At sunset on the last day of the conference, the experts found themselves circling a familiar puzzle known as the “zombie problem.” Suppose that you’re a scientist studying octopuses. How would you know whether an octopus is conscious? It interacts with you, responds to its environment, and evidently pursues goals, but a nonconscious robot could also do those things. The problem is that there’s no way to observe consciousness directly. From the outside, it’s possible to imagine that the octopus is a “zombie”—physically alive but mentally empty—and, in theory, the same could be true of any apparently conscious being. The zombie problem is a conversational vortex among those who study animal minds: the researchers, anticipating the discussion’s inexorable transformation into a meditation on “Westworld,” clutched their heads and sighed.