Fukushima happened in a highly regulated (and correspondingly expensive) environment. Improving the regulation doesn't necessarily just mean adding more of it or tightening targets far beyond any useful range. Also, arguably the evacuations as a response to Fukushima caused more harm than the radiation would have, an example of paranoia as opposed to effective regulation.
(Chernobyl also happened because of paranoia but that was largely the Soviet's obsession with saving face to the point of hiding a design flaw from the operators).
Though, I will push back on the idea that this is exclusively why it's so expensive: Nuclear has plenty of other issues with cost beyond excessive safety targets.
> Nuclear has plenty of other issues with cost beyond excessive safety targets.
What are they?
> the evacuations as a response to Fukushima caused more harm than the radiation would have
This is a convenient opinion, not a true fact. Not different than saying, "I broke my leg after running from a tiger, so not running from a tiger would had been safer".
People try to convert that opinion on truth by the procedure of repeating it again and again, but we are still waiting to see the facts and the logical thinking that sustain this claim.
The radiation could have done a lot of harm, and nobody in march 2011 knew if the plant could be stabilized. They toke the 120% correct choice with the available data. And would be still the option taken in 2026 under similar circumstances. Everything else is reckless negligence.
To those that repeat in 2026, "but the plants never exploded". remember that they didn't had the luxury of knowing this at the time. Also that the inconvenient facts (like cancer ratio increasing, or newborns with troubles) were hidden under the rug and ignored for as much time as they could, so any comparison is biased from the start.