Artificial intelligence governance frequently begins with a technical question: How capable is the system? That question matters. A more capable system may produce better predictions, recommendations and decisions. But consequential AI requires a second question that is fundamentally different: Who …
