OpenAI Says a Secret AI Model Cracked Hundreds of Open Math Problems in One Prompt—Mathematicians Want Receipts

What happened

Only 162 of the 722 papers have a Lean-formalized main result, and OpenAI warns that some unformalized results could have issues. MIT's Andrew Sutherland says the one-prompt, single-agent claim is unverified until the model is released, and the release omits the prompts that an advisory group at the Institute for Advanced Study recommended disclosing. OpenAI is an artificial intelligence company based in San Francisco.

OpenAI published 722 math manuscripts on GitHub on Tuesday, all produced by an internal model the company has not released. OpenAI posted 722 AI-written math manuscripts from an unreleased model, saying most came from a single prompt.

Decrypt News Artificial Intelligence OpenAI Says a Secret AI Model Cracked Hundreds of Open Math Problems in One Prompt—Mathematicians Want Receipts OpenAI posted 722 AI-written math manuscripts from an unreleased model, saying most came from a single prompt. In brief OpenAI published 722 math manuscripts in 372 result families on GitHub from an unreleased internal model. An OpenAI spokesperson said almost everything came from a single prompt handed to a single AI agent (AI that carries out multi-step tasks rather than answering one question), though some may have taken multiple attempts.