Policy Is Not Capability: Some Safety Benchmarks Measure Which Definition of "Unsafe" You Trained On.
arXiv:2610.25060v1 Announce Type: cross Abstract: A guardrail classifier is reported as if unsafe were a property of text. It is not: it is a property of a policy, and the policy enters the model through its labels. We exploit an accident of an industrial depl
arXiv cs.AIResearch source·