New York's RAISE Act puts OpenAI, Anthropic under new oversight

New York has opened the registration portal that will require large frontier AI developers, including Anthropic and OpenAI, to sign up with the state by January 1, 2027, under the RAISE Act, the state’s AI safety and transparency law that takes effect at the start of the year. The law covers companies that bring in upwards of $500 million a year, and it requires three things of them: a safety framework posted publicly on their site, notice to the state within 72 hours of a critical safety incident, and a transparency report on each model they release.

A newly formed state office, the Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT), will implement the law and oversee the labs it covers. Violations can draw penalties of up to $1 million for a first offense and $3 million for repeat ones, and the state can also pursue enforcement against companies that never register — though a spokesperson wouldn’t say what that enforcement might look like.

Governor Kathy Hochul paired the rollout with the launch of the New York Center for AI Responsibility and Research, based at Binghamton University. The center, in Hochul’s words, will give the state independent eyes on how advanced AI systems actually perform on safety and security, feeding its findings into DIGIT’s review of the labs’ safety frameworks; the governor first floated the center back in January, backed by $55 million in donations and capital. “We’re building a national model for smart, sensible AI regulation, a model that puts communities above corporations and people above powerful technology,” she said.

The RAISE Act joins a growing stack of New York AI rules. New York already enforces a separate law requiring AI companion apps to periodically disclose that users are chatting with a bot, not a person, and to take reasonable steps toward spotting signs of suicidal thinking or self-harm. October also brought a separate law for minors that bars chatbots from behaving like a deceptive human companion, deploying emotionally manipulative or overly flattering responses, or nudging a child toward things like suicide or an eating disorder; that one takes effect June 1, 2027. Other states have moved on similar lines: California’s SB 53, in force since January, requires frontier labs to publish safety frameworks, protect whistleblowers and report critical incidents within 15 days, while Illinois’ SB 315, effective at the start of 2027, adds independent safety compliance audits.

Why it matters

New York is now attaching binding transparency and incident-reporting duties, with real financial penalties, to frontier AI labs — and pairing that regulation with an independent research center gives California and Illinois, which are moving on similar bills, a template to watch as they finalize their own rules.