> Scaling CoTs to e. g. millions of tokens or effective-indefinite-size context windows (if that even works) may or may not lead to math being solved. I expect it won't.
> (If math is solved, though, I don't know how to estimate the consequences, and it might invalidate the rest of my predictions.)
What does it mean for math to be solved in this context? Is it the idea that an AI will be able to generate any mathematical proof? To take a silly example, would we get a proof of whether P=NP from an AI that had solved math?
I think "math is solved" refers more to AI performing math studies at the level of a mathematics graduate student. Obviously "math" won't ever be "solved" but the problem of AI getting to a certain math proficiency level could be. No matter how good an AI is, if P != NP it won't be able to prove P=NP.
Regardless I don't think our AI systems are close to a proficiency breakthrough.
Edit: it is odd that "math is solved" is never explained. But "proficient to do math research" makes the most sense to me.
> (If math is solved, though, I don't know how to estimate the consequences, and it might invalidate the rest of my predictions.)
What does it mean for math to be solved in this context? Is it the idea that an AI will be able to generate any mathematical proof? To take a silly example, would we get a proof of whether P=NP from an AI that had solved math?