Partially Verified65% confidenceEventExact time
OpenAI publicly released 722 AI-generated mathematical manuscripts covering 372 result families, derived from testing an internal model on approximately 4,000 mathematical research problems.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
10/7/2026
First Seen
Valid until: 1/5/2027
Sources
Related Entities
Related Claims
UnverifiedOpenAI published a batch of mathematical results produced by internal frontier models, comprising 722 manuscripts; the models were posed approximately 4,000 open research questions, with each result consuming an average of about 3 hours of ChatGPT Pro reasoning compute.80% similarUnverifiedOpenAI recently disclosed its models' performance on mathematical problem solving, involving attempts at reasoning and proving complex mathematical propositions.76% similarUnverifiedOpenAI released a 249-page research collection claiming that an internal version of its next-generation model family, Astra, has achieved ten major breakthroughs in mathematics and theoretical computer science.73% similarUnverifiedAccording to Buckmaster, OpenAI mathematics team lead Sébastien Bubeck told him that an OpenAI model had autonomously generated a complete roughly hundred-page proof of forced Navier-Stokes blowup, and proposed arranging authorship credit for Buckmaster.71% similarUnverifiedOpenAI's AI systems have already solved more than 100 open problems in mathematics70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/984054API
curl https://kongchang.com/api/v1/knowledge/claims/984054MCP
get_claim(id=984054)