It doesn't. The impact of radiation is a rise in cancer rates which even if they substantially shorten lifetimes aren't X risk.

Your answer is basically 'radiation causes rise in cancer which can substantially reduce lifetimes is not a risk'.

I don't see how that's reasonable, I suggest you might have typed something incorrectly.

There are still quite a number of risks inherent in the Nuclear process that require only a few operational failures for disaster.

If very smart countries and operationally efficient countries like Japan screw up the risks, then anyone can. And that there were mitigating circumstances does not change the fact that Japan still had great governance and scientists and persisted with either risky models or a failure to measure the risk.

Basically, we'd need to suck the risk out of every part of the process in order for it to work well.

I think it's possible.

What's the existential risk? The absolute worst case that I can imagine is that all the waste somehow ends up in the ground water. That would be extremely bad, but localized to the area where the waste is stored.

Existential is as you mentioned: ground water pollution of the entire NE seaboard. Terrorist dirty bomb in NYC that affects 30 million people as the weather carries particles etc.

inverse square law and lack of any volume of any highly active source makes dirty bombs a nonissue. they are scarier than they are effective.

Again, waste storage is incredibly secure and safe. It wouldn't leech into the water. Nuclear subs have sank into the ocean, and, surprise, there wasn't a global disaster.