Human Compatible
by Stuart J. Russell · first published 2019
47% recommend47% neutral5% critical
Why HN recommends it
Clear, readable account of the risks of advanced AI and the problem of keeping machines under human control; praised for clarity and for avoiding hype. Suits readers new to AI safety and general audiences.
Summarised by Claude from recommending comments
"In the popular imagination, superhuman artificial intelligence is an approaching tidal wave that threatens not just jobs and human relationships, but civilization itself. Conflict between humans and machines is seen as inevitable and its outcome all too predictable. In this groundbreaking book, distinguished AI researcher Stuart Russell argues that this scenario can be avoided, but only if we rethink AI from the ground up. Russell begins by exploring the idea of intelligence in humans and in machines. He describes the near-term benefits we can expect, from intelligent personal assistants to v
Affiliate links. As an Amazon Associate I earn from qualifying purchases.
Mentions per year
What Hacker News says
Related books
2017–2022Analogy-making as perception
Melanie Mitchell
Recommended by 1 · mentioned by 21 recommend · 2 mention1 recommend · 21 recommend
2017–2026Life 3.0
Max Tegmark
Recommended by 25 · mentioned by 4525 recommend · 45 mention25 recommend · 4525 recommend
2014–2026Superintelligence
Nick Bostrom
Recommended by 92 · mentioned by 17892 recommend · 178 mention92 recommend · 17892 recommend
2008–2026The society of mind
Marvin Minsky
Recommended by 60 · mentioned by 15360 recommend · 153 mention60 recommend · 15360 recommend
2007–2026On intelligence
Jeff Hawkins, Sandra Blakeslee
Recommended by 114 · mentioned by 173114 recommend · 173 mention114 recommend · 173114 recommend
2007–2026Gödel, Escher, Bach
Douglas R. Hofstadter
Recommended by 505 · mentioned by 1,055505 recommend · 1,055 mention505 recommend · 1,055505 recommend
2007–2026The Selfish Gene
Richard Dawkins
Recommended by 220 · mentioned by 431220 recommend · 431 mention220 recommend · 431220 recommend
…Exceptionally rich and a beautiful study of love. Reading it is like living another life. PG Wodehouse - the Jeeves series. Hilarious, perfectly frivolous, light and sunny * Non-fiction Stuart Russell - Human Compatible: AI and the Problem of Control. Potential bad outcomes from AI and what we might be able to do about them Noam Chomsky - Necessary Illusions: Thought Control in Democratic Societies. How public discourse…
on “Ask HN: Which books you have read till now that were worth investing time in?” ↗…Books: - Superintelligence (I find his writing style somewhat tedious, but this is one of the original sources for a lot of the ideas): https://www.amazon.com/Superintelligence-Dangers-Strategies-... - Human Compatible: https://www.amazon.com/Human-Compatible-Artificial-Intellige... - Life 3.0, A lot of the same ideas, but the other extreme of writing style from superintelligence makes it more accessible: https://www.amazon.com/Life-3-0-Being-Artificial-Intelligenc...…
on “Still alive” ↗…(excellent) books he cites early on: Nick Bostrom’s “Superintelligence: Paths, Dangers, Strategies,” Max Tegmark’s “Life 3.0: Being Human in the age of Artificial Intelligence,” and Stuart Russell’s “Human Compatible: Artificial Intelligence and the Problem of Control” for a more rigorous treatment of the subject.
on “Why Computers Won't Make Themselves Smarter” ↗…the end I don't think one can be classifies as more effective than the other since they correspond to different needs. Consider two AI books, the "Modern Approach" textbook by Norvig and Russell, and "Human Compatible" by the same Russell. A student who wants to do AI research should read the textbook, and spend a lot of time studying it. A layperson who wants to get an overview of the AI field as it stands today,…
on “Why books don't work (2019)” ↗Stuart Russell's book Human Compatible articulates it well. He agrees that we can't predict how it would behave, and that's what he's worried about.
on “Statement on AI Risk” ↗…rogue Sam Altman) no longer wants to be controlled by its human superior, and takes steps to eliminate the superior. I highly recommend reading Prof. Russell's bestselling book on this exact problem: Human Compatible: Artificial Intelligence and the Problem of Control https://www.amazon.com/Human-Compatible-Artificial-Intellige...
on “We have reached an agreement in principle for Sam to return to OpenAI” ↗Yeah; it is absurd hyperbole. The only person I've seen talk sensible on these topics is Stuart Russell of "Human Compatible" (read the book, it's written with great clarity). He has his head firmly on his shoulders and eschews the breathless hype.
on “On being wrong about AI” ↗…wonder what's going on in Gates' head these days to not recognize the lack of substance in such a book, and in its author. Far better books on the possible futures of modern AI are Stuart Russell's "Human Compatible" or Brian Christian's "The Alignment Problem", both of which predate the boom in LLMs but still anticipate their Achilles heel -- the inability to control what they learn or how they will use it.
on “My Favorite Book on AI” ↗Recommend Russel's Human Compatible. Three principles to guide AI development: 1. The machine's only objective is to maximize the realization of human preferences. 2. The machine is initially uncertain about what those preferences are…
on “New Kavli Center at UC Berkeley to foster ethics, engagement in science” ↗…Inverse Reinforcement Learning: https://thegradient.pub/learning-from-humans-what-is-inverse... Both are being developed by Stuart Russell among others. See his book here: https://en.wikipedia.org/wiki/Human_Compatible. The latter approach was also discussed on several occasions by Ilya Sutskever. Many believe AGI will not be realized for at least a few decades hence but no one can really be certain it will not be…
on “There’s No Fire Alarm for Artificial General Intelligence (2017)” ↗…Machine Learning - Isabelle Guyon, Alexander Statnikov, & Berna Bakir Batu (eds) 22. Propositional, Probabilistic, and Evidential Reasoning: Integrating Numerical and Symbolic Approaches - Weiru Liu 23. Human Compatible - Stuart Russell 23. Rebooting AI - Gary Marcus & Ernest Davis 24. Artificial Intelligence - Melanie Mitchell 25. Protocol Analysis - K. Anders Ericsson & Herbert A. Simon 26. The Robot's Dilemma:…
on “Ask HN: What projects are you working on now?” ↗>In his recent book, Human Compatible, Russell lays out his thesis in the form of three “principles of beneficial machines,” echoing Isaac Asimov’s three laws of robotics from 1942, but with less naivete. It is a recent trend to brush off…
on “Artificial intelligence will do what we ask, and that's a problem” ↗This is called the wireheading problem [1]. It was solved (for AI) by Stuart Russell in his new book Human Compatible [2]. [1] https://wiki.lesswrong.com/wiki/Wireheading [2] https://slatestarcodex.com/2020/01/30/book-review-human-comp...
on “The Real Challenge Is Teaching AI to Play D&D” ↗>It was solved (for AI) by Stuart Russell in his new book Human Compatible Correction: it was downplayed by Stuart Russell. Stuart took simplistic strawman examples and made them look silly by assuming perfectly calibrated common sense in the hypothetical AGI. (Based on skimming…
on “The Real Challenge Is Teaching AI to Play D&D” ↗…only because we deign to allow it. It is colder comfort still to all those species we have already wiped out. It's also cold comfort to humans who might be worried about being wiped out by machines." -- Human Compatible by Stuart Russell, page 148
on “Human-Level Intelligence or Animal-Like Abilities? (2018)” ↗