PERSON
Paul Ekman
The psychologist who spent six decades mapping the universal grammar of the human face—and whose framework for emotion recognition became the contested foundation of a multibillion-dollar industry he lived long enough to distrust.
Paul Ekman is the man who handed the machines their method. Born in Washington, D.C. in 1934, he flew to the New Guinea highlands in the late 1960s with a set of photographs and a question that now governs airports, classrooms, and hiring interviews: do all people, everywhere, make the same face when they are afraid? His answer—that a small set of basic emotions is accompanied by universal, biologically rooted facial expressions—became the intellectual foundation of
affective computing, and the six or seven emotion categories his research established are the categories every emotion-recognition classifier still inherits. Yet the most important thing to understand about Ekman in the AI age is the gap between the framework the industry absorbed and the framework he actually built: a framework that distinguished the measurable face from the felt interior, that warned against confident inference from a single expression without context, and that, in his late dialogues with the Dalai Lama, turned steadily inward toward the felt