Forty years ago this month, reactor No. 4 at Chernobyl exploded. I grew up about 350km away, in Mogilev, Belarus, in an area affected by Chernobyl fallout. I am not writing this to relitigate Chernobyl. I am writing it because the nuclear debate of 2026 is being conducted with the same quality of argument that surrounded 1986: loud, tribal and largely disconnected from the evidence.
The nuclear renaissance is back. Twenty-five countries backed the declaration at COP28, and the total later rose to 33. The EU Taxonomy includes nuclear energy under specific conditions. Microsoft signed a deal to support the planned restart of Three Mile Island Unit 1. Goldman Sachs calls it a new nuclear age. The political momentum is real, and the enthusiasm is understandable. Climate targets are slipping, energy security is a crisis, and nuclear’s life-cycle emissions of 12g of CO₂ per kilowatt-hour are genuinely among the lowest of any electricity source.
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
But here is the question that the narrative does not answer: who is actually building this renaissance, with whose fuel, on whose timeline, and in whose geopolitical interest?
The answer is uncomfortable for everyone involved. Of 40 reactor construction starts globally in 2020–24, 97% were in China and Russia. Western vendors – Westinghouse, EDF, the companies whose flags flew at COP28 – had zero new starts. The countries declaring the renaissance are not the countries building it. The gap between declaration and delivery is not a temporary inconvenience. It is structural, a product of two decades of dismantled industrial capacity, first-of-a-kind cost failures, and post-Fukushima regulatory paralysis that no policy declaration has yet reversed.
The climate argument for nuclear is real but arrives too late. A reactor that takes 15 to 20 years to build in the West displaces fossil fuels in the 2040s. The climate window for preventing the most severe warming outcomes is the 2020s and 2030s. Battery storage costs are falling fast enough that in several scenarios, the case for new nuclear weakens further – not because nuclear is wrong in principle, but because it arrives too late. Life extension of existing plants – Belgium’s Doel reactors and France’s ageing fleet – is the only nuclear climate action available on a timeline that matches the emergency. It is systematically underinvested relative to the rhetoric around new builds.
The energy security argument is equally paradoxical. European governments present nuclear as liberation from dependence on Russian energy. Meanwhile EU nuclear fuel imports from Russia increased in 2023 and 2024, despite the invasion of Ukraine. Five EU member states operate Russian-designed VVER reactors with no fully qualified Western fuel alternative. Hungary is building two new Russian reactors under a €12.5bn Russian state loan. EU sanctions did not include nuclear energy or nuclear fuel in the same way as many other sectors. The independence argument is real for some reactor types and fuel chains. For others, it is the substitution of one dependency for another, potentially for 60 years per reactor.
Then there is Zaporizhzhia. Europe’s largest nuclear plant has been under military occupation since 2022 – where an operational nuclear facility has been used as an instrument of military coercion. There is no dedicated international legal framework that clearly addresses this specific wartime scenario. This risk grows with every new reactor built in a geopolitically contested region.
None of this means nuclear energy is finished, unnecessary, or uniquely dangerous. The evidence does not support that conclusion either. What it means is that the right question is not the one being asked. The debate between pro-nuclear and anti-nuclear camps – and I have spent considerable time auditing both sides’ arguments – is mostly a contest of selective evidence. Pro-nuclear advocates conflate cheap existing plants with expensive new ones. Anti-nuclear advocates deploy LCOE comparisons stripped of system costs, and proliferation arguments that inadvertently favour Russian exports over Western ones. Both camps are conducting a debate that is less rigorous than the question deserves.
The question that matters is not whether to support nuclear in the abstract. It is this: under what concrete conditions can nuclear deliver its climate contribution reliably and quickly enough to matter, and are those conditions currently being met? On current evidence, they are not. Western construction costs have not been brought under control. The fuel supply chain is not diversified. The build rate is a factor of three below what climate scenarios require. And the Zaporizhzhia precedent has introduced a category of risk that the governance architecture of the nuclear renaissance has not begun to address.
The renaissance becomes real when concrete is poured on schedule, when fuel comes from diversified sources, when the build rate matches the climate timeline, and when the governance architecture matches the security rhetoric. Until then, it remains what the evidence shows it to be: a political declaration in search of an industrial delivery mechanism.
Forty years after Chernobyl, contaminated areas in Belarus and Ukraine remain. The decisions that created them were made elsewhere, by people who were confident they knew what they were doing. That pattern – confident declarations and distant consequences – is worth remembering as the next nuclear age is announced.
Hanna Skryhan, PhD, is an independent researcher in environmental policy and climate adaptation based in Warsaw, Poland.
