SB 2 Connecticut Senate · 2025 Regular Session

AN ACT CONCERNING ARTIFICIAL INTELLIGENCE.

SB 2 regulates artificial intelligence systems used in major life decisions like hiring, loans, housing, and healthcare access. It defines "high-risk AI" as systems making "consequential decisions" (such as employment, credit, or housing) and requires developers and deployers to prevent algorithmic discrimination. The law specifically excludes video games, internal business tools, research activities, and non-discriminatory pattern analysis from regulation. It applies to companies operating within the state using such AI systems for covered decisions.
Bill status passed 3 of 5 stages cleared
Introduction
Jan 2025
Committee Review
May 2025
Senate Passage
May 2025
House Passage
Governor
Introduced Jan 8, 2025 Last action May 16, 2025
Maddy AI version diff · 2 comparisons

What changed between versions

Committee Bill GL Joint Favorable Substitute · 5 edits
MODERATE
The bill was revised from a Committee Bill to a Joint Favorable Substitute, incorporating changes to definitions and scope. Key modifications include redefining how 'consequential decisions' are made, updating the definition of 'high-risk artificial intelligence systems' with more specific exclusions, and clarifying roles for developers, integrators, and deployers. These changes aim to refine the regulatory framework for AI systems while reducing certain exemptions and clarifying applicability.
Scope change
The bill's scope was refined to more clearly define which AI systems are considered 'high-risk' and which are excluded from regulation, with specific attention to employment, education, and other consequential decision-making contexts.
DEFINITION

The definition of 'consequential decision' was modified to emphasize 'access to' employment, education, and government services, clarifying that the bill applies to decisions affecting access rather than just decisions made about those areas.

The definition of 'high-risk artificial intelligence system' was significantly expanded with seven specific exclusions, including anti-fraud technology without facial recognition, video game technology, cybersecurity tools, internal management systems, document classification systems, decision pattern detection tools, and natural language communication tools.

The definition of 'deploy' was changed from making consequential decisions to simply putting a high-risk AI system into use, broadening the scope of what constitutes deployment.

The definition of 'Integrator' was updated to clarify that integrators neither develop nor substantially modify the AI system but integrate it into products or services offered to others.

The definition of 'General-purpose artificial intelligence model' was restructured to specify criteria for generality, task range, and downstream application integration, while excluding models used only for development, prototyping, and research before market release.

Floor votes · Senate May 14, 2025

How they voted

324
Passed
Total votes 36
May 14, 2025
D Democratic25
25 Yea
100% Yea
R Republican11
7 Yea 4 Nay
63% Yea
Vote distribution
All Yea All Nay Mixed No data
Full legislative history

Actions timeline

Total actions
30
Key actions
8
Committee
7
Amendments
2
May 14, 2025
Upper · Passed
SEN. PASSED, SEN. AMEND. SCH. A,B
upper
May 14, 2025
Upper · Passed
SEN. ADOPTED SEN. AMEND. SCH. B
upper
May 14, 2025
Upper · Passed
SEN. ADOPTED SEN. AMEND. SCH. A
upper
May 12, 2025
Upper · Passed
Joint Favorable
upper
May 8, 2025
Upper · Passed
IMMEDIATE TRANSMITTAL TO COMMITTEE
upper
May 6, 2025
Upper · Passed
Joint Favorable
upper
Mar 21, 2025
Upper · Passed
Joint Favorable Substitute
upper
Feb 20, 2025
Committee
REF. TO JOINT COMM. ON General Law
upper
Feb 19, 2025
Upper · Passed
DRAFTED BY COMMITTEE
upper
Jan 8, 2025
Committee
REF. TO JOINT COMM. ON General Law
upper
0 primary · 0 co-sponsors

Sponsors

No sponsor information available.