Disagreements Remain on Need for AI-Specific Robocall Rules
Commenters disagreed on whether the FCC should require additional disclosures relative to AI calls, in reply comments to an NPRM that commissioners approved 5-0 in August. Consumer and public interest groups urged a smart approach, targeting calls that will most likely confuse consumers. Industry commenters said no new rules are needed now.
Sign up for a free preview to unlock the rest of this article
Communications Daily is required reading for senior executives at top telecom corporations, law firms, lobbying organizations, associations and government agencies (including the FCC). Join them today!
While FCC Commissioner Brendan Carr, President-elect Donald Trump's for chair (see 2411180059), voted for the NPRM, he warned about overregulation (see 2408070037). “I don’t believe in no regulation of AI, and at the same time … there’s a risk of overdoing it early on,” he said at the time.
“Commenters agree that the Commission should not (i) adopt a specific definition of ‘AI-generated’ call or (ii) impose pre- and on-call disclosure and consent requirements to use AI,” CTIA said in comments posted Monday in docket 23-362. Proposals in the NPRM “risk confusing consumers, callers, and providers alike and chilling the use of innovative solutions,” CTIA said. That stance was consistent with initial comments of the U.S. Chamber of Commerce and others last month (see 2410110039).
Consumer and public interest groups urged a requirement of prior express consent for some calls that use AI. “The distinguishing characteristics of calls that should trigger the specific consent requirement and the in-call disclosure relate to the extent to which the calls are likely to deceive or misrepresent to recipients the origin or the party responsible for the call, or to confuse the recipient regarding whether they are speaking to a real human being or not,” the groups said.
Calls most likely to deceive or confuse recipients are those that are “either interactive, or in which AI is used to clone a human’s voice to misrepresent the true caller,” the groups said: “AI-assisted phone scams have moved from being annoying to being terrifying.” The National Consumer Law Center (NCLC), the Consumer Federation of America, the Electronic Privacy Information Center and Public Knowledge signed the filing.
The FCC should rely on “the plain language” of the Telephone Consumer Protection Act “rather than adopt a definition of AI-generated calls,” said USTelecom. “The record confirms the need for the Commission to take a cautious and targeted approach.” USTelecom said it agrees with NCLC and others that “calls should be regulated based on the degree to which the AI may create a risk of harm or confusion for the recipient.”
NCTA questioned the need for new rules. “There is nothing inherently harmful about using AI to make calls, including using AI-generated voices,” NCTA said: "There are a wide variety of uses of AI in the calling context, ranging from extremely beneficial to very harmful.” Should the FCC proceed, “avoid discouraging beneficial uses of technology and narrowly tailor any rules only to those circumstances that present enhanced risk of harm,” cablers urged.
The Edison Electric Institute warned that the FCC's definition of AI-generated call is “overly broad and could lead to confusion and unintended consequences.” Any rules “should target specific areas where bad actors might misuse AI to evade existing regulations,” EEI said. It called for protections for calls and texts from utilities to their customers.
The American Bankers Association raised concerns on behalf of its members. The proposed rules “could be interpreted as imposing new consent and disclosure requirements on a broad swath of consumer-benefitting calls placed by banks, including suspicious activity alerts, one-time passcode requests, and callbacks placed by the bank at the customer’s request,” ABA said. If the FCC adopts the “expansive,” proposed definition of AI-generated call, “consumers may also be confused and decline to consent to receive these important ... calls,” bankers said.
NTCA said, “Like communications providers of all kinds, NTCA members routinely utilize robocalls and/or robotexts to contact customers” after obtaining consent if required “for purposes such as reminding them of an upcoming appointment with a technician or notifying them of network outages.” Requiring that carriers obtain new AI-specific consent “would not only be overly burdensome … it could cause consumer confusion.”
The Chamber said it continues to view rules as premature. “The impact of these new rules, if adopted on the communications industry and main street businesses, will be substantial, and the Commission should heed the concerns raised in the record.”