United States – New York AI Regulation & Governance

New York's cross-sectoral AI framework rests, in force today, on two instruments. NYC Local Law 144 of 2021 prohibits employers and employment agencies from using an automated employment decision tool unless an independent bias audit has been conducted. Separately, the Artificial Intelligence Companion Models Law (General Business Law Article 47), in force since 5 November 2025, requires an operator of an AI companion available to New York users to disclose at the start of each interaction that the user is interacting with AI, and to repeat that reminder at least every three hours during extended sessions. Enacted but not yet in force is New York's Responsible AI Safety and Education Act (RAISE Act), finalized when Governor Kathy Hochul signed a chapter amendment into law on 27 March 2026. The RAISE Act is codified as General Business Law Article 44-B and, from its 1 January 2027 commencement, requires developers of frontier AI models to publish transparency reports and report critical safety incidents within 72 hours. It applies to a "large frontier developer" defined as one with prior-year revenue exceeding $500 million where a frontier model is trained using more than 10^26 floating-point operations or with training compute costs exceeding $100 million, and before deploying such a model a large frontier developer must implement, publish and retain a written safety-and-security protocol and submit test records to state agencies. Also enacted but not yet in force is the Prohibition on Unsafe AI Companion Features for Minors Act (General Business Law Article 48), enacted via Senate Bill S9051B as Chapter 368, under which an operator of a covered AI companion available to New York users may supply any of 12 listed "unsafe" features only where the user is verified as not a covered minor, from 1 January 2027.

New York's prohibited-practices position is centred on an enacted instrument not yet in force. The Prohibition on Unsafe AI Companion Features for Minors Act (General Business Law Article 48) bars an operator of a covered AI companion from supplying any of 12 listed unsafe features to a user it cannot verify is not a covered minor, taking effect 1 January 2027. Secondary legal reporting on the statute indicates the same prohibition will be enforced exclusively by the New York Attorney General, with civil penalties reported at up to $25,000 per violation.

New York's only in-force high-risk-systems regime is NYC Local Law 144, which prohibits an employer or employment agency from using an automated employment decision tool unless it has had a bias audit within the prior year, a summary of the audit and its distribution date are published, and candidates are given at least 10 business days' written notice before the tool is used. The underlying instrument records show the law taking effect from 1 January 2023, with enforcement commencing from 5 July 2023.

In force since 5 November 2025, the Artificial Intelligence Companion Models Law (General Business Law Article 47) requires an operator of an AI companion available to New York users to disclose that the user is interacting with AI, to repeat that notification at least every three hours during continuing interaction, and, per the underlying reporting, to implement a protocol to detect and address expressions of suicidal ideation or self-harm. Enacted but not yet in force is the RAISE Act, which defines a "large frontier developer" as one with prior-year revenue exceeding $500 million and a frontier model as one trained using more than 10^26 floating-point operations or with training compute costs exceeding $100 million. Before deploying a frontier model, such a developer must implement, publish and retain a written safety-and-security protocol and submit test records to state agencies, and from 1 January 2027 must disclose a critical safety incident to the New York Attorney General and the Division of Homeland Security and Emergency Services within 72 hours of determining one occurred, or within 24 hours where there is an imminent risk of death or serious physical injury.

New York's transparency regime for AI companions is in force under the Artificial Intelligence Companion Models Law (General Business Law Article 47), effective from 5 November 2025. An operator of an AI companion available to New York users must disclose at the start of each interaction that the user is interacting with AI and remind users at least every three hours during extended sessions. The same provision, as reported, also requires the operator to implement a protocol to detect and address expressions of suicidal ideation or self-harm.

The only stored substrate for this module is an enacted provision of the RAISE Act not yet in force: before deploying a frontier model, a large frontier developer must implement, publish and retain a written safety-and-security protocol and submit test records to state agencies, a duty that commences with the Act's 1 January 2027 effective date.

In force today, NYC Local Law 144 prohibits an employer or employment agency from using an automated employment decision tool unless an independent bias audit has been conducted within the prior year, a summary and the audit's distribution date are published, and candidates receive at least 10 business days' written notice. Enacted but not yet in force is New York's RAISE Act, finalized by chapter amendment on 27 March 2026 and taking effect 1 January 2027, which creates a new frontier-AI oversight office within the Department of Financial Services (NYDFS); a large frontier developer with prior-year revenue of $500 million or more must write, implement, comply with and publish a Frontier AI Framework. Also enacted but not yet in force is the SAFE for Kids Act (General Business Law Article 45), which bars a covered online platform from providing an addictive algorithmically personalised feed, or nighttime notifications, to a user it knows or should know is under 18 unless the operator confirms the user is not a minor or obtains verifiable parental consent; the underlying instrument record places its effective date at 25 January 2027. Last in standing order is a proposal rather than enacted law: at its 9 October 2026 meeting the New York State Gaming Commission considered proposed rules under 9 NYCRR Part 5404 that would prohibit licensees and their affiliates, agents and vendors from using automated decision systems, algorithmic profiling, AI, machine learning or generative systems in promotions, odds or wager recommendations. These wagering rules have not been adopted and remain at the proposal stage.

New York's safety-and-incident-reporting regime is wholly prospective, resting on the RAISE Act (General Business Law Article 44-B), which commences 1 January 2027. Before deploying a frontier model, a large frontier developer must implement, publish and retain a written safety-and-security protocol and submit test records to state agencies, and must publish transparency reports and report critical safety incidents within 72 hours. A developer must disclose a critical safety incident to the New York Attorney General and the Division of Homeland Security and Emergency Services within 72 hours of determining one occurred, or within 24 hours where there is an imminent risk of death or serious physical injury, and separately must report such an incident to NYDFS's new oversight office within the same 72-hour window (24 hours for imminent risk), with penalties reported at up to $1 million for a first violation and $3 million for subsequent violations. A large frontier developer with prior-year revenue of $500 million or more must also write, implement, comply with and publish a Frontier AI Framework.

In force today, the NYC Department of Consumer and Worker Protection (DCWP) may impose civil penalties of up to $500 for a first automated employment decision tool violation and between $500 and $1,500 per day for each subsequent or continuing violation under NYC Local Law 144. Enacted but not yet in force is the RAISE Act's enforcement architecture: finalized by chapter amendment on 27 March 2026 and taking effect 1 January 2027, it creates a new frontier-AI oversight office within the Department of Financial Services (NYDFS), to which frontier developers must report critical safety incidents, with penalties reported at up to $1 million for a first violation and $3 million for subsequent violations. The Act, codified as General Business Law Article 44-B, separately requires transparency reports and critical-incident reporting within 72 hours, and empowers the New York Attorney General to bring a civil action against a large frontier developer for reporting failures or false statements, with penalties reported at up to $1,000,000 for a first violation and up to $3,000,000 for each subsequent violation.

The only stored relief provision is contained in the RAISE Act, enacted but not yet in force: Section 1426 excepts accredited New York colleges and universities conducting academic AI research, and the Empire AI consortium, from the Act's frontier-developer duties, an exception that takes effect alongside the Act's 1 January 2027 commencement.