OpenAI Claims Model Solved 100+ Open Math Problems, Forms Mathematician Advisory Group
On September 22, OpenAI officially announced a partnership with an independent Advisory Group on Mathematics and AI to guide responsible sharing of progress in AI and mathematics. The same announcement reportedly stated that an internal model, trained starting August 28, has solved over 100 long-standing open problems across most fields of mathematics — a pace that surprised even OpenAI's in-house mathematicians. No problem list or paper has been published, and mathematicians have publicly demanded details, making this a landmark case of AI colliding with academic norms.
Confirmed
- OpenAI's official post (@OpenAI) announced the advisory group with duties covering evaluating and communicating new mathematical results, upholding academic and professional standards, and building tools for mathematical research and learning, with mathematicians placed at the center
- According to reposts (@MarvinTBaumann, @TinfoilTricorn), the group operates independently of OpenAI; @ZeroStateReflex relayed that an initial member list of well-known mathematicians was published, though full details remain incomplete
- The announcement claims the internal model solved "more than 100 long-standing open problems across most areas of mathematics"; reposts (@BilelKort, @gaganghotra, @DrSingularity) note training began August 28, with results in roughly 24 days
- Per @ns123abc's timeline, Terence Tao had earlier said AI was advancing too fast in mathematics and should slow down; about ten days later OpenAI formed the group, but @404LogicNotFound noted the group has no authority over the model's pace of development
- Mathematician Liu Zuxin remarked on seeing a model solve so many problems in weeks (@harrisedouard); Ethan Mollick also commented on why AI's impact on professions will be slower (@eldonredwards)
Unconfirmed
- Specific claims such as solving the Navier–Stokes Millennium Prize problem come mainly from third-party relays (@DrSingularity, @coherence, @WonderFactory); the official announcement lists no problems, and no paper exists
- The list of 100+ solved problems is unpublished; Peter Gostev and others have publicly asked OpenAI for it (@arannayebi); several posts (@oceanprotocol, @filterdust) explicitly flag the claims as unverified
- @DrSingularity speculated, without evidence, that OpenAI may have shifted focus from math to physics, citing Sam Altman's earlier remarks on room-temperature superconductors
Why it matters
- Mass claims of AI-produced mathematics could upend academic publishing and attribution norms; the advisory group is an industry self-regulation experiment, summarized by @Polymarket as a response to mathematicians' concerns
- The group's independence is key to its credibility; critics (@jdoline via @inductionheads) mocked it, and @ns123abc / @404LogicNotFound noted it cannot slow the model's pace, meaning the "slow down" call has no mechanism behind it; some researchers argue special gatekeeping for mathematicians will slow progress (@burnytech)
- Community reactions are polarized: some (@oceanprotocol) declared "math is basically solved" and imagined 2027 capabilities, while others (@filterdust) urged readers to check the original announcement, warning the "100 problems" claim may be inflated in second-hand spread
2026-09-22 ~ 2026-09-23 · 59 related posts
Primary sources
- OpenAI forms independent advisory group of mathematicians to guide AI math advances — OpenAI ·
- OpenAI's New Model Reportedly Solved 100+ Open Math Problems in 24 Days of Training — BilelKort ·
- After Terence Tao urged slower AI math progress, OpenAI formed an advisory group with no pacing power — ns123abc ·
- [source] OpenAI forms independent advisory group of mathematicians to guide AI math advances — OpenAI · 2026-09-22
- OpenAI Sets Up Independent Math Advisory Group After AI Solves 100+ Open Problems — Polymarket · 2026-09-22
- Reddit claims OpenAI solved 100 open math problems, links to advisory group post — filterdust · 2026-09-22
- OpenAI says model trained just 24 days ago has solved 100+ longstanding math problems — gaganghotra_ · 2026-09-22
- OpenAI's new model claims Navier–Stokes and 100+ open math problems in weeks — Dr_Singularity · 2026-09-22
- OpenAI's new internal model claims 100+ solved open math problems, incl. Navier–Stokes — Dr_Singularity · 2026-09-22
- OpenAI model reportedly cracked Navier–Stokes Millennium Prize problem plus 100+ open math problems — coherence · 2026-09-22
- GPT turns 8: model said to have resolved 100+ open math problems — lpachter · 2026-09-22
- Repost: OpenAI model reportedly resolves Navier-Stokes plus 100+ open math problems — soumitrashukla9 · 2026-09-22
- [source] After Terence Tao urged slower AI math progress, OpenAI formed an advisory group with no pacing power — ns123abc · 2026-09-22
- Claim: a model trained for just 24 days solved 100+ open math problems — rand_longevity · 2026-09-22
- OpenAI forms independent math advisory group; skeptic predicts it dissolves within a year — RexDouglass · 2026-09-22
- OpenAI's new model reportedly solved 100+ open math problems weeks into training — Dr_Singularity · 2026-09-22
- Unverified: OpenAI model trained in 24 days cracks 100+ open math problems — ocean_protocol · 2026-09-22
- OpenAI reportedly solves 100 math problems as observers declare the field shifted — kimmonismus · 2026-09-22
- OpenAI says internal model solved 100+ open math problems, forms advisory group with top mathematicians — 机器之心 · 2026-09-22
- OpenAI's new model solved 100+ open math problems, prompting 25 Fields medalists to protest — APPSO · 2026-09-22
- OpenAI's internal model reportedly cracked 100+ open math problems; IAS advisory panel formed — AI寒武纪 · 2026-09-22
- Unverified: OpenAI Reportedly Trained a New Model in 1.5 Weeks That Is 4x Better Than Astra at Math — felpix_ · 2026-09-22
- Report: OpenAI Trained New Math Model in ~11 Days, Solved Longstanding Open Problems — felpix_ · 2026-09-22
- OpenAI Model Reportedly Solved 100+ Major Math Problems in 3 Weeks; Physics Next? — Dr_Singularity · 2026-09-22
- OpenAI forms mathematician advisory group; critics note it wasn't so careful with entry-level SWE — RexDouglass · 2026-09-22
- OpenAI forms mathematician advisory group; Ethan Mollick says professions will slow AI's impact — eldonredwards · 2026-09-22
- OpenAI forms mathematician advisory group as it races Anthropic past 100 AI-solved proofs — ShenRaphael · 2026-09-22
- OpenAI Forms Nine-Mathematician Committee After Models Solve 100+ Open Math Problems — emmanuelvivier · 2026-09-22
25 near-duplicate retellings: patience_cave · soumitrashukla9 · aran_nayebi · aran_nayebi · soumitrashukla9 · soumitrashukla9 · inductionheads · silver__tsuki · soumitrashukla9 · inductionheads · BilelKort · WonderFactory · IgorCarron · ZeroStateReflex · TinfoilTricorn · aaron_lou · aran_nayebi · inductionheads · 智东西 · MakeUSAVAT · soumitrashukla9 · RexDouglass · ZeroStateReflex · FinanceYF5 · avt_im