The AI mathematical model from Google DeepMind has achieved a breakthrough in this year's International Mathematical Olympiad (IMO). According to official sources, the formal mathematical reasoning model AlphaProof and the improved geometric solving model AlphaGeometry2 jointly tackled four out of the six challenging problems in the current IMO, reaching a level equivalent to that of a human silver medalist for the first time.
In this human-versus-machine competition, AlphaProof demonstrated exceptional capabilities, not only successfully solving two algebra problems and a number theory problem but also providing rigorous proofs for its answers. Notably, AlphaProof also conquered the most challenging problem in the competition, which was only successfully answered by five human contestants during the actual event. Meanwhile, AlphaGeometry2 performed excellently in the proof of geometric problems.
The combined performance of these two AI systems is remarkable; they ultimately scored 28 points, achieving full marks on each solved problem, a result comparable to the highest score of the human silver medal group. However, the two combined problems remained unconquered by the AI systems, indicating that there is still room for improvement in certain mathematical domains.
This achievement not only showcases the tremendous potential of artificial intelligence in advanced mathematical reasoning but also opens up new possibilities for AI applications in mathematical research and education. Nevertheless, the unsolved problems also remind us that in the complex world of mathematics, the insights and creative thinking of human experts remain indispensable.