Last Friday, Meta made big news: three massive nuclear energy deals worth 6.6 gigawatts of power by 2035. That’s enough electricity to light up 5 million homes. And the market went absolutely wild.
Vistra’s stock jumped 10%. Oklo surged 8%. By some accounts, the pre-market action was even crazier, with both stocks climbing nearly 17-18%. These aren’t quiet, boring utility stocks anymore. They’re the new AI play.
The context
Here’s what’s really happening: Meta just became one of the biggest corporate buyers of nuclear energy in American history. They’ve signed 20-year agreements with three companies: Vistra (an energy giant that operates actual nuclear plants), Oklo (a startup backed by OpenAI‘s Sam Altman), and TerraPower (Bill Gates‘ nuclear venture).
But why? Because AI is eating electricity for breakfast, lunch, and dinner.
US power demand for data centers is expected to climb at least 30% by 2030, with most of the new demand coming from data centers. Meta’s building something called the Prometheus supercluster in Ohio—a 1-gigawatt AI data center that needs to come online this year. And one gigawatt is no joke. It can power 750,000 homes.
The problem? There isn’t enough juice on the grid to keep up. Solar and wind are great, but they’re unreliable. Gas plants contradict climate commitments. The only solution that provides 24/7 clean baseload power? Nuclear.
The deal breakdown
Meta’s approach is fascinating because it hedges all bets:
With Vistra, they’re buying 2.1 gigawatts from existing nuclear plants in Ohio and Pennsylvania, plus funding upgrades that’ll add another 433 megawatts. This is the immediate play—power that actually flows today.
With Oklo and TerraPower? That’s the long game. These companies are building Small Modular Reactors—the next generation of nuclear tech. Oklo’s campus in Pike County, Ohio, could come online as soon as 2030. TerraPower’s Natrium reactors target 2032.
Financial terms weren’t disclosed, but building 6 gigawatts of new advanced nuclear could require more than $120 billion in capital costs. And that electricity won’t be cheap either—nuclear could cost Meta $141-$220 per megawatt-hour versus just $50-60 for gas or renewables.
Why this matters
First, the race is on. Microsoft effectively restarted Three Mile Island. Google signed deals with Kairos Power. Amazon’s partnering with multiple nuclear companies. This isn’t one tech giant making a bet—it’s the entire industry signaling that nuclear is essential infrastructure for AI supremacy.
Second, look at those stock movements. Oklo shares gained over 360% in last one year, while Vistra shares gained 17% in 2025 after rising over 200% in 2024. Utilities—traditionally the most boring sector in finance—have become AI volatility plays. That’s bonkers.
Third, this creates a fascinating investment thesis across multiple timeframes. Want immediate exposure? Buy existing utility operators like Vistra. Want high-risk, high-reward? Bet on SMR startups like Oklo. There’s uranium miners, reactor manufacturers, nuclear engineering firms—an entire ecosystem is suddenly investable.
The catch
But pump the brakes before you go all-in. Small modular reactors have never been commercialized at scale. Oklo went public via SPAC in 2024, and while it has landed a large deal with data center operator Switch, it has struggled to get its reactor design approved by the Nuclear Regulatory Commission.
Nuclear projects take a decade to develop. Data centers can be operational far more quickly. That creates a timing mismatch. Plus, there’s the small matter of whether these promised cost reductions will actually materialize. TerraPower estimates $50-60 per megawatt-hour eventually, but the first reactors will cost significantly more.
And here’s the uncomfortable bit: ratepayers in the mid-Atlantic are already paying higher electricity bills to support new and proposed data centers. Someone’s footing this bill, and it might be you.
The bottom line
Meta’s nuclear bet isn’t just about keeping Instagram running. It’s a declaration that whoever solves the energy equation wins the AI wars. Mark Zuckerberg has been crystal clear: power is the bottleneck limiting AI growth.
The energy giants—Vistra, Oklo, TerraPower—aren’t just getting deals. They’re getting validation. Wall Street is now pricing nuclear power companies as AI infrastructure plays, not boring utilities. That’s a fundamental shift in how markets value energy.
Is energy the next big AI bet? The stock market just answered with a resounding yes. The question now is whether these promises of clean, abundant nuclear power will actually materialize, or whether we’re inflating another speculative bubble.
Either way, the AI revolution just became an energy race. And that makes utilities—yes, utilities—one of the most fascinating sectors to watch in 2026.
Sonia Boolchandani is a seasoned financial writer She has written for prominent firms like Vested Finance, and Finology, where she has crafted content that simplifies complex financial concepts for diverse audiences.
Disclosure: The writer and her his dependents do not hold the stocks discussed in this article.
The website managers, its employee(s), and contributors/writers/authors of articles have or may have an outstanding buy or sell position or holding in the securities, options on securities or other related investments of issuers and/or companies discussed therein. The content of the articles and the interpretation of data are solely the personal views of the contributors/ writers/authors. Investors must make their own investment decisions based on their specific objectives, resources and only after consulting such independent advisors as may be necessary.