NEW YORK — Gov. Kathy Hochul is expected to sign a landmark artificial intelligence bill establishing New York’s first guardrails for tech heavyweights to mitigate catastrophic risks.
The deal comes after intense talks between the governor and sponsors Assemblymember Alex Bores and state Sen. Andrew Gounardes over the details. Hochul redlined much of the Responsible AI Safety and Education Act last week and replaced it with language largely matching a comparable California law, while the bill’s sponsors pressed to keep as much teeth in the legislation as possible.
In its final iteration, the RAISE Act requires developers of the industry’s most advanced AI tools, called frontier models, to report critical safety incidents within 72 hours and imposes a $1 million penalty for the first violation and $3 million for subsequent ones. It creates an oversight office within the Department of Financial Services to assess frontier models.
“New York is once again leading the nation in setting a strong and sensible standard for frontier AI safety, holding the biggest developers accountable for their safety and transparency protocols,” Hochul said Friday in a statement. “This law builds on California’s recently adopted framework, creating a unified benchmark among the country’s leading tech states as the federal government lags behind, failing to implement common-sense regulations that protect the public.”
New York finalized its AI safety law just before California’s SB 53 law goes into effect on New Year’s Day and as states are confronting President Donald Trump’s executive order targeting state-level regulation of the fast-evolving AI industry.
“In New York, we defeated last-ditch attempts from AI oligarchs to wipe out this bill and, by doing so, raised the floor for what AI safety legislation can look like,” Bores, a Manhattan Democrat and candidate for Congress, said in a statement. “And we defeated Trump’s — and his donors — attempt to stop RAISE through executive action greenlighting a Wild West for AI.”
New York’s RAISE Act will regulate the largest players in the fast-evolving AI industry, and require safety protocols and reporting to prevent large-scale harm like the creation of bioweapons. The models that stand to be subject to enforcement include ChatGPT, developed by OpenAI; Claude by Anthropic; LLaMA by Meta; Gemini by Google; and Copilot by Microsoft.
Hochul pushed for New York’s approach to mirror California’s SB 53 law and help set a national standard on AI guardrails. Tech industry leaders have warmed to California Gov. Gavin Newsom, who like Hochul is a Democrat, and the AI law he signed in September.
What ultimately became law in California is laxer than earlier iterations of SB 53, which is known as the Transparency in Frontier AI Act. New York’s RAISE Act will be stricter than California’s regulations, which give frontier model developers 15 days to report critical safety incidents and cap penalties at $1 million per violation. But it’s less stringent than what state legislators passed in June. As an example, the earlier version of the RAISE Act included penalties of up to $10 million for the first violation and $30 million for subsequent violations.
“Big tech oligarchs think it’s fine to put their profits ahead of our safety — we disagree,” Gounardes, a Brooklyn Democrat, said in a statement. “With this law, we make clear that tech innovation and safety don’t have to be at odds. In New York, we can lead in both.”
The negotiations between the governor and legislators were so fraught that it looked one week ago like the RAISE Act could be vetoed. Hochul’s version of the RAISE Act — returned to legislators last week as so-called chapter amendments — included language from SB 53 and eliminated much of the bill’s original text, according to two people who viewed the document. The people who spoke with POLITICO for this article were granted anonymity to speak about sensitive negotiations.
Tech behemoths have lobbied for nationally consistent regulations rather than a state-by-state “patchwork” of laws. Bores has called for the same and applauded SB 53. But he has also said he sees the California law as a starting point or floor, not the ceiling for accountability for deep-pocked AI interests.
Bores is a congressional contender seeking to succeed Rep. Jerry Nadler in Manhattan. He’s also a computer engineer who formerly worked for data tech firm Palantir and has staked his House bid on his fight against the unchecked growth of powerful tech companies.
He is the first political target of the $100 million super PAC Leading the Future, which is funded by OpenAI’s president and venture capital firm Andreessen Horowitz. His own network of donors includes people affiliated with tech giants that compete with those companies.
One of Bores’ chief rivals in the House race, Nadler protégé and Assemblymember Micah Lasher, is a cosponsor of the RAISE Act.
