tl;dr: Mathematician Frank Calegari argues that OpenAI's October 6 release of 700+ AI-generated math papers—which reportedly cracked many major open problems—marks a watershed moment that the math community should embrace rather than mourn, comparing the shock to the 1980s classification of finite simple groups. He proposes concrete reforms: evaluate all work as if AI-assisted, prioritize in-person interviews over paper files in hiring, demand better exposition from journals (or abandon them), emphasize breadth in graduate training, and reward problem selection and vision over mere proof production.