Bill > S06953

This bill, known as the Responsible AI Safety and Education Act (RAISE Act), establishes comprehensive regulations for large developers of advanced artificial intelligence (AI) models, called "frontier models," in New York State. The bill defines a frontier model as an AI system trained with over 10^26 computational operations or costing more than $100 million to develop, and imposes several key requirements on developers. These include creating and publishing a detailed safety and security protocol that outlines potential risks, implementing safeguards to prevent critical harm (defined as causing death or serious injury to 100+ people or over $1 billion in damages), conducting annual reviews of safety protocols, and promptly disclosing any safety incidents to state authorities within 72 hours. Large developers must also retain records of their testing procedures and are prohibited from deploying a frontier model that could create an unreasonable risk of critical harm. The bill empowers the New York Attorney General to bring civil actions against violators, with potential penalties of up to $10 million for a first violation and $30 million for subsequent violations. Importantly, the bill does not create a private right of action, meaning individuals cannot sue directly under this law. The regulations apply to any frontier models developed, deployed, or operating in New York, and are designed to be cumulative with existing legal obligations.

Budget and Finance, Business and Industry

https://www.nysenate.gov/legislation/bills/2025/S6953