Human Compatible: Artificial Intelligence and the Problem of Control

What it is about

Stuart Russell argues that the standard goal of building machines that optimize fixed objectives becomes dangerous as systems grow more capable, because human preferences are complex, uncertain, and easily misspecified. He proposes assistance-oriented AI that remains uncertain about what humans want and learns through interaction rather than pursuing a rigid target. I need this book close whenever I start thinking the question has become simple.

Why I keep it

Russell on control. The problem stated by someone qualified.

Key ideas

Beside it on the shelf · AI & Technology