New York Agencies Must Publish AI Assessments Under New Law

Agencies of the state of New York will have to disclose their use of artificial intelligence technologies under a new law signed by the state’s governor, Kathy Hochul.

Hochul signed the bill as part of a year-end round of addressing legislation voted on by state lawmakers earlier in the year.

The law, which was strongly backed by labour unions, limits how AI can be used to replace employed state workers.

It also bars the use of AI to make automated decisions such as whether someone receives unemployment benefits or child care assistance, unless the system is being consistently monitored by a human.

AI assessments

State agencies will be required to perform assessments of any software that uses algorithms, computational models or AI techniques and submit those reviews to the governor and top legislative leaders, as well as publishing them online.

Senator Kristen Gonzalez, one of the bill’s sponsors, said she has seen few comparable bills in the US, especially with regard to preventing job losses.

She told The New York Times the law was aimed at giving the state “clear processes and standards” when introducing new automated decision-making systems.

The New York State AFL-CIO union claims AI tools are already being used to make determinations on applications for public housing, public assistance and unemployment insurance.

Job protection

“The priority for the labour movement is to protect the jobs and the rights and privacy of workers,” said Mario Cilento, the organisation’s president.

Large tech companies lobbied against the law, saying it could slow down the state’s ability to test and use new technologies.

“My concern is that it takes us too far away from allowing government to experiment and create opportunities to provide better services,” said Julie Samuels, the president of Tech:NYC, an industry association that represents Google, Microsoft, Meta and other large tech firms.