
The time scope of this question is unlimited, it will close when the next Millennium Problem is solved.
Millennium Problems: https://en.m.wikipedia.org/wiki/Millennium_Prize_Problems
People are also trading
Hello! I was not active until these days when I started receiving emails about the posts made here.
To be honest I don't feel certain about the resolution of this right now, as far as I know the Clay Institute has not yet accepted the proof and they may take a while before accepting it. I feel that if I closed this right now I would be basing that conclusion primarily on press coverage rather than on the mathematical community’s acceptance of the proof.
Also, there could be the debate about if they stole the work of Alpöge and Buckmaster. In my opinion even if they stole it I think that OpenAI made sufficient advancement so as to count this as solved by an AI, but I am really open to debate on that.
So yeah, those are my thoughts right now, I'll be happy to hear yours
@calour If you ping mods again in 48 hours, I'll give a judgment. You'll get extra speed if I'm included in the ping.
@MikeAckermann That no or little work was done by humans, at least that's what I was thinking when I wrote the question
@TobyBW I think that if the AI made progress and solved the problem by itself it still counts as the AI solving it, regardless of the advancement of the solution prior to the AI solution. For example if a mathematician takes all there is to know about one of the problems, and then solves it from there, I think we would still say that he solved it. That's the same way in which I see the AI solution.
