Uranium ETF Gains Nearly 4% as Nuclear Build Stocks Diverge From AI Power Demand Story
The URA uranium ETF rose 3.94% on 2026-08-05 even as VIX climbed and individual nuclear equities continue to show sharply uneven results.
The URA uranium ETF climbed 3.94% to $42.49 on 2026-08-05, a move that sits in contrast with the choppy performance of nuclear equities over the past several months. VIX rose 4.04% to 16.50 on the same session, suggesting broader risk appetite is tightening even as nuclear exposure remained in demand.2,7
Behind that divergence lies a sector that has produced spectacular returns for some investors and painful drawdowns for others. Oklo Inc., the Santa Clara-based advanced reactor developer, gained 593% over the twelve months through mid-May 2026, according to TheStreet Pro data, yet fell 9.38% on Monday (2026-05-18) alone when a broader selloff hit nuclear names. SMR, based in Corvallis, Oregon, dropped more than 9% on that same session after gaining 156% year-to-date and 310% over twelve months.2
BWX Technologies, by contrast, fell just 1.8% on Monday (2026-05-18), a relative outperformance traders attributed to its more diversified defence and government reactor servicing business rather than pure-play exposure to the commercial small modular reactor pipeline. The gap between defence-linked and merchant nuclear names has widened as execution risk on new builds has become harder to ignore.2
The underlying demand narrative driving all these names remains the AI data centre buildout. Quick Read Capital has been rotating into energy companies supplying power to hyperscalers, with nuclear and renewable baseload generation seen as the cleanest solutions to a constraint that is increasingly concrete. Fluence Energy's stock rising 98.2% in a single week after it disclosed master supply agreements with two hyperscalers and a record $5.6 billion backlog — shares closed at $24.16 on May 8, 2026 — illustrated how sharply markets are repricing companies perceived to be in the AI power supply chain.1
But Fluence's story also captures the hazard. Shares remain down roughly 39% year-to-date despite the backlog announcement, and stockholders' equity stood at negative $265.88 million with cash of just $36.59 million, according to the same data. A record order book does not resolve a stretched balance sheet.1
On the upstream side, Westinghouse's adjusted EBITDA jumped 33% to $122 million in Q1 2026, though revenue of $606.30 million missed consensus by 26% and earnings per share of $0.33 came in fractionally below the $0.34 estimate, per AOL reporting. Adjusted net earnings nearly tripled to $145.59 million. The demand backdrop for Westinghouse's services remains strong, but the headline miss signals that converting order book growth into recognized revenue is taking longer than analysts had pencilled in.6
Policy is moving to match the investment thesis, if slowly. Italy's lower house of parliament approved legislation on Thursday (2026-06-04) that opens the path for nuclear reactors to enter the power mix from the 2030s, Montel reported. Italy had been nuclear-free since a 1987 referendum. The law does not commit to specific projects or timelines beyond enabling the regulatory framework, and the distance between legislative approval and first concrete is considerable.5
In the United States, Guggenheim Securities analyst Jim Schaefer has argued for financing models that aggregate heavy power users — particularly large technology firms — to fund nuclear fleets through risk-sharing partnerships, an approach aimed at reducing the single-developer exposure that has plagued previous build programmes in the West, according to The Economist. Goldman Sachs has separately added small modular reactors to its demand forecasts, a signal that SMR deployment is now part of mainstream energy scenario planning rather than a niche research question.3,4
Yet the gap between planning and commissioning remains wide. The Economist's May 2026 analysis noted that demand conditions for nuclear are now being recreated in the West, but under circumstances different enough from the original build era that financing structures, supply chains, and regulatory timelines all require substantial rebuilding. The DeepSeek-driven selloff in January and the recovery that followed — with some nuclear names gaining over 100% year-to-date by mid-May 2026 — showed how quickly sentiment can shift when AI spending assumptions are questioned.2,3
The 2026-08-05 uranium ETF move resolves none of those uncertainties. The nearer-term signal worth tracking is whether Westinghouse and other established nuclear services companies can close the gap between their adjusted earnings and headline revenue figures over the next two reporting quarters, since that is where the distance between the AI power narrative and actual nuclear build economics will be most legibly priced.6