AI Regulation Tracker  /  Consumer protection and telecom

FCC Confirms AI-Generated Voice Calls Fall Under the TCPA and Need Prior Consent

In a Declaratory Ruling (FCC 24-17, CG Docket No. 23-362) adopted on February 2, 2024 and effective on release February 8, 2024, the Federal Communications Commission confirmed that calls using AI to generate or clone human voices are an artificial or prerecorded voice under the Telephone Consumer Protection Act. That means callers need prior express consent, must identify who is responsible for the call, and must offer opt-out on any call that advertises or telemarkets. This is an in-force federal interpretation, not a proposal. It has been settled law since 2024.

By Anthony Guerriero  ·  The Leveraged Years Regulatory Research  ·  Published July 20, 2026  ·  Last updated July 20, 2026

Here is the plain reading. The TCPA has long prohibited initiating a call to a residential line using an artificial or prerecorded voice without the prior express consent of the called party, unless an exemption applies, and it restricts artificial or prerecorded voice calls to wireless numbers as well. The open question, as AI voice tools got good and cheap, was whether a synthetic or cloned voice counts as an artificial or prerecorded voice. The Commission answered it directly. In the ruling, the FCC states that it confirms the TCPA restrictions on the use of "artificial or prerecorded voice" encompass current AI technologies that generate human voices. Because the statute does not define those terms, the Commission read them to cover the technology that most obviously fits the words.

What callers actually have to do

The ruling does not invent a new rulebook. It plugs AI voice calls into the robocall rules that already exist, and those rules have three moving parts.

First, consent. The Commission is explicit that callers that use such technologies must obtain the prior express consent of the called party to initiate such calls absent an emergency purpose or exemption. If the call includes or introduces an advertisement or constitutes telemarketing, the consent has to be prior express written consent under the Commission rules, which is a higher bar than a verbal yes.

Second, identification. Every artificial or prerecorded voice message has to state, at the start of the message, the identity of the business, individual, or other entity responsible for initiating the call. An AI voice does not get a pass on saying who is calling.

Third, opt-out. In every case where the message includes or introduces an advertisement or constitutes telemarketing, the caller has to offer specified opt-out methods so the person can ask you to stop calling that number. These duties apply, in the Commission words, to any AI technology that initiates any outbound telephone call using an artificial or prerecorded voice to consumers.

How this is distinct from the FTC impersonation rule

It is easy to blur this together with the Federal Trade Commission action on AI voice cloning, and they are not the same thing. The FTC impersonation rule is a different agency operating under a different statute, and it targets the fraud itself: it makes it unlawful to impersonate a business or government agency, and the FTC has moved to extend that to impersonation of individuals, which is what AI voice cloning enables. That is a deception-and-impersonation mechanism aimed at bad actors.

The FCC ruling here is not about fraud at all. It is a consent regime. It says that any caller using an AI voice, including a completely legitimate business making ordinary marketing or service calls, is placing an artificial or prerecorded voice call and therefore owes the full set of TCPA duties. You can be honest, well-intentioned, and clearly identified, and still be in violation if you dialed with an AI voice and did not get consent. A compliance program needs to account for both instruments, because they cover different failure modes.

Why a US professional should care

This one has direct, immediate US effect, which is exactly why it matters more than a lot of the AI-law headlines. The TCPA is a private-right-of-action statute. It is one of the most heavily litigated consumer laws in the country, with statutory damages that stack per call, and a plaintiffs bar that watches for exactly this kind of technology shift. A business that switched its outbound dialing to an AI voice to save money, without revisiting its consent posture, has quietly increased its exposure, not reduced it.

For attorneys, the advisory work is concrete. If you have clients running outbound calling, the questions are whether they are using or piloting AI or synthetic voices anywhere in the stack, whether their consent records actually support artificial or prerecorded voice calls and, for anything promotional, prior express written consent, whether their scripts open with a compliant identification, and whether telemarketing calls carry a working opt-out. Vendors matter too, because a business is generally on the hook for calls made on its behalf, so the AI calling platform in the contract is a liability question, not just a procurement one. The safe posture is simple to state: treat an AI or cloned voice on an outbound call as a robocall, because under FCC 24-17 that is what it is.

Browse the full AI Regulation News tracker

Informational analysis for working professionals, not legal advice. This briefing summarizes an in-force FCC Declaratory Ruling interpreting the TCPA. Confirm how it applies to your specific calling practices with qualified counsel.