[coldtea]

>Um. This isn't something where you can just run your reactor in different configurations at test time and run time.

No, but lying and having experts being, ahem flexible, about the expected safety is very easy, and par for the course when selling multi-billion dollar projects...

It's also very easy to ignore "black swan" event cases, and the potential impact to millions of lives, just because you think you've covered everything there is to cover.

[saulrh]

Chernobyl was just about the worst possible outcome of a nuclear disaster: The majority of the radiological material was swept up in a cloud of graphite dust and lofted into the upper atmosphere by a fire that burned, uncontrolled, for over a week. Fukushima happened in one of the most densely populated areas on Earth. And yet, even counting those, Nuclear is still the safest energy source we have. We can talk about black swans and fudging the estimates, and I'll admit that your line of argument would be stronger if we were talking about the very first reactors to be built, when they were still just blueprints.

But that's not the case. We have real-world evidence that we can use to calibrate our expectations against reality. The fact is that, even if you count Chernobyl and Fukushima, our existing reactors are safer than fossil fuels. And I think that you would have an exceedingly difficult time arguing that newly-built reactors would be less safe than existing reactors.