An OpenAI model has disproved a central conjecture in discrete geometry (openai.com)

OpenAI's recent breakthrough has led to the resolution of a nearly 80-year-old question in discrete geometry known as the planar unit distance problem, originally posed by mathematician Paul Erdős. Traditionally believed to be constrained by square grid configurations, the new findings, generated by a general-purpose reasoning model rather than one specifically trained for mathematics, reveal an infinite family of configurations that improve the maximum number of unit-distance pairs among points. This result, validated by external mathematicians, signifies the first instance where an autonomous AI has solved a prominent open mathematical problem, demonstrating the model's deep reasoning capabilities. The implications of this achievement are far-reaching, not only for mathematics but also for the future of AI research. The proof utilizes unexpected connections from algebraic number theory to address an elementary geometric question, suggesting that AI can unearth insights that may encourage mathematicians to explore new pathways. This achievement may herald a new era of collaboration between AI and human researchers, where advanced models could become essential partners in tackling complex problems across various fields such as biology, physics, and engineering. As AI systems evolve to support intricate reasoning and knowledge synthesis, their role in research could transform significantly, affirming the need for human oversight in directing the path of future inquiries.