Chernobyl was the result of turning off all safety mechanisms and running the reactor outside of its design range to see what would happen.

That was more-or-less my impression.

Edit: See https://www.damninteresting.com/chernobyl/

"But I think Chernobyl was the result of under-investment and poor management"

Chernobyl was extremely poorly engineered, Soviet Union style. Its major flaw was no containment vessel.

Even then, it took real effort to make it fail, and it was all totally unnecessary. It was probably the worst testing-related disaster in history.

The other major flaw of RBMK reactors is a dangerously high positive void coefficient of reactivity, which self reinforces the nuclear reaction in case of a loss of coolant accident.

https://en.m.wikipedia.org/wiki/Void_coefficient

Isn't the point that even with the current mixture of managment, it's safer? However, I don't know if these numbers include the long-term risk from storing/processing the spent fuel.

Yes, we've demonstrated that you can basically mismanage plants into meltdown, and even then nuclear still ends up killing fewer people.

Over a period of 50 years, you can mismanage a couple of plants and still kill fewer people. But not the entire industry.

I don't know what you are trying to say.

50 years of operation of hundreds of plants. We could have had a repeat of all of the accidents every year and it would still be one of the safest alternatives.

Heck, we could have Chernobyl every year, and even based on the current worst case estimates of reduced life expectancy and counting every one of them as a death, we'd still not compensate for 1 Banqiao failure, or more than 1-2 years worth of coal.

But we're not having a Chernobyl every year. What we are having is vast numbers of preventable deaths due to other types of power generation.

We're talking millions dead over the last 50 years due to other types of plants.

I think you have to be careful here. It's entirely unclear how small dose exposure over long times damage humans. There might be a threshold effect: Some people speculated that low doses actually help, as they trigger the DNA self-repair systems. Slightly more, and that self-repair can't help anymore. So 1 Chernobyl might be fine for people further way, let's say, in the US, 2 or 10 or ... and you suddenly get a lot of Cancer deaths.

If you look at what medicine deemed as safe, the limits were reduced every couple of years.

I'm not saying you are wrong - I actually want to believe this statistic, but has anyone actually compiled the data to back this up?

Specifically - "we could have Chernobyl every year ... we'd still not compensate for more than 1-2 years worth of coal"

Where is it managed well?

I have no idea. I don't even know if it is practically possible to manage well enough to guarantee zero disasters.

But there are many nuclear plants around the world that have operated with no problems. As I said, I'm no expert.

new design exist that in case of disaster self contains under most conditions, trouble is it's very hard to obtain certifications needed for operation today as opposed to decades ago, so we're mostly stuck with old designs.

The UK seems pretty good at it: https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_Ki...

by design already. all recent reactors have containment tanks to prevent radioactive leakage in a meltdown event.