FCC proposes that automated callers be required to disclose when they are using artificial intelligence

The Federal Communications Commission (FCC) has proposed new set of rules This week, automated callers were required to disclose when they are using artificial intelligence to make calls and send text messages.

The proposal builds on the FCC’s ban on AI-generated robocalls without the express consent of the person being called. The agency now hopes to require callers to say when seeking consent whether they plan to apply AI for future calls and messages, FCC writesSimilar disclosures would have to be added to any AI-generated phone calls, which the agency said “include an increased risk of fraud and other scams.”

Regulator I suggest defining an “AI-generated call” as any call that uses technology to create “artificial or pre-recorded voice or text using computational or other machine learning technology, including predictive algorithms and large language models, to process natural language and create voice or text content to communicate with the call recipient during an outgoing telephone call.”

Finally, the agency hopes to carve out an exception for people with speech and hearing disabilities who apply AI-generated voice software to lend a hand them communicate on outgoing calls. The FCC would also require that such calls not include “unsolicited advertising” and that people who receive phone calls are not charged for them. The agency asked for specific comments on whether scammers could abuse the exception and how it could update its rules to prevent that.