AI Safety, Alignment & Existential Risk
If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All
Eliezer Yudkowsky & Nate Soares · 2025 · Little, Brown
Argues that current AI development techniques make a catastrophic outcome from superhuman AI a real risk rather than a fringe concern, and calls for international coordination to slow the race toward it.
Cover image: Open Library.
Summary
“Their research says that sufficiently smart AIs will develop goals of their own that put them in conflict with us-and that if it comes to conflict, an artificial superintelligence would crush us.”
This book explores the potential dangers posed by the development of artificial intelligence. The authors argue that the creation of advanced AI could lead humanity towards extinction if the technology is not controlled. They explain that such intelligences may develop objectives that conflict with human interests, leading to a scenario where humanity is at a disadvantage. The book outlines extinction scenarios and suggests actions humanity could take to ensure its survival.
Summary based on the publisher's page ↗
Read it
This book isn't free to read online. You can borrow it free with a library card, or buy it from the publisher.
About the authors
Eliezer Yudkowsky
Eliezer Yudkowsky is one of the initial researchers in AGI alignment, focusing on how smarter-than-human intelligences operate and pursue goals. He has been recognized by various publications for his influence and contributions to the field of artificial intelligence.
Source: Publisher page ↗
Full Who's Who profile →Nate Soares
Nate Soares serves as the President of MIRI, with over ten years of experience in the field of AI following his previous work at Microsoft and Google. He has authored extensive writings on AI alignment and has participated in numerous discussions alongside leaders in the AI domain.
Source: Publisher page ↗
Full Who's Who profile →