How does this square up with literally what Terence Tao (TFA) writes about O1? Is this meant to say there's a class of problems that O1 is still really bad at (or worse than intuition says it should be, at least)? Or is this "he says, she says" time for hot topics again on HN?
o1-preview is still quite a specialized model, and you can come up with very easy questions that it fails embarassingly despite it's success in seemingly much more difficult tests like olympiad programming/maths questions.
You certainly shouldn't think of it like having access to a graduate student whenever you want, although hopefully that's coming.
