Anthropic's Unreleased AI Model Pushes Riemann Hypothesis Verification Past Two-Thirds

Anthropic announced August 10, 2026 that an unreleased AI model raised the verified proportion of Riemann zeta zeros satisfying the Riemann hypothesis from 41.6% to 67.2% — the largest advance on the 150-year-old problem in decades. The result, formalized in Lean and confirmed by two in-house mathematicians, extends a lower bound, not a full proof. The Clay Institute's $1 million prize remains unclaimed.