Mayor Zohran Mamdani’s administration is proposing a new rule cracking down on companies using automated chatbots for customer service, requiring businesses to disclose when they use such systems and to allow customers the option of talking with a person instead.

The new proposed rule is the first of its kind in the country, according to the Mamdani administration. It would also require such automated customer service systems to provide help in a “reasonable and timely manner,” and avoid repeatedly directing consumers to menus that don’t answer their questions or respond with irrelevant or incorrect information.

“Companies have hid behind technology to ignore customer complaints, stymie refunds, and avoid fixing building errors,” Samuel Levine, the commissioner of the Department of Consumer and Worker Protection, said in a virtual briefing with reporters on Tuesday morning. “Our worry in New York City is that AI is only going to make this relationship worse. So… we’re going to try to fix this. We’re drawing a line to save people time and money, and ensure people can get actual customer service.”

The proposal comes as public polling shows skepticism is growing about artificial intelligence and amid mounting calls to regulate the industry, including from within. Mamdani echoed those calls at an event in Manhattan on Monday evening, saying that the federal government isn’t doing enough to regulate AI. In September, his administration banned AI use in public elementary and middle schools.

There will be a public hearing about the proposed rule on Nov. 19. When the rule goes into effect will depend on the volume of comments the agency receives, according to Levine, and after it goes into effect, there will be a transition period to allow companies to implement the necessary changes.

The rule will apply to any companies doing business in New York City, Levine said, not just those based in the city. The rule applies to any customer service system in which all or “substantially all” the communication is made by technology, rather than a human being. The proposed policy relies on the city's Consumer Protection Law. Violators would be required to pay $525 for their first violation, $1,050 for their second violation, and $3,500 for their third and any subsequent violations.

Legislators across 37 states have introduced over 120 bills proposing regulations on chatbots, according to the technology governance nonprofit the Future of Privacy Forum . Several states have passed regulations on AI chatbots, largely requiring disclosure of their non-human status and addressing safety concerns, particularly for children; however, few if any appear to zero in on customer service chatbots, as New York City’s proposed regulation does.

Levine said he sees the proposed rule as a model for other cities and states.

“We think every jurisdiction should do this,” Levine said. “We want this idea to spread.”