New York Moves Forward on Landmark AI Safety Law as Federal Action Stalls

New York Gov. Kathy Hochul is positioning her state as a stark contrast to the federal government's approach to artificial intelligence regulation, moving forward with implementation of the state's landmark AI safety law even as Washington remains largely gridlocked on the issue. On Monday, Hochul is expected to announce next steps for implementing the RAISE Act, New York's AI safety legislation set to take effect in January, along with the first major hire for the state office responsible for overseeing the law, according to details shared first with POLITICO. A State Response to Federal Inaction The announcement comes amid growing concern over the potential risks of rapidly advancing AI technology. While some lawmakers in Washington have pushed for restrictions to slow AI development and strengthen safeguards, President Donald Trump has resisted those efforts, dismissing concerns about AI risk as a hoax. Over the weekend, Trump suggested the federal government could play some role in monitoring AI for potential dangers, though he offered few specifics. Hochul said last week she would continue looking for additional actions New York could take at the state level to lead in the absence of federal leadership, though she didn't detail what further steps might follow. She's pointed to her administration's earlier AI-related efforts, including signing the RAISE Act in December and implementing a one-year moratorium on new data centers over the summer, as evidence New York has been proactive on the issue well before it became a national flashpoint. In a statement, Hochul said that while Trump and Washington Republicans remain inactive as AI grows increasingly unpredictable, New York does not intend to follow suit, emphasizing that companies building powerful AI systems should be transparent about safety measures and held accountable when problems arise. What's Changing Starting in November Beginning in November, frontier AI developers will start registering through a state portal to prepare for compliance ahead of the RAISE Act's January 1 deadline. Marc Gilman has been named deputy director for the RAISE Act, marking the first hire for New York's newly created Office of Digital Innovation, Governance, Integrity and Trust (DIGIT). Gilman previously served as general counsel and vice president of compliance at Theta Lake, an AI security and compliance company, and currently serves as an adjunct law professor at Fordham University. This marks his first role in government. Department of Financial Services Acting Superintendent Kaitlin Asrow, whose department houses DIGIT, said Gilman's hands-on experience working with technology companies and deep legal background in AI made him a strong fit for the position. The Hochul administration said additional full-time staff hires will be announced soon, though it hasn't specified how large the office will ultimately be. Hochul's budget allocates $10.4 million toward RAISE Act implementation, covering IT infrastructure and initial staffing costs, according to Asrow. What the RAISE Act Requires The RAISE Act, short for the Responsible AI Safety and Education Act, requires large frontier AI developers operating in New York to publish publicly available safety frameworks detailing how they're mitigating risk. Developers must also report critical safety incidents that could lead to catastrophic outcomes — defined as death or serious injury to 50 or more people, or at least $1 billion in damages — to the state within 72 hours, along with quarterly risk assessments. The state attorney general's office will lead enforcement of the law. Developers will also be required to pay an assessment fee to fund DIGIT's oversight work, similar to how the Department of Financial Services charges banks operating in New York. Gilman said in a statement that his goal is to bring greater transparency to the AI industry while supporting responsible development practices. Calls for DIGIT to Go Further Even before the RAISE Act formally takes effect, some lawmakers are urging the new office to expand its scope. Democratic state Sen. Andrew Gounardes, one of the bill's original sponsors, said DIGIT has the authority to propose additional regulations or legislation to strengthen AI safety standards going forward, and expressed hope that the office develops into a genuine source of expertise for shaping future policy. Gounardes, along with state Assemblymember Alex Bores, led the original push for the RAISE Act, which Hochul later negotiated into a more industry-friendly version closely aligned with California's AI legislation. Bores, who is expected to join Hochul at Monday's announcement, said he'd like to see DIGIT request more data from tech companies, implement universal monitoring of deployed AI systems, require third-party auditors with employee-level access, prohibit developers from releasing models that fail their own internal safety tests, and expand whistleblower protections — several provisions that appeared in earlier drafts of the legislation but didn't make the final version. Bores said he has consistently believed federal action on AI is necessary, but that the federal government has so far proven unable to deliver it, leaving states as the only level of government actively regulating the technology. While he maintains the RAISE Act needs to go further, he said he's encouraged to hear the governor emphasizing the role of states, and New York specifically, in advancing AI safety. Asked about DIGIT's plans beyond implementing the RAISE Act, Asrow described AI regulation as a fast-moving space and said the office will coordinate with other state partners and the governor's office to determine where additional authority might be needed, both within AI policy and beyond. A Broader Trend Among States Other governors have recently taken steps to implement their own AI protections, including Govs. Gavin Newsom of California and Ron DeSantis of Florida. The issue has also become a talking point for gubernatorial candidates in the closing stretch of their campaigns. Asrow said the Hochul administration has been coordinating with California, Illinois, and Massachusetts on AI regulatory approaches, describing the current moment as one where states are learning from each other in real time. She said she remains grateful for the collaboration with other states as New York continues refining its approach to AI safety.