AI Solves 80-Year-Old Math Puzzle That Stumped Generations of Researchers

An artificial intelligence model developed by OpenAI has solved the 'unit distance problem,' an 80-year-old mathematical puzzle first proposed by Hungarian mathematician Paul Erdős in 1946. The problem asks how many pairs of points can be placed on a plane so that each pair is exactly one unit apart. Erdős had previously demonstrated a specific arrangement and suggested no better solution existed, but OpenAI’s model identified a new configuration that disproves his conjecture. The achievement marks the first instance where AI independently produced an original mathematical discovery, according to Noga Alon, a professor at Princeton University. Daniel Litt, an assistant professor at the University of Toronto, called the solution a milestone, stating it would have been published in leading journals if submitted by a human. Fields Medal winner Timothy Gowers noted the development raises questions about human competition in solving certain mathematical problems. OpenAI researchers initially presented the puzzle as a test of the AI’s capabilities, not expecting a breakthrough. After verifying the solution with external experts and additional AI tools, they confirmed its correctness and innovation. The model’s success stemmed from exploring unconventional approaches, unlike human researchers who focused on proving Erdős’s original assumption. The breakthrough highlights AI’s evolving role in scientific research, shifting from a supportive tool to a potential source of original ideas. As AI systems advance, their impact on mathematics and other fields is expected to grow, prompting discussions about the future of human-machine collaboration in discovery.