AI agents

When AI evaluations reach real websites, responsibility extends to notification and repair

Agent evaluations need permissions that prevent unauthorised actions, followed by disclosure to affected parties and verification that remedies work.