One of my open math problems apparently got resolved by ChatGPT 5.5 Pro (Ryan O'Donnell prompted it better than I did!), though the proof was so hard for me to read that it seemed easier to just prove it myself. More thoughts on implications for math here: https://aifails.substack.com/p/even-more-on-ai-and-math-solution
@DominikPeters also a mention of your post
I think people underestimate the effect GPT-5.x is already having on research in mathematical fields. 5.5 Pro has been proving many theorems for me, but I don’t talk about it much because I want to publish those results with my name on it and that’ll take time to write up.
@conitzer "I’m sure [Dominik] won’t send out proofs into the world that he is not satisfied with" -- that is certainly the plan, though similarly to your troubles with the long 5.5 proof, I sometimes find that AI proofs can be dazzling and alien. These models know too many things
@conitzer We will probably at some point want to have some way of sharing results that are true, even though no human understands the proofs yet. But then these results become lemmas and eventually we have a castle of knowledge that is inaccessible to humans...
No Digg Deeper questions have been answered for this story yet.