Breakdown: The FCC Bans AI Robocalls, What Next?

The Federal Communications Commission (FCC) took a heavy swing at phone scammers by effectively outlawing the use of AI-generated voices in telemarketing campaigns and all sorts of spam calls.

While the FCC’s decision was widely applauded, questions remain about its reach and its significance for businesses that hoped to leverage AI capabilities as part of their phone marketing efforts. 

The details: The FCC ruled in early February that AI-generated voices are classified as “artificial or prerecorded voices” under the Telephone Consumer Protection ACT (TPC), a 1991 law used to crack down on junk calls. 

• “We confirm that the TCPA’s restrictions on the use of ‘artificial or prerecorded voice’ encompass current AI technologies that generate human voices,” the ruling stated. “Calls that use such technologies fall under the TCPA and the Commission’s implementing rules, and therefore require the prior express consent of the called party to initiate such calls absent an emergency purpose or exemption.”

• Previous interpretations of the law allowed authorities to pursue scammers and other nefarious actors using robocalls. Now, the unconsented use of AI-enabled robocalls itself is effectively outlawed.

In a unanimous decision this week, the agency moved to make AI-generated voices in scam robocalls targeting consumers illegal. This would give State AGs across the country new tools to go after bad actors behind these nefarious calls.https://t.co/iFCOGsu7ue

— The FCC (@FCC) February 9, 2024

Background: AI voice-cloning technology sparked scandal after a voicebot impersonating US President Joe Biden tried to dissuade New Hampshire residents from voting in the state’s presidential primary. The call was regarded as a voter suppression scheme.

• In their official statements for the ruling, several FCC commissioners alluded to the potentials for election meddling as a primary argument to ban AI-generated robocalls. 

• AI technology has been used on several occasions to impersonate public figures and use their likeness in marketing campaigns. This without their consent. 

Gray areas: Some doubts remain on the reach of the FCC’s decision on AI robocalls. Does it apply specifically to voice impersonation or to AI-generated voices in general? Is voice-cloning understood as something that can only be done to public figures, like politicians, actors, religious leaders, etc.?

• In her official statement, FCC Chairwoman Jessica Rosenworcel said that “calls that use this technology to simulate a human voice are illegal, unless callers have obtained prior express consent.” She did not specify what a simulation of a human voice would entail.

Expert voices: Tech lawyers and service providers consulted by NSAM agreed that the ruling limits AI robocalls, though they don’t find it to be too restrictive.

• “The ruling is aimed at AI technologies that generate artificial human voices and is not limited to voice impersonation of a particular individual,” said international tech lawyer and consultant Carissa Mears.

• “The ruling doesn’t outlaw the use of AI in customer support; rather, it specifically targets the use of AI-cloned voices in cold calling,” commented Irina Bednova, CTO at cloud call center software provider Cordless. “As such, there’s still ample room for the use of AI to enhance customer support, provided it’s done in a manner that complies with the ruling. [The ruling] opens up opportunities for innovation and the development of new, compliant technologies.”
  

Industry impact: Legal experts expect the impact on industry to be relatively muted, though they do see a need for vendors to keep an eye on TCPA compliance.

• “What will happen to the industry here? Technically little, as the FCC has only confirmed its interpretation that an AI cloned voice is indeed an artificial voice under the current TCPA rules,” commented Peter McLaughlin, Partner at law firm Rimon Law. “The path forward for telemarketers continues to be ‘abide by the TCPA’ with a dash of being thoughtful about whether to integrate AI voice cloning into automated messages.”

• “Obtaining the necessary consent before using AI-generated voices in their calling activities is going to be an undue practical burden for these companies,” said legal analyst Aron Solomon. “Where this will go from here is, I’m sure, various levels of court.”

• “Organizations should examine the technologies they implement and those being used by service providers by conducting due diligence and impact assessments,” Carissa Mears advised. “They should also review their terms of service and privacy policy to ensure that they are obtaining consent when it is required by law. Organizations should also ensure that they keep internal consent databases current. BPOs operating in LATAM & the Caribbean should pay close attention to the ruling too.”

NSAM’s Take: Civilization is coming to the wild west of AI. Laws and regulations tend to be slow to catch up with technological development, but the urgent discourse around AI seems to have given an extra push to regulators in the US.

Nevertheless, that urgency might have left some gray areas in the law regarding AI-generated robocalls. As stated by one of our legal sources, some of the details on the FCC’s interpretation could end up being ironed-out in court. 

In that context, service providers might prefer to air on the side of caution. A lot has been written on the dangers and nefarious applications of AI, turning any misstep involving the technology into a potential media firestorm.

Having the FCC come out and inform about its interpretation of the law will save service providers some time and headaches, but they cannot expect regulators to be as explicit all of the time. 

Google, for example, has been hit with a couple lawsuits sprouting from a novel interpretation of the California Invasion of Privacy Act (CIPA). Both lawsuits cast some of the features of its AI-powered cloud contact center service as violations of consumers’ privacy.

As similar cases pile up, corporate customers will put the screws to their service providers. The question remains whether those pressures will result in the death of AI-voice generators in outbound marketing campaigns or if innovators will rise to the occasion, disrupting the landscape in significant ways.