For most of the 2010s, nuclear power looked like a technology in managed decline. Fukushima had spooked regulators and publics alike. Germany was shutting reactors down with something close to ideological fervor. Natural gas was cheap. Solar and wind were getting cheaper by the month. The received wisdom was that nuclear's moment had passed, that it was too slow to build, too expensive to run, and too politically toxic to save.
That received wisdom is now being stress-tested in real time.
From California to South Korea to Poland, governments are either extending the lives of existing reactors, contracting for new ones, or both. The geopolitical shocks of the last three years, particularly Russia's weaponization of natural gas supplies after the invasion of Ukraine, exposed something uncomfortable about modern energy policy: grid reliability and energy sovereignty are not problems you can solve with intermittent generation alone. Nuclear, suddenly, looks less like a relic and more like insurance.
The Diablo Canyon Turning Point
No single story captures the shift better than Diablo Canyon. California's last operating nuclear plant, sitting on a bluff above the Pacific Ocean near San Luis Obispo, was supposed to close by 2025. Pacific Gas and Electric had agreed to retire it years earlier under pressure from environmental groups. The state had a plan. Then reality intervened.
California's grid started struggling with extreme heat events. The state's ambitious offshore wind timelines kept slipping. In 2022, the state legislature voted to keep Diablo Canyon open, providing $1.4 billion in state loans to support the extension. PG&E then applied to the Nuclear Regulatory Commission for a 20-year license extension, a process that is now actively underway. The plant's two units generate roughly 2,200 megawatts, enough to power about 3 million homes, and losing that capacity before replacement generation was ready simply stopped making sense.
What happened in California is being replicated, in different forms, across multiple countries. Japan restarted reactors it had idled after Fukushima and passed legislation allowing plants to operate beyond 60 years under certain conditions. Belgium reversed a planned nuclear phaseout. France, which had been quietly allowing its fleet to age without committing to major new investment, announced plans to build six new EPR2 reactors, with options for eight more. The political consensus that had been drifting away from nuclear reversed direction with striking speed.
Why Now, Specifically
The timing is not accidental. Three forces converged almost simultaneously.
First, the energy security crisis triggered by the Ukraine war made European governments acutely aware of what it means to depend on a hostile supplier for baseload power. Gas prices spiked to levels that would have seemed absurd two years earlier. Countries that had nuclear capacity used it. Countries that had shut it down wished they hadn't.
Second, grid stress from electrification is accelerating. Electric vehicle adoption, data center buildouts driven by AI workloads, and the electrification of heating and industrial processes are all adding load to grids that were designed for a different era. The U.S. Energy Information Administration projects that electricity demand, which had been essentially flat for two decades, will grow significantly through 2035. Intermittent renewables help, but every grid operator knows you need something that runs when the wind doesn't blow and the sun doesn't shine.
Third, the cost calculus for existing nuclear shifted. Running a plant that's already paid off is often the cheapest electricity available, with operating costs frequently below $30 per megawatt-hour. The expensive part of nuclear is building it. Keeping an existing plant alive is a different financial equation entirely.
New Builds Are a Different Story
Here's where nuclear's revival gets complicated. Life extensions for existing plants make clear economic sense. New construction is a much harder argument to make.
Georgia's Vogtle Units 3 and 4, the first new nuclear reactors built in the United States in roughly 30 years, came in catastrophically over budget and schedule. The project was originally estimated to cost around $14 billion. Final costs exceeded $35 billion. The delays stretched years beyond initial projections. Westinghouse, the reactor vendor, went bankrupt in the process. The story was similar in Finland, where the Olkiluoto 3 EPR reactor came online 13 years late after massive cost overruns.
Supporters of new nuclear argue, with some justification, that these were first-of-a-kind projects in countries that had lost their nuclear construction expertise over decades. The learning curve is brutal when you've forgotten how to do something. South Korea and China, which have continued building reactors without long gaps, consistently come in closer to budget and on schedule. South Korea's APR-1400 units have been completed in the 7-to-9-year range at costs that would be considered remarkable by Western standards.
The policy response in the U.S. has been to throw money at the problem. The Inflation Reduction Act included a production tax credit for existing nuclear plants and investment tax credits for new construction. The Department of Energy has funded small modular reactor development through companies like NuScale, TerraPower, and X-energy. Congress passed the ADVANCE Act in 2024 to streamline NRC licensing processes. Whether any of this produces new operating capacity before 2035 is genuinely uncertain.
Small Modular Reactors: Promise and Patience Required
Small modular reactors, or SMRs, have become the optimistic answer to the cost and construction time problem. The theory is that factory-built, standardized reactor modules in the 50-to-300-megawatt range can be produced more cheaply and deployed more quickly than giant conventional plants. Several designs are in various stages of development and licensing.
NuScale Power, which received NRC design approval for its VOYGR SMR, was considered the front-runner in the U.S. Then, in late 2023, its flagship Utah project collapsed after projected costs roughly doubled to around $89 per megawatt-hour, making it uncompetitive. The company laid off a significant portion of its workforce and has been restructuring since.
That setback doesn't mean SMRs are finished, but it does illustrate that the gap between concept and commercially viable product is still real. TerraPower's Natrium reactor, backed in part by Bill Gates, is under construction in Wyoming and aims to demonstrate its technology in the late 2020s. GE Hitachi's BWRX-300 has attracted interest from utilities in Canada and the U.S. The next five years will tell us a great deal about whether factory economics can actually tame nuclear construction costs.
The Geopolitics of Nuclear Fuel
One dimension of the nuclear revival that gets less attention than it deserves is fuel supply. Most Western reactors run on enriched uranium, and the enrichment supply chain has historically run through Russia to a significant degree. Until recently, a substantial share of enriched uranium used by U.S. utilities came from Russian sources.
The U.S. Congress passed the Prohibiting Russian Uranium Imports Act in 2024, which phases out Russian uranium imports with certain waiver provisions. That's geopolitically sensible but creates real near-term supply challenges. Building out domestic uranium enrichment capacity, from Urenco's facility in New Mexico to Centrus Energy's American Centrifuge technology, takes time and capital. Uranium prices have already risen sharply in response to supply concerns, with spot prices more than doubling from their 2020 lows.
For countries betting heavily on nuclear expansion, securing fuel supply chains is as strategically important as building the plants themselves. France is working to rebuild its uranium procurement relationships. South Korea imports essentially all of its uranium but has been diversifying suppliers. This is energy security thinking applied all the way down the supply chain, which is exactly the kind of systemic thinking that was absent from energy policy for too long.
What This Means for the Grid, and for You
If you live in a region with existing nuclear plants, the most immediate implication is that those plants are probably going to run longer than anyone expected a few years ago. That's generally good news for electricity prices and grid stability. Nuclear's fuel costs are low and relatively stable compared to gas, which means it acts as a hedge against fossil fuel price volatility.
For consumers watching electricity bills climb alongside EV adoption and heat pump installations, nuclear's reliability premium matters. A grid with substantial nuclear baseload can absorb more intermittent renewables without requiring as much expensive battery storage or long-distance transmission. It's not an either/or choice; nuclear and renewables complement each other more than they compete.
The harder question is whether the political and industrial will exists to actually build new capacity in countries that need it. The United States, United Kingdom, and several European nations are making commitments that will take 15 to 20 years to materialize as operating plants. That's a long time to maintain policy consistency across election cycles and public opinion shifts.
Nuclear's return isn't guaranteed. But the conditions that made it easy to dismiss have changed substantially, and the countries that dismissed it most enthusiastically are now, quietly, reconsidering. That reversal is significant, and it's happening faster than most analysts predicted just three years ago.
