In this first-of-its kind enforcement action, the FCC has proposed forfeitures not only against a party for allegedly instigating a deepfake, robocall campaign, but also against the voice provider for alleged STIR/SHAKEN violations that facilitated the misuse of generative AI technology. Specifically, the Commission has proposed, in separate Notices of Apparent Liability, penalties of $6 million against political consultant Steve Kramer and $2 million against voice service provider Lingo Telecom, LLC.

The proposed fines stem from an alleged illegal robocall and spoofing campaign in which prospective voters for the Democratic Presidential Primary Election in New Hampshire received robocalls carrying a deepfake generative AI audio recording of President Biden’s cloned voice telling them not to vote in the upcoming primary election. The caller ID information on the calls indicated that such calls came from the phone number of a New Hampshire Political Operative. Instead, the NAL alleges that it was Kramer who was responsible for the spoofed robocalls, made without the knowledge or consent of the individual identified in the caller ID.

According to the NAL, Kramer engaged an intermediary carrier, which used the services and equipment of another intermediary carrier to transmit the calls. The intermediary carrier transmitting the calls used voice service provider Lingo to originate the traffic onto the public switched telephone network. Lingo allegedly sent 2,000 of these spoofed calls through its network to potential New Hampshire voters and, according to the FCC, did not make any effort to verify the caller ID information or whether the intermediary carrier was authorized to use the NHPO number before applying an A-level STIR/SHAKEN attestations to the calls. The A-level attestation made it less likely that other providers would detect the calls as potentially spoofed.

All five FCC Commissioners approved the action against Kramer. The three Democrat Commissioners voted in favor of the action against Lingo, while the two Republican Commissioners, Carr and Simington, concurred with the decision. Carr noted that he wanted to ensure that the “FCC does not undertake ‘rulemaking through enforcement’ by creating new, substantive obligations that go beyond the standards” that are set forth in the FCC’s rules. Simington similarly argued that the Commission “must immediately act to establish clear standards within which the industry can operate.” Despite these concerns, as AI tools become more accessible and are increasingly used for harmful purposes, the Commission seems willing to continue its ramp-up of enforcement and rulemaking with regard to the emergency of AI tools, including deepfake voice cloning technology.

Just in February, the Commission unanimously adopted a declaratory ruling confirming that AI generated messages or voice calls fall within the TCPA’s restrictions and the FCC’s rules on artificial or pre-recorded voice calls. In the declaratory ruling, the Commission emphasized that voice cloning, and other uses of AI were being used in ways that can uniquely harm consumers, similar to the incident at hand. Additionally, on May 22, FCC Chairwoman Rosenworcel shared with her colleagues a proposal that, if adopted, would evaluate whether the Commission should require disclosure of AI-generated content in political ads on radio and TV.

While the Commission did not specifically determine that the illegal robocall campaign instigated by Kramer violated TCPA requirements, the calls apparently violated the Truth in Caller ID Act of 2009 and the FCC’s rules by maliciously spoofing the number of a prominent local political consultant. Regardless, voice providers should continue to be vigilant regarding their STIR/SHAKEN practices and their “Know Your Customer” protocols.