I think alignment work is more promising than control work — LessWrong

Summary * The primary ToC for control makes the case that control is compelling even if it does not scale to ASI. I think it is underdiscussed that…

x I think alignment work is more promising than control work — LessWrong AI Frontpage 98 I think alignment work is more promising than control work by Alec Harris 3rd Jul 2026 10 min read 15 98 Summary The primary ToC for control makes the case that control is compelling even if it does not scale to ASI. I think it is underdiscussed that this is also true for alignment (for all the same reasons). Even though control does not need to scale to ASI, the further control does scale, the better. This is also true of alignment, and alignment seems more likely to scale further. This is my mainline con

saved by

• Lydia Nottingham
• Kaustubh Kislay
• Eric R

related reading

• The Case Against AI Control Research — LessWronglesswrong.com
• The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
• Reading Listblog.redwoodresearch.org
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org
• Introduction to AI Control - by Sarah - BlueDot Impactblog.bluedot.org
• Introduction to AI Control - by Sarah - BlueDot Impactbluedot.org
• An overview of areas of control workblog.redwoodresearch.org
• Research Areas in AI Control (The Alignment Project by UK AISI) — AI Alignment Forumalignmentforum.org
• The case for ensuring that powerful AIs are controlledblog.redwoodresearch.org
• How useful is AI control? @ Trackstracks.xlabtracks.workers.dev
• Plans A, B, C, and D for misalignment risk — LessWronglesswrong.com