New York’s mayor, Eric Adams, has confessed to employing audio deepfakes to communicate with residents in languages unfamiliar to him. Through this innovative use of technology, Adams has been audible in Spanish and Mandarin, aiming to broaden his reach.
In a recent English press briefing, the Democratic mayor explained that these “robocalls” were mainly purposed to spread the word about local events, such as recruitment fairs and concerts. Adams, who identifies as a “techie,” stated that the feedback from locals had largely been affirmative, with many excitedly remarking on hearing their mayor converse in their native language. Expanding on this, Adams remarked, “We are becoming more welcoming by using technology to speak a multitude of languages.”
Addressing potential skepticism over the calls’ authenticity, he jovially concluded the conference with a Mandarin greeting, “ni hao.”
According to The City, a local news outlet, these multilingual calls have reached over four million individuals, inclusive of thousands in Spanish, and hundreds in Mandarin, Cantonese, and Yiddish. The deepfake voice technology used is attributed to ElevenLabs, a firm which previously faced controversy when its software generated celebrity voices for inappropriate content. Notably, Spotify is exploring similar tools for translating podcasts into multiple languages.
Post-conference, criticism ensued, with campaigners labeling these calls as “unethical” due to the absence of any disclosure about the artificial voice generation. Albert Fox Cahn of the Surveillance Technology Oversight Project labeled it as a “creepy vanity project,” adding, “This is deeply unethical, especially on the taxpayer’s dime. Using AI to convince New Yorkers that he speaks languages that he doesn’t is deeply Orwellian.”
