Part of The AI Control Crisis · 11 stories · since Sep 4 · newest 30m ago
OpenAI's math scoops spark mathematician revolt over 'stolen' proofsOpenAI claims its model cracked the Navier-Stokes Millennium Prize
7 Sep 12 4:00 AM · 12d ago · 6 articles · 17 posts · 3 comments · 6 sources · development 7 of 8
OpenAI announces its model solved the century-old Navier-Stokes Millennium Prize Problem; mathematicians, already unsettled by the credit disputes, call the announcement premature and self-serving.
“It seems like immature playground boasting writ large, underpinned by billions of dollars & the potential for significant environmental damage”
Prof James RobinsonTristan Buckmaster NYU mathematics professorLevent Alpöge Mathematician, Buckmaster's collaboratorAndreas Thom MathematicianOpenAI AI companyJames Robinson Mathematician, University of Warwick
The whole story articlespostscomments the bright band is this development · numbered dots are the others · click one to jump
Reported in the same hours no headline names this development itself — these 3 claims were published in its stretch
-
first by CBC, 12d ago · also The Verge
1 more headline
- OpenAI just wants to win Mastodon · 12d ago
-
first by Guardian Science, 12d ago · also The Guardian
-
first by Mastodon, 13d ago · also Gizmodo
What people said 9 voices · best of 10 · verbatim
-
F
“if frontier labs see a useful path to a discovery, they can spend tens of millions of dollars using LLMs to beat the original researchers to a proof — a dynamic that will incentivize secrecy”
-
OpenAI just wants to win https://www. theverge.com/ai-artificial-int elligence/994255/openai-millennium-prize-problem-tristan-buckmaster-competition?utm_source=flipboard&utm_medium=activitypub Posted into Tech News @ tech-news-theverge
-
What's interesting is that my company has developed SO many AI tools in the past year, and I was tasked with using a bunch of them as part of my workflow and to evaluate them from my experience. Some tools are annoying. Some tools are wildly good and useful. Some tools really do help reduce something that would take a few weeks down to a few…
-
>whether researchers can trust $CLOUD_SERVICE_OR_PROVIDER with $PROPRIETARY_SECRETSOf course not... are you crazy‽ Cloud is TRUST ME BRO information security! It absolutely dumbfounds me why anybody at all subscribes to a cloud or SaaS or IaaS, or anything which isn't under only their direct control!Then again they (individuals and organizations)…
-
O
Twenty-five leading mathematicians signed an open letter arguing that AI labs are threatening their intellectual work. #Techcrunch #breakingnews #news #othur
-
J
RE: https:// mathstodon.xyz/@andreasthom/11 7240535270608201 We can assume unreleased models under active development are being trained and retrained constantly, that's what developing them means, so they are exposed to new data that enters the training set. The high capacities of these models mean that they are absolutely capable of reproducing…
-
So a system trained on stolen data reportedly is stealing data from its customers? Shocking
-
P
OpenAI fought dirty on career-making math problem, says NYU mathematician | TechCrunch https:// techcrunch.com/2026/09/08/open ai-fought-dirty-on-career-making-math-problem-says-nyu-mathematician/
-
M
RE: https:// mathstodon.xyz/@tao/1172536299 67855195 Y a des moments, tu te dis que les youtubers "philosophe" et IA-hypers devraient, un peu prendre des pincettes
All 8 developments of OpenAI's math scoops spark mathematician revolt over… →
Hacker NewsMastodonXNewswiresBlueskyGoogle NewsReddit