A plain-English overview of how the TCPA, the FCC’s ruling on AI voices, Do Not Call rules and state recording laws affect AI calling agents in the United States.
This article is general information, not legal advice. Rules change and depend on your situation, so have your own counsel review your calling programme.
Answering calls from your own customers is far less regulated than placing calls. Most compliance questions arise when an AI agent calls people, especially for marketing or sales.
The Telephone Consumer Protection Act restricts calls made with an “artificial or prerecorded voice”. In 2024 the FCC confirmed that voices generated by AI technology count as artificial voices under the TCPA. In practice this means outbound AI-voice calls generally need the recipient’s prior express consent, and marketing calls need prior express written consent.
Telemarketing calls must respect the National Do Not Call Registry, state registries and your own internal opt-out list, and must stay within 8 a.m. to 9 p.m. in the recipient’s local time. Callers must identify themselves and honour opt-out requests.
Federal law allows recording with one party’s consent, but several states, including California, Florida, Illinois, Pennsylvania and Washington, generally require every party to consent. Because calls often cross state lines, the safe practice is to disclose recording at the start of every call.
Disclose that the caller is speaking with an automated assistant, store proof of consent, scrub against do-not-call lists, respect calling hours per time zone, offer an easy opt-out and make a human handover available. ZahrionTech builds these controls into the agent, and you should confirm the legal basis for any outbound programme with your own adviser.