FCC is weighing an exemption that would let political groups robocall cell phones with AI voices

Today the gate is consent. The Telephone Consumer Protection Act bars most automated calls to mobile numbers unless the subscriber agreed first ^[1]. In our view, that makes the consent record the first check a voice-AI dialer runs against any US mobile number.

The petition replaces that check. Club for Growth asks for an exception covering non-commercial political calls, according to AP as relayed by mezha.net ^[3]. The conditions described are two. The recipient must be able to opt out of further calls, and one organisation may not call the same person more than three times in 30 days ^[4]. For a builder, a permission lookup before the first call becomes an opt-out write that has to land during or after the call, plus a rolling 30-day counter per recipient.

The report does not define "organisation". If it means the committee, a vendor dialing for several clients keeps one counter per client. If it means the vendor, the cap binds the platform across all of them. The FCC gave AP no separate comment on the petition ^[7].

The ad data shows the pace of adoption. It does not size call volume. Wesleyan Media Project counted at least 164 AI-made or AI-enhanced political ads by early September, with nearly $80 million spent on placing them ^[8]. In May the same researchers counted 33 ads and about $20 million ^[9]. That is roughly five times the ads ^[16] and four times the spend ^[17].

The researchers drew on media mentions, AI disclaimers and their own review of political ads across broadcast and cable TV, streaming and digital advertising ^[10]. They say the true count may be higher, because AI use is not always visible ^[10]. AP found that 69% of the ads carried no AI notice ^[11]. For that growth to carry over to phone calls, campaigns would have to adopt synthetic voice on the phone as fast as they adopted it in ads. The count does not include calls at all.