Insurance policies underwritten against the AIUC-1 audit results, giving enterprises deploying AI agents a financial backstop when an agent misbehaves.