a16z Podcast - Submarines and the Future of Defense Manufacturing

Episode Date: March 25, 2026

David Ulevitch speaks with Chris Power, founder and CEO at Hadrian, and Vice Admiral Robert Gaucher, the Pentagon's first direct reporting portfolio manager for submarines, at the opening of Hadrian's Factory Four in Cherokee, Alabama. They discuss the state of America's submarine industrial base, why the Navy now needs more than five times the manufacturing capacity it had a decade ago, and how software-driven factories and a new workforce can close the gap.   Resources: Follow Chris Power on LinkedIn: https://www.linkedin.com/in/powerc/ Follow VADM Robert Gaucher on LinkedIn: https://www.linkedin.com/in/robertgaucher/ Follow David Ulevitch on X: https://x.com/davidu Stay Updated:Find a16z on YouTube: YouTubeFind a16z on XFind a16z on LinkedInListen to the a16z Show on SpotifyListen to the a16z Show on Apple PodcastsFollow our host: https://twitter.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

The real advantage that submarines bring is our stealth and access. We can pretty much go anywhere in the world undetected. We can carry nuclear missiles on our ballistic missile submarines, and that ensures that we continue our decades of peace without nuclear war that any country that tried to attack us with nuclear weapons would be destroyed. At the end of the Cold War, we walked away from manufacturing. The amount of work that we need now to replenish our fleet is on the order about 70 million hours. The power of combining the new workforce, American software, American steel,

and American spirit is you have to get this productivity jump somehow with advanced factories. It's not a money problem. We have to get this productivity uplift by fusing workforce trading and software together to go a lot faster. In the mid-1980s, the United States built four nuclear submarines a year. Then the Cold War ended, production collapsed, and nine out of ten manufacturing jobs vanished. An entire generation was told to skip the factory floor. Four decades later, the Navy needs more than five times the capacity it had a decade ago. The Columbia class program requires roughly 70 million labor hours. The workers who could fill them aged out, and nobody replaced them. This is not a budget problem. The money exists. The people do not.

The question is whether software-driven manufacturing can compress a decade of training into something the country can scale. David Ullovich speaks with Chris Power, founder and CEO at Hadrian, and Vice Admiral Robert Goucher, the Pentagon's first submarines are. I am very, very, very, very lucky to have two incredible folks joining me on stage. This is going to be a conversation you do not want to miss. So in addition to having Chris Power, the founder and CEO of Hadrian, who you heard earlier, we also have Admiral Robert Goucher, who is in charge of all submarine production for the United States Navy. Let me give you a quick background. For those you that are not familiar with the defense industrial base, the Navy is an extraordinary force.

They operate across the largest domain on Earth. They cover more than 70% of the planet, and the Navy is responsible for projecting American power, maintaining deterrence, and ensuring stability across that entire space. This is a service that has been operating continuously since 1775. That's 250 years already.

It's adapted into new technologies, new threats, new missions, and it remains one of the most capable institutions in the world. And that's why we're here in Cherokee. That's why we're opening this facility. We're opening of Hadrian's Factory 4, a 2.25 million square foot advanced manufacturing facility that's going to support the Columbia and Virginia class submarine programs. So let's going to start out, actually, with you, Admiral.

For people that are not familiar with the real world problems, what problems do we solve with submarines? Why do we have a submarine program and why is it important that we rebuild our submarine program? I think the real advantage that submarines bring is our stealth and access. We can pretty much go anywhere in the world undetected, and so that becomes a very big threat for our enemies or adversaries. And as I look at the two biggest missions that we satisfy, we have fast attack submarines that go out and operate, and they make sure that our seaways and waterways remain free and open to anybody. who would try to shut them down.

And you heard Chris talk about how Australia needs to get supplies from external. Well, the submarine force can make sure that those supplies make it when they need to and where they need to. The other mission, which is our number one mission in the Department of War, is strategic deterrence. So we carry nuclear missiles on our ballistic missile submarines like the Columbia class that we're building. That ensures that we continue our decades of peace without news. nuclear war because the submarines are unable to be found so that assures that any country that tried to attack us with nuclear weapons would be destroyed. And that's why we call them the survivable leg of our nuclear triad. That is that second strike capability. So nobody attacks us

because they know the submarines are out there and they can attack back. When we talk about industrial capacity, I'll start with you, Admiral. Now I'll go to Chris. What does it mean to say that we need industrial capacity. Yeah, so what it means, just a quick history lesson, at the end of the Cold War, we walked away from manufacturing. We only built about three submarines in the 1990s until we started up the Virginia class program again. And even when we were building at a rate of one per year, that was only about 13 million hours of work that was required to build a single Virginia class submarine. The amount of work that we need now to

to build our 2 plus 1 to replenish our fleet and to replace our Ohio-class ballistic missile submarines with Columbia is on the order of about 70 million hours. So that's more than five times as much as where we were just a little over a decade of go. And so we're still on that trajectory. Because of the focus on shipbuilding and maritime dominance, we're really trying to supercharge that

with investments like we have here at Hadrian. And Chris, how do you think about industrial capacity and what it is and what it means? I think about a very similar to the Vice Admiral. At the end of the day, the capital equivalent can be purchased by Hadrian and all of Hadrian's investors or the United States Navy. But the real thing that we lost post the Cold War

was the skilled workforce. And that's what we talk about when we say hours. And because we cut the jobs down by, you lost nine out of ten jobs. And then we told all the kids in the 80s and 90s that can manufacturing go get a four-year journalism degree. So the big guard is that really smart people at work and through the rest of the enterprise,

we need 10, 20 times more of them to even catch up and meet the vice appellal's goals. But most of them are in their late 50s, mid-60s because we lost this demographic of skilled people. For us, the power of combining the new workforce, it's kind of American software, American steel, an American spirit is you have to get this productivity jump somehow with advanced factories.

Otherwise, it's not a money problem. We could spend $10 billion hiring this man, two million welders that he needs. They just don't exist in the country. We have to get this productivity uplift by fusing workforce training and software together to go a lot faster. That is the main problem.

It is their people problem. So, Admiral, you stepped into a new role. You've served in the Navy for a long time, but you are the Pentagon's first direct reporting portfolio manager for submarines. That means you report directly to Deputy Defense Secretary's Steven Feinberg, people have colloquially called you the submarine czar. What is a submarine

czar? And why was this role created? Can't we just order submarines and get submarines? So the reason that the Deputy Secretary of War decided to create the direct report portfolio managers, or we call them derpums. Submarines czar is a much better title than derpo. So there's three that were created, and they're all related to strategic deterrence missions. You have Golden Dome, which is meant to defend the country against missile attacks. Then you have the critical major weapons systems, which are the Air Force nuclear triad programs, so the bombers and the ground base intercontinental ballistic missiles. And then the third leg of the strategic triad is submarines.

You see that nuclear strategic focus where we said, we've got to get these programs right. And so the deputy wanted to make sure that they got elevated so that any bureaucracy that could be inserted, along the way could just be short-circuited so that we really focus on the outcome that we want, which is to build more submarines, to stand up Golden Dome, and then to fix our ground-based leg and air-based leg of the triad. So how do you think about success over the next three to five years?

We heard about this factory being a 50-year-plus initiative, but how do you think about success in the near term? I look at successes, being on the cadence to deliver the submarines we need to deliver. we may not be at the rate in three years, but we have to be closing on it, and we have to have all the levers that we need to pull moving in the right direction. Obviously, a huge portion of that

is this outsourced work as we build capacity because we don't have enough of it in our private shipyards that traditionally build submarines. We need to spread that throughout the rest of the country in order to be able to get that capacity up and to be able to hit the cadence that we need. So this next question, I'll start with Chris, But then I'll go to you, Admiral.

What is the hardest part of working with the government, working with the Navy, coordinating across primes, working with all the other folks you work with, working with legislatures? How hard is it to actually restart this industrial base? I would say three years ago was incredibly difficult, and now honestly to pull something like this off for the last six, seven months with Deputy Secretary Feinberg,

Secretary Phelan, honestly, compared to the size and scale of something like this, if you're in the commercial sector, is pretty fast. But it's because Congress, both, the Senate in the House, the Navy, the Department of War, has really hit the go-fast button and removed a lot of these blockers. And secondly, it's cultural. One of the smartest things we could be doing is, hey, there is a single accountable person that just runs this instead of 20 people trying to, to contribute. And now that that's all getting cleaned up, not just in submarines, but in

drawn dominance, in Golden Dome, you know, unfortunately, the vice admiral's got one of the toughest jobs at the company, but there is now a single person in command to control of the entire enterprise. It can make fast decisions, take risks, place multiple bets. We're going to be successful here at Hadrian with the enterprise, but we need to put like all of the bets in the ground now because the three or four years for us to pay off. So let's just do multiple things. That's not possible unless there's a single man or woman in charge. And that's been the biggest sea change of how easy or how it is to deal with legislature,

the Navy, the Department of War. And Admiral Goucher, you have a lot of different stakeholders that you're hurting together, primes, suppliers, other folks across the Pentagon. What are you finding to be the most difficult parts to coordinating success? I think you know that there are a lot of stakeholders. And I kind of joke when there's a law of conservation of authority,

When I get it, somebody else loses it. So you've still got to work through all that. And the truth is, we're really in the transition program. I mean, I'm really about a month into the job right now. So we want to do that in a controlled manner. A good example is this Hadrian deal was on the cusp of being solved. And so I went and talked to Jason Potter and the secretary and we said, hey, don't jump in the middle of this. Let's get this across the goal line.

And so things like that. But as we kind of move forward and I am starting to take things on, we just got to make sure that we have. have a smooth transition. So hopefully within a couple months, I'll be able to say that, hey, I've passed that an amount of bigger problems of how do we build submarines faster and then we're moving that ball down the field. And Admiral, I'm going to ask you a question about drones and missile interceptors because we see them in the news all the time in the fight in Iran. We see lots of talk about drones and missile interceptors. It's a pretty rare opportunity to have someone like you who has been a submarine commander who can talk about the silent

fleet to a certain extent. Talk about where submarines still fit in even a modern conflict today, to the extent that you give us some insight into how we should think about the criticality of submarines beyond just the second strategic deterrence and the nuclear triad. Well, I think as long as we keep thinking forward and pacing our adversary, submarines will always have a place. I look at just today our ability to get off of another country's coast undetected. We can choose what we wanted to from there. I mean, I can shoot a missile or I can shoot a torpedo. And as we look to underwater drones, we can start to tailor those. The real advantage there is that we can then build the payload and whatever I wanted to deliver outside of the submarine. And then I just have to figure out

how to load it on. And I could pick my payload. So whatever the conflict is of the future, that gives us an opportunity. The other thing that we can do, because we're self-sustaining, we have our own defense, we have our own command of control capability, is I could control other things. So I could see a world where a submarine goes far forward, it connects with the network of drones to give the signal. So you still have a human in the loop to make the decision or to see what's happening and provide that.

And the submarine may never have to shoot a weapon. Amazing. I want to shift to the factory here in Alabama. Chris, this is more than just a factory. It's really going to be a modern, advanced manufacturing facility. Talk to us a little bit about how a modern software-driven factory

works differently than a traditional factory. There are two big things. A big thing number one is we don't have enough machinists, we don't have enough quality inspectors, we don't have enough welders in the country. So at that task level, we have to get them 90% more productive and make the rest of the time human in the loop easier

because if you want to train a Navy welder today, it's going to take you a decade. So we have to augment the skills with software and also make it more accessible for training new workforce. That's number one because if we translate the 70 million hours target into people, with software we can make that 50% less people, 70% less people. The second thing is the flexibility that constructing something like a submarine demands is incredibly what we call high-mix low volume, right?

So if you're building a Toyota Camry, you're making 20,000 of them a year, you can set up production lines that are just cranking out, nuts, bolts, components, assemblies, in traditional kind of factory automation. Like Foxcombe with Apple, right? You can easily automate a million iPhones. What the software enables is running at that level of factory productivity, but with the flexibility that's something like a submarine demands, because you don't necessarily need a hundred of the same thing.

you need one and you need a slight variant, especially with sustainment and maintenance and even submarine construction, you don't know when you need stuff anyway. So having this flexible manufacturing system, apart from the raw productivity, it really is the speed and flexibility to reconfigure lines to give the yards what they need kind of on demand versus, oh, no, I realize that I need a spare to keep submarine construction going

and you go back to the supply chain and it might take 12 months. And the whole thing's just stalled out. So it's really the velocity and agility. and reducing the amount of people because we just don't have the people in the U.S. We have to get this productivity uplift with software and a new workforce. Yeah, I've seen some of your other factories

and your other facilities and people are going to come to start to recognize the importance of being able to change out the product line, increase throughput, have better visibility into the entire production line. It's really tremendous. Admiral, when you look at a facility like this, I know you're still ramping up to speed in your new job,

but you look at the potential of a facility, facility like this and you start to dive into the supply chain, where do you feel we can make the most impact around capacity? Where do you feel like there's the biggest bottlenecks in areas where we might find other opportunity to help increase our rate of production? There's a couple things that I'm looking at right away for this facility. One thing is our in-service submarines, so submarines that have already been built but have to go through their periodic maintenance, maybe to repair a pump or a valve. I think the opportunity to take obsolete parts where the companies have gone out of business, turn it over to Chris, have them figure out how to manufacture

or use at the facility here, obsolete parts, and then turn them back over so we can keep the existing submarines running will be a huge early opportunity for us. The other part of this is one of the big holders. I'm going to riff a little bit off what Chris was just saying with the factory. There's really three things you need to build a submarine, right? You got to have people, right? At some point, you have to turn some wrenches. Okay. Those people have to be, productive. And what I would tell you is, in some areas today, we are less than 50% productive. So essentially for every two people, I'm getting one hour, right? And that's no one's fault. That's just where we are. It has to do with experience. It is an exceptionally difficult process.

And it's always been that way. But I think even if we were operating at our peak performance of the mid-1980s when we were pumping out four submarines a year, we were still only at about 0.7 or 0.8. I mean, only so productive you can get with a human. So I think by you use, using machines, we have an opportunity to actually get that productivity up. Not only are we able to run it around the clock, we're able to be more effective in doing it. And then the last thing that you need is you need the actual parts, which we've been talking about. And the biggest thing that we find is you walk around our yards, it slows them down

and stops them is what's called sequence critical material, right? You get to the point where I've got one more part I've got to have and I can't finish the step. And I've got to get to that and be able to manufacture it. As we get spun up past the first articles, and we really show the capability of this plant, I think those will be the types of things that we push down here. And Chris and I have already talked about things. It's things like air flasks.

It's things like our hatches for escape trunks and things like that, that he's very confident we'll be able to do here in this plant. That's fantastic. I think, yeah, having software, as Chris mentioned, play a role in making sure that, you know, if you only have 95% of the parts, that's not enough to make a submarine. You need 100%. and having software help drive some of that production line and provide visibility is really key.

So Chris, turning it back over to you, how are you going to prove to the Navy that you can bring up this capability and deliver quickly? Firstly, we've already proven it, but I will say this is a like generational huge lift, right? This is going to be six, seven hundred people just setting the thing up. You know, what I think is we've got a really good plan. We've got really good partners in the Navy.

We've got really good partners in the primes that own the designs. and I think everyone understands that we're going to go super fast and shove things through the qualification system as fast as possible to make sure we're de-risking it as fast as possible. And then building a submarine is complicated. Some of this is going to come online fast in seven months

and some of it's going to take two years. The other thing, frankly, is that we're dedicated to is the way we've structured this is basically we can't be successful out of the Navy but we're also capital at risk. So we are strongly incentivized to basically, I'm sorry, it comes down to culture, sweat bullets until we pull this off, not just because of the mission, because of how we've structured this entire enterprise.

Okay, so we're going to wrap up here in a few minutes. I want to talk about a couple last questions. The first, we talk about a Columbia-class submarine, costs $16 billion, takes almost a decade to build, carries the nuclear deterrent, and then in my world, in the venture capital world, we talk about companies like Serronic and Anderil that are building very small, much more lower cost, autonomous vessels, underwater drones. They cost a fraction of that, and they're designed to be attritable, to be expendable. How do these things fit together, Admiral?

I think that the process of manufacturing some of those drones has some similar corollaries because you're having to build an integrated system. And so when you build a submarine, that's a similar process. You'd actually, some of the experience that companies like Ander will get from building these unmanned systems can translate to building modules for submarines. To be honest, I had that conversation two days ago with Anderl and how we get to that end state. I do think there are some corollaries, but you also have to remember that those drone capabilities are typically a single mission. And part of the value of whether it's a submarine or a surface ship

with humans on board is that they are multi-mission systems. They're not just a niche tool. They're a responsive, versatile tool that can do any number of things. I like your comment earlier about also. The submarines can be the conventional. man in control for a whole fleet of autonomous systems. I thought that was very cool. Chris, I want to kind of start to wrap up with you here. This is Factory 4. You're endeavoring to have factories 5, 6, 7, 8, and beyond.

What has to go right to make this model repeatable across the country? And what's one misconception that you want to make sure that everyone here today and everyone who's listening at home when we release this? What's one misconception that you want to correct for them so that they know about building and manufacturing here in the U.S.? I think the biggest misconception is, hey, it's all automated. We don't need the people. The amount of software engineers, you might have smart people from Tesla and SpaceX and GE that we need to set the things up.

And the amount of workforce development, it has to be a national mission. And it's all about the people and the productivity combined. We've already proven that we can do this in different states. I think people do not understand how complicated a submarine is to construct and design. and secondly, if you think about easy manufacturing and difficult manufacturing and maybe you put Starship at the top of the list, submarines in terms of tolerancing, your precision and quality and welding are easily number one. So to do something like this for the first time ever that's not the private shipyards or the Navy itself and do that knowledge transfer is a huge deal.

Because it's not just making this stuff. The level of precision and quality we have to operate here is as hard, if not. more difficult as building a rocket or something like that. And this is like a very, very serious endeavor because this man has to send a bunch of sails underwater and like this thing has to go 100% right for 30, 40 years. It's huge. Admiral Gautcher, we are very, very lucky to have you here. I believe these are your first sort of public remarks that you've taken on this new role. You've been a submarine commander. We are just incredibly fortunate to have you here. We're incredibly fortunate to have you serving our country. Both of you gentlemen,

I want to thank you for being here for everything you're doing for our country. And thanks for doing this. Thank you, David. Appreciate it. Yeah, thanks very much. Thanks for listening to this episode of the A16Z podcast. If you like this episode, be sure to like, comment, subscribe, leave us a rating or review and share it with your friends and family.

For more episodes, go to YouTube, Apple Podcast, and Spotify. Follow us on X and A16Z and subscribe to our Substack at A16Z.com. Thanks again for listening. And I'll see you in the next episode. As a reminder, the content here is for informational purposes only. Should not be taken as legal business, tax, or investment advice, or be used to evaluate any investment or security and is not directed at any investors or potential investors in any A16Z fund.

Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see A16Z.com forward slash disclosures.