OpenAI Claims AI Breakthrough, Solves 80-Year-Old Math Problem

OpenAI has made headlines with a purported breakthrough in artificial intelligence, asserting that an unreleased AI reasoning model has successfully tackled and resolved a mathematical problem that has puzzled mathematicians for almost eight decades. The challenge in question is the "Planar unit distance problem," first introduced by the renowned mathematician Paul Erdos in 1946. This problem involves determining the maximum number of times a given distance can occur among a set of points in a plane. The AI model, according to OpenAI, produced an entirely original mathematical proof. This proof effectively disproved a long-held geometry conjecture by demonstrating a more efficient configuration of points than human experts had previously conceived. This development underscores the rapid progress in AI's ability to engage in abstract and complex mathematical reasoning, a domain traditionally considered a bastion of human intellect. The announcement follows a trend of increasing AI proficiency in mathematics. For instance, in 2024, Google DeepMind introduced Alpha Geometry, an AI system that achieved performance levels comparable to International Mathematical Olympiad contestants. Both OpenAI and Google DeepMind's unreleased models reportedly showed similar impressive results in the 2025 math competition. Experts like Fields medalist Tim Gowers have lauded OpenAI's result as a "milestone in AI mathematics," suggesting that AI is helping to explore the vast landscape of mathematical knowledge more thoroughly. While the specific details of the model and its proof are yet to be fully disclosed, the claim signifies a potential shift in how complex mathematical problems might be approached and solved in the future, with AI playing an increasingly central role in generating novel insights and proofs.