Nuclear Revival Meets a Chronology Problem as Data Centers Pile In
Policy and capital are flowing into nuclear, but the build-out timeline risks leaving the next decade's power deficit unserved.
European Commission President Ursula von der Leyen called Europe's retreat from nuclear power a "strategic mistake" in March 2026 and pledged €200 million for small modular reactors. Washington wants to quadruple U.S. nuclear capacity by 2050 and fast-track advanced reactor permits. The direction of travel is clear.1
The IEA expects more than 70 GW of new nuclear capacity to come online by the mid-2030s, one of the strongest pipelines in three decades. But that pipeline will not power data centers next year, nor provide the near-term backup capacity a resilient grid requires. The gap between stated ambition and construction reality is where the market question lives.1
The gap is widening fast. U.S. data-center power demand alone is projected to more than triple over the next decade, from 34.7 GW in 2024 to 106 GW by 2035, according to IEA data. Microsoft, Amazon and Google have all signed nuclear supply deals, signaling that corporate offtakers are betting on atomic power as part of the solution.1
Yet the construction reality remains stubborn. Global investment in data centers reached roughly $580 billion in a recent year, dwarfing the projected nuclear spending level the IEA expects to top $100 billion annually under stated policies. Even a 70% increase in nuclear investment over the past five years leaves a substantial scale gap against that data-center load trajectory.1
The regulatory environment has shifted decisively in favor of nuclear across Europe, as surging electricity demand from artificial intelligence and data centers collides with volatile global energy markets and the political need for structural energy independence.3
China is not waiting. It is rapidly expanding nuclear capacity domestically and internationally as part of a broader geopolitical push, while U.S. and European discussions still center on demonstration projects, future concepts and theoretical deployment timelines. That contrast is not abstract — it shows up in construction starts and grid connection dates.4
Germany's position remains an ambivalent factor. The collapse of the Franco-German-Spanish Future Combat Air System coincided with reports of German interest in expanding its F-35 order, capturing the hedging at the heart of Berlin's strategic posture. That same hedging extends to energy: Germany's post-Fukushima nuclear phaseout leaves it structurally exposed to the price and security risks the nuclear revival is meant to address.2
What makes this cycle distinct from previous nuclear false dawns is that the evidence is showing up in policy, capital markets and corporate power purchase agreements simultaneously. Governments and utilities want nuclear to work. The operative question is whether they can build it fast enough to matter for the 2030s power crunch.1
The IEA expects nuclear spending to top $100 billion annually under stated policies — real money, but set against the tripling of data-center load and the 70 GW pipeline target, enough to start construction rather than fill the 2030-2035 power gap.1
For traders, the forward signal worth monitoring is whether any major reactor reaches grid connection before 2032. If the build-out slips past that line, gas-fired generation and renewable-plus-storage will shoulder the incremental load for longer, keeping downward pressure on gas basis costs and upward pressure on carbon prices. The nuclear revival has arrived in the policy books — the timing question will be answered in the construction yards.1