When the top two executives of a nuclear startup vanish without explanation, that's not a personnel change - that's a red flag the size of a cooling tower.
The CEO and CFO of Fermi, a Texas-based startup building nuclear power for AI data centers, have suddenly departed the company. No explanation given. No transition plan announced. Just... gone.
The AI power crisis is spawning ambitious solutions, but this one just hit a wall.
Fermi was co-founded by former U.S. Energy Secretary Rick Perry - not exactly a small name to attach to a venture. The company went public via IPO at Nasdaq in October 2025, with CEO Toby Neugebauer ringing the closing bell in what must have seemed like a triumphant moment.
Six months later, both he and the CFO are out, and the company isn't saying why.
The nuclear-for-AI pitch has been one of the hottest narratives in infrastructure investing. AI training and inference require massive, reliable power. Data centers are already straining electricity grids. Nuclear offers carbon-free baseload power at a scale that could actually meet demand.
On paper, it's compelling. In practice, nuclear projects are notoriously difficult - technically complex, regulatory nightmares, and capital-intensive beyond what most startups can handle.
Fermi has reportedly "faced headwinds with its AI campus in Texas," which is diplomatic language that could mean anything from permitting delays to fundamental technical or financial problems.
What makes this departure particularly concerning is the silence. When executives leave startups on good terms, there's usually a statement thanking them for their contributions and announcing the transition plan. When they leave and the company says nothing, it suggests the departure wasn't planned or amicable.
In nuclear energy, that's especially worrying. This isn't a software company where you can iterate fast and break things. Nuclear projects require regulatory approval, safety certifications, and investor confidence that the leadership knows what they're doing.
Losing your CEO and CFO simultaneously undermines all of that.
The broader AI power story is real. Microsoft is restarting Three Mile Island. Google and Amazon are investing in small modular reactors. Nuclear power for data centers isn't a crazy idea - it's becoming industry consensus.
But consensus around the destination doesn't guarantee any particular company will get there. Nuclear startups are littered with ambitious promises and failed execution. Without knowing why Fermi's leadership departed, it's impossible to assess the company's viability.
Investors should be asking hard questions. Employees should be updating resumes. And anyone who pre-ordered power from Fermi's future AI campus should probably have a backup plan.
The technology is impressive - in theory. The question is whether this company can actually build it. And right now, that question just got a lot harder to answer.
