Inner Alignment: Explain like I'm 12 Edition - LessWrong
(This is an unofficial explanation of Inner Alignment based on the Miri paper Risks from Learned Optimization in Advanced Machine Learning Systems (which is almost identical to the LW sequence) and t…
x Inner Alignment: Explain like I'm 12 Edition — LessWrong Best of LessWrong 2020 Inner Alignment Mesa-Optimization AI Frontpage 189 Inner Alignment: Explain like I'm 12 Edition by Rafael Harth 1st Aug 2020 AI Alignment Forum 15 min read 47 189 Ω 59 (This is an unofficial explanation of Inner Alignment based on the Miri paper Risks from Learned Optimization in Advanced Machine Learning Systems (which is almost identical to the LW sequence ) and the Future of Life podcast with Evan Hubinger ( Miri / LW ). It's meant for anyone who found the sequence too long/challenging/technical to read.) Note
related reading
- Inner Alignment: Explain like I'm 12 Edition — LessWronglesswrong.com
- Alignment remains a hard, unsolved problem — LessWronglesswrong.com
- Risks from Learned Optimization: Introduction — LessWronglesswrong.com
- The Inner Alignment Problem — AI Alignment Forumalignmentforum.org
- How To Go From Interpretability To Alignment: Just Retarget The Search — LessWronglesswrong.com
- Narrow Misalignment is Hard, Emergent Misalignment is Easy — LessWronglesswrong.com
- Self-Other Overlap: A Neglected Approach to AI Alignment — LessWronglesswrong.com
- Model Organisms of Misalignment: The Case for a New Pillar of Alignment Research — AI Alignment Forumalignmentforum.org
- Outer vs inner misalignment: three framings — LessWronglesswrong.com
- (My understanding of) What Everyone in Technical Alignment is Doing and Why — LessWronglesswrong.com
- SLT for AI Safety — LessWronglesswrong.com
- Why AI alignment could be hard with modern deep learningcold-takes.com