Home | Substack

I think alignment is necessary, but not enough for AI to go well.

I think alignment is necessary, but not enough for AI to go well. By alignment, I mean us being able to get an AI to do what we want it to do, without it trying to do things basically nobody would want, such as amassing power to prevent its creators from turning it off.1 By AGI, I mean something that can do any economically valuable task that can be done through a computer about as well as a human can or better, such as scientific research. The necessity of alignment won’t be my focus here, so I will take it as a given. This notion of alignment is ‘‘value-free’’. It does not require…

related reading

• Dario Amodei — The Adolescence of Technologydarioamodei.com
• AGI Ruin: A List of Lethalities — LessWronglesswrong.com
• The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
• The Case Against AI Control Research — LessWronglesswrong.com
• The case against AI alignment — LessWronglesswrong.com
• Introduction to AI Control - by Sarah - BlueDot Impactblog.bluedot.org
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• I think alignment work is more promising than control work — LessWronglesswrong.com
• The Artificiality of Alignmentjoinreboot.org
• Dario Amodei — The Adolescence of Technologydarioamodei.com
• The case for ensuring that powerful AIs are controlledblog.redwoodresearch.org
• The Alignment Problem — LessWronglesswrong.com