N.Y. mayor ‘speaking’ Mandarin and more

His voice is contorted into several languages for robocalls, adding to concerns about artificial intelligence.

ALBANY, N.Y. — New York City Mayor Eric Adams has been using artificial intelligence to make robocalls that contort his voice into languages he doesn’t speak — a practice that poses new ethical questions about the government’s use of the rapidly evolving technology.

The mayor told reporters about the robocalls on Monday and said they’d gone out in languages such as Mandarin and Yiddish to promote city hiring events. The calls haven’t included disclosures that he speaks only English or that the calls were generated using AI.

“People stop me on the street all the time and say, ‘I didn’t know you speak Mandarin’ .... The robocalls [use] different languages to speak directly to the diversity of New Yorkers,” the Democratic mayor said.

The calls come as regulators struggle with how to ethically and legally navigate the use of artificial intelligence when deepfake videos or audio can make it seem that anyone anywhere is doing anything a person with the required technical skills wants.

The New York watchdog group Surveillance Technology Oversight Project has slammed Adams’ robocalls as an unethical use of AI that is misleading city residents.

“The mayor is making deepfakes of himself,” said Albert Fox Cahn, executive director of the organization. “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 outright Orwellian. Yes, we need announcements in all of New Yorkers’ native languages, but the deepfakes are just a creepy vanity project.”

The growing use of AI and deepfakes, especially in politics and for election misinformation, has prompted calls and moves by government and media companies for greater regulation.

Google was the first big tech company to say it would label deceptive AI-generated ads that could fake a candidate’s voice or actions for election misinformation. Instagram and Facebook parent Meta doesn’t have a rule specific to AI-generated ads but has a policy restricting “faked, manipulated or transformed” audio and imagery for misinformation.

A bipartisan U.S. Senate bill would ban “materially deceptive” deepfakes relating to federal candidates, with exceptions for parodies and satire. Two Democratic members of Congress have written to the heads of Meta and X (formerly Twitter) to express concerns about AIgenerated political ads on their social media platforms.

A number of tech companies, including the music and podcast streaming service Spotify, have shown off AI tools in recent weeks that can dub a person’s speech in another language, making it sound as though that person is speaking in that language. In Spotify’s case, the tool translates podcasts into multiple languages in the podcaster’s voice.

Adams defended himself against ethical questions about his use of AI, saying his office is trying to reach New Yorkers through the languages they speak.

“I’ve got to run the city, and I have to be able to speak to people in the languages that they understand, and I’m happy to do so,” he said. “And so, to all, all I can say is a‘ ni hao’ ” — which is Mandarin for “hello.”