81% of the most important math discoveries in the last three years were released today.Blue = human discoveredRed = AI-discovered, released before todayGreen = AI-discovered, released today
>>17069047Another vibeGOD victory
>>17069047OK, assuming it's not slop:What is the most interesting potential application for each of these results?
>>17069144Its math bro. Its always been useless lmfao
81% of all posts on /sci are ai bot posts
>>17069047They want to go fast and over the top to break the limit of humanity with energy over time. Short bursts.
Has the NS been confirmed or not
>>17069209why should mathematicians verify it? openai produced the slop, let them verify itwhat, they want to even steal verification effort?
>>17069211Openai works in conjuction with mathematicians, including witten.
>>17069144>>17069147Yeah I suddenly see videos from mathematicians "ok but these Ai results have no real world apication". Yeah kek as if modern math has that>>17069211>we refuse to do our jobDude you get paid you can't just do the work you like
>>17069144The most important application of all time: more mathematics
>>17069047This list is slop.
>>17069047Call me crazy but I'd consider unramified GLC more significant than NS or quasi RH. People act like it's not as significant just because it's not one of the millennium problems
>>17069462wait did llms steal it too?https://arxiv.org/html/2605.23167v1
>>17069462I think most people who know what all three are would agree with you.
>>17069211their mathematicians already did tho
>>17069051>fastDelusion>strongBuggy but the people using it do not see them because they are clueless on integers.It is the greatest scam and cult in human history.
>>17069882Short it then cuck
>>17069051this is not going to make the lives of anyone better, but it is going to make the lives of a lot people worse. why celebrate it?
>>17069917How dare those robot makers try to provide humanity with abundant energy, information, and physical resources.
>>17069211Anyone who asks this kind of a question isn't a real mathematician. If there's a result you've been interested in for years, of course you want to understand it. Even if it's horribly written. Some people will give up as there is a lot of unreadable slop in there and wait for a human to digest it (it often happens with human written math as well, there's a genius who's a horrible writer so most people don't get what they even wrote, and they wait for someone else to "translate"), but everyone should give it an honest try.
>>17069047>discoveriesNot really. They were discovered already. Le AI just filled in the gaps in proofs. It's not very interesting. Why should I care?
>>17070015This, Newton didn't give a shit about proofs, neither does Kontsevich
