New York RAISE Act Takes Effect: OpenAI, Anthropic Must

New York launched a registration portal on November 10, 2026, for its landmark RAISE Act, setting the stage for one of the nation's most comprehensive AI safety regimes. Frontier AI developers including OpenAI and Anthropic must register by January 1, 2027, when the law officially takes effect. The state's new Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT) will oversee compliance and enforcement.

Registration Requirements and Deadlines

The RAISE Act targets large frontier AI developers, defined as companies generating more than $500 million in annual revenue. These organizations must register through the newly opened state portal before the January 1, 2027 deadline. Governor Kathy Hochul emphasized during a press briefing that the legislation establishes "a national model for smart, sensible AI regulation, a model that puts communities above corporations and people above powerful technology."

Companies failing to register face enforcement actions, though specific penalties for non-registration were not disclosed by state officials. Once registered, organizations must maintain ongoing compliance with transparency and safety reporting obligations.

Safety Framework and Incident Reporting Mandates

Under the RAISE Act, covered companies must publish comprehensive safety frameworks on their public websites. These frameworks detail how organizations identify, assess, and mitigate risks associated with their AI systems. The law also mandates reporting of critical safety incidents within 72 hours of discovery, a significantly shorter window than California's 15-day requirement under SB 53.

Additionally, frontier developers must issue regular public model transparency reports. These documents provide visibility into model capabilities, limitations, and deployment contexts. The reporting requirements aim to give researchers, policymakers, and the public better insight into advanced AI system operations.

Enforcement Powers and Financial Penalties

The Office of DIGIT holds substantial enforcement authority under the legislation. First-time violations can result in fines up to $1 million, while subsequent violations carry penalties as high as $3 million. These financial consequences represent some of the steepest AI-related penalties enacted at the state level.

Supporting DIGIT's oversight mission, Governor Hochul also launched the New York Center for AI Responsibility and Research at SUNY Binghamton on November 10, 2026. The center, backed by $55 million in donations and capital announced in January, will independently evaluate advanced AI systems for safety, security, and operational integrity. It will work directly with DIGIT to review companies' submitted safety frameworks.

Availability and Broader Regulatory Context

Officially launched on November 10, 2026, the registration portal marks the operational start of New York's AI oversight infrastructure. The RAISE Act joins a growing roster of state-level AI regulations. New York already enforces a 2025 law requiring AI companion operators to disclose bot interactions and detect suicidal ideation. In October 2026, the governor signed additional chatbot safety legislation for minors, restricting manipulative interactions and harmful content encouragement, set to take effect June 1, 2027.

Other states have enacted parallel measures. California's SB 53 took effect in January 2026, requiring safety frameworks, whistleblower protections, and incident reporting. Illinois's SB 315, launching January 2027, adds independent safety compliance audits to similar requirements.

What This Means

New York's RAISE Act establishes a precedent for state-level AI governance that balances innovation with public accountability. By requiring frontier labs to operate transparently and report safety incidents rapidly, the law addresses growing concerns about AI system risks without imposing pre-deployment approval processes. The collaboration between DIGIT and the academic research center at SUNY Binghamton signals a model that combines regulatory oversight with independent technical evaluation. As other states watch New York's implementation, the RAISE Act may influence the shape of AI governance across the United States, potentially creating a patchwork of compliance requirements for companies operating nationally.