New York state authorities launched a registration portal that large frontier AI developers, including OpenAI and Anthropic, must use to register by 1 January 2027. The deadline coincides with the entry into force of the RAISE Act, the state’s AI safety and transparency legislation, according to the announcement.
Obligations under the RAISE Act
The law applies to large frontier developers, which it defines as companies with more than USD 500 million in annual revenue. From the start of 2027, these firms will be required to publish safety frameworks on their websites, report critical safety incidents within 72 hours, and issue public transparency reports for their models.
A newly formed body, the Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT), will implement the act and oversee large frontier AI labs. Penalties can reach USD 1 million for a first violation and USD 3 million for subsequent violations. The state may also bring enforcement action against companies that fail to register through the portal, although a spokesperson declined to provide details on how this would be applied.
Independent evaluation centre
Alongside the portal, the state’s governor launched the New York Center for AI Responsibility and Research, based at Binghamton University, part of the State University of New York. The centre will independently evaluate the safety, security, and operation of advanced AI systems, and will work with DIGIT to review the safety frameworks submitted by companies. The initiative was first announced in January 2026, backed by USD 55 million in donations and capital.
At a press briefing, the governor described the state’s approach as an effort to build a national model for AI regulation, one that places communities above corporations.
A broader state-level regulatory picture
The RAISE Act is part of a wider set of AI rules in New York. In 2025, the state began enforcing a law requiring operators of AI companions to regularly notify users that they are interacting with a bot and to make reasonable efforts to detect suicidal ideation and self-harm. In October 2026, the governor signed a separate chatbot safety law focused on minors. It restricts deceptive human-like interactions, emotionally manipulative or sycophantic responses, and the encouragement of behaviours such as suicide or disordered eating. That law takes effect on 1 June 2027.
Other US states have adopted comparable frameworks. California’s SB 53, in force since January 2026, requires frontier AI labs to maintain and publish safety frameworks, put whistleblower protections in place, and report critical safety incidents within 15 days. Illinois’ SB 315, which takes effect at the start of 2027, contains similar requirements and adds independent safety compliance audits.
For developers operating across these jurisdictions, the requirements are not uniform. New York’s 72-hour incident reporting window is considerably shorter than California’s 15-day requirement, while Illinois introduces an external audit component. With the New York and Illinois regimes both starting at the beginning of 2027, large AI developers will face overlapping state-level transparency and reporting obligations from the start of the year.
