China Outbuilds America 230-to-1. Saronic Has a Plan

I am thrilled—really thrilled—because Saronic is a defense company that has been doing amazing work, and we're so lucky to have the co-founders here on the program. Welcome to the program, Dino Mavrukas. How are you, sir?

Doing well. Thanks for having us.

And your co-founder here, Vib Alka car. Let's talk a little bit about Saronic. We're going to get into everything, but you were in the news because you're making autonomous ships, and one of them was used recently in the Strait of Hormuz, I understand.

I have to start there, Dino. Tell us about the mission. Tell us what happened. This is the first time an autonomous ship has ever rescued people in the field, and then we'll get back into your history and everything. But I think we have to start with this story because it's incredible.

It's just a huge story. I can't get into too many specifics of the operation and the mission for obvious reasons, but before we get into what it means for the country or our company, the most important thing is that those 2 pilots came home safely, regardless of how they got there, right?

What we were able to demonstrate through this partnership with the Navy—for us, the Navy actually trusted us in this moment. Lives are on the line. We're going to send this autonomous platform to rescue these people. This is not a patrol. This is not maritime domain awareness. This is the highest stakes of the highest stakes.

There are downed American pilots in the Strait of Hormuz, and there's Corsair, which you actually see right over my right shoulder. That's our 24-foot, fully autonomous speedboat, for those who don't know. He said, “We're going to use that to rescue the pilots.” That's actually a milestone moment for the country.

I can tell you a personal story firsthand of what this means. Leaving people behind in the U.S. military just isn't a thing. The U.S. military does not do it. I was in the Navy SEAL teams for 11 years, from 2004 to 2015. My last 5 years were on SEAL Team 6. I personally experienced this in 2005: There were 4 SEALs trapped on a mountainside in Afghanistan, and the helicopter was shot down while going to rescue them.

What we were able to prove in this moment was that you can go and rescue people without putting additional soldiers in harm's way. That's the impact of what this had. That's what it means for the country. For our company, it was an extremely proud moment where the Navy trusted us in this high-stakes environment.

Absolutely incredible. Obviously, thank you for your service. I know you can't get into too many details, but I've been reading all about it. I have to wonder: When did these 2 pilots know that this was an autonomous ship?

When the ship comes rolling up, did somebody send them a message saying, “Hey, you're going to get picked up by the autonomous ship. Grab the side of it. Climb on board”? What's the protocol for the autonomous ship picking up 2 people? Are they swimming out to it? I just want to get a little color on this before it becomes a movie.

It's the first time that it happened, right? When a boat pulls up to you and there's nobody on it, you can very quickly deduce that it's fully autonomous. These pilots were in the water for—I don't even know how long. They were in a completely contested environment. We're in a conflict with Iran. They're in the Strait of Hormuz, and we can go through the dynamic there.

But they looked at this platform and said, “That's my way home. That's how I get out of here,” right?

I love it. All right, let's step back a bit here. You're a Navy SEAL. Vib, you've worked in this space for a while, and you previously worked at Anduril, I understand.

The mission, as I understand it, is to greatly increase America's ability to produce ships. For people who don't know, the amount of tonnage and the number of ships we're able to make in America and most Western countries is very low. But there's an adversary in the Pacific that is able to produce a lot of ships. It's known as the Chinese Communist Party.

Maybe from first principles, you could explain what you're building and why it's important, and just how insane the incompetence of the West in terms of building ships has become. We are really bad at making ships right now. Am I correct?

We're beyond bad. Let me take a step back and highlight where we're at, and then I'll let Vib go through what we're doing about it and how we're changing that dynamic.

But, Jason, what you were saying—just to put it in context—the United States can build 100,000 gross tons of ships every year. A gross ton is this volumetric measure of a ship. If you build a 200-foot ship, that's different from an 800-foot ship, but let's compare apples to apples: 100,000 gross tons.

The Chinese, which you were referencing, can build 23 million gross tons. So they can outbuild the U.S. 230 to 1. It wasn't always the case. 30 years ago, the Chinese didn't have this capability. They had 5% of the world's shipbuilding capacity. Today, they have 57% of the world's shipbuilding capacity.

That problem is actually getting worse because the Communist Party is constantly subsidizing the entire industry. They're undercutting the commercial market on price. They're not just subsidizing the direct cost to build; they're subsidizing raw material and labor costs. So a ship built in the United States is actually 5 to 6 times the cost of that same ship built in China. They keep getting more and more orders, so that 57% is just going up.

How did they get that capability? Is it because, as a manufacturer to the world, they needed to ship out products, so they did this in a commercial way first? Or was it always very intentional—that they needed to own the South China Sea?

They grew around the commercial market. Then I'll talk about the status of our Navy—where the naval fleet is today, what that means, and then where the Chinese Navy is and where that's going.

Because, again, 30 years ago, the Chinese Navy was a littoral backwater navy. Today, they have nuclear submarines, aircraft carriers, and hypersonic missiles. There's been a massive change over the last 30 years.

What's happened in the U.S. shipbuilding industry is a complete decline. The U.S. naval fleet has 296 ships today. There's a congressional mandate—a statutory minimum of 355 ships—that was set by Congress in 2018. That was 8 years ago. Last year, we built 9 ships, but we retired 19 naval vessels. We're going in the completely wrong direction.

The Chinese delivered somewhere in the ballpark of 30 ships. Their fleet size is around 370. They'll be at 450 before you know it. But the commercial market is where that discrepancy is even larger, right? They delivered over 1,000 commercial ships last year. The United States built 5.

Five. 1, 2, 3, 4, 5.

1, 2, 3, 4, 5. One hand.

Wow.

So if a conflict actually kicks off—and we saw this in the U.S. in World War II, when we had the production capacity and were building thousands of vessels every year—that commercial capacity isn't going to go to building cargo containers anymore. It's all going to flip to defense capacity, and it's going to be building for what the Chinese Communist Party needs to win a conflict.

That's what we have to be building toward. That's what we're doing at Saronic. That's why we're thinking, “How do we build mass? How do we build scale in whatever it is that we do?”

And so, how do you do it, Vib? I would think—and listen, I'm a neophyte here—but if these are autonomous ships, you can remove things. You don't need beds, a commissary, or latrines. I don't know what percentage of a Corsair or battleship is there to support humans on the ship, but I have to think half the mass of it is to keep humans on it. Am I directionally correct?

I think half is a little excessive, but—

Okay.

There's a significant percentage of the components, the cost, and the labor that actually go into catering to human beings on board. At a high level, you're really reducing the complexity of the entire product. You don't need to worry about a separate electrical system for humans. You don't need to worry about, as we mentioned, doors, bathrooms, stairs, et cetera.

At the end of the day, you can start to end-to-end optimize a lot of different things. Even if you look at Corsair, for example, if you were to buy a commercial 24-foot boat, it's unlikely to go 1,000 nautical miles. It's also unlikely to drive and have the control authority of the boat that we built here because, again, you don't have to worry about a person taking shots in the middle of the ocean.

So you can optimize across a bunch of different areas.

You can optimize naval architecture—the speed, range, and payload of the craft itself—and reduce a lot of the complexity by removing subsystems that are meant for humans.

Yeah, and the Corsair, which is my favorite ship just based on the name—I believe from my research it was originally meant as a pirate ship or a privateering ship—but is it now standard lingo that it’s a 24-foot boat with a certain profile? Is that how these designations come about, the names of these ships?

I wouldn’t say that Corsair is now the universal 24-foot autonomous surface vessel. It’s the name of one of our product lines, and we chose it very purposefully; we chose all our names purposefully. Even the name of the company—we were talking about our Greek heritage before the show and where that came from.

The Saronic Gulf is a gulf off the coast of Athens where the Battle of Salamis happened. It’s one of the most famous naval battles in history. It came after the Battle of Thermopylae, the famous stand of the 300 Spartans, and it was the naval battle between the Greeks and the Persians where the Greeks were wildly outmatched. The Persians had much, much larger ships and a much larger fleet, and the Greeks were able to lure them into this narrow strait and use smaller, more maneuverable platforms to ultimately defeat the Persians.

Now, cost is part of this in terms of these ships in the field and the number of ships. My understanding of contemporary China is that they don’t build a lot of big ships; they build a lot of ships. So, they’ve got a lot of smaller and very varied numbers of ships, if I’m correct there. Then the cost becomes the issue, because building a bunch of aircraft carriers—I keep hearing that supersonics are going to take out all these aircraft carriers in the first 15 minutes.

Dino, is that true, what we hear? Do we need smaller, faster ships? And what is the cost of something like the Corsair that you used on this incredible mission?

Yeah, I would say, at a high level, the goal is mass—large-scale, attritable mass. If an aircraft carrier takes 10 years to build and $13 billion, you’re not really achieving that. I think there’s always going to be a role and responsibility for a craft like that, but ultimately, you’re right: if you’re big and very obvious to find, and people can find you from space, it’s a real risk.

When you reduce the overall cost of just doing things in the ocean, that’s ultimately where the autonomous piece comes in. You can deploy at scale and monitor a larger body of water with fewer people. Ultimately, what you’re really getting to is removing people from harm’s way. So, when we think about designing and thinking through the price points of these things, the comparison to manned craft isn’t necessarily even comparable as you get into larger ships.

For us, it’s really a question of what is the cost that you would typically have to use humans for, and how much is that actually being reduced by having a large autonomous fleet?

Corsair’s a million bucks or something, do you know?

Yeah, I want to go back to one thing you said earlier: China is building both very large ships and a lot of ships. They’re doing it all, right? When we think about how you actually change the paradigm for the U.S. Navy, let’s boil it down to unit economics. Let’s think about this like a commercial company.

You mentioned payloads. What are you actually moving on the ocean? One of the most important metrics is VLS tubes—vertical launch systems. It’s basically how many big Tomahawk missiles you can carry on a ship; it’s how much firepower you’re fielding. So, let’s use an example of that: a naval destroyer costs about $3 billion.

$3 billion?

$3 billion. And I’m not even getting into the sustainment and maintenance costs over the life cycle of the destroyer. I’m just talking about what it costs to buy. If we do that, it’s $10 billion over 30–40 years. It costs $3 billion and takes 6–8 years to make. You can field about 96 VLS tubes, so I’m going to round up and make that 100 for easy math. On average, you’re fielding somewhere between 10 and 15 VLS tubes per year at a cost of $30 million per tube.

Marauder—I won’t give you the exact price or cost—can reduce that cost by a significant amount, and we’re going to be building 20 Marauders per year at our shipyard in Louisiana. Now, we can carry the equivalent of 16 VLS tubes on the back of that 180-foot, fully autonomous ship. So now you can field 320 VLS tubes per year as opposed to 15, and you could do it at a fraction of the cost.

This is how you actually save the taxpayers hundreds of billions of dollars and get more capability into the field faster.

That’s just extraordinary when you think about it. What do you lose by not having humans on it? Are we going to be faced with a moment in time where these are kind of like drones, where there are no pilots? They’re just pilots in a shipping container dropped somewhere, either in the middle of America and doing it remotely, like we’ve done.

Is that the future of warfare with these ships? What do you lose when you don’t have a team there on the boat? And what is the criticism—valid or invalid—that people have of unmanned ships?

Well, I’m going to let him talk about the way that we’re using software, because I think the criticism, to your question, is: What happens if we’re in the middle of the ocean and the ship breaks down? Who’s going to fix it? It’s not that we don’t need this technology; it’s folks who aren’t used to technology saying, “How’s the technology actually going to work?” And that’s what we’re proving out every single day.

To your point of what do you lose by not having people on them, it all boils down to the mission. This is why we partner so closely with the Navy. This is why we work with actual operators: What are you using? What are you delivering? If the mission is long-range fires, we’re packing a bunch of VLS tubes on the back of Marauder. You actually don’t lose anything. You’re just delivering those capabilities into the field while keeping people out of harm’s way, and that’s a very, very important part of the equation.

But if our ships—let’s be clear, if our boats and ships aren’t accomplishing the mission—then they’re not relevant. They’re not going to be used in the first place. And there is going to be some mix. I don’t know what that mix is. I don’t think anybody really knows what that mix is of manned and unmanned ships.

So, what you’re going to see in the Navy is a hybrid Navy. The things that you’re missing from having people on the ships—forward-deployed decision-makers in the environment—you can actually have out there on manned ships, but you’re using unmanned ships to put into combat areas and conflict areas to keep those people very safe.

Fab, take us through the production of these. How do you achieve speed at it? A friend of the pod, Elon Musk, with the Teslas, over time eventually built the full stack. In the beginning, if you had a Tesla, you would look at the shifter and say, “Hey, that looks like my Mercedes.” That’s because they had their steering column from Mercedes, one of the early investors. The HVAC was from somebody else.

But eventually, he had materials coming in one side of the factory and cars coming out the other, and he made every single thing in that car. He was no longer dependent on any third-party manufacturers or a supply chain. Famously, the HVAC in the Model 3 and Model Y, when they came out, reduced the drain on the battery massively, because it was purpose-built for that mission.

So, I’m assuming you’re doing that from first principles, but what’s the secret to this rapid, lower-cost production?

Yeah, listen, I think within the shipbuilding industry at large, there are a bunch of different challenges as to why production isn’t as high as it should be. One of those core reasons, and the Navy has talked about this extensively, is investing in the supply chain.

When you only build 5 commercial ships a year, your genset manufacturers and marine engine manufacturers also don’t have to produce at a very high rate. One of the things that we get the luxury of doing by being vertically integrated is that we get to work with these manufacturers directly, do nonrecurring engineering work, and invest in digitizing a lot of the components that historically have been very analog.

You asked earlier, “What do people—what do the people do on the ships? What are they going to do if the ships are autonomous?” When you think about, as Dino mentioned, the life cycle of these ships being 30 years, and you think about a ship that was built 30 years ago, you have people doing extremely manual things in the engine room that today should be extremely trivial to automate through software.

A lot of that work actually gets completely absorbed by the software stack, so you don’t have to worry about it. As we choose components, we invest in the supply chain; we invest in the industrial base. We actually help bring all of these analog components that, for a long time, were meant to be operated by humans into the modern era. We give them software APIs, make sure that they can be controlled by MCPs or a software stack somewhere else, and can be remotely accessed and whatnot.

So, the complexity is going to go down massively when you're building one of these, yeah?

Absolutely, yeah. And frankly, I don't think that the person who's sitting in the engine room is super thrilled about it anyway.

No, not a great day.

You're actually changing the—you’re democratizing the decision-making, right? You have more mass out in the ocean, and more sailors are going to be making decisions about what that mass is going to be doing and where assets are going to be, instead of worrying about coolant pressure in an engine, right?

And so, when we think about the actual design, there's a lot of different variables there. There are some components that are commercial off-the-shelf, right? Elon, in the early days, from a compute perspective, had to invest in ASICs, right? The compute stack wasn't there to run large models, run inference, and meet the power budgets that they wanted to. In 2026, it's a totally different game, right?

There's a lot of different parts of the supply chain where we're looking at investing in the industrial base and making sure that our engine manufacturers and genset manufacturers are keeping up. On the other side, there are particular components where we're like, "Hey, the marine industry actually hasn't had this level of investment in a long time, so we're going to have to build this in-house."

A couple of examples around that are a lot of electronics related to the sensors and compute, but there's a bunch of different things there that help us build and assemble quickly.

Let's talk a little bit about Dino on the business side, how you work with the Department of War, formerly known as the Department of Defense. Emil Michaels and a bunch of the folks over there are really focused on getting startups, essentially, like you are, more contracts. But you also are moving out, I assume, of this cost-plus world, and you're just figuring out, "Hey, what's the right price for this?"

Take us inside that paradigm shift. You have these primes, and then you have these upstarts. It seems like, based on history, the incumbents sometimes don't take this lying down. So, take us inside the paradigm shift from cost-plus to what these—I guess they call them the new primes—like yourself and Anduril are doing.

Yep. I'll just define cost-plus for everybody listening. Basically, it means, "Hey, we're going to go build you something, Mr. Government, and however long that takes, however much it costs, you're just going to pay us 10% to 15% on top of that."

So, I'm actually incentivized—and I'm not saying there's malicious intent—but I'm incentivized to make it take longer and cost more, because if it costs $100 million, then I only make $15 million. But if it costs $1 billion, then as a company, I make $150 million. So, there's this perverse incentive to drive up costs.

Now, that's not the only reason the industry is where it is today, but we're very encouraged by the push from the administration to adopt things at speed and scale, to do that quickly and differently, and to move away from cost-plus.

You had it: We invest our own capital. Today, we're just about 4 years old. We've raised $2.5 billion in private capital. We're pouring that into R&D.

I remember when we decided to build Marauder, again, a 180-foot ship. Viv and I were talking, and he was like, "Who's crazy enough to go and build a ship on IRAD?" And he was like, "Well, we are," because that's what the country needs. That's what it demands. That's what we're going to go do. And our investors are giving us the capital to go do it.

They're actually entrusting us with that capital. That's the reverse of what's happening in the large primes, where their investors are kind of demanding share buybacks. They're saying, "Give us our capital back."

So, the country very much needs companies like us to build for the future. When you look at what the administration is doing, even just in shipbuilding, it's recognizing that shipbuilding is a critical strategic industry, right? The president, one of the first things he did was put out the maritime executive order. That was followed by the maritime action plan, right? There's the SHIPS Act sitting in Congress.

You had the Secretary of War give a whole speech on defense acquisition reform, talking about how we're going to push the department to be better, to move faster, to work with companies like us that are building on firm-fixed-price contracts, that are focused on delivering at speed and scale, and driving down cost for the taxpayer.

So, all of that is extremely encouraging, but I also want to be clear here on the flip side. We're in the early days, Jason. Like—

Yeah.

Very early days. When you look at the overall Department of War's budget, 1% goes to autonomous systems. That's not nearly enough. Make that 5%, make that 10%—I don't know the right mix—but it needs to go up, and it needs to go up quickly.

I'll just add one more thing. Historically, shipbuilding—the design is bifurcated from the builder, right? So, the design—someone gets paid to do the design and the ground-up general assembly and all that—and it's handed off to a totally different person to go and build a ship.

There's a statistic out there that something like 70% of sequence-critical components on some Navy ships are sole-source suppliers, like single-source suppliers. And so, to Dino's point, it's not just that you're incentivized to take longer; it's that you're not incentivized to change that necessarily, because you know that you're still going to get paid.

Whereas, from our perspective, when you own the design, you own the shipyard, software engineers, shipbuilders, naval architects, and welders can all sit down and eat lunch together in the yard. There are end-to-end optimizations that you get that the system, from a historical perspective, just wasn't set up to gain.

Yeah, let's do a little look at what happened in the Strait of Hormuz. We weren't able to control it completely. It doesn't seem like anybody in history has been able to control it.

If we were to fast-forward and you had your entire fleet—let's assume not unlimited, but massive capacity—how would you be able to secure something like the Strait of Hormuz with these autonomous vehicles? And what might that look like? Take us into a scenario 5 years from now. What would it take?

That's the type of scenario that we're building for. I think before talking about the scenario, let's talk about the Strait of Hormuz itself, right? The environment, the part of the world—what neighborhood is it in, right?

It connects the Persian Gulf, which is 200 miles wide, to a 20-mile-wide strip, which is the Strait of Hormuz, and then connects to the Indian Ocean and connects these large oil-producing countries to global markets. On one side of the Persian Gulf, you have Iraq, Kuwait, Saudi Arabia, the UAE, Qatar, and Oman. And on the other side of the Persian Gulf, the north side is all controlled by Iran.

So now Iran is saying, "Well, how are we going to disrupt traffic in the Strait of Hormuz?" Well, it's only a 20-mile-wide stretch. I can use missiles, I can use fast-attack vessels, I can use naval mines, I can use cheap drones, because I don't have a large distance to cover. I can launch these things anywhere from shore.

So, it becomes this completely chaotic environment where sending naval destroyers into it, or trying to get commercial ships past, becomes extremely dangerous. Using large numbers of autonomous vessels, you can combat these threats in completely different ways. You can offer scale, persistence, and risk reduction in ways that manned vessels just can't.

And again, I won't get into too many of the specifics on the types of operations that we're building toward, but autonomy is a way to counter these effects—these effects that really aren't meant to be countered by traditional manned ships alone. We can counter that with scale and technology and keep people out of harm's way.

Well, you have some small ships. Yeah, you've got the Cutlass and you've got the Spyglass in this buildout. Those are 6-foot, 15-foot, or whatever.

We actually started with those, and I'll give you an interesting story. We actually don't build those—we haven't productized those. Those were great subscale demonstrators, but our real flagship product is Corsair.

And when you talk about, again, the scenario in the Strait of Hormuz and what type of volume we can bring there, it's not only 5 years in the future. We could build 2,000 Corsairs per year right now at our manufacturing facility in Austin. I know you live in Austin. We'd love to show you around.

When you see it—when you see it, it actually comes to life. You're like, "Oh, wait. You can actually build thousands of these vessels every single year."

How many would you need? 1,000? 5,000 in the Strait of Hormuz?

I don't think you need... It's a 20-mi wide strait.

This is their offensive strategy. They have these speedboats that they use. I don't know what they're called.

They call them fast attack vessels.

Fast attack vessels. These guys are essentially kamikaze, right? They just zip in, with no regard for safety or their lives, and they will attack.

They're using fast attack vessels. They're using cheap drones, so you have to be able to have counter-UAS capabilities. They're using mines. They're just floating mines into the strait without regard for what those mines may set off. So, it's a really complicated environment that you need to bring a lot of capabilities toward.

Yeah, but it sounds like with artificial intelligence, which is the next card that I think is going to turn over, you put these Corsairs out there. Could they, with AI, be able to use satellite data from Planet or whoever's providing it, or military satellites, and just zip out there and know, say, somebody's putting out some mines? Somebody's putting out some of these fast attack vessels. Just intercept them.

And now they have an actual cost: human life and a limited amount of resources. We're America, and we have you guys building ships at scale. They don't have this ability to build AI-based ships. So, talk about what AI on these ships could do eventually.

Yeah, so I would say today, there's a bunch of different machine-learning applications that run on our boat today. There's heavy amounts of compute on all of our craft that allow people to have tons of capability, whether it's navigation and the autopilot or self-driving of the craft itself, or compute for payloads, whether that's counter-UAS, as Dino mentioned, or a bunch of other things, really. There's a variety of different mission sets.

The end state here is that you have a distributed fleet across the ocean that, invariably, because the ocean is a treacherous place by default—the ocean's still bad even without a war zone. Salt water hits your antennas, and you lose comms. So, there are a bunch of different radio communications on this. There are a bunch of different sensors. All of those things are networked together with machine learning and a bunch of different software components of the software stack to make sure that you can basically translate intent from a commander or sailor and then result in a mission outcome.

Let's talk, Dino, a little bit about the adversary, China. They have started to put weapons on robots. I'm assuming that they're doing the same if they're willing to put weapons on one of those dog robots, like the Spot robots. They're putting weapons on them, and they're demonstrating it. Whatever they're demonstrating, they've already got in production. Let's just call it what it is.

They're not signing some UN treaty not to put autonomous and AI weapons into use. This is a different type of adversary. I'm sure North Korea and other bad guys are doing similar things. I believe we have treaties where we're not allowed to put weapons and use AI-based weapons. Am I correct about that in the West here?

There's an autonomous weapon system standard called 3009 that basically outlines exactly how AI can be used on an autonomous system.

Tell us all about that, because this is the RoboCop scenario. This is the robotic soldier or the robotic boat that's able to engage the enemy without a human if it loses comms, or to do a kamikaze mission, which seems like something we might have to do. These are things we might have to do, Dino, versus the Chinese, because they're not signing these treaties. They're not thinking ethically. They are not respecting the Declaration of Human Rights. They've got their own playbook, right?

I think that the important thing to articulate here, especially when you use the RoboCop example, is just to quell people's fears a little bit. Even in that example, where we're sending boats off the coast of Taiwan and we're in the Taiwan Strait and they're engaging, we're still using artificial intelligence to bifurcate between an enemy vessel and a friendly vessel, and between a noncombatant and a combatant.

People are still making the decisions and setting the mission intent. It's just government policy that's setting, “Okay, when are you authorized to engage versus not?” Artificial intelligence actually just gives the military the opportunity to operate at a greater scale and then operate more efficiently. These aren't robots actually just making decisions on when to go to war or not.

Got it. So, it says, “Hey, this is a bad guy. Our recommendation is to neutralize it.” Then a human has to say, “Okay, neutralize it.”

But if you're up against an adversary who says, “No, no, we built this thing. Anybody who comes into the blockade in Taiwan”—which we think is a pretty significant nonzero probability in our lifetime—“I'm just going to have these ships kill anything that's not our ship.” Period. Full stop. Then we lose, don't we?

No, because all you're doing is putting policy thresholds and approval thresholds into the technology. In a real-world doomsday scenario, we go to World War III, the Taiwan Strait is a complete conflict. Exactly what you said: our commanders are going to make that same judgment, right? They're going to be like, “Hey, there are no fishing boats out here right now.”

Yeah.

There's no cruise ships coming through the Panama Canal. If you're not a friendly vessel, you're an enemy vessel, and that's that. Our technology actually enables that.

What we were talking about earlier—the autonomy, the artificial intelligence, and everything else—is there, and you're building the policies into the software based on government procedures and when those authorizations need to be in place. Those authorizations will change based on the threat environment that we're in.

Is the plan for you to sell to our allies? And what are the rules there? Because it sounds like we need as many ships for our Navy as we can get. But if you can get up to capacity, then the French, the Germans, and the English have essentially just stopped making ships, right? They have very depleted militaries, is my understanding. The only people who are really doing a great job at this are the folks in South Korea, which I believe are building only 1 ship for us.

Maybe you could talk about the other top countries in the world—democracies and our allies—and what their capability is, and then your ability to help arm them and sell to them.

We're very focused on our allies and partners. If you talk to the Department of War, if you talk to the United States Navy, they actually want our allies and partners to have these capabilities as well. This is going to be a combined effort, right? We mentioned earlier that China has 57% of the world's shipbuilding capacity. That's not limited to just that one industry. We could go industry by industry and talk about their manufacturing might and their manufacturing dominance.

You really do need a combined effort. You need our allies and partners to have this technology. We're talking to our allies in Asia, obviously, and Europe and the Middle East right now about getting not only our products there, but also standing up domestic production in those countries. How do we give them sovereign capability? How do we have production lines overseas so that they're already forward-deployed if we get into a conflict in the first place?

Pretty amazing stuff. And you have an announcement today, I understand, that we're going to share with the All-In audience for the first time.

That's right. So, we've been working on Port Arthur for a long time now. Port Arthur, just for the listeners, is our shipyard of the future. This is going to bring American shipbuilding back to a place that we haven't seen since World War II.

To be clear, we're already operating a shipyard. We have a shipyard in Franklin, Louisiana. That's where we're building Marada. We're ramping that up to 20 Marada. We can bring that up to 50 Maradas per year, every single year. That is a 100-acre shipyard.

What we're doing at Port Arthur is 10x-ing that. We're looking at a shipyard that's well over 1,000 acres. We've been doing this search for over a year now: What is the ideal location for this shipyard? Where are we going to find the right partner, the right land, the right workforce, and the right community to be a part of?

We're proud to announce that we found that in Brownsville, Texas. We're headquartered in Austin, and we're going to be basing Port Arthur in Brownsville. We're going to start on an 800-acre plot of land, which, in and of itself, is now the largest shipyard in the United States.

Amazing.

Just to give you a sense of scale, we have the opportunity over time to scale that to 4,000 acres. We're going to invest billions of dollars into this project. We're going to create 10,000 jobs over the next 10 years. We're going to bring shipbuilding in this country back again to a place that we haven't seen since World War II.

We're just grateful for the partnership from the state, from Governor Abbott, from Cameron County, and then from the city and the Port of Brownsville for making this happen.

I mean, this is one of the great things, Viv, about being based in Texas. They let us build things in Texas. As a recent resident here, it's pretty amazing to live in a state where the state's like, “What do you want to build? How can we help?” as opposed to, “Hey, F off,” like they told Elon in California, and he got the message, received it, and replied to that lunatic.

So, maybe we can talk a little bit about operating Saronic in Texas and the crazy support you get from Abbott and everybody.

One of the questions that we used to get back in the day was, “You guys are a maritime company. There’s no water in Austin. What are you guys doing there?” I would say Austin has this magical nexus of both digital and physical labor forces. You can find software talent, AI talent, and designers, but you can also tap into the industrial base from San Antonio and the oil and gas industry in Houston.

When you marry those two together, you can actually start to co-design these things across the entire stack. I mentioned earlier that typically ship design and shipbuilding are totally independent. With Port Alpha, we get the luxury of greenfield-designing what the manufacturing plant for ships should look like.

You can optimize the design of the ship for the yard and the design of the yard for the ship. It’s very similar to how companies today do hardware-software co-design on chips. You design the compiler, and they fit together to optimize both for the other. We have the luxury of being able to do that for ships here, and I can’t think of a better place to do it than Texas.

Well, Starbase is right down the block—SpaceX, obviously. Elon’s made that work, and I think it’s a 4-, 5-, or maybe 6-hour drive down to the coast, right? You guys can either zip there on a PJ or just drive, so it’s pretty straightforward.

It’s straightforward. It’s easy for logistics. What SpaceX has done in Brownsville is incredible. I think what Brownsville has done for SpaceX, helping them build out the community down there, is incredible.

We were very impressed and very excited to be basing Port Alpha there. At the end of the day, it’s the people who are going to make this work. It’s the labor force; it’s the workforce. Viv mentioned it earlier, and we can talk about why the American shipbuilding industry has declined so much, but it’s the supply chain and the workforce that have suffered the most.

One of the things we focus on is how we rebuild that workforce. How do we upskill existing talent for advanced manufacturing? How do we take somebody who’s building a car at Ford and get them building ships very efficiently and very quickly?

That all starts with the design. If you design a ship from the ground up, from a first-principles approach, to be more simplistic and to be designed and manufactured at scale, then you can put in the training, the work instructions, the processes, and everything else to make it much faster. Let’s train and build the workforce.

Then, let’s be honest: You have to make it cool for young people to go work in a shipyard again. You have to pay them good wages. Every one of our employees—every welder and every pipefitter—has good wages, the same benefits as our Silicon Valley software engineers, and equity in our company.

They’re fired up. I mentioned our shipyard in Louisiana; people are fired up down there. It’s not just the workforce—it’s the community. Some of the stories I hear when I’m down there are so encouraging. People say, “I’m wearing my hat in the grocery store, and people are coming up to me saying, ‘You guys are kicking butt. Keep building ships.’” Nobody else in the country is doing what we’re doing.

These are great salaries. People are getting paid significant amounts of money, obviously, to work in manufacturing at your company and down at Starbase. Now you have these 2 incredible companies. I don’t know if there are any other ones in Brownsville, but this is a small community. There are only about 200,000 people down there. This is going to be an incredible boomtown.

It’s going to be incredible. It’s going to be awesome. It’s going to be awesome.

I mentioned earlier the cost of building ships in the United States. The United States just isn’t competitive in the commercial market. Again, we can go through the specifics of what happened over time, but it costs 5 to 6 times as much to build a ship in the United States as it does in China.

We think that by investing in products and processes and new shipyards, we can cut the cost of a ship in the United States in half. If you have a ship that costs $300 million, we can get that under $150 million. That’s not by simply paying people less. We’re paying people more, but we’re investing in our product and our process. We’re designing the product to be built that way in the first place.

We’re making the investments needed for long-term sustainability and to rebuild the workforce.

All right, if people want to come work with you, Vib, how can they apply? Is there a careers page, and what are you looking for? What talent do you need to come do this incredibly patriotic duty and go work at Saronic?

Saronic.com. We have a careers page. You can reach out on Twitter, LinkedIn, or wherever you want. We’re hiring pretty aggressively right now.

The biggest thing we look for, generally speaking, is being fairly neuroplastic, especially in 2026, when ChatGPT and Claude are getting so much better every year. It’s important right now to be ready to adapt to what the future looks like. Computing is changing. There are going to be agents that can help you on the manufacturing floor to assemble faster if you have an issue. All these digital systems are changing the paradigm here.

When we first got to Gulf Craft down in Louisiana, a lot of the processes were pretty much pen and paper. That’s common across most shipyards. Bringing that into the 21st century quickly is going to require people to be thoughtful, adapt to the future, and be passionate and excited about building the future.

Go to Saronic.com and join the mission. There are 300 jobs listed on the website. There’s something there for everybody.

As a fellow Greek, Dino, we were obviously the cradle of innovation, where science, math, technology, art, and everything happened. But we were also great warriors, and we patrolled the sea. Obviously, Greece has a lot of coastline, so you’re doing our Greek brothers and sisters and ancestors proud.

Greece has a lot of coastline, and they’re actually big in the shipping industry. Now I hear, more often than not, “Oh, it’s good to see a Greek building ships.”

I can sleep at night now. I was very worried about the Strait of Hormuz. I was worried about the Pacific theater and the South China Sea. Dino, now that I know my Greek brother is on top of this, my head’s going to hit the pillow. I’m going to sleep like a baby tonight. I know you’ve got us covered.

Thanks so much, guys. Sincerely, on behalf of all Americans, thank you for doing this work. This is something where we really need technologists and entrepreneurs to help us innovate because we’re falling behind, and we’re the good guys in this. We’re the ones spreading democracy and freedom and protecting them.

The only thing in the world that isn’t trending in the right direction—if you look at lifespans, infant mortality, and people’s standard of living, everything is trending in the right direction except the spread of democracy, which has been on the decline in our lifetime. Let that sink in, folks. There are more people living under authoritarians than under democracies.

I’m sure you’ve seen the statistics, Vibin and Dino. We need to have a great military and advanced technology if we’re going to maintain the number of people living in democracies, in freedom, and hopefully increase it. There are a lot of bad guys out there.

Thank you so much for doing this work.

That’s right. Well, thank you for having us. I’ll end on a positive note: We’re seeing that start to turn around in a really big way. It’s not just Saronic. You mentioned Anduril; there’s SpaceX and Palantir before that. Now there’s a big wave, a big push into defense tech, because people realize the importance and magnitude of the moment we’re living in right now.

I tell our team, “This is our generation’s version of the space race.” Make no mistake about it: We cannot lose. That’s how hard we’re working at Saronic. That’s how we’re working with a sense of urgency, and there are a lot of other companies coming into the space that are working with that same sense of urgency.

I am very optimistic. I know this country can do it. When we put our minds to something and get a collective effort behind it, nobody can beat us.

Nobody can, yeah. Nobody can beat us. And listen, if you’re thinking about going and optimizing an ad network at some big tech company—no dig to anybody—that’s incredibly boring. This is incredibly meaningful. You’ll get 20 or 30 years into your career and say, “What did I do?”

Go work at Saronic. You’ll get a couple of years into your career and say, “I’m doing important work.”

That’s right.

That's right. Every person who walks through the door every single day is quite literally changing the world. We remind them of that, and they know it.

Yeah. All right, Dino. Thank you so much for coming on.