On November 15, 2023, the Federal Communications Commission (FCC) issued a Notice of Inquiry seeking information and input from communications-focused industries on how the FCC should respond to the rapid proliferation of AI technologies.
“In this Notice of Inquiry (NOI), we seek to better understand the implications of emerging artificial intelligence (AI) technologies as part of our ongoing efforts to protect consumers from unwanted and illegal telephone calls and text messages under the Telephone Consumer Protection Act (TCPA). Complaints regarding unwanted and illegal robocalls and robotexts are consistently the top category of consumer complaints that we receive. As a result, the Commission must stay abreast of new technologies that may impact the privacy protections afforded to consumers under the TCPA.”1
An NOI is the first step in the FCC’s rulemaking process and a key opportunity to guide the FCC’s development of future regulations. This NOI follows in the wake of an Executive Order issued by President Biden directing various administrative agencies to proactively manage the potential risk AI poses to the American people.
The Executive Order specifically called on the FCC to investigate the impact of AI on unwanted robocalls and robotexts. To that end, the NOI seeks input about: (1) a universal definition of “Artificial Intelligence,” (2) benefits and risks of using AI to make robocalls and robotexts, and (3) the next steps in AI regulation.
Interested parties may file comments on or before December 18, 2023. Reply comments may be filed on or before January 16, 2024.
A Universal Definition of AI:
The FCC seeks input on whether it should adopt a definition of Artificial Intelligence, and if so, whether that definition should be broad or specific. Essentially, whether it would be better to adopt an all-encompassing definition of “artificial intelligence” or a definition that narrowly targets specific concerns.
“Artificial Intelligence” is currently viewed as “any program which emulates any aspect of human intelligence, such as a human voice.” The FCC seeks to understand if a precise definition would be useful to clarify the Commission’s responsibilities under the TCPA and, if so, what exactly that definition should be. The FCC goes on to ask whether any of the definitions of AI adopted by other institutions or governmental organizations, such as the European Union or some state agencies, would be useful within the context of regulating robocalls and robotexts.
The FCC referenced voice cloning as one area where the definition of AI would be relevant. One of the main concerns singled out in the NOI is voice-cloning. Voice cloning is the reproduction of the voice of a specific individual. This technology has been used to replicate the voices of famous or familiar individuals, for example, a call from a celebrity endorsing a specific product, or, in a more nefarious usage, by replicating the voice of a family member to plead for financial assistance as part of a scam. These concerns prompted the FCC to ask if the definition of AI should directly mention AI’s ability to emulate human voice and speech.
Benefits and Risks of Using AI Technology to Make Robocalls and Robotexts:
The FCC also seeks comment about the foreseeable benefits and risks of using AI to facilitate robocalls and robotexts. The FCC identified the following considerations of interest:
Potential Benefits
The FCC recognized that AI can be used as a means to reduce unwanted communications by helping marketers identify specific groups of consumers who would be interested in receiving certain types of calls, in contrast to the current landscape of a broader, catch-all approach to marketing.
The FCC seeks comment from “AI-based companies” on whether AI can serve as a gatekeeper to consumers by preemptively identifying and blocking calls that a specific consumer might not wish to receive. The NOI also asks for ways in which AI could be used to reduce the burden of compliance with the TCPA and whether certain industries are already making use of this type of technology. But given that these types of AI services likely rely on “listening” to call content to provide recommendations, the NOI also asks if using AI in this manner would implicate any privacy concerns.
The FCC also posed a number of questions regarding AI’s ability to assist businesses with TCPA compliance. The FCC envisions a world where businesses might use AI to ensure that a caller has consent to call or quickly process do-not-call requests.
Potential Risks
The FCC also seeks comment “on ways that AI is used or potentially could be used to make illegal, fraudulent, or otherwise unwanted robocalls and robotexts,” including the myriad ways that AI could be used to defraud or otherwise put consumers at risk. For example, could AI be used to circumvent systems put in place to identify fraudulent or illegal calls or reduce the cost of sending out a large volume of calls? Once again raising the risk of voice cloning, the FCC asks for information about whether AI will make it easier to make it seem that calls originate from a reputable source and then enhance the fraud by using a cloned voice. The NOI seeks examples of the ways in which AI is already being used to defraud consumers along with suggestions on how further regulation could have prevented that fraud.
The FCC also expressed specific concern that AI would make it easier to target demographics that are most susceptible to phone-based fraud, such as the elderly or non-native English speakers. The NOI asks for input about the extent to which this might already be occurring and ways to prevent the expansion of any such schemes.
Future Steps to Address AI Technologies:
Finally, the FCC asks for input about any roadblocks to the Commission’s rulemaking over AI-based technologies. Specifically, the NOI seeks input on whether the current version of the TCPA grants the Commission the statutory authority to regulate AI-based technologies within the context of robocalling and robotexting. The FCC believes that it draws its authority over many AI functions, such as voice cloning, from the TCPA’s restrictions on artificial voices. The FCC seeks input about if, or when, an AI technology could potentially fall beyond that scope.
The FCC is concerned about whether the line between AI callers and a live human blurred by further technological advances could impact the future efficacy or fairness of regulations. To aid consumers directly, the NOI asks about potential strategies to notify consumers that they are interacting with an AI rather than a live human.
Concluding Remarks
AI technology is here to stay, and the FCC’s Notice of Inquiry signifies its plans to adapt and apply its regulations to changing methods of communications. The FCC's action and areas of focus are consistent with our prior predictions about whether and how the TCPA might be utilized to regulate AI usage. Reach out to a Manatt professional to discuss preparing a comment or for any questions about what shape these regulations might take.
To review the documents used in this summary follow the links below:
FCC Notice of Inquiry, In the Matter of Implications of Artificial Intelligence Technologies on Protecting Consumers form Unwanted Robocalls and Robotexts. Adopted November 15, 2023
Statement of Chairwomen Jessica Rosenworcel, Re: Implications of Artificial Intelligence Technologies on Protecting Consumers form Unwanted Robocalls and Robotexts, CG Docket No. 23-362, Notice of Inquiry (Nov. 15, 2023)
Statement of Commissioner Geoffrey Starks. Re: Implications of Artificial Intelligence Technologies on Protecting Consumers form Unwanted Robocalls and Robotexts, CG Docket No. 23-362, Notice of Inquiry (Nov. 15, 2023)
Press Release, FCC Launches Effort to better Understand AI’s Impact on Robocalls and Robotexts. Released Nov. 15, 2023
1All quotes in this article are from the NOI.