Google and OpenAI's AI models achieved a historic milestone by winning gold medals at the International Mathematical Olympiad, solving five out of six problems using natural language reasoning, indicating significant advancements in AI's mathematical capabilities and potential for solving complex research problems.
DeepMind's AI system, AlphaGeometry, has achieved a gold-medal performance on geometry problems from the International Mathematical Olympiad, demonstrating the ability to rigorously prove geometric facts at a star-student level. Unlike previous language-based AI models, AlphaGeometry uses a custom language for writing geometry proofs with a rigid syntax, making its output machine-readable and easy to check for accuracy. By combining statistical guesses with symbolic reasoning, the AI system shows promise in solving complex mathematical problems, although it is not yet autonomously engaging with modern research-level mathematics.
Computer scientist Trieu Trinh has developed AlphaGeometry, an A.I. system that solves geometry problems from the International Mathematical Olympiad at nearly the level of a human gold medalist. The system, which was developed during Trinh's residency at Google, joins the company's A.I. systems known for tackling grand challenges. AlphaGeometry's success represents a significant advancement in A.I.'s ability to solve complex mathematical problems.