BY HARPER FREEMAN
When Robert Croghan first became a New York City public employee over 55 years ago, workers used Dictaphones to record conversations and crafted reports and memos on typewriters inside stuffy rooms without air conditioning. At the time, Croghan said in an interview last week, “computers, of course, were unheard of,” and the pace and tenor of technological change he witnessed over five decades was considerable but, he feels, ultimately for the better.
Croghan, president of the Organization of Staff Analysts since 1985, is now staring down what may be the biggest and most impactful technological change of his long career, one that he said brings to mind the science fiction books he’s adored since his youth. Artificial Intelligence is already being deployed in many agencies across the city’s government, leading to concerns over job security, the possibility of mistakes like a SNAP eligibility review that affected 30,000 New Yorkers and other, novel mishaps that have raised the alarm of union leaders.
“People are more frightened about AI than any technological advancement in a while,” Croghan said. “And I can see why.”
‘Threatening the livelihood of thousands of members’
Odetty Tineo, the political director of District Council 37, the largest union of municipal workers, testified at a State Senate committee hearing in January that AI introduced by management across the city at a rapid clip has already undermined city workers.
“Managers in city agencies such as [Health and Hospitals] and the Department of Social Services are aggressively introducing AI in work that deals directly with the health and welfare of the public sector and New Yorkers,” she testified to the Senate Standing Committee on Internet and Technology. The results of that implementation, Tineo added, have been mixed.
AI has been used to surveil workers in the Department of Parks and Recreation driving vehicles, she said, and AI tools in the Administration of Children’s Services were being trained on data over a decade old. The most egregious error caused by AI tools to date, Tineo said, came when Maximus, a contractor utilized by the state “messed up” the approval process for SNAP benefits that affected more than 30,000 individual New Yorkers just before Thanksgiving last year.
Eligibility specialists and other city workers were pushed to work overtime to correct those mistakes, and Tineo fears that further reliance on Maximus, which proudly describes its use of AI, and other AI-friendly private contractors, could lead to job loss in the civil service.
“A lot of our members hold titles that are highly vulnerable to automation, such as data entry and customer service,” Tineo testified at the hearing. “The city’s drive for efficiency will be targeting these roles, threatening the livelihood of thousands of members.”
A spokesperson for Maximus did not respond to a request for comment.
Local 1549 at the forefront
All of the examples cited by Tineo affected members of Local 1549, the NYC Clerical-Administrative Employees union, whose leadership sat behind Tineo as she gave her testimony in January. The union of around 14,000 city workers is one of the largest DC 37 affiliates and represents civil servants across numerous city agencies who are at the forefront of AI use.
“There’s no one actually holding AI accountable for when it makes mistakes, like when it made that mistake causing over 30,000 cases to be delayed,” Elson Primus, an executive board member of Local 1549, told The Chief last week. Each one of those individuals whose benefits were delayed could have affected the food and health benefits of an entire family, he pointed out.
Management is deploying AI faster than the union can respond to it in many cases, Primus said, but he argued that its crucial protections are put in place to secure workers’ jobs and reduce surveillance.
DC 37 is pushing for lawmakers to pass bills that would guarantee public-sector unions the right to bargain with the city over AI use, limit AI surveillance of workers, mandate human oversight of AI tools and protect against displacement from AI.
The union has been tight-lipped about what exact AI protections it’s hoping to secure in bargaining with the city – DC 37 Executive Director Henry Garrido didn’t respond to a request for comment – and bargaining for the upcoming contract cycle has yet to begin.
Primus feels that without limits or regulations on its use, the displacement by AI will continue unabated.
“Where does it end?” he asked. “It starts with New York City workers today, but where does it end?”
Croghan, the OSA president, said his union has yet to field bargaining topics from members, but said that when union leadership does, he expects AI to be a concern that some workers have, at which point the union will develop a bargaining strategy for it.
“This is a new situation which has the potential to be extremely wonderful and helpful but also has the potential to be dangerous and destabilizing, and also sometimes downright stupid,” the longtime union leader said. “But I don’t see how you can stop the onrush of science.”