A general-purpose AI has cracked Paul Erdős's 1946 unit distance problem by producing geometric configurations that exceed long-standing conjectured bounds. The solution proves at least n^(1+δ) unit-distance pairs exist for some δ > 0, with Princeton mathematicians verifying the result. Tim Gowers and Arul Shankar called the breakthrough a significant advance in mathematics.
Erdős's Enduring Challenge
Paul Erdős introduced the unit distance problem in 1946 while studying combinatorial geometry. The deceptively simple question asks how many pairs of points can be exactly one unit apart in a set of n points on a plane. For decades, mathematicians grappled with establishing tight lower bounds for these configurations.
Breaking the Bound
The AI generated point arrangements that surpass all previously conjectured limits for unit-distance pairs. Its solution mathematically establishes at least n^(1+δ) pairs for a fixed positive δ, shattering assumptions that had constrained the field for generations. The configurations achieve this density through precise spatial arrangements no human had previously constructed.
Princeton Verification
Researchers at Princeton University rigorously tested the AI's output. Their independent review confirmed the solution's validity and the proof's soundness. The team documented the verification process through multiple rounds of computational checks and peer review within their department.
Math Community Response
Fields Medalist Tim Gowers and Princeton's Arul Shankar both described the AI's result as a significant advance. Their assessment centers on how the solution redefines what's mathematically possible for unit-distance problems. The work arrives without accompanying explanations for how the AI derived the configurations.
Mathematicians now aim to determine the exact maximum δ possible for the bound.
