You are ignoring the elephant in the room, which is the distribution of the risk, both statistically and geographically. Coal kills consistently, and mostly wh

You are ignoring the elephant in the room, which is the distribution of the risk, both statistically and geographically. Coal kills consistently, and mostly where it is used and mined. If you profit from cheap coal energy, you have a higher risk of pulmonary disease. If better filters are installed, your risk is reduced, incentivizing you to trade off risk and reward. If a country builds unmanagebly big hydro dams, a disaster will also affect mostly its own population. With nuclear energy you have a few catastrophic incidents that render entire areas uninhabitable but also contaminate countries that don't even benefit from it (e.g. Austria got fallout from Chernobyl, having no nuclear power itself). At the same time, by pure luck, many reactors identical to Chernobyl did not blow up and provided the surrounding areas with ridiculously cheap energy. This is a Black Swan type of risk that completely disconnects risk taking and reward. Even if you are very neglegient, the risk is still very low, allowing you to reap massive benefits. Even worse, countries that don't benefit but may be affected by a disaster may pay you for upgrading your security measures (this has happened in the EU). I'd rather have technology that is more dangerous but where risk actualization mostly and consistently affects those who are rewarded if everything works out well.