Hours after Anthropic claims a Claude proof of the Riemann hypothesis, Stanford releases an all-human competing draft
Hours after Anthropic claims a Claude proof of the Riemann hypothesis, Stanford releases an all-human competing draft
Anthropic mathematician Levent Alpöge announced that, working with Claude, he had found a proof of the Riemann Hypothesis and the Generalized Riemann Hypothesis, released only in the Lean programming language across 14 data packages. Hours later, Stanford's Ravi Vakil posted an independent, all-human draft proof of the GRH by Alessandro Masullo, reviewed for months by Brian Conrad, then updated that several people had reported problems in it. Alpöge said Claude supplied the key insight of chaining as many transformations as the square root of the sequence length, and expects to write a human-readable version in coming months.
It captures a same-day collision between an AI-assisted proof and a human one, plus the verification caveats on both sides.
Source: The Decoder · the-decoder.com