoia Capital. Welcome to the stream. Third time appearance on TVPN. How are you doing? I think we have a question. By the way, he's like leaning down. These guys are much taller in first and then you Oh, yeah. Yeah. Yeah. Uh yeah. Yeah. At one. Yeah.

We'll do six minutes then we'll bring him in. Great. Uh great to see you. Who's the VIP at one? Josh Wolf is coming on. Legend, great guy. Good buddy. You guys could share your mic. I think that'd be entertaining. Uh what what's the initial reaction? Is this your third Hill in Valley? Fourth Hill and Valley.

Something like that. Third. You know, I actually met you for the first time at the last hill in Valley. Yeah, I missed I missed the first one which was a big mistake. Got the invite. I think I got food poisoning or something before, but uh this thing has grown every year. It just gets crazier and crazier. Yeah. Yeah.

Uh are you do you like the framing of like re-industrialization being the focus? Last year was a lot of like Tik Tok stuff. Is there something else we should be focused on or is reindustrialization really the heart and soul of uh the agenda with Hillen Valley?

Whatever the agenda is, I think the conference is focused on how tech meets Washington and like how we play nice together in the future. And I think that's the right focus and it's, you know, what technologies are going to matter in the future, what policies can happen that will help us get there.

And I think like re-industrialization is critical, but I I'm already hearing, you know, there's a lot of talk of a lot of other things, too. Yeah. Yeah. like energy, you know, all sorts of things. Yeah. Yeah. Yeah.

It does it does feel like even defense tech, if defense tech ends up being involved in anything, it ends up overshadowing it because it's just exciting. You have these planes and drones and dominate like great videos. Even better demos. Even better demos. Yeah, the nearest demo is wild. I got that one early on.

Saurin's a champion drone pilot. So, sort of great drone uh great demo. Uh, in terms of energy, uh, are should every venture capitalist be on the hunt for the Elon of energy? It feels like there is Elon energy might end up being Elon. You think he's going to go into it?

I mean, we I can't read his mind, but if he does, just tell me where to wire the money, Elon. Is there any limit? Elon is Elon.

That's kind of my reoring resoring semis but like there has to be an upper limit to like you know how much industrial work he can do at some point or I mean look on so there have been to the credit of the tech industry there have been probably 10 to 20 pretty promising vision startups in the last five six years and probably five to 10 fusion startups I haven't invested in any of those which you know means I'm not doing my heart to just like raise the tide of all and my kind of the challenge for me has been it's even though I believe so much in the category I have like I believe in the macro I have to get the micro right.

I made that comment before and it's it's been hard for me to pick like who's going to win in that category and I I think this is something that we kind of struggle with and I think some sectors Yeah.

there should be a lot of government support like for all these if if one ends up as like you know number 10 in the space but is still like a public good and is like is a positive payback for the government. Anyways, I think we need more government.

I just feel like there's an opportunity for a startup that's like, "Yeah, SMRs are cool. People are building those. There's new designs that are great, but our startup is just copy pasting Diablo Canyon. We're going to copy paste it a 100 times. " 500 times. Exactly.

But that does seem hard to Who on uh We've had a lot of people on from the valley side of Hill and Valley. Who on the hill side have you been, you know, excited about talking with that you think really gets it? I mean, I was just on a panel with Senator Todd Young, and I I think he deeply gets it.

And for example, one comment he just made was like he doesn't want to like re onshore everything, only like a small number of very strategic industries. He doesn't want to onshore garments. Yeah. You know, and he's fine with things that are not there's different levels of like how strategic something is. Yeah.

And you know, it's fine if the medium strategic things go to friendly nations. Yeah. And so anyways, I think Todd has had very nuanced opinions.

I think up right now or next is Richie Torres who is someone on the left who I I've generally admired and felt like has very even though I disagree with him a lot of times I think he's a very principled person who really deeply cares about America and is willing to say unpopular things that he believes in.

For me, anyone that is willing to be bipartisan, like not just purely tribal, but take a on like some topic that is core dogma of their party and they're willing to go against that dogma, like that's someone that I generally admire. Yeah. Yeah. I mean, that should be a good fit for Silicon Valley.

Like the whole search for contrarianism is like pervasive in Silicon Valley, right? And so pervasive in Silicon Valley, a little less pervasive in Washington. Yeah. But it still exists and there's pockets of it that you have to seek out. Seek out. Um, what is your read on the new defense budget?

We're hearing this $150 billion number. Uh, for a while it seemed like there was going to be a freeze, maybe very bad for challenger defense tech companies.

Then it seemed like the floodgates open and there's all these huge numbers for even just smaller series A companies are throwing out, oh, there's a billion dollars hanging out for my thing. Uh, how optimistic are you about that?

And how optimistic are you about uh not just like the SpaceX's of the world but the actual upandcomers, the next generation of defense tech? You know, unpopular opinion. I simultaneously believe that defense tech is one of the most important categories in the world.

But I also believe that there's going to be a very small number of winners. You know, call it in the west. This is including Europe and Israel and America. Roughly 10 companies that go on to be really big next gener next generation companies call it companies with 10 billion plus market caps.

And I think there will be multiple with greater than hundred billion dollar market cap. Wow. And I think that the defense tech companies of the next generation are going to be way better businesses than the last generation. So I think it's kind of focus on efficiency. Focus on efficiency. Focus on vertical integration.

If you go look at like I think I think one of the core things people got wrong about Tesla in the public markets going back to like 2017 2018 is they were comping it to auto companies. Yeah. And they're coming to GM or Ford whatever. And people didn't realize that GM and Ford don't really make anything anymore. Mhm.

They basically design cars and and are, you know, customer acquisition channels and maybe some financing, but they subcontract out everything and and so they would look at, oh, GM or Ford have these market caps and Tesla's already here, but they would ignore NXP and ignore like all these supply chain companies. Yeah.

And I think we've kind of been doing the same with the defense primes. Interesting. You know, Boeing, Lockheed, they subcontract out almost everything. that goes into their only a slice of the market cap.

You're only getting a slice of the market cap and I think these next generation companies are going to have way more vertical integration, higher margin. I actually think that we're going to see a change with how procurement happens to move away from cost plus. I think we're already starting to see that.

Um, but basically the reason it's unpopular is I think there's going to be a very small number of winners that win mega hard and then a lot of, you know, losers or companies get rolled up into the big, you know, the big winners.

Were you surprised that I think the number was $500 million was earmarked for autonomous systems and when you compared that to some of the other categories it felt pretty low. Do you do you have any insight into into into how they got around to that number or you know general reaction?

So like there's different philosophies. I think one philosophy would be that like we're trying to push a boulder up a hill in terms of like changing the defense budget and moving budget towards next generation emerging industries.

And if that's something where like every year it's predictably getting bigger and bigger and bigger and if 10 years from now 40 50% of the budget is going towards kind of next generation strategic technologies then I'll be very happy. Yeah.

Um, what would be very bad is if it's like 500 this year and then zero next year. That's a really great take.

It's like it's like if a company raises, you know, $100 million in VC and they have to spend it all in the first year, like they're going to probably spend it really inefficiency inefficiently versus like, hey, let's like stage this out on certain milestones. And so, yeah, I think that's the right.

And this has been the look, I learned this lesson the hard way with my cyber security company that focus on defense. our first big contract, we got we got awarded a $10. 5 million contract and then government sequestration happened.

Like they couldn't pass a congressional bill and we basically got told like either we accept the contract with 60 cents on the dollar or no contract at all.

So obviously we took the 60 cents on the dollar but and we were lucky that this was actually our first big contract because it taught us from day one how unpredictable government funding can be. But it's that predictability like if it's if you know it's going to be 500 this year and a billion next year etc.

you can invest strategically if you have no idea what it's going to be in the future. It's just very hard to plan. Totally. Uh do you have Yeah. You guys get Josh last last question 60 seconds. Uh do you have a re you know a a a take on this idea of whether we're in an AI race but is it a is it a is it an infinite game?

Is it just sort of a perpetual game? or is it uh uh is AI war even the right framing? Right. I feel like this has been debate. Oh man, I I'm so glad you asked this. I've been obsessing over this for a couple years.

I have a pretty strong opinion that the foundation model companies are going to capture way more value than people realize. My I think it's a combination of operating systems plus cloud which means this will be incredible businesses. Let me just make this point.

So like if you think on the operating system level, there's Windows, there's Apple, there's Android by Google, there's you know Linux and open source and these companies like the value capture of Microsoft or Apple, it's not on the operating system, it's on all the applications that they control and built on top and same with Android.

Um but by control I think the operating system are going to be the foundation models themselves. I think there's incredible kind of capex and development to get to the point where you are the like one of the top operating system that has eyeballs. Yeah.

Yeah, but then I think the lock in and the value capture will be with the applications that are kind of verticalized on top and I think there will be an additional moat which will be like the mode of cloud the cloud has done a very poor job of the application lock in the way the operating systems did but they've done a very good job of having a hardware mode and a capex mode and uh and I think it'll I think I think these AI foundation model companies are going to have both they're going to have like a capex lockin mode and they're going to have like the usage and application mode plus value.

And then just the last thing I'll say on this and and like if you look at Linux, Linux probably has five times as many installs as like Windows. But the value capture of Linux is like 1/100th of Microsoft. Um but Red Hat's still a big business. But Red Hat's still a great business.

But I I personally think this is where we're going to go with AI is that like the open source deepseeek and all that stuff will be great and it'll be there will be more open source models used in the world. I think the value capture will be lower by that than XAI, OpenAI, Anthropic, some Chinese income, etc.

Get Josh Wolf on here. The best. Let's go. Yeah, we got to do a whole deep dive on that. We'll see you back in LA. Peace, guys. Let's bring in Josh Wolf. And I got a post that we are live. Um, so much to dig into. 60C question. Jordy, you you you hit him with the the the deepest possible question.

Deepest possible question. I wanted to take more on uh on this idea of, you know, should we be investing in Chinese AI companies? Yes. Yes. Yes. Well, well, we have Josh Wolf, the man himself. Great to see you. Welcome to the stream. Uh, how you doing? Please.