← Back to live feed1 story
OpenAI released 722 math manuscripts grouped into 372 families, including a claimed proof of the quasi-Riemann hypothesis with a 7/8 bound and partial progress on the Riemann, Hodge, and BSD Millennium Prize problems, as well as a claimed resolution of Hilbert’s tenth problem over the rationals and a result raising the chromatic number of the plane lower bound to six. The findings were produced by an internal, unreleased OpenAI model and include papers, proof artifacts, and reasoning summaries, with consultation from the Institute for Advanced Study's Advisory Group on Mathematics and Artificial Intelligence.
Sign in to suggest edits
Key sources
- SUPPORT@hesamation“>september: OpenAI publishes a Navier-Stokes Millennium Prize solution >25 Fields medalists sign an open letter”x.com
- SOURCE@openai“We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results.”x.com
- SUPPORT@kimmonismus“OpenAI says the standard procedure used an average of three hours of ChatGPT Pro thinking compute per result.”x.com
- SUPPORT@mathemagic1an“Partial progress on 3 millenium prize problems: Riemann, Hodge and BSD”x.com
- SUPPORT@cremieuxrecueil“This 7/8 quasi-Riemann hypothesis proof looks like one of the most important advances in analytical number theory in decades.”x.com
- SUPPORT@willdepue“Finite-time blowup for 3D incompressible Navier–Stokes with smooth forcing (Clay statements C/D), OpenAI internal multi-agent model, Lean-formalized, 2026”x.com
- SUPPORT@spicey_lemonade“It’s somewhat unclear to me how the math board influenced the release. Without the math board, how would they have released this?”x.com
- SUPPORT@petergostev“I probably spent 300 billion tokens on Hadwiger–Nelson problem (GPT-5.6 and Astra) and got absolutely nothing - and here it is, five-colours is no go, lower bound has moved to 6 or 7.”x.com