US communications body bans automated AI-generated robocalls
The U.S. Federal Communications Commission ruled on Thursday that robocalls with AI-generated voice clones are breaking the law under the Telephone Consumer Protection Act of 1991
The U.S. Federal Communications Commission unanimously declared on Thursday that automated calls featuring voice replicas generated using artificial intelligence are in violation of the Telephone Consumer Protection Act of 1991.
The telecommunications legislation, enacted over three decades ago, now encompasses some of the most cutting-edge artificial intelligence systems in use today. The ruling, effective immediately as of Feb. 8, represents the FCC’s most forceful action to date in its ongoing endeavors to suppress AI-facilitated fraudulent activities and dissemination of false information in anticipation of the 2024 election period.
“This may seem like a concept from a distant future, but it is a reality that is already among us,” remarked FCC Chairwoman Jessica Rosenworcel in a statement released alongside the declaratory ruling.
“This technology has the capacity to deceive us through auditory, visual, and interactive means, leading us to believe that various fabricated claims are genuine.”
The FCC’s comprehensive prohibition comes shortly after reports of a voter suppression campaign targeting thousands of New Hampshire residents prior to the state’s presidential primary. The robocalls traced back to a Texas-based group, featured a replicated voice of President Joe Biden advising residents against voting in the Jan. 23 primary.
Scammers have already utilized AI technology for various purposes, ranging from generating deep fake celebrity videos to peddling counterfeit medical benefit cards and mimicking the voices of individuals’ loved ones for fictitious kidnapping schemes. In November, the FCC initiated a public Notice of Inquiry concerning the use of AI in fraudulent activities, as well as potential strategies for employing the same technology to combat malicious actors.
According to Rosenworcel, Thursday’s announcement aims “to take a step further.” Enacted in 1991, the Telephone Consumer Protection Act originally covered unwanted and “junk” calls containing artificial or prerecorded voice messages. Upon examination, the FCC unsurprisingly concluded that AI-generated voice replicas essentially represent more advanced versions of the same spam tactics, and therefore fall under the same prohibitions.
“We are all aware that unwanted robocalls are a blight on our society. However, I am particularly troubled by recent harmful and deceptive uses of voice cloning in robocalls,” remarked FCC Commissioner Geoffrey Starks in a supplementary statement. Starks went on to label generative AI as “a new threat” in the context of voter suppression efforts leading up to the U.S. election season, thus necessitating immediate action.
Source: Newsroom
