New YorkAI regulation

Regulation Status

• New York is among the most active US states on AI regulation, having enacted a broad set of targeted AI laws rather than a single omnibus statute. Its most consequential measure is the Responsible AI Safety and Education (RAISE) Act, which imposes safety-framework, transparency, disclosure, and critical-safety-incident reporting obligations on large frontier developers—developers of frontier models trained above 10^26 operations with more than $500 million in prior-year annual gross revenues—with penalties up to $1 million for a first violation and $3 million for subsequent violations.
• New York City's Local Law 144 prohibits employers and employment agencies from using automated employment decision tools for screening unless the tool has undergone a recent bias audit and a public audit summary is available; it also requires candidate or employee notices, information about assessed qualifications and data practices, and an opportunity to request an alternative selection process or accommodation. At the state level, New York has legislated heavily on synthetic media and likeness, including election-deepfake disclosure duties, extension of privacy rights to AI-generated uses of a person's picture, likeness, or voice, coverage of deepfake intimate images, and digital-replica rules for service contracts, fashion models, deceased performers, and synthetic performers in advertising.
• New York has also moved on AI in government and consumer-facing systems. The LOADinG Act requires state agencies to list automated decision-making tools, conduct and publish impact assessments, preserve meaningful human review for benefits and rights-affecting uses, and restrict automated systems in certain internal employment decisions; related measures require state AI inventories and protect public employees' collective-bargaining and employment rights. Consumer measures include AI companion self-harm-detection and non-human disclosure requirements, Safe By Design Act protections for minors including default-off AI companions on covered platforms, personalized algorithmic-pricing disclosures, and a prohibition on algorithm-facilitated landlord price coordination.