When an industrial camera or robot acts, how do I know the system was authorized to cause that physical effect?

Once AI crosses from detection into causing a physical action, model accuracy is not sufficient. The question is whether the system was authorized to cause that effect, then evidence of what occurred. Live industrial control is scoped during evaluation.

Deployment scope is determined during evaluation. Hood does not prove that AI will always be safe.

If this problem matches your environment, START with the systems, effects and authority boundary you need to control.