Summary:
La nouvelle loi texane sur la régulation de l’IA, TRAIGA, vise à protéger les consommateurs tout en favorisant l’innovation. Ce cadre axé sur l’humain impose des exigences de transparence aux entreprises et crée un conseil dédié à l’évaluation de l’utilisation de l’IA. Bien que conçu pour responsabiliser les entreprises, le projet de loi permet également un environnement d’innovation contrôlé. En raison de cette régulation, Texas pourrait devenir un modèle pour d’autres États, malgré des défis pour préserver son attractivité en tant que hub technologique.
Original Link:
Original Article:
New AI Regulation Law Protects Consumers, but Could Slow Texas’ Tech Growth
Human-centered framework could become a model for other states
By Julianna Plewes, Fri., July 18, 2025
While California failed to pass a similar law, Texas is launching more robust regulation of AI.
From small startups to established tech Goliaths, companies in California are rushing to Texas to operate in a business-focused, anti-regulation state – or so we thought. Texas lawmakers surprised companies and politicians alike by passing the Texas Responsible Artificial Intelligence Governance Act, or TRAIGA, which overwhelmingly protects consumers while implementing thorough expectations for businesses using artificial intelligence.
“From the pro-consumer, pro-individual citizen, pro-privacy standpoint,” said Matthew Murrell, AI law expert, “this bill is shockingly good.”
Murrell designed and taught UT law school’s first-ever class on the law of AI and will assume a new position as an assistant professor of law at the University of New Mexico this August. He said that Texas is pioneering the future of AI legislation, with only a handful of states passing similar comprehensive bills.
Murrell said that this bill, which takes effect in January, places the idea of human-centeredness at its core, a concept used in other AI legislation, such as that passed by the European Union. This means that when companies develop AI systems, they must first consider the implications for the members of the public and their consumer impact.
TRAIGA does so through several mechanisms. For one, the bill emphasizes the importance of transparency for companies developing AI, requiring disclosure when used. The act also prohibits certain types of AI activity, such as encouraging self-harm and utilizing biometric data.
In addition, the act creates a Texas AI Council tasked with evaluating the use of AI, ensuring companies design AI with the public’s best interest in mind, offering guidance to the Legislature, and issuing reports, among other things. The council falls under the Texas Department of Information Resources and will comprise members of the public appointed by the governor, lieutenant governor, and speaker of the House.
“This bill is shockingly good.” – UT-Austin AI Law Expert Matthew Murrell
While designed to hold businesses accountable, the bill provides leeway for innovation, said Murrell, as long as companies are willing to jump through some hoops. TRAIGA accounts for the fact that artificial intelligence is still a new and evolving field, creating a regulatory sandbox program to allow controlled innovation. This program provides legal protection to companies designing and testing AI systems under conditions of government oversight and program evaluation.
Despite these new requirements, the average consumer likely will not notice any changes. Large companies, on the other hand, will “have to take a hard look at what the company does and what the requirements of the law are,” said Murrell.
These expectations created by TRAIGA challenge Austin’s position as what some have called the new Silicon Valley. Emerging as a technology hub over the past 20 years, Austin and other major cities in Texas have grown closer to the tech world as companies relocate to capitalize on the state’s perceived anti-regulation, pro-business attitude.
Especially after California Gov. Gavin Newsom vetoed the state’s own version of a comprehensive AI regulation bill, Murrell said: “It runs counter to the narrative that all these companies are fleeing California because there’s no regulating in Texas.”
While states like Texas and California create and vote on AI regulation, the federal government has stayed quiet. Congress’ current version of Trump’s “big beautiful bill” includes a federal moratorium on artificial intelligence, preventing nationwide legislation on AI for 10 years.
The idea, Murrell said, is to allow AI to develop and grow and to enable our companies to compete fully with those in other countries. However, politicians and legal experts on both sides of the aisle have raised concerns over the decision. “I think that it’s a little too simple,” Murrell said. “I think it’s short-sighted.”
With the moratorium on the table, states must develop their own laws if they hope to regulate artificial intelligence. That means other states will likely look to TRAIGA as a model for their own AI legislation.
It’s a good thing, too, says Murrell. “If you’re a consumer focused on privacy, this is a win.”