wrapper-minds are the enemy — LessWrong

This post is a follow-up to "why assume AGIs will optimize for fixed goals?". I'll assume you've read that one first. …

x wrapper-minds are the enemy — LessWrong Goal-Directedness AI Frontpage 108 wrapper-minds are the enemy by nostalgebraist 17th Jun 2022 AI Alignment Forum 9 min read 43 108 Ω 28 This post is a follow-up to " why assume AGIs will optimize for fixed goals? ". I'll assume you've read that one first. I ended the earlier post by saying: [A]gents with the "wrapper structure" are inevitably hard to align, in ways that agents without it might not be. An AGI "like me" might be morally uncertain like I am, persuadable through dialogue like I am, etc. It's very important to know what kind of AIs would o

related reading

• why assume AGIs will optimize for fixed goals? — LessWronglesswrong.com
• Please don't throw your mind away — LessWronglesswrong.com
• Why Tool AIs Want to Be Agent AIs · Gwern.netgwern.net
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• The Case Against AI Control Research — LessWronglesswrong.com
• The case against AI alignment — LessWronglesswrong.com
• The Alignment Problem — LessWronglesswrong.com
• Deep Deceptiveness — LessWronglesswrong.com
• Beliefs are Chosen to Serve Goals — LessWronglesswrong.com
• Optimality is the tiger, and agents are its teeth — LessWronglesswrong.com
• What Is The Alignment Problem? — LessWronglesswrong.com
• The Problem — LessWronglesswrong.com