Human Compatible

Archival Summary & Scope

Distinguished AI researcher Stuart Russell challenges the common fear that superhuman artificial intelligence inevitably threatens civilization. In this groundbreaking book, Russell argues that catastrophic conflict can be avoided if we rethink AI from the ground up. He proposes designing machines to be inherently uncertain about human preferences, making them provably humble, altruistic, and committed solely to human objectives. By establishing this new foundation for AI, Russell shows how we can create machines that are provably deferential and beneficial. Hailed as "the most important book on AI this year" (The Guardian) and "the book we've all been waiting for" (Sam Harris), this vital work offers a new path for humanity's successful coexistence with increasingly intelligent machines.