Advanced
Open
Pro
Design the Privacy Controls for an AI Voice Agent That Records Calls and Can Clone a Voice
Your company will deploy an AI voice agent to handle customer calls in several countries. Product also wants to (a) keep recordings to improve the ASR, (b) let the agent speak in a branded voice, and (c) offer a future "your own voice" premium feature that clones the customer's voice for audio messages. Legal asked you for a technical privacy design (you are not giving legal advice).
- What data does the system create, and which of it is sensitive?
- Design disclosure, consent, redaction, retention and access controls.
- What additional controls does voice cloning need?
- How would you verify the controls actually work?
Share this question