OpenAI announced it has solved ten major problems in mathematics and theoretical computer science, including the first construction of a non-sofic group and a proof that the multicolour Ramsey number R(3,3,...,3) grows superexponentially in k. The announcement came in early August 2026, marking a significant milestone in the application of large language models (LLMs) to complex mathematical challenges, according to gowers.wordpress.com.
The breakthroughs were revealed through a series of talks and discussions following OpenAI's announcement. The construction of a non-sofic group addressed one of the most important unsolved problems in group theory, while the Ramsey theory proof tackled a major open problem previously considered unlikely to be solved within a human lifetime. These results highlight the advanced capabilities of LLMs in formal mathematical reasoning, as detailed on gowers.wordpress.com.
These achievements demonstrate a new level of proficiency for AI in abstract mathematics, a field traditionally resistant to automation. The solutions to these longstanding problems position OpenAI at the forefront of AI research in theoretical computer science and mathematics. The work also challenges prior assumptions about the limits of LLMs, showing their potential to contribute to foundational scientific knowledge, as noted on gowers.wordpress.com.
The announcement has sparked widespread interest in the AI and mathematics communities, with experts analyzing the implications of these solutions. OpenAI's success in solving these problems was documented on August 12, 2026, on gowers.wordpress.com, providing a detailed account of the mathematical significance and the AI methodologies involved.