The alignment definition problem: extensional views are retrospective only
zetalyrae argues that the common extensional definition of alignment—judging by observed behavior—only works retrospectively, and prospective judgments require a harder intensional definition based on intrinsic properties.
2026-09-11 ~ 2026-09-11 · 3 related posts
- zetalyrae: extensional definitions of alignment only work retrospectively — zetalyrae · 2026-09-11
- 'Aligned is as aligned does' only works retrospectively; we need an intensional definition — zetalyrae · 2026-09-11
- Alignment's intensional definition problem: you must define 'agent', 'goals' first — zetalyrae · 2026-09-11