Where I agree and disagree with Eliezer — LessWrong

Paul writes a list of 19 important places where he agrees with Eliezer on AI existential risk and safety, and a list of 27 places where he disagrees.…

x Where I agree and disagree with Eliezer — LessWrong Best of LessWrong 2022 AI Risk AI World Modeling Curated 933 Where I agree and disagree with Eliezer by paulfchristiano 19th Jun 2022 AI Alignment Forum 22 min read 224 933 Ω 218 ( Partially in response to AGI Ruin: A list of Lethalities . Written in the same rambling style. Not exhaustive. ) Agreements Powerful AI systems have a good chance of deliberately and irreversibly disempowering humanity. This is a much more likely failure mode than humanity killing ourselves with destructive physical technologies. Catastrophically risky AI systems

saved by

• Alex Zhao
• Nibir Sankar
• Julian H
• Arjun Khandelwal
• Will Anderson
• Carolanne Jiang
• Lawrence Feng
• Anaya Mandal

related reading

• AGI Ruin: A List of Lethalities — LessWronglesswrong.com
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• Critical review of Christiano's disagreements with Yudkowsky — LessWronglesswrong.com
• What failure looks like — LessWronglesswrong.com
• The Case Against AI Control Research — LessWronglesswrong.com
• The Problem — LessWronglesswrong.com
• Consider chilling out in 2028 — LessWronglesswrong.com
• My AI Opinions - by Scott Alexander - Astral Codex Tensubstack.com
• I Would Have Solved Alignment, But I Was Worried That Would Advance Timelines — LessWronglesswrong.com
• The Alignment Problem — LessWronglesswrong.com
• A positive case for how we might succeed at prosaic AI alignment — AI Alignment Forumalignmentforum.org
• The Artificiality of Alignmentjoinreboot.org