Texas TRAIGA Compliance Checker (2026)

The Texas Responsible AI Governance Act (HB 149) was signed by Governor Greg Abbott on 22 June 2025 and takes effect on 1 January 2026. Its duties are codified at Texas Business & Commerce Code Chapter 552 and are enforced by the Texas Attorney General under § 552.201.

Any person that develops, deploys, or promotes an AI system for consumers or residents of Texas. This includes SaaS vendors selling AI features into Texas, employers using AI in hiring, healthcare providers using AI in patient interactions, and government entities using AI on Texans. There is no small-business exemption.

Five categories of prohibited AI use: (1) social scoring by a Texas government entity, (2) biometric identification of individuals without consent, (3) manipulation of human behavior to cause self-harm or harm to others, (4) generation or promotion of child sexual abuse material (CSAM), and (5) production or distribution of non-consensual deepfake sexual imagery. These violations are uncurable and per-se enforceable.

Under Tex. Bus. & Com. Code § 552.202 you have an affirmative defense to a TRAIGA enforcement action if you (a) implement and maintain a documented risk management program that reasonably conforms to the NIST AI Risk Management Framework — or an equivalent framework such as ISO/IEC 42001 — and (b) cure any curable violation within 60 days of written notice from the Attorney General. This is the single highest-leverage compliance step under TRAIGA.

Curable violations are subject to civil penalties of up to $10,000 per violation after a 60-day cure period. Uncurable violations — the § 552.052 prohibited practices — are subject to civil penalties of up to $200,000 per violation. Enforcement is exclusive to the Texas Attorney General under § 552.201. There is no private right of action.

Yes, if the AI system interacts with consumers. Section 552.051 requires a clear, conspicuous, consumer-comprehensible disclosure that the consumer is interacting with an AI system. The disclosure should appear at or before the beginning of the interaction. Chatbots, voice agents, AI-generated customer support, and AI-driven decision engines all trigger this duty.

TRAIGA layers on top of the existing Texas Capture or Use of Biometric Identifier Act (Tex. Bus. & Com. Code § 503.001, CUBI). AI systems that capture biometric identifiers must satisfy both statutes: CUBI consent requirements and TRAIGA § 552.053 duties. TRAIGA also strengthens the consent rule by treating unconsented biometric identification as a prohibited practice under § 552.052.

The AI Regulatory Sandbox (Chapter 552, Subchapter G) is a program administered by the Texas Department of Information Resources (DIR) that allows AI developers to test novel systems in Texas with temporary regulatory relief. Participants file quarterly reports on system performance, incidents, and consumer complaints. Failure to report suspends the sandbox protection.

Yes. Employers using AI as a substantial factor in employment decisions (hiring, promotion, termination) must notify affected candidates or employees, disclose the categories of data considered, and provide a mechanism for human review of adverse decisions. NYC LL-144 employers already familiar with bias-audit rules should still expect additional TRAIGA duties when hiring Texas candidates.

Colorado SB 24-205 targets "high-risk" AI systems making "consequential decisions" and imposes duties on both developers and deployers, including impact assessments and post-deployment reviews. Texas TRAIGA is broader in scope (any AI interacting with Texas consumers) but narrower in prescriptive requirements (mainly disclosure, prohibited-practice list, and NIST safe harbor). Multi-state operators typically need to satisfy both, plus NYC LL-144 for AEDT hiring tools.

Federal contractors operating in Texas remain subject to TRAIGA unless expressly preempted. National security and defense AI is generally excluded, but consumer-facing AI used in Texas by federal contractors is not. Additional obligations from the White House AI Executive Order (revoked in 2025) and OMB memoranda M-24-10 / M-24-18 may still apply to federal use cases.

HB 149 bill history and enrolled text: https://capitol.texas.gov/BillLookup/History.aspx?LegSess=891&Bill=HB149. The codified statute is at Texas Business & Commerce Code Chapter 552. The Texas Attorney General enforcement page: https://www.texasattorneygeneral.gov/consumer-protection/artificial-intelligence.