A philosopher and mathematician argues that frontier AI solving famed problems undermines the "checklist" view of mathematical progress, and proposes a "generative encounter" model in which human purpose and open-ended inquiry remain central to mathematics.