I mean, the problem isn't that an accident is extremely unlikely, it's the fact that it only needs to happen ONCE for it to be a major disaster.
Like, I feel this way about nuclear power. A nuclear power plant is extremely safe, and when properly run shouldn't have any major accidents that would lead to a devastating meltdown. But what if, across decades of use, there's some people who run it who shouldn't be and they make mistakes (Three Mile Island, Chernobyl), or some freak accident or disaster eventually occurs (Fukushima)? Then what? You have a colossal disaster that you can't take back. If you have something that's going to be running for years and years eventually something is going to go wrong, and if that something is a thing that is going to be so absurdly damaging, that's when you need to start having the conversation about if it's worth doing at all. I don't like the idea of simply going, "Well, we don't have to worry about a grave catastrophe negatively impacting thousands of lives and costing us billions of dollars as long as a very unlikely accident doesn't occur within a 40-year time frame". Like, that's long enough for it to happen.
sure, but if in both cases you make the status quo optimal system worse than the colossal fuck up over large timescales, you've just completely lost
ie what foolmo said
---
List the ominous stern whisper from the delphic cave within: They enslave their children's children who make compromise with sin