New York launches RAISE Act implementation — mandatory frontier-AI developer registration and 72-hour incident reporting

On September 21, 2026, Governor Kathy Hochul announced the next steps for implementing the Responsible AI Safety and Education (RAISE) Act, New York's land-mark AI safety statute that takes effect January 1, 2027. Starting in November 2026, large frontier AI developers will be directed to register with the State through a portal and prepare for compliance with strict transparency, safety, and incident-reporting requirements — including a 72-hour deadline to report critical safety incidents. Beginning January 2027, covered AI companies must comply with RAISE Act standards and file regular reports (including quarterly catastrophic-risk assessments) with the new Office of Digital Innovation, Governance, Integrity and Trust (DIGIT) housed in the NY Department of Financial Services. Hochul also named Marc Gilman as DIGIT deputy director overseeing implementation and publicly floated possible future requirements including an AI 'kill switch.'

This is the first concrete operationalization of a binding US state statute that imposes registration, safety-planning, and incident-notification duties directly on frontier AI developers — a compliance clock that starts in November 2026, before the law's effective date. New York is coordinating with California, Illinois, and Massachusetts on AI regulation, signaling a growing multi-state enforcement bloc for AI deployers regardless of federal policy.

Large frontier AI developers operating in or directing products at New York should prepare to register with the DIGIT portal from November 2026, and stand up 72-hour critical-safety-incident reporting procedures plus quarterly catastrophic-risk assessments in advance of the January 1, 2027 effective date.