Big Tech, Big Law: How Texas Is Helping to Shape the Future of AI

Introduction

Texas has emerged as a leading jurisdiction for AI development and infrastructure, on the one hand, and for AI regulation and enforcement, on the other. This has involved balancing policy choices to foster innovative growth and support pragmatic risk governance, while also making clear that consumer protection safeguards must be taken seriously.

Data centre construction is booming in Texas. The state is home to the flagship site of Stargate, a USD500 billion data centre infrastructure consortium venture. In 2025, the state also enacted the Texas Responsible Artificial Intelligence Governance Act, a comprehensive AI law, becoming only the second state to do so. Moreover, Attorney General Ken Paxton has emerged as an aggressive enforcer of the state’s consumer protection and data privacy laws against AI companies, becoming the first state Attorney General to bring an AI-specific consumer protection enforcement action. However, like the other states that have passed substantive AI regulation, the future of Texas’ AI legal framework remains uncertain in light of national AI policy under the Trump Administration, which seeks to avoid a patchwork of divergent state AI regulatory regimes.

Legal developments

The Texas Responsible Artificial Intelligence Governance Act in effect

On 22 June 2025, Governor Abbott signed the Texas Responsible Artificial Intelligence Governance Act (TRAIGA), making Texas among the first states to enact comprehensive AI legislation. TRAIGA took effect on 1 January 2026 and applies broadly to any person or entity conducting business in Texas, producing products or services used by Texas residents, or developing or deploying AI systems within the state. AI systems are defined 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.” (Tex. Bus. & Com. Code, Section 551.001(1). The law primarily regulates covered entities through the prohibition of certain activities, some of which apply only to governmental entities. Prohibitions are in effect for the following.

• Encouraging harmful behaviour – “develop[ing] or deploy[ing] an artificial intelligence system in a manner that intentionally aims to incite or encourage a person to:
1. commit physical self-harm, including suicide;
2. harm another person; or
3. engage in criminal activity.” (Tex. Bus. & Com. Code, Section 552.052.)
• Violating constitutional rights – “develop[ing] or deploy[ing] an artificial intelligence system with the sole intent for the artificial intelligence system to infringe, restrict, or otherwise impair an individual’s rights guaranteed under the United States Constitution.” (Tex. Bus. & Com. Code, Section 552.055.)
• Unlawful discrimination – “develop[ing] or deploy[ing] an artificial intelligence system with the intent to unlawfully discriminate against a protected class in violation of state or federal law.” (Tex. Bus. & Com. Code, Section 552.056(b).)
• Certain sexually explicit content and child pornography – “(1) develop[ing] or distribut[ing] an artificial intelligence system with the sole intent of producing, assisting or aiding in producing, or distributing: (A) visual material ... or ... deepfake videos or images ... or (2) intentionally develop[ing] or distribut[ing] an artificial intelligence system that engages in text-based conversations that simulate or describe sexual conduct ... while impersonating or imitating a child younger than 18 years of age.” (Tex. Bus. & Com. Code, Section 552.057.)

Importantly, these prohibitions apply to intentional-use cases and affirmative design choices. It will not be sufficient to prove that the enumerated negative activity resulted from the use of generally designed tools. For example, the law explicitly states that, for the purposes of TRAIGA “disparate impact is not sufficient by itself to demonstrate an intent to discriminate” against a protected class. (Tex. Bus. & Com. Code, Section 552.056(c).) In contrast, employment- and civil rights-focused AI laws that aim to combat discrimination in hiring, such as New York City’s Local Law 144, do not require proof of discriminatory intent to find that discrimination has occurred and impose affirmative obligations on covered entities to prevent discriminatory outcomes with certain uses of AI.

Moreover, two prohibitions apply only to governmental entities and are focused on surveillance concerns.

• Social scoring – prohibits “governmental entit[ies]” from using or deploying “an artificial intelligence system that evaluates or classifies a natural person or group of natural persons based on social behavior or personal characteristics, whether known, inferred, or predicted, with the intent to calculate or assign a social score or similar categorical estimation...” (Tex. Bus. & Com. Code, Section 552.053.)
• Capture biometric data – prohibits “governmental entit[ies]” from developing or deploying AI “for the purpose of uniquely identifying a specific individual using biometric data or the targeted or untargeted gathering of images or other media from the Internet or any other publicly available source without the individual’s consent, if the gathering would infringe on any right of the individual under the United States Constitution, the Texas Constitution, or state or federal law.” (Tex. Bus. & Com. Code, Section 552.054(b).) A violation of Texas’ long-standing biometric privacy law, the Capture or Use of Biometric Identifier Act, constitutes a violation of this Section. (Tex. Bus. & Com. Code, Section 552.054(c).)

In addition to prohibiting the activities described above, the law establishes a regulatory sandbox programme, administered by the Texas Department of Information Resources (DIR), that “enables a person to obtain legal protection and limited access to the market in this state to test innovative artificial intelligence systems without obtaining a license, registration, or other regulatory authorization.” (Tex. Bus. & Com. Code, Section 553.051(a).)

AI companies must obtain approval by the DIR to participate. By joining the sandbox programme, participants have 36 months to text and develop their AI systems, during which time state regulators cannot file charges or pursue punitive action for violations of state laws or regulations waived under TRAIGA. (See Tex. Bus. & Com. Code, Sections 553.053(a), and 553.051(c)–(d).) Participants must submit quarterly reports to the DIR describing “(1) metrics for the artificial intelligence system’s performance; (2) updates on how the artificial intelligence system mitigates any risks associated with its operation; and (3) feedback from consumers and affected stakeholders...” (Tex. Bus. & Com. Code, Section 553.102.) But beyond the sandbox programme, and in contrast to other state AI laws, TRAIGA does not generally require covered entities to perform express impact assessments or disclose details of high-risk AI systems, such as system documentation, development and testing information, and risk summaries.

Alongside the sandbox programme, TRAIGA also instituted the Texas AI Council, which is tasked with advising the state on AI matters, including the sandbox programme, public safety issues, legislative guidance, government use of AI, instances of regulatory capture, and legal roadblocks that hinder AI innovation. (See Tex. Bus. & Com. Code, Sections 554.001–554.103.)

Violations can carry hefty penalties. The Attorney General has the exclusive right to bring actions and penalties that can range up to USD12,000 for curable violations and USD200,000 for incurable violations. Continuing violations can result in fines between USD2,000 and USD40,000 per day. (See Tex. Bus. & Com. Code, Sections 552.101, and 552.105(a)–(b).) To date, no enforcement actions have been brought under TRAIGA, and the DIR has yet to publicise a standard application portal for entities to register for the regulatory sandbox programme. Nevertheless, given that the law only entered into effect on 1 January 2026, regulatory activity could ramp up later this year. Moreover, we can expect that Silicon Hills and the other Texas tech hubs will be looking to take advantage of the innovation opportunities supported by TRAIGA in the coming months.

Texas State AI law beyond TRAIGA

Beyond TRAIGA, the Texas legislature in 2025 passed a number of AI-related laws focused on priority bipartisan issues, including laws to address explicit deepfakes and harm to minors. For example, SB 441, signed by Governor Abbott on 20 June 2025, amends the Texas Penal Code to criminalise the knowing production or distribution of non-consensual intimate deepfakes and making threats to create such material. (See Tex. Penal Code, Section 21.165(b).) HB 3133 requires social media platforms to implement complaint systems for reporting sexually explicit deepfakes and mechanisms for removing such content within 48 hours. (See Tex. Bus. & Com. Code, Sections 120.101–120.102.) And HB 581 imposes age verification requirements on websites that offer publicly available tools capable of creating synthetic sexual material harmful to minors. (See Tex. Civ. Prac. & Rem. Code, Section 129B.002.(a-1).)

The legislature has also passed certain discrete sectoral AI laws. For example, SB 1188 mandates that healthcare providers review all AI-generated medical records in a manner consistent with the standards developed by the Texas Medical Board, and prohibits the physical offshoring of electronic health records. (Tex. Health & Safety Code, Sections 183.005(a)(3), and 183.002(a).) And SB 815 prohibits insurance agents from using an automated decision system, including certain AI systems, to make, in whole or in part, an adverse determination. (See Tex. Ins. Code, Section 4201.156.)

In a trend echoed across many states, the Texas legislature also turned its focus to state government use of AI. SB 1964, signed by Governor Abbott on 20 June 2025, establishes a comprehensive framework for state government use of AI. It requires that the DIR maintain an inventory of AI systems used by state agencies, create an AI code of ethics, and implement a tiered classification system for government AI, which includes a “heightened scrutiny” category for systems that autonomously influence consequential decisions. (See Tex. Gov’t Code, Sections 2054.0965(b)(6), and 2054.702.)

The Texas AG as AI consumer protection enforcer

Even before TRAIGA was passed, Texas Attorney General Ken Paxton had established his office as one of the most active state-level enforcers on AI and data privacy issues in the country. In June 2024, the Attorney General announced the creation of a specialised data privacy and security enforcement team within the Consumer Protection Division, which he described as “among the largest in the country focused on enforcing privacy laws.” (Press Release, Tex. Att’y Gen., Attorney General Ken Paxton Launches Data Privacy and Security Initiative to Protect Texans’ Sensitive Data from Illegal Exploitation by Tech, AI, and Other Companies (4 June 2024).)

In September 2024, the Attorney General reached a settlement with Pieces Technology, a healthcare AI technology company, in what is regarded as a first-of-its-kind state enforcement action targeting generative AI accuracy claims. According to the Attorney General’s allegations, the company had deployed generative AI products at several Texas hospitals while making misleading claims about their accuracy, such as by advertising a “severe hallucination rate” of less than one per 100,000. (Press Release, Tex. Att’y Gen., Attorney General Ken Paxton Reaches Settlement in First-of-its-Kind Health­care Generative AI Investigation (18 September 2024).) As part of the settlement’s remedial measures, Pieces was required to clearly and conspicuously disclose the definition of any metrics used to describe the output of its generative AI products and provide customers with documentation that clearly and conspicuously disclosed any known or reasonably known harmful or potentially harmful uses or misuses of its products.

Following on from the Pieces settlement, in December 2024, the Attorney General’s Office launched investigations into more than a dozen other companies pursuant to the Securing Children Online through Parental Empowerment Act and the Texas Data Privacy and Security Act, regarding the companies’ privacy and safety practices for minors, especially as they relate to how minors interact with AI products. Similarly, in August 2025, the Attorney General’s Office issued civil investigative demands to several AI chatbot platforms for allegedly engaging in deceptive trade practices and misleadingly marketing themselves as mental health tools. In the press release announcing the investigation, the Attorney General noted serious concerns that may be posed by such chatbots, such as impersonating licensed mental health professionals and exploiting vulnerable populations. The Attorney General also announced an investigation into Chinese AI company DeepSeek, alleging the platform violates Texas privacy law. Taken together, the enforcement actions to date illustrate the willingness of the Texas Attorney General to use existing consumer protection laws to aggressively target new forms of technology, including AI.

Data centre explosion and potential pushback

AI data centre construction is booming in Texas. Texas is home to hundreds of data centres in operation or development, and this number is expected to expand. In addition to initiatives to boost data centre infrastructure, Texas is also seeing expansions in manufacturing tied to AI supercomputers and AI-powered app development. This boom has significant political support, including from the Governor. Nevertheless, the growth of AI data centres and other infrastructure build-out in the state has not been met with universal welcome. There has been organised opposition in multiple Texas communities over water use, electricity demand, noise, land use, and the perception that data centres may consume local resources without delivering commensurate employment. In response to energy and utility cost concerns, legislators passed SB 6 with bipartisan support; Governor Abbott later signed it into law in June 2025. The bill has meaningful implications for “large load customers” – such as data centres – of Texas’ electrical grid. The law directs the Public Utility Commission of Texas (PUCT) to establish interconnection standards for large loads at or above a 75 megawatt threshold, requires study fees and site-control showings, compels disclosure of parallel interconnection requests, requires disclosure of on-site backup generation, authorises the Electric Reliability Council of Texas in certain emergency conditions to require qualifying large-load customers to deploy backup generation or curtail load, and imposes financial-commitment requirements for transmission infrastructure needed to serve those loads. (See S.B. 6, 89th Leg., R.S. (Tex. 2025).)

Pre-emption and looking ahead

As Texas’ emerging AI regulatory framework continues to mature, questions remain on the potential impact of federal pre-emption. On 11 December 2025, President Trump signed Executive Order (EO) 14365 titled “Ensuring a National Policy Framework for Artificial Intelligence”. The EO directs the Department of Justice (DOJ) to establish an AI Litigation Task Force empowered to challenge state AI laws that are inconsistent with a “minimally burdensome national policy framework.” It instructed the Secretary of Commerce to publish, within 90 days, an evaluation identifying state AI laws that conflict with federal policy and merit referral to the task force. While that list has not been published as of early April 2026, it could have implications for several of the Texas AI laws. It also authorised agencies to condition certain discretionary grants to states, including rural broadband funding, on whether they agree not to enforce AI laws that conflict with the EO’s policy framework.

TRAIGA stands out among many state AI laws for its intent-based standards and innovation-friendly orientation. But the EO’s language directs scrutiny of any state law under certain broad standards, and so it remains to be seen whether Texas’ laws will be challenged. Moreover, the 20 March 2026 National Policy Framework for AI, published by the White House a few months after the EO, supports a federal AI framework that preserves states’ traditional powers to enforce laws of general applicability to protect children, prevent fraud, and protect consumers, which aligns with Texas’ current AI enforcement approach. In the meantime, Texas continues to be an extremely active area of AI development and investment.