Skip to main content

OK this is probably a dumb question but why did we all decide that bio risk was the most scary thing AIs could do? Did someone write up a justification of that somewhere?
in reply to Daniel Filan

Ben Millwood
Ben Millwood
•
I don't have the impression that it's the most scary thing, just that it's the thing that's maybe the simplest story for why and how it's scary?
in reply to Ben Millwood

Daniel Filan
Daniel Filan
•
I guess I see a lot of evals and papers that only make sense if it's one of the top scary things.
in reply to Daniel Filan

Ben Weinstein-Raun
Ben Weinstein-Raun
•
I think it's maybe the scariest thing if you only believe in misuse risk, and lots of people seem to only believe in misuse risk for some reason.
in reply to Daniel Filan

Daniel Ziegler
Daniel Ziegler
•
I think what we decided was more like: it might be the *first* way we get an actual catastrophe
⇧