A computer can pass the bar exam, but until very recently it could not answer a question a five-year-old gets right: did the rooster's crow cause the sun to rise, or did the sun cause the rooster to crow? Machines were great at correlation and lousy at causation. The man who closed that gap is the subject of this episode: Judea Pearl, the Israeli-American computer scientist whose work on Bayesian networks and causal inference taught artificial intelligence how to ask why.

We trace his career at UCLA, the introduction of Bayesian networks for probabilistic reasoning under uncertainty, and the bigger leap (the do-calculus and structural causal models) that lets a machine draw physical arrows between variables, isolate confounders, and answer counterfactual questions. We unpack why this matters everywhere from epidemiology and drug tria ...