2025 was confirmed as the hottest year in recorded human history, continuing a streak that has now run for over a decade. The science is not the contested part anymore. What deserves more attention is the gap between what is being done and what the models say is necessary.

Global renewable energy deployment is genuinely accelerating. Solar installation in particular has beaten nearly every projection made ten years ago — costs have fallen faster and adoption has spread wider than analysts expected. This is good news and it is real. The problem is that it is happening alongside continued fossil fuel use rather than replacing it at the rate required to hit agreed temperature targets.

Nuclear power was supposed to be a fading technology. Expensive, politically toxic after Fukushima, outcompeted by renewables. The reversal now underway is genuine and worth understanding.

The driver is AI. Data centers powering large language models and the infrastructure they require consume enormous and rapidly growing amounts of electricity. Unlike residential or commercial demand, these loads are constant — 24 hours a day, 365 days a year. Solar generates during the day. Wind generates when the wind blows. Nuclear generates all the time, regardless of conditions. For a tech industry trying to guarantee power availability at scale, nuclear has become newly attractive precisely because of the attribute that made it economically awkward in a grid context: it does not stop.