OpenAI has claimed that its unreleased AI reasoning model solved a decades-old mathematical problem that had remained unsolved for nearly 80 years. The model produced an original mathematical proof disproving a famous geometry conjecture, according to a blog post published on Wednesday, May 20. At the center of the purported breakthrough is the 'Planar unit distance problem,' first posed by Paul Erdos. The problem asks a deceptively simple question: if points are placed on a flat two-dimensional plane, how many pairs of points can be exactly one unit apart?

The claim has drawn attention from prominent figures in the mathematical and AI communities, including Thomas Bloom and Tim Gowers. The model's ability to generate a proof for a problem that had resisted resolution for decades highlights the potential of AI in advancing mathematical research. The breakthrough has also raised questions about the broader implications of AI in fields traditionally reliant on human insight and creativity.

The problem, first posed in 1946, had remained unsolved for nearly 80 years. OpenAI's model produced a proof that disproves the conjecture, marking a significant milestone in the intersection of AI and mathematics. The blog post details how the model's reasoning process led to the discovery, offering a glimpse into the future of AI-assisted mathematical exploration. The achievement has been described as a step toward more fully exploring the 'cathedral of mathematics' that has been built over the centuries.

The implications of this development are far-reaching, particularly in terms of cost, lock-in, and governance. As AI systems become more capable of solving complex problems, questions arise about the reliance on proprietary models and the potential for vendor lock-in. The cost of accessing such advanced AI tools may also influence their adoption and impact across various industries. Additionally, governance frameworks will need to evolve to address the ethical and practical challenges associated with AI-driven discoveries.

The claim is still under development, with OpenAI continuing to refine and validate the model's capabilities. The blog post published on Wednesday, May 20, marks the beginning of a broader conversation about the role of AI in mathematical research. As the model's performance is further evaluated, its potential to contribute to other unsolved problems in mathematics and beyond will become clearer. The development underscores the potential of AI in advancing scientific and mathematical knowledge.