OpenAI's latest model, GPT-5.2, is demonstrating significant advances in mathematical reasoning, solving complex problems at the college level and beyond. In a notable recent achievement, the model generated original proofs for several decades-old Erdős problems, which were subsequently verified by prominent mathematicians.
This breakthrough is part of a broader industry trend where AI is moving beyond pattern-matching to autonomous proof-generation. Competitors like Google are also reporting major successes, with their own AI systems achieving gold-medal level performance at the International Mathematical Olympiad. This leap in AI's logical capabilities is poised to accelerate discovery across science, engineering, and finance by tackling complex optimization and modeling tasks.