Enable the user to provide feedback indicating their preferences during regular interaction with the AI system.

Empower the user to steer how the AI evolves over time by enabling them to provide explicit feedback on AI system outputs and behaviors. For example, for an AI-powered entertainment service that recommends content, give the user controls to guide the content towards what they are truly interested in (e.g., by encouraging them to rate each content recommendation).

Gathering user feedback can also help monitor and ensure that an AI system is working as intended.

Use Guideline 15 design patterns to encourage granular feedback:

• G15-A: Encourage explicit feedback on individual system outputs
• G15-B: Request explicit feedback on selected system outputs
• G15-C: Reporting inappropriate content
• G15-D: Use existing public rating or interaction data as feedback for the system