OpenAI publishes solutions to more than 370 outstanding math challenges. The results divide mathematicians but most agree: math will never be the same

OpenAI has recently released AI-generated solutions to over 370 mathematical challenges, some of which are considered grand challenges in the field. This massive collection of results has left mathematicians both excited and divided. Some see it as a groundbreaking development, while others view it as an assault on mathematics as an academic discipline.

OpenAI achieved these results using an unreleased internal AI model, with the average computation time per solution being around three hours. The breakthroughs include full or partial solutions to many of the most important mathematical problems mathematicians have been grappling with. This includes full or partial results for some of the Millennium Prize problems, with progress made on three of them, although not fully solved yet.

The release of these solutions has sparked a debate within the mathematical community. While some are enthusiastic about the new areas for exploration and the potential for learning logical reasoning skills that can be applied to other domains, others are concerned. They fear that the perception that AI has "solved math" may lead to a withdrawal of funding for mathematical research and discourage promising young mathematicians from pursuing the field.