Satirical Take: Letting a Homogeneous Few Set the Rules Counts as 'Alignment'
adamamcbride · x · 2026-09-16
A satirical retweet comments on Anthropic and METR's alignment work: it really is a solution to the alignment problem if a very small set of ideologically homogeneous people decide the rules for everyone and impose them unapologetically.
The author adds, darkly, that maybe a lot of people will die — perhaps even most — but when it's over you'll have an AI aligned with the remaining humans, so it's a kind of solution.
More from AGI Musings
- Nicholas Thompson: AI extinction talk is overblown, but alignment research still deserves priority — GaryMarcus · 2026-09-17
- The Substitution Problem: How AI Quietly Decides Which Thinking Skills Students Practice — DrKavner · 2026-09-17
- Andrew Ng: software engineering most affected by AI, yet still can't find enough engineers — dawnsongtweets · 2026-09-17
- After 35 years coding, retiree now lets AI write all his code, only tests and debugs — Sirtemed · 2026-09-17
- Dean Ball flags anti-safetyists' mood-affiliation arguments on AI risk — deanwball · 2026-09-17
- Writer loses $2m book deal after his African dialect was mistaken for AI text — theindependentonline · 2026-09-17