Claude Attempts Riemann Hypothesis, Improves Zeta Function Lower Bound to 67.2%

sudoraohacker · x · 2026-08-11

An Anthropic staff member recently challenged an unreleased research version of Claude with the Riemann hypothesis, one of the most famous unsolved problems in mathematics.

Although the model failed to prove the hypothesis, it made unexpected mathematical progress. It improved a longstanding lower bound for the fraction of zeros of the Riemann zeta function that satisfy the hypothesis, increasing it from 41.6% to 67.2% by drawing on extensive prior mathematical research.

Two mathematicians at Anthropic validated Claude's paper and produced an informal note for experts. Claude also generated a formally verifiable proof of its result. While Anthropic doesn't expect these specific techniques to fully prove the Riemann hypothesis, the experiment highlights the rapid progress in AI mathematical reasoning capabilities.

Related event: Anthropic's Claude Achieves Breakthrough in Tackling Riemann Hypothesis(21 posts)→

Original post →

More from Models

Models channel →