New York has opened the registration portal for one of the strongest AI safety laws in the United States — and the biggest AI labs now have a hard deadline. Under the RAISE Act (Responsible AI Safety and Education Act), large frontier AI developers must register with the state by 1 January 2027, publish their safety frameworks, and report critical safety incidents within 72 hours. The penalty for a first violation can reach $1 million, and $3 million for later ones.
Who has to comply
The law covers "large frontier developers" — companies with more than $500 million in annual revenue that build advanced AI models. TechCrunch names OpenAI and Anthropic as the companies in scope. Smaller labs and startups are not covered; this law is aimed squarely at the frontier labs whose models millions of people use every day.
What the law actually requires
- Register with the state through the new portal, by 1 January 2027.
- Publish a safety framework on the company's own website — the rules it follows to keep its models safe.
- Report critical safety incidents to the state within 72 hours of discovering them.
- Issue public model transparency reports explaining how their models work and what they can do.
Who enforces it
New York created a new office to run the programme: the Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT). It will implement the law and oversee the large labs directly. On the same day the portal opened, Governor Kathy Hochul also launched the New York Center for AI Responsibility and Research, based at the State University of New York's Binghamton University, which will independently review the safety frameworks companies submit. The state can also bring enforcement actions against companies that fail to register, though a spokesperson declined to say exactly how that would work.
Why this matters beyond New York
New York is not the first state to move — California and Illinois have passed their own AI measures — but the RAISE Act is the most direct attempt yet to make frontier labs answerable to a state government. If it works, expect other states to copy the model: registration, published safety rules, fast incident reporting and real fines. And the rules do not stop here: a separate New York chatbot safety law for minors, signed in October, takes effect on 1 June 2027, restricting deceptive human-like behaviour, emotionally manipulative responses and encouragement of harmful behaviour in chatbots used by children.
Neither OpenAI nor Anthropic had publicly responded to the registration requirement when the portal opened. For users, the practical effect is simple: the companies behind ChatGPT and Claude will have to show their safety homework in public — and tell the state quickly when something goes wrong.
Key takeaways
- New York opened its RAISE Act registration portal on 10 November 2026; the deadline is 1 January 2027.
- It covers large frontier developers — over $500M revenue — including OpenAI and Anthropic.
- They must publish safety frameworks, report critical incidents within 72 hours and release public transparency reports.
- Fines run up to $1 million for a first violation and $3 million for later ones.
- A new office, DIGIT, enforces the law, backed by an independent research centre at SUNY Binghamton.
Source: TechCrunch — OpenAI, Anthropic face new oversight as New York's AI safety law takes effect (10 November 2026). Hero photo: the New York State Capitol in Albany, by Beyond My Ken, CC BY-SA 4.0, via Wikimedia Commons.
