Worse than that, it had a negative learning effect. Nuclear did have some learning but over the course of it positive effect (Wright's law) was more than counte

Worse than that, it had a negative learning effect. Nuclear did have some learning but over the course of it positive effect (Wright's law) was more than counterbalanced with discovering many, many things that can go wrong, requiring very expensive design changes and even more expensive retrofits to existing reactors, and eventually required designers to become so paranoid about new designs they became prohibitively expensive to implement because of extended safety measures even in areas where they were probably unnecessary. Simply put: too many things can go wrong, price of mistake is too damn high, and any mistake is probably unfixable because even if a reactor is fine, retrofitting an existing reactor to accommodate safety-related change is frequently impossible because it's guts are forever radioactive even after a short runtime. It was a political mistake to ever move with the nuclear power to outside of it's initial role: production of plutonium for nukes. Ironically, plutonium-producing reactors are much much safer because they don't need to produce high-pressure steam so can remain at atmospheric pressure cooled by vast amounts of cooling water, and they never develop too much radioactivity within themselves because fuel is irradiated there for 20x-100x shorter durations than in energy-producing reactors (to ca. 1 MW thermal*day per kg) because doing more increases posioning with Pu.240, so even a potential accident on an industrial reactor is much less dangerous, and much less likely.