AI hard takeoff before 2100 might count as an example for me, although the issue is rather complex.
Also, I don’t see any reason why things like quantum chromodynamics (or, for that matter, special relativity) shouldn’t count, unless you’re just trying to rule out a large class of predictable answers.
Ruling out a large class of predictable answers is part of it, but it’s also arguably more a case of “impressions don’t point anywhere” than “impressions point the other way”. If challenged on that point I’d have to think about whether I could make it more precisely, but I was kind of hoping people would intuitively agree about the spirit of the question (which is why I didn’t include the second paragraph at first).
AI hard takeoff before 2100 might count as an example for me, although the issue is rather complex.
Also, I don’t see any reason why things like quantum chromodynamics (or, for that matter, special relativity) shouldn’t count, unless you’re just trying to rule out a large class of predictable answers.
Ruling out a large class of predictable answers is part of it, but it’s also arguably more a case of “impressions don’t point anywhere” than “impressions point the other way”. If challenged on that point I’d have to think about whether I could make it more precisely, but I was kind of hoping people would intuitively agree about the spirit of the question (which is why I didn’t include the second paragraph at first).