First of all, these numbers for the Fukushima cleanup don't make sense. 500B is the annual budget of medicare, or roughly the market cap of a FAANG. Your source appears to be wikipedia, so I looked into it, and their figure is from a scientific american article in which the Japanese government is quoted as saying 15 billion USD for the cleanup and 60 billion USD for relocation damages. Quite a lot of money, but definitely an order of magnitude less than the cost of the earthquake and tsunami themselves. A "private think tank" in Japan then wildly conjectured this number to actually be between 400 billion and 700 billion USD, and for some reason, this is the figure that Wikipedia uses. So, interestingly, I am claiming that this is the same phenomenon as before. This think tank might even be the CATO institute, for all we know.

Second, I said for reliable power. That was sneaky of me, but gas, coal, nuclear, and to some extent hydro, are the only reliable power sources I am aware of. Wind is not completely random, but cannot be relied upon, and solar cannot be run at night. Hydro is reliable, but represents a much bigger population risk than nuclear, with no real possibility of safe failure. I like hydroelectric power, I think it's neat, but it's definitely a sleeping dragon.

Assuming your lower estimate of 15B + 60B, that remains two orders of magnitude above the insurance policy you claim "is extremely unlikely" to be exceeded, so arguably my point stands.

"Reliable power" is well.. this keeps coming up, and it's always this black-and-white thing; sometimes the wind doesn't blow, so you can't rely on it.

And then the other side will counter that the downtime of wind is predictable, unlike that of nuclear (see the need to raise the price ceiling in EU day-ahead markets to deal with the unplanned nuke outages in France going on right now). This side will say the predictable 45% capacity factor of wind is preferable to the unpredictable 90% capacity factor of nukes.

In reality, a stable energy system is built by combining energy sources of different strengths. Wind and solar provides bulk cheap power most of the time, with hydro and nukes ramping to match production changes. This is identical to how the grid is already ramping to match demand changes every day.

We see this dynamic every day in EU markets, with wind+hydro slotting together, hydro flexing up and down as wind ebbs and flows. You can see this live in the "Origin of Electricity" charts on electricitymap: https://app.electricitymap.org/zone/SE

Honestly curious about how you come to the conclusion that the risk of hydro is that much bigger than nuclear. One would assume that this is a pure engineering problem and therefore "mitigated many times over" just as well as in nuclear?