A leading AI researcher argues that machines designed to pursue fixed objectives are dangerous, and proposes instead systems that are uncertain about human preferences and learn them by observing people.
This is an external publication. AI Governance LK did not write it; the summary and notes on this page are our own.
Why it matters
One of the best book-length introductions for readers from any background, written by the co-author of the standard AI textbook.
What you will learn
- Explain the “standard model” of AI and its problems
- Describe assistance games as an alternative
Before you start
Suggested background. This is editorial guidance, not a requirement.
- None
How to cite
Russell, S. (2019). Human Compatible: Artificial Intelligence and the Problem of Control. Viking.