Ensure the AI system’s language and behaviors do not reinforce undesirable and unfair stereotypes and biases.
Enable the user to provide feedback indicating their preferences during regular interaction with the AI system.
Inform the user when the AI system adds or updates its capabilities.