Texas Responsible Artificial Intelligence Governance Act

July 02, 2025

On June 22, 2025, the Texas Governor signed the Texas Responsible Artificial Intelligence Governance Act (HB 149), which will go into effect on January 1, 2026. Although much of the new law will focus on government agencies, some provisions could have an impact on companies.

Artificial intelligence

The new law defines an “artificial intelligence system” as “any machine-based system that, for any explicit or implicit objective, infers from the inputs the system receives how to generate outputs, including content, decisions, predictions or recommendations, that can influence physical or virtual environments.”

Biometrics

The law uses a narrow definition of biometric identifiers: “retina or iris scan, fingerprint, voiceprint or record of hand or face geometry” but set forth some prohibitions on the uses of that information:

an individual has not been informed of and has not provided consent for the capture or storage of a biometric identifier of an individual for a commercial purpose based solely on the existence of an image or other media containing one or more biometric identifiers of the individual on the Internet or other publicly available source unless the image or other media was made publicly available by the individual to whom the biometric identifiers relate.

The law contains exceptions for the uses of biometric identifiers for:

  • Voiceprints for financial institutions
  • A broad exception for training AI: “the training, processing or storage of biometric identifiers involved in developing, training, evaluating, disseminating or otherwise offering artificial intelligence models or systems, unless a system is used or deployed for the purpose of uniquely identifying a specific individual.” But if that data is subsequently used for a commercial purpose, that would be a violation of this law.
  • Fraud presentation, responding to security incidents, or preventing, detecting, protecting against or responding to any other illegal activity

Discrimination

The new law has several restrictions on government agencies, but also prohibits any person who:

  1. Promotes, advertises or conducts business in this state
  2. Produces a product or service used by residents of this state
  3. Develops or deploys an artificial intelligence system in this state

from developing or deploying an artificial intelligence system “with the intent to unlawfully discriminate against a protected class in violation of state or federal law.” The law contains exceptions for insurers that are providing insurance services “subject to applicable statutes regulating unfair discrimination, unfair methods of competition, or unfair or deceptive acts or practices related to the business of insurance.” There is another exception for federally insured financial institutions “if the institution complies with all federal and state banking laws and regulations.”

Additional restrictions

In addition to the discrimination restrictions noted above, the new law also identifies several prohibited uses of AI that are not likely to be applicable to most businesses. These additional prohibited uses include developing or deploying an AI system (1) to manipulate human behavior to incite harm (self-harm and to another person) or criminal activity, or to circumvent an informed decision; (2) that intentionally results in political viewpoint discrimination or violates a person’s freedom of association or ability to express their beliefs or opinions; or (3) with the sole intent of producing or assisting in the production of explicit videos, images and child pornography.

The law also imposes certain restrictions on government agencies, including requiring the disclosure to consumers that they are interacting with an AI system, and prohibiting the use of an AI system for social scoring if it could lead to the unfavorable treatment to that person or infringes that person’s civil rights.

Regulatory sandbox

Although the details will be provided in regulations, the new law outlines a regulatory sandbox that “allows a person, without being licensed or registered under the laws of this state, to test an artificial intelligence system for a limited time and on a limited basis.” The person “must obtain approval from the department and any applicable agency before testing an artificial intelligence system.“ The law provides that the testing period cannot exceed 36 months, unless the Texas Department of Information Resources “finds good cause for the test to continue.” The program participant must file quarterly and annual reports. The quarterly report, which under the law is confidential, must contain:

  1. Metrics for the artificial intelligence system's performance
  2. Updates on how the artificial intelligence system mitigates any risks associated with its operation
  3. Feedback from consumers and affected stakeholders that are using an artificial intelligence system tested under this chapter

The law would prohibit the Attorney General or state agency from taking action against the company relating to the artificial intelligence during the testing period.

Penalties

The law does not contain a private right of action. Only the Attorney General has the power to enforce its provisions. The law provides for monetary civil penalties:

  1. For each violation the court determines to be curable or a breach of a statement submitted to the attorney general under Section 552.104(b)(2), not less than US$10,000 and not more than US$12,000
  2. For each violation the court determines to be uncurable, not less than US$80,000 and not more than US$200,000
  3. For a continued violation, not less than US$2,000 and not more than US$40,000 for each day the violation continues

The Attorney General can also recover attorney's fees and reasonable court costs or other investigative expenses. Note that the law places a burden of showing that the defendant did not use reasonable care on the Attorney General: “There is a rebuttable presumption that a person used reasonable care as required under this chapter.” State agencies have limited enforcement rights, with a maximum penalty of US$100,000.

Takeaways

The new Texas law is far simpler than Colorado’s AI law, but Texas companies should keep monitoring this Act and the regulations, and explore the opportunities of utilizing the “sandbox” concept the test the AI.