Make it easy to edit, refine, or recover when the AI system is wrong.
Engage in disambiguation or gracefully degrade the AI system’s services when uncertain about a user’s goals.
Enable the user to access an explanation of why the AI system behaved as it did.