New York Launches First State Program to Independently Test AI Systems

New York is launching the nation's first state program to independently test AI systems and train AI safety experts, Politico reported, ahead of a public announcement on Tuesday by Gov. Kathy Hochul.

The New York Center for AI Responsibility and Research at Binghamton University will run the work. It draws on more than 80 AI scientists and researchers from across the state, plus computing power from the Empire AI consortium.

"The stakes are too high to simply take AI companies at their word," Hochul said, according to Politico.

Who leads the center and how it is funded?

Jeannette Wing, Columbia University's executive vice president for research, and Greg Morrisett, vice provost of Cornell Tech, will lead the program. Wing said detecting and preventing harm from autonomous AI agents forms part of the challenge. "We need new techniques that can be applied to this new kind of system," she said.

Funding totals $55 million. The center, first announced by Hochul in January, rests on a $30 million gift from donors led by Tom Secunda, a Bloomberg LP co-founder who chairs Empire AI's board, and a $25 million research capital investment from SUNY.

Secunda described the mission in January: the center would "test systems, measure performance, and explain how advanced models behave."

What can the testers actually examine?

The program carries a hard limit. Researchers can examine only what AI companies are forced to disclose publicly. Models still in internal development stay private.

That constraint matters because the program starts before the RAISE Act takes effect on Jan. 1. The law requires AI companies operating in New York to submit detailed safety frameworks. The governor's office hopes the center's researchers will scrutinize those submissions.

Hochul framed the initiative as a response to federal inaction. "While the federal government fails to act, New York is leading the way on AI safety and holding big tech accountable," she said.

The testing program is not the state's only collision with AI infrastructure. New York lawmakers are also wrestling with the technology's physical footprint; MLQ.ai covered the state's data center pause on Nov. 2.

The center's reach will depend on the disclosure regime. Once the RAISE Act forces companies to file safety frameworks in January, the Binghamton researchers gain a concrete body of material to audit. Until then, their scrutiny stops at what companies choose to publish.