American author, philosopher, and neuroscientist
It is sobering to admit that chaos seems a probable outcome even in the best-case scenario, in which the AGI remained perfectly obedient. But of course we cannot assume the best-case scenario. In fact, “the control problem”—the solution to which would gu
AI researcher who popularized the idea of friendly artificial intelligence
Yudkowsky argues that as AI systems become increasingly intelligent, new formal tools will be needed in order to avert default incentives for harmful behavior, as well as to inductively teach correct behavior.