PƒPPrompts for Progress
← Problems indexbenchmark

mathematics · research mathematics

First Proof research problems

Ten research-level mathematics problems used to assess AI-generated proof attempts with problem-author feedback.

This benchmark stays aggregated until each problem and its changing expert assessment can be represented faithfully.

Open reference source ↗

Attempt ledger

What has been tried?

A complete attempt outcome does not automatically change the editorial status of the underlying problem.

01
Feb 14, 2026 · campaignOpenAI attempts all ten First Proof problems

An internal model produced attempts for ten research-level problems; expert assessments varied, and one attempt initially thought likely correct was later acknowledged as incorrect.

Mixed campaignExpert-reviewed evidence