OpenAI and Anthropic logos on screens, symbolizing new New York AI safety law taking effect, increasing oversight.

Editorial illustration for OpenAI, Anthropic face new oversight as New York AI safety law takes effect

OpenAI, Anthropic face new oversight as New York AI...

By AI Daily Post Edited by Brian Petersen, Editor-in-Chief

November 10, 2026 • 3 min read

New York opened a registration portal Tuesday that will require OpenAI, Anthropic and other large frontier AI developers to sign up with the state by January 1, 2027. The requirement comes from the RAISE Act, a safety and transparency law set to take effect at the start of the year. It applies to companies pulling in more than $500 million in annual revenue, forcing them to post safety frameworks publicly, disclose critical safety incidents within 72 hours, and release transparency reports on their models.

Enforcement falls to a newly created state office called DIGIT, short for the Office of Digital Innovation, Governance, Integrity, and Trust. Companies that skip registration or violate the law's terms face fines up to $1 million for a first offense and $3 million for repeat violations. A state spokesperson declined to detail what enforcement actions might look like beyond the fine structure.

Governor Kathy Hochul framed the law as a pushback against industry claims that AI development can't be slowed down. She made the comments Tuesday at a press briefing announcing the registration system alongside a separate initiative tied to SUNY Binghamton.

The RAISE Act, New York’s safety and transparency bill, kicks in at the start of the year. It mandates that large frontier developers — companies with more than $500 million in annual revenue — have to publish safety frameworks on their websites, report critical safety incidents within 72 hours, and issue public model transparency reports.

OpenAI, Anthropic face new oversight as New York’s AI safety law takes effect - TechCrunch AI

Why this matters

New York just became the state frontier labs can't shrug off. The registration portal is live, the January 1, 2027 deadline is fixed, and the $500 million revenue threshold means this squarely targets OpenAI, Anthropic, and whoever else is playing in that tier. For builders and founders watching from outside New York, don't assume this stays local.

California tried something similar last year; New York's version has teeth around published safety frameworks, not just promises. If you're raising money or shipping frontier-scale models, "we'll self-regulate" is no longer a sufficient answer to investors or regulators asking what your safety disclosures actually look like on paper.

Hochul's framing, that "collateral damage" rhetoric from tech moguls isn't acceptable, signals where state-level politics is heading: less patience for inevitability arguments, more appetite for paper trails. Compliance teams at frontier labs should be budgeting for this now, not in late 2026. Watch whether other states copy New York's registration-portal model, that's the real signal of how fast this spreads.