Solving a 31-year-old grand challenge in mathematics – the KLS conjecture
AI Summary: Researchers including Xinzhi Zhang of Microsoft Research have proved the 31‑year‑old Kannan‑Lovász‑Simonovits (KLS) conjecture, establishing that for log‑concave high‑dimensional distributions the optimal flat (hyperplane) cut approximates the optimal arbitrary cut within a universal constant independent of dimension. The proof demonstrates a dimension‑free geometric principle that simple flat cuts capture the narrowest bottlenecks of such distributions, tightening previous bounds that grew with dimension. The breakthrough resulted from a human‑directed AI‑assisted research process in which AI systems generated mathematical candidates that researchers evaluated, pruned, and redirected. Independent researchers have validated the result and a machine‑checked formal verification confirms the proof, removing a major theoretical obstacle for the analysis of randomized algorithms and high‑dimensional sampling.