Breakthroughs Weekly — 25 October 2026
An AI-produced proof that the Riemann zeta function has no zeros near the edge of the critical strip is drawing serious expert endorsement.
An AI-written proof of the quasi-Riemann hypothesis wins over number theorists
Quanta Magazine · 19 Oct 2026 · Number theory
OpenAI released two papers showing that the nontrivial zeros of the Riemann zeta function cannot lie close to the edges of the critical strip: a 199-page version dated 30 September confines them to real parts between 1/8 and 7/8 and has been checked in Lean, and a shorter 49-page version around 5 October narrows that to between 1/12 and 11/12 and has not yet been formalised. This is a weakened cousin of the Riemann hypothesis, not the hypothesis itself, but it breaks a barrier that human methods had not crossed. It was first posted on 6 October as part of a large batch of AI results, and became news for mathematics this month when Quanta reported that the experts it consulted largely believe it is correct, with Stanford’s Jared Duker Lichtman calling it the most significant AI achievement in mathematics so far. Terence Tao’s blog has since been working through the arguments. The papers are preprints with no journal review, NYU’s Maksym Radziwiłł says both are too poorly written for a journal as they stand, and Peter Sarnak will not call it a theorem until people have checked every detail.