New York City wants to force companies to tell consumers when they’re interacting with artificial intelligence and other computerized systems, part of a proposal to cut down on customer service “doom loops.”
The city’s Department of Consumer and Worker Protection unveiled a proposal Tuesday covering all automated systems — including rules-based chatbots and AI — that handle customer service calls. It would also require a one-touch option to connect customers with a human representative.
Automated customer service has long been a feature of modern life, often leading to consumer complaints about “Kafkaesque” loops of misunderstanding and failed attempts to get help, DCWP Commissioner Samuel A.A. Levine said on a Tuesday call with reporters.
“Our worry in New York City is that AI is only going to make this worse,” he said.
States such as California and New Jersey have laws requiring companies to inform customers when they’re dealing with an automated system. New York City’s proposal goes further and should be a model for other jurisdictions, Levine said.
New York City’s proposal comes as the Trump administration boosts the AI industry and federal lawmakers sit back, Levine said.
“We are not waiting for Washington to come to the rescue of working people,” he said.
Tuesday’s proposed rule wouldn’t cover companies subject to exclusive federal jurisdiction, such as airlines and national banks. But it would apply to other industries that do business with New York City’s more than 8 million residents.
The proposal seeks to hold companies responsible if an AI customer service bot provides deceptive or incorrect information. Companies that promise customer service but don’t deliver meaningful assistance would also be subject to penalties.
Companies would face claims under the “failed promise of customer service” provision if their automated systems direct consumers to menus that “fail to adequately address such consumer’s inquiries” or provide “unresponsive, irrelevant, or erroneous information,” according to the proposal.
“Whether you’re trying to resolve a problem with your bank, get a refund or return a package, you deserve access to real help,” Mayor Zohran Mamdani (D) said in a statement. “This rule makes clear: customer service cost-cutting can’t come at New Yorkers’ expense.”
The proposal, should it take effect as written, leaves a lot of open questions and could lead to significant costs for companies, said Steven Fulop, the president and CEO of Partnership for New York City, a nonprofit business advocacy group.
“My question would be, how would you decide if a chatbot was actually solving a problem versus not,” he said in a telephone interview.
Without a firm definition of what constitutes a failure to provide customer service, companies face an increased litigation risk that could exacerbate the affordability crisis, Fulop said.
‘Mamdani Time’
The proposal comes with a 30-day comment period and the DCWP is set to hold a hearing Nov. 19.
The city is looking to move on “Mamdani time” to finish the rule, Levine said, citing the approximately six months it took the DCWP to finalize its landmark click-to-cancel rule earlier this year.
“I have seen from many years of privacy enforcement that companies are perfectly capable of providing one set of protections in California and one set of protections in Oregon,” Levine, the former head of consumer protection efforts at the Federal Trade Commission, said.
Companies that violate the city’s rule would face potential restitution to harmed consumers and civil penalties starting at $525 per violation, according to the proposal.
While the DCWP plans to hire more investigators to police the chatbot proposal once it takes effect, customer complaints will be the key driver for agency actions, Levine said.
“New Yorkers are our eyes and ears,” he said. “The way the vast majority of our rulemakings, lawsuits start is with one consumer filing a complaint.”