Connecticut SB 2 - An Act Concerning Artificial Intelligence

• Establishes comprehensive AI regulation framework effective October 1, 2024, requiring developers and deployers of high-risk AI systems to use reasonable care to protect consumers from algorithmic discrimination beginning July 1, 2025.

• Mandates developers of high-risk AI systems provide documentation, impact assessments, and risk disclosures to deployers; requires deployers to complete impact assessments and maintain annual monitoring of AI systems for discriminatory bias.

• Requires general-purpose AI model developers to create and maintain technical documentation, establish copyright compliance policies, and make public summaries of training data by January 1, 2026.

• Establishes AI disclosure requirements: generative AI systems must mark synthetic content in machine-readable format; deployers must disclose AI use to consumers before making consequential decisions affecting employment, housing, credit, education, or justice remedies.

• Creates educational and economic development initiatives including AI training programs, advisory councils, workforce development partnerships, and grant programs to support AI innovation and responsible use across state government and healthcare sectors.