In the wake of an AI-generated robocall that featured a deepfake of President Biden’s voice, the Federal Communications Commission unanimously adopted a Declaratory Ruling confirming that calls that use cloned voices or other AI technologies that simulate a human voice fall under the TCPA and generally require prior express consent of the called party to initiate such calls.

TAKEAWAYS

• The FCC issued a cease-and-desist letter to a company it suspected of originating fake, artificial intelligence (AI)-generated presidential calls to New Hampshire residents.
• The Do Not Disturb Act introduced in the House would require a disclosure for any robocall that uses AI to emulate a human.

The federal government in recent weeks has taken steps to protect Americans from robocalls and robotexts that use artificial intelligence (AI)-generated content. Bad actors are increasingly using AI to fool consumers into thinking that the caller or texter is a real person, and certain AI technologies have become sophisticated enough to hold a conversation with the caller. To this end, the Federal Communications Commission (FCC) issued a Notice of Inquiry in an effort to better understand how AI technologies are currently being used in robocalling and robotexting and how they might be used in the future. In the most recent high-profile example of using AI-generated content in a robocall, New Hampshire residents received a call purportedly from President Joe Biden telling them to stay home and not vote in the state’s primary election. The call was not authorized by President Biden or his campaign, nor did it include a legitimate message from the president but instead was a so-called deepfake using the president’s voice. The FCC acted swiftly in the wake of the New Hampshire incident by issuing a cease-and desist letter to the company that it suspected of originating the illegal calls and by adopting a Declaratory Ruling to clarify the applicability of certain of its rules to robocalls and robotexts that use AI-generated content. U.S. Rep. Frank Pallone, Jr. (D-NJ), ranking member of the U.S. House Committee on Energy and Commerce, which has jurisdiction over the FCC and telecommunications issues, introduced legislation that seeks to require a disclosure for any robocall that uses AI to emulate a human.

Notice of Inquiry and Declaratory Ruling
Recognizing that AI technologies could be used to deceive consumers, the FCC in November 2023 issued a Notice of Inquiry (NOI) to examine and understand the use of AI in robocalling and robotexting and how the FCC could protect consumers using its authority under the Telephone Consumer Protection Act (TCPA). The NOI also recognizes that harnessing the power of AI technologies may help protect consumers from harmful robocalls and robotexts. The NOI asks commenters to weigh in on, among other things, how the FCC should define “Artificial Intelligence,” what are the potential benefits and risks to the technologies, and whether AI technologies could aid disabled persons when communicating with other parties. The NOI also notes that the FCC believes “voice cloning” falls within the restrictions in the TCPA on using artificial or prerecorded voice messages. The comment period for the NOI closed on January 16, 2024. At this point, options for the FCC include issuing a Notice of Proposed Rulemaking seeking comment on specific rules related to AI and robocalling or taking no further action.

As a response to the NOI and the fake presidential calls in New Hampshire, the FCC on February 2, 2024, unanimously adopted a Declaratory Ruling confirming that the consumer protections of the TCPA apply to calls that use cloned voices or other AI technologies that simulate a human voice, as such calls would be captured by the definition of an “artificial or prerecorded voice.” Any such call would require prior express written consent from the called party, unless it fell under one of the TCPA exceptions. The Declaratory Ruling acknowledges that not all calls that use AI-generated voices are harmful or unwanted, but that the called party should control whether to receive those types of calls.

Cease-and-Desist Letter
About two weeks after New Hampshire residents received the fake calls, the FCC issued a cease-and-desist letter to Lingo Telecom (Lingo), the company it suspected of originating the traffic. The letter demanded that Lingo investigate the apparently illegal traffic identified by the FCC and block or cease accepting all of the apparently illegal traffic within 14 days of the letter if the company’s investigation determines that it was part of the call chain for the identified traffic or substantially similar traffic. The FCC also issued a Public Notice notifying all U.S.-based voice service providers to the suspected illegal traffic coming from Lingo and authorizing them, at their discretion, to block or cease accepting traffic from Lingo without liability under the Communications Act of 1934 if Lingo failed to effectively mitigate the suspected traffic. If Lingo fails to act, the FCC may impose stricter penalties, including requiring downstream voice service providers to block all traffic coming from Lingo.

H.R.7116: Do Not Disturb Act
Introduced on January 29, 2024 by Rep. Pallone, Jr. and four Democratic cosponsors, the Do Not Disturb Act, if enacted, would, among other things, update the current TCPA definition of “automatic telephone dialing system” to “robocall” to better reflect current usage, direct the FCC to study authentication and traceback of text messages, adopt new requirements specifically for VoIP service providers, and require any robocall that uses AI to emulate a human being to carry a disclosure at the beginning of the call. The legislation doubles the forfeiture or criminal fine for violations of the TCPA when AI is used to impersonate a person or entity with the intent to defraud, cause harm or wrongfully obtain something of value. The legislation has been referred to the House Committee on Energy and Commerce.

Taken together, these measures show that the federal government is engaged in the fight against illegal and unwanted robocalls and robotexts.

[View source.]