Hochul Pitches New York's RAISE Act as the Template for Federal AI Rules Under a Democratic Congress
At a Midtown press conference, the governor said Congress should adopt New York's frontier AI safety law, which takes effect Jan. 1, 2027.

New York Gov. Kathy Hochul said Tuesday that the incoming Democratic-controlled Congress should use New York's frontier AI safety statute as the model for national regulation. Speaking at a press conference at her Midtown Manhattan office, she said the incoming Congress would "restore some sanity to Washington."
Hochul told reporters that Congress should adopt New York's Responsible AI Safety and Education Act, even though dozens of competing AI enforcement bills are circulating on Capitol Hill. In her words: "When they start looking for ways to regulate AI, the first place they should look is New York."
What the RAISE Act requires
The statute is the compliance event that matters most for companies in the near term. Starting in November, the state will direct large frontier AI developers to register with the state and prepare for requirements on transparency, safety and incident reporting, including reporting critical safety incidents within 72 hours. The law takes effect Jan. 1, 2027, and targets large frontier developers.
According to the governor's office, developers will report to the state's Office of Digital Innovation, Governance, Integrity and Trust (DIGIT), which sits within the Department of Financial Services. The obligations listed in the September 21 release include publishing safety and transparency frameworks on the developer's website, filing quarterly assessments of catastrophic risk, and registering with DIGIT, filing a disclosure statement at least every other year starting Jan. 1, 2027, and paying assessments.
On enforcement, the Attorney General can bring civil actions against large frontier developers that fail to submit required reporting or make false statements, with penalties of up to $1 million for a first violation and up to $3 million for subsequent violations.
The Hochul administration says at least two companies have agreed to sign on, but it declined to say which ones.
Federal money and the wider ask
AI was one item on a longer list. "This is one element of our very high expectations on this Congress," Hochul said, naming federal dollars for healthcare, transit projects such as the Interborough Express, and offshore wind among her priorities. She also played down the risk that those asks could stall if Sen. Charles Schumer or Rep. Hakeem Jeffries lost leadership bids, saying leadership challenges are routine at the start of a session.
The Washington picture
Reuters reported Nov. 8 that Democrats are planning a flood of legislation targeting AI systems and data center development when they take control of Congress in January. Planned bills would shield consumers from rising electricity costs tied to data centers, require independent testing of advanced AI systems and clarify when AI companies can be held liable for harm caused by their products. Schumer, expected to become majority leader, said protecting consumers from higher energy costs linked to data center expansion would be among his priorities next year.
The obstacle is the White House. Reuters reported that any legislation would face steep odds while President Donald Trump remains in office, as he has repeatedly dismissed calls for mandatory AI rules. The Center Square reported that Trump signed an order last December that limited states' ability to regulate AI, and that Hochul was proceeding despite threats of legal challenges from the industry.
Hochul has also signaled more state action. The governor's September 21 release said she will explore ways to build on the RAISE Act.
What this means for in-house counsel
If your company develops or fine-tunes large frontier models, confirm the status of the registration process the state said it will direct starting in November, and treat the Jan. 1, 2027 effective date as a live deadline. The statute's 72-hour incident reporting clock needs an internal escalation path that reaches legal quickly.Published safety and transparency frameworks create a public record, and quarterly catastrophic-risk assessments filed with the state's DIGIT Office can be shared with other government authorities. Counsel should review those documents with the same care applied to securities disclosures, because the Attorney General's enforcement authority extends to false statements.
Do not assume New York's text will be the national standard. Hochul's push depends on a Congress that has not yet convened, on leadership races she acknowledged could be contested, and on a president who has resisted mandatory AI rules. Counsel may want to consider whether building to the strictest state requirements, New York and California among them, would leave the company better placed whichever way federal preemption or federal legislation goes. Keep an eye on the executive order's effect on state AI laws when you assess exposure.
Finally, plan for oversight beyond statute. Reuters reported that House Democratic committee leaders already asked OpenAI, Anthropic, Meta, Google and SpaceX in October to detail their security practices and report cases of AI tools acting in ways their designers did not intend. Reuters also reported that Democrats could compel testimony. Counsel at companies that are not yet in contact with government affairs on these topics may want to prepare for possible document requests and testimony invitations, and align state compliance records with what you would be willing to say to a congressional committee.
Free weekly email
The Counsel House Brief
The week's most important legal news for general counsel and in-house teams, in one short email.
Counsel House corrects errors promptly. Spot one? Email daniel@counselhouse.org.