Alignment Is a Training Problem. Deployment Is a Verification Problem.
The AI safety conversation has a center of gravity, and it sits inside the training run. Alignment, in the dominant framing, is something you do to a model: shape its objectives, tune its refusals,