Status: Signed into law by Governor Ned Lamont on May 27, 2026 as Public Act No. 26-15. The Connecticut General Assembly passed SB 5 on May 1, 2026 (Senate 32-4, House 131-17), and the Governor's office announced the signing at a ceremonial event on June 2, 2026. Connecticut becomes the seventh state to enact a comprehensive AI governance statute and the most expansive AI law enacted in the 2026 session.

Citation: Connecticut Public Act No. 26-15 (SB 5, 2026 Regular Session) — the Connecticut Artificial Intelligence Responsibility and Transparency Act (“AIRT Act”)

Jurisdiction: Connecticut

Category: AI Governance, Algorithmic Transparency, Employment Decision Systems, Frontier Model Safety, AI Companion Chatbots

Industries: All Connecticut employers using AI in personnel decisions; foundation-model developers; consumer-facing AI chatbot operators; state agencies; K-12 education; physical security and surveillance technology providers operating in Connecticut

Enacted Date: May 27, 2026 (Public Act No. 26-15; signing announced June 2, 2026)

Effective Date: Phased — October 1, 2026 (employment AI provisions and frontier model whistleblower protections); January 1, 2027 (AI companion chatbot rules); October 1, 2027 (remaining provisions including state agency AI inventories and impact assessments)

Summary: SB 5 is a 39-section omnibus statute that sets disclosure, transparency, accountability, and consumer-protection obligations across the AI lifecycle. It regulates four distinct domains in a single bill: (1) automated employment decision technology, including notice and recordkeeping obligations for employers and developers; (2) AI companion chatbots that foster emotional attachment, with disclosure-at-start and recurring-hourly notice plus suicide and self-harm response protocols; (3) frontier AI model safety, with whistleblower protections for employees of developers training models above the 10²⁶ floating-point-operation threshold; and (4) public-sector AI governance, requiring state agency AI inventories, impact assessments, and a new Connecticut AI Academy. SB 5 also adds an AI-related disclosure to layoff notices filed with the Connecticut Department of Labor under WARN, integrates AI into K-12 computer-science standards, and clarifies that use of an automated employment decision tool is not a defense to a discrimination claim. Following Colorado’s repeal-and-replace earlier in 2026, Connecticut becomes the new high-water mark for state-level AI law.

Key Requirements:

Automated Employment Decision Technology (effective October 1, 2026): Developers of AI tools used as a “substantial factor” in hiring, promotion, discipline, or discharge must give deployers compliance-related information sufficient to evaluate algorithmic discrimination risk. Deployers must notify affected employees and applicants of the technology’s use, purpose, data categories, and data sources, and must retain documentation.
WARN AI Disclosure (effective October 1, 2026): Layoff notices filed with the Connecticut Department of Labor must disclose whether AI contributed to the decision to reduce or eliminate positions.
Frontier Model Whistleblower Protections (effective October 1, 2026): Developers training foundation models using more than 10²⁶ floating-point operations must protect employees who report concerns that the developer or model may contribute to a catastrophic risk, and must maintain internal reporting channels.
AI Companion Chatbot Rules (effective January 1, 2027): Operators of chatbots that foster emotional attachment must declare at the start of each interaction (and at least hourly during ongoing sessions) that the user is interacting with an AI system. Operators must implement protocols to detect expressions of suicidal ideation, self-harm, or imminent violence and refer users to crisis resources.
State Agency AI Governance (phased through October 1, 2027): State agencies must inventory AI systems in use, complete impact assessments, and publish summary information. The Connecticut AI Academy is created to train state employees on responsible AI use.
K-12 Integration: AI literacy is added to Connecticut’s K-12 computer-science curricular standards.

Enforcement & Penalties: Violations of the automated employment decision provisions are deemed unfair or deceptive trade practices under the Connecticut Unfair Trade Practices Act (CUTPA), enforceable solely by the Connecticut Attorney General. There is no private right of action. For violations occurring on or before December 31, 2027, the Attorney General may issue a 60-day cure notice before filing suit; after that date the cure period is no longer available. Civil penalties under CUTPA apply.

Affected Entities: All Connecticut employers using AI as a substantial factor in personnel decisions; consumer-facing AI chatbot operators serving Connecticut residents; foundation model developers above the 10²⁶ FLOP threshold (regardless of headquarters location, where they impact Connecticut residents or employees); Connecticut state agencies; and any vendor selling AI-enabled products into Connecticut that touch employment, education, or consumer markets.

IntelliSee Relevance: High. SB 5 reinforces the regulatory direction that has made Connecticut a strategic state for AI physical security adoption: transparency and accountability are required, but biometric identification, facial recognition, and emotional-attachment AI patterns are the regulated risks. IntelliSee’s platform sits cleanly on the favored side of this divide. The system uses no facial recognition and no biometric identification; it analyzes visual events — a brandished firearm, a fall, an aggressive-behavior pattern, an unauthorized intrusion — rather than analyzing or identifying people. Connecticut customers in healthcare, K-12, higher education, commercial real estate, and state and municipal facilities can deploy IntelliSee’s AI gun detection, fall detection, and aggressive-behavior monitoring as engineering controls without triggering the employment-decision, biometric-identification, or chatbot-disclosure obligations that drive most SB 5 compliance work. Connecticut Department of Labor WARN filings that name AI-influenced layoffs do not apply to IntelliSee deployments because the platform does not make personnel decisions. Connecticut state agencies completing the SB 5 AI inventory will benefit from the platform’s clear documentation: a defined functional purpose (visual threat detection), no PII collection, no facial recognition, and verifiable, time-stamped alert records that support agency impact-assessment requirements. Security and risk leaders in Connecticut should treat SB 5 as a positive signal: regulated risks are concentrated in identity-based and decision-making AI, not in event-based visual analytics.

Compliance Timeline

May 1, 2026 — Connecticut General Assembly passes SB 5 (Senate 32-4, House 131-17).
May 27, 2026 — Governor Lamont signs SB 5 into law as Public Act No. 26-15; ceremonial signing announced June 2, 2026.
October 1, 2026 — Automated employment decision technology provisions, WARN AI disclosure, and frontier model whistleblower protections take effect.
January 1, 2027 — AI companion chatbot rules take effect.
October 1, 2027 — State agency AI inventory, impact assessment, and Connecticut AI Academy provisions fully operational. Attorney General 60-day cure-notice period for automated employment violations expires on December 31, 2027.

Related Legislation & Resources

Bill text and history: Connecticut General Assembly — SB 5 (2026) | Governor Lamont signing announcement (June 2, 2026)

Connecticut Artificial Intelligence Responsibility and Transparency Act (SB 5, 2026)

Compliance Timeline

Related Legislation & Resources

Turn Legislative Mandates Into a Working Safety Program