Make it easy to invoke or request the AI system’s services when needed.
Make it easy to dismiss or ignore undesired AI system services.
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.
Maintain short-term memory and allow the user to make efficient references to that memory.