Google DeepMind's AlphaProof (reinforcement-learning system using formal reasoning in Lean) and AlphaGeometry 2 (neuro-symbolic geometry solver) jointly scored 28 of 42 points on the 2024 International Mathematical Olympiad, matching a silver medalist performance. AlphaProof proved two algebra and one number-theory problem; AlphaGeometry 2 solved the geometry problem in 19 seconds; two combinatorics problems remained unsolved. Problems were manually translated to formal language before solving, with solution times ranging from minutes to three days. This demonstrates verifiable gains in multi-step formal reasoning on problems judged by IMO officials and independent mathematicians including a Fields Medal winner.
283 other tracked developments touching the same claims.