Bill > HB2157

This bill establishes new regulations for the development, deployment, and use of "high-risk artificial intelligence systems," which are defined as AI systems intended to make or significantly influence "consequential decisions" that have a substantial impact on individuals, such as decisions about parole, employment, housing, or financial services. Developers of these systems must take reasonable care to prevent "algorithmic discrimination," which is defined as unlawful differential treatment or impact based on protected characteristics like race, sex, or disability, and must provide detailed documentation about the system's limitations, intended uses, and measures taken to mitigate discrimination. Deployers, or those who use these systems to make consequential decisions, must also implement risk management policies, conduct impact assessments before deploying or significantly updating a system, and disclose to consumers when they are interacting with an AI system, explaining its purpose, nature, and how it influences consequential decisions. The bill also includes provisions for identifying synthetic content generated by AI, outlines exceptions for certain types of AI systems and activities, and allows for civil action against developers or deployers who violate these regulations, with a potential affirmative defense if violations are discovered and cured within 45 days.