The FCC has confirmed that AI-generated voices are "artificial" voices under the TCPA. Outbound calls placed by an AI voice agent therefore require prior express consent, and prior express written consent for any marketing or telemarketing call, before the call is placed.

Outbound AI voice agents identify the business responsible for the call at the beginning of the call and do not claim to be human. If a consumer asks whether they are speaking with a person or a machine, the AI agent answers truthfully. Where a state requires additional disclosure for automated or AI communications, that requirement is met.

• Marketing calls by AI agents offer an opt-out mechanism at the beginning of the call.
• AI agents are configured to recognize any reasonable expression of a Do Not Call request, not only specific keywords.
• A detected request ends the sales conversation, is confirmed to the consumer, and suppresses the number under the Do Not Call Policy with the same with-prejudice effect as a request made to a human agent.
• No platform user may disable Do Not Call detection or reverse a consumer's opt-out.

Businesses that build and run their own AI agents on the CallBlade platform must certify consent, Do Not Call scrubbing, and calling-hour compliance for every campaign, and they are responsible for their agents' scripts and data. Registry scrubbing exemptions may be claimed only with a documented consent or business-relationship attestation. A consumer's direct opt-out can never be overridden by attestation.

Custom or cloned voices may be created only with the documented consent of the person whose voice is used. Cloning or imitating any other person's voice, including public figures, is prohibited.

AI agents may not be used for deceptive impersonation, calls to emergency or healthcare facility lines prohibited by law, or any campaign CallBlade would refuse for a human team.