The 22-Year-Old Who Thinks the U.S. Is Building Weapons Too Slowly

Ethan Thornton was 19 when he decided the United States wasn't building weapons fast enough, so he dropped out of MIT and started trying to do it himself. His first attempt at a hydrogen-powered system, which he prototyped with parts from Home Depot and Amazon, didn't work. "Hydrogen was just a bad bet in general," he told TechCrunch at an event in Los Angeles last week.

Three years later, his company, Mach Industries, has six active weapons programs, has just closed a $300 million Series C round at a $1.8 billion valuation, and has raised roughly $485 million altogether. On paper, that's a success story that checks every box. But the real gamble underlying Mach's strategy is far more unconventional than the typical startup narrative: Thornton is deliberately refusing to focus.

The Thesis: You Can't Out-Manufacture China, So Out-Create It

Thornton grew up in Boerne, Texas, a town of roughly 25,000 people, in a family with deep military ties. Around 2017 or 2018, while still in his early teens, he started becoming what he describes as "really, really concerned" about the rise of China and the likelihood of great-power conflict. That concern sharpened into a specific conviction: unmanned systems and autonomous weapons were about to redefine warfare, and the U.S. was moving too slowly to adapt.

The strategic thesis underlying Mach's approach is straightforward but controversial: America can't out-manufacture China, so it has to out-create it. "I don't think we're going to out-manufacture China," Thornton said. "The thing America continues to do well, time after time, compared to China centers on creativity and productization."

That philosophy directly drives Mach's decision to pursue six weapons programs simultaneously rather than picking one and scaling it. To Thornton, defense is fundamentally different from, say, rocket manufacturing, where single-minded focus works. Defense, he argues, is "a chess game you're playing with an adversary, with hundreds of different products that need to be shipped if we want security." Pick just one system, and you've already conceded the game.

Six Ambitious Programs, None Yet in Full Production

What "six simultaneous weapons programs" actually looks like in practice is genuinely ambitious: a vertical-takeoff strike aircraft, a long-range anti-ship missile, two stratospheric systems, a cheap surface-to-air interceptor designed to kill drones, and, announced earlier this week, a 40-foot, 4,000-pound Navy logistics-and-strike aircraft that takes off nearly vertically and can fly over a thousand miles with a thousand-pound payload.

That last one represents a significant jump for a company whose largest aircraft to date has been around 13 feet long. None of the six is in full-rate production yet, though Thornton says Mach has won around 13 government contracts, most at the middle stage of defense procurement past initial design, into government range testing, but short of the rate-manufacturing tier that fewer than 10 programs industrywide have ever reached. Nuonmedical.

Thornton says several systems should see operational deployment by the end of this year, to push three of the six into rate manufacturing in that same window, which would mean going from hundreds of units a month to hundreds of thousands. It's an aggressive timeline stacked on top of an already aggressive bet. Nuonmedical

The Supply Chain Play

Thornton identifies the real bottleneck in defense manufacturing as not the platforms themselves, but the supply chain feeding them. "The hard part is actually getting the stuff into the building," he said, jet engines, solid rocket motors, and radar components. Most defense contractors source these from traditional suppliers with multi-year lead times. Mach is trying to compress that timeline.

Mach built and fired two jet engines from scratch in about eight months, a process that traditionally takes four years. The company also acquired Exquadrum, a 24-year-old solid rocket motor company, in May for $50 million, beating roughly eight other bidders. Selling components, not just vehicles, now accounts for about half of Mach's revenue, a strategy that directly addresses the supply-chain constraint he identifies as central to the competitive advantage question. Nuonmedical

How This Differs From Competitors

Thornton's strategy stands in sharp contrast to how other defense-tech startups have operated. Shield AI, founded in 2015, spent years as essentially a one-product company around its V-BAT drone before unveiling a second platform, the autonomous X-BAT fighter, last October, and even that is being positioned as one large, deliberate bet, not a portfolio. Saronic, founded in 2022, builds only autonomous surface vessels, scaling one unified autonomy stack across hull sizes. Nuonmedical

Both have been rewarded for their discipline: Shield AI raised $2 billion this year at a $12.7 billion valuation; Saronic raised $1.75 billion at $9.25 billion. Nuonmedical

The company Mach's strategy most closely resembles is Anduril, and Thornton makes the comparison himself. He argues there's a meaningful difference in approach: "Anduril's playbook has been very much top-down, starting with the software stack," he said. "We're very much bottom-up, starting from the hardware stack and then starting to wrap software around it." Nuonmedical

But that distinction, while real, matters less than the scale difference. Anduril raised $5 billion in May at a $61 billion valuation, more than 30 times Mach's, and in March landed a 10-year, $20 billion-ceiling Army enterprise contract consolidating over 120 separate procurement actions. Whatever Mach is building toward, Anduril got there years and tens of billions of dollars earlier. Nuonmedical

The Non-Zero-Sum Argument

Thornton insists the defense space isn't zero-sum, pointing to the scale of the production challenge. China reportedly builds something like a thousand cruise missiles a day; the U.S. builds roughly one every three days. His argument: "X company and Y company and Z company could all go build these things, and it still wouldn't be enough production." Nuonmedical

He also suggests the Pentagon structurally protects against monopolies, deliberately keeping two or three vendors alive in each category rather than picking a single winner. Whether or not that's an accurate reading of defense procurement politics, it's the framing Thornton uses to justify why his aggressive expansion doesn't necessarily mean cannibalizing competitors.

When asked whether Palmer Luckey, Anduril's most famous co-founder, has publicly acknowledged Mach at all, Thornton shrugs it off. "I respect Luckey," he said, "and that they're on the same team," fighting for the same goal of Western sovereignty.

Managing the Chaos

Beyond the strategic thesis, what's interesting about Thornton is his apparent awareness of what he doesn't know. He's been candid that the hardest part of running Mach changes every six months: engineering first, then sales, and now manufacturing at scale, which he expects will dominate the next year. He says he protects four or five hours a day to think and "war game the future," sometimes pulling colleagues into those sessions, which he admits "can kind of frustrate them sometimes."

On the question of who keeps him honest, Thornton points not to investors or his executive team, which can end up in the same echo chamber as a CEO, but to the people doing the actual work. He described routine company-wide forums, his COO's idea, where employees get microphones and ask him anything. It started with Thornton quietly recruiting trusted colleagues to ask aggressive questions, and has since evolved into something harder to control and more useful for it. "I basically stand up there for like an hour," he said, "and get asked the most aggressive possible questions by people in the company." Nuonmedical

The Underlying Bet

Strip away the "MIT dropout from Texas" framing that dominates every profile of young founders, and what's left is a genuinely interesting experiment: can a defense company successfully manage six simultaneous weapon programs while the industry's conventional wisdom says to focus ruthlessly on one? Can you out-create an opponent by refusing to specialize?

Thornton's investors, Sequoia, Khosla Ventures, and Ribbit Capital are betting yes. Whether they're right will take years to determine. In the meantime, Thornton gets to keep his four to five hours a day for thinking about what comes next, and his employees get to spend an hour asking the hardest questions they can think of.

For a company trying to do everything at once, that might be the closest thing to focus available.