OpenAI has released a range of new mathematical results produced by an internal frontier model, publishing them in a GitHub repository with protocols for paper revisions and citations. The release follows advice from the Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study. OpenAI is also sharing Lean formalizations of many proofs, plus details on how results were obtained, including reasoning summaries and compute estimates. The average result used roughly three hours of ChatGPT Pro thinking.
No score is assigned. Sources and their independence are shown in the citation chain below.