Yesterday, after facing opposition from state lawmakers and AI safety advocates, the U.S. Senate voted to remove a provision from President Trump’s tax cut and spending bill that would have imposed a 10-year moratorium on state regulation of artificial intelligence (AI) systems. The ban was struck from the bill with a vote of 99-1. Senator Ted Cruz (R-Texas) voted in favor of the removal despite previous opposition to state regulation of AI. Initially, Cruz and some lawmakers had proposed to deny states access to federal funding if they imposed regulations on AI. The One Big Beautiful Bill Act now heads back to the House of Representatives and possible final passage.
The striking of the moratorium on state AI regulation is good news for many states, including Texas, that have already enacted laws to police AI. In June, Governor Greg Abbott signed into law the Texas Responsible Artificial Intelligence Governance Act (Act). The Act, presented as HB 149, established regulations for the use of AI within Texas. The Act was designed to facilitate and advance “responsible development of [AI] systems,” protect individuals from foreseeable risks associated with AI, and create transparency in the development or use of AI, including such use by state agencies. In amending Subtitle D of the Texas Business & Commerce Code, the Act establishes some general definitions and contains some of the following key provisions:
Disclosure: Governmental agencies must disclose when consumers are interacting with an AI system. Such disclosure must be clear, conspicuous, and written in plain language without use of a dark pattern.
Manipulation of human behavior: An AI system cannot be developed or deployed that intentionally aims to motivate an individual to engage in self-harm, cause injury to another, or participate in criminal activity.
Social scoring: Governmental agencies cannot use an AI system that evaluates or classifies people based upon “social behavior or personal characteristics…with the intent to calculate or assign a social score” that may be detrimental or infringes upon any state or federal constitutional right.
Capture of biometric data: Governmental agencies are prohibited from capturing biometric data for the purpose of identifying a specific individual.
Constitutional protection: An AI system cannot be developed with the sole intent of infringing upon an individual’s constitutional rights or unlawfully discriminating against a protected class.
Certain sexually explicit content and child pornography: A person may not develop an AI system with the sole intent of producing or distributing certain sexually explicit content and child pornography, including deep fake videos or images and text-based conversations that replicate or depict sexual content.
The Act places enforcement of its provision under the exclusive authority of the attorney general. Pursuant to the Act, the attorney general is required to create and maintain an online mechanism through which consumers can file complaints. Upon receiving a complaint, the attorney general may issue a civil investigative demand to determine whether a violation has occurred. Upon finding a violation, the attorney general shall notify the offending party and provide an opportunity to cure. If the violation is not cured, the person would be liable for civil penalties in varying amounts.
The Act goes into effect on January 1, 2026.