The mathematical community has experienced an unprecedented "earthquake." OpenAI has officially announced that its internal general reasoning model independently overturned and solved a classic problem in combinatorial geometry that had puzzled the academic community for 80 years: the Erdős unit distance problem.

1. The Challenge of the Problem: A Seemingly Simple "Maze"

This problem was proposed by the legendary mathematician Paul Erdős in 1946: Given n points on a plane, what is the maximum number of pairs of points that have a distance exactly equal to 1?

  • The Collapse of Long-standing Consensus: For the past 80 years, top mathematicians generally believed that the optimal solution should be a grid-like arrangement, with the number of point pairs growing almost linearly.

  • The AI's "Inspired Stroke": Instead of sticking to traditional geometric approaches, OpenAI's model used algebraic number theory (including class field towers and the Golod–Shafarevich theorem, among others) to construct a completely new way of arranging point sets, proving that the growth rate of unit distance point pairs can indeed exceed linear growth.

2. Why Is This Called a "Milestone in AI Mathematics History"?

Timothy Gowers, a Fields Medal winner and renowned mathematician, commented on this event: "This is undoubtedly a milestone in the history of AI mathematics. If this paper were written by a human and submitted to the Annals of Mathematics, I would unhesitatingly recommend it for acceptance."

  • From Assistant to Autonomy: This is not just a simple computation; it marks the first time AI has pioneered a new path in the "no-man's land." It not only provided an answer but also introduced a geometric perspective and interdisciplinary bridge previously unexplored by humans.

  • Peer Review: The achievement spans 125 pages and was quickly verified by top mathematicians worldwide after being made public. Multiple experts unanimously agreed that its logic is rigorous and its innovation is extremely high.

3. Industry Signal: AI Evolves from "Calculating Fast" to "Thinking Deep"

This breakthrough reveals a deep shift in the logic of AGI (Artificial General Intelligence) development:

  1. A New Paradigm for Scientific Discovery: AI is no longer limited to "organizing data" within existing knowledge systems, but has shown the potential of an independent researcher, capable of exploring autonomously, proposing hypotheses, and conducting logical arguments.

  2. The Advantage of Cross-disciplinary Thinking: The model successfully connected the geometric problem with advanced algebraic number theory. This ability to link disciplines is precisely what human experts have long prided themselves on—intuition and deep insight.

  3. Verifying Future Potential: This proof process demonstrates that general reasoning models are now capable of maintaining rigorous logical chains. This means that the same architecture could potentially replicate this "scientific discovery" capability in more complex research areas such as physics, materials science, and biomedical engineering in the future.

Conclusion