Earlier this week, New York state lawmakers took a major step in the right direction by passing comprehensive artificial intelligence regulations. They passed their RAISE Act, a bipartisan bill that imposes transparency requirements on advanced AI models. The bill with the catchy name would apply only to a select group of companies—specifically, the largest technology firms in the world. It features California heavyweights, such as OpenAI and Google, and Chinese companies like DeepSeek and Alibaba. If signed into law by Governor Kathy Hochul, the RAISE Act would mandate that these companies publish detailed safety and security reports on their AI models.
We were elated when our RAISE Act received overwhelming bipartisan support. New York state Senator Andrew Gounardes and New York Assemblymember Alex Bores were the co-sponsors of the bill. After clearing in the state’s Senate floor, the bill is now on the governor’s desk awaiting his signature. At this point, Governor Hochul can either sign the bill as-is into law, ask for amendments or veto the bill entirely.
Senator Gounardes emphasized the urgency of regulating AI technologies, stating, “The window to put in place guardrails is rapidly shrinking given how fast this technology is evolving.” This sentiment illustrates the increasing alarm from legislators about the risks that might come with AI’s rapid and largely unregulated advancements.
The RAISE Act takes further aim at the effective implementation of California’s notorious “AI safety bill” SB 1047. If enacted, this law would give New York’s Attorney General unprecedented powers. They should be allowed to apply civil penalties of at least $30 million for non-compliance with known standards. The bill definitely does not require AI model developers to include a “kill switch” in their models. It does not hold companies accountable for other material harms that might result from post-training tweaks to frontier AI models.
Safety advocates are strongly supporting the bill. Among the most prominent supporters is tech pioneer Geoffrey Hinton, a Nobel Prize laureate, as well as fellow AI research pioneer Yoshua Bengio. Their support underscores the need for timely, critical oversight of AI development as the technology rapidly advances.
As Anjney Midha, a general partner at Andreessen Horowitz recently, said of state-level AI regulation, “I think this is a terrible idea. He criticized the RAISE Act, labeling it “yet another stupid, stupid state level AI bill that will only hurt the US at a time when our adversaries are racing ahead.” Midha’s remarks mirror fears from industry executives as to what obstacles to innovation may be unintentionally created by a regulatory framework.
In addressing the potential impact of the RAISE Act on businesses, Assemblymember Bores remarked, “I don’t want to underestimate the political pettiness that might happen, but I am very confident that there is no economic reasons for them to not make their models available in New York.” His statement underplays and misleads on the very real positive prospects that we all should have if tech companies actually have to adapt to regulatory changes.
The implications of the RAISE Act are enormous. If it becomes law, New York will establish America’s first set of legally mandated transparency standards for frontier AI labs. If successful, this move could serve as a model for other states to follow. If skepticism around AI safety stays high, other states will be tempted to follow suit.
Nathan Calvin is the Vice President of State Affairs and General Counsel at Encode. Specifically for the RAISE Act, he noted that the goal is to address criticisms levied at previous AI safety legislation. He noted that this legislation is intended to balance encouraging innovation while protecting the public.
As New York gears up to attract one of those prizes, we hope that the RAISE Act’s future will be more secure. The outcome will likely impact how AI companies operate within the state and could influence regulatory approaches across the nation.