The Erdős Breakthrough

OpenAI
OpenAIMay 20, 2026

Why It Matters

The breakthrough proves AI can generate novel, deep mathematical insights, potentially transforming research speed and methodology across scientific disciplines.

Key Takeaways

  • AI solved Erdős distinct distances problem, first major math breakthrough
  • Model improved known construction using deep algebraic number theory tools
  • Human researchers doubted output before AI proved solution validity
  • AI explored countless proof paths, surpassing human decision‑making limits
  • Breakthrough signals AI’s potential across mathematics, science, and engineering

Summary

The video announces that an artificial‑intelligence system has solved the Erdős distinct distances problem, a landmark unsolved question in combinatorial geometry. It is hailed as the first clear instance of AI delivering a genuine mathematical breakthrough.

The AI model not only reproduced the known construction but identified a substantially better one, leveraging sophisticated algebraic number‑theory techniques that had eluded human mathematicians. Researchers noted the solution required navigating an intricate web of decisions, which the AI explored exhaustively, something humans found too delicate to manage.

One of the scientists described his reaction: “I couldn’t believe it, I lost sleep for nights,” underscoring the shock of seeing a machine outperform expert intuition. The team also said the result compresses the timeline for future discoveries, hinting at a “golden era” for mathematics driven by AI.

If AI can routinely crack problems of this depth, it could accelerate innovation across mathematics, physics, biology and engineering, reshaping how research is conducted and shortening the path from conjecture to proof.

Original Description

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946.
For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better.
This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.
The proof came from a general-purpose reasoning model, not a system built specifically to solve math problems or this problem in particular, and represents an important milestone for the math and AI communities.
This result points to something larger: AI systems are becoming capable of holding together long, difficult chains of reasoning, connecting ideas across distant fields, and surfacing paths researchers may not have explored.
We believe those same abilities will soon accelerate work in biology, physics, engineering, and medicine.
That future still depends on human judgment. Expertise becomes more valuable, not less. AI can help search, suggest, and verify. People choose the problems that matter, interpret the results, and decide what questions to pursue next.

Comments

Want to join the conversation?

Loading comments...