Historic Breakthrough in Mathematics by AI
OpenAI announced that its general-purpose reasoning model disproved a central conjecture concerning the "unit distance problem" in discrete geometry, which had remained unsolved for approximately 80 years since it was proposed by the renowned mathematician Paul Erdős in 1946. This achievement is notable as a milestone demonstrating AI's ability not merely to process existing knowledge but to autonomously generate new discoveries for difficult scientific problems that have eluded human solution for a long time. Until now, the mathematical community generally believed that a square grid arrangement of points was close to optimal; however, OpenAI's model discovered an entirely new structure that surpassed it. This discovery has also been verified by an independent group of mathematicians, confirming its correctness.
Technical Details: What is the Unit Distance Problem?
The "unit distance problem" is a seemingly simple yet profoundly deep problem that asks, "When n points are placed on a plane, what is the maximum number of pairs of points whose distance is exactly 1?" For many years, it was conjectured that the upper bound for this number of pairs was approximately linear with respect to the number of points n (technically, n to the power of 1+o(1)). However, OpenAI's model overturned this conjecture, demonstrating the existence of point configurations with more than n to the power of 1+δ pairs (where δ is a positive constant). Surprisingly, to solve this pure geometry problem, the AI applied advanced mathematical concepts such as the "infinite class field tower" from algebraic number theory and discovered an ingenious approach—projecting high-dimensional structures onto a plane—which would be difficult for humans to conceive. This suggests that AI possesses the ability to find unexpected connections between different mathematical fields and pave new paths for proofs.
Impact and Outlook for Engineers
This achievement demonstrates to Japanese engineers that AI is not merely a tool for streamlining existing tasks like programming and data analysis. Going forward, AI is expected to take on the role of a "research partner" collaborating with humans in complex R&D fields such as materials science, drug discovery, and physical simulation. By autonomously generating and testing hypotheses and proposing novel solutions that humans might have overlooked, AI could dramatically accelerate the pace of scientific and technological discovery. Engineers will need to increasingly develop skills to view AI not merely as a tool to "use," but as an "interlocutor" for exploring unknown problems, understanding its reasoning process, and guiding it appropriately.
📦