7h ago

OpenAI's Sébastien Bubeck and academic Daniel Litt say GPT-5.5 disproved the mathematical sum-product conjecture

2 top authors

Boris Alexeev guided the proof using a unit distance solution

Original post

Sebastien Bubeck@SebastienBubeck

Expert prompting was done by Boris Alexeev btw 😅

6:01 PM · Jun 3, 2026

Sentiment

Pos100%

Neg0%

Users are positive about GPT-5.5 disproving the Sum-Product Conjecture because it supports advancing toward a Kolmogorov Appendix style of AI explanation.

1 comment with sentiment.

2 more posts

Sebastien Bubeck@SebastienBubeck·7hReply
@littmath totally. I am however a big fan of moving into the direction of "Kolmogorov Appendix" that I mentioned a while ago, where you show the "minimal prompt" that allows the model to recover your result. Details TBD (and there are many!)
View on
Jimmy Apples 🍎/acc@apples_jimmy·6hQuote tweet
the public doesn’t see that current models can do so much, you don’t need the internal locked behind lab doors. Also not sure I agree with Sebastien on letting the math community slowly work out these problems themselves, they’ve been notorious gatekeepers.