On January 1, 2026, Texas became the first state with a broad AI governance law on the books. The Texas Responsible Artificial Intelligence Governance Act — TRAIGA for short — applies to anyone who builds or uses AI systems in Texas. That includes law firms. If your firm uses ChatGPT, Claude, or any AI-powered research or drafting tool, this law applies to you, even if you've never heard of it.
What Is TRAIGA?
TRAIGA started as House Bill 149. Governor Greg Abbott signed it on June 22, 2025, and it took effect January 1, 2026. It's the most comprehensive AI law passed by a U.S. state so far, and it sets rules for how AI systems can be built and used in Texas.
The law defines an 'artificial intelligence system' broadly: any machine-based system that takes inputs and generates outputs — content, decisions, predictions, or recommendations — that can affect the real world. That definition covers general AI assistants, AI legal research tools, AI drafting software, and AI-powered intake chatbots.
Why This Applies to Your Law Firm
TRAIGA splits responsibility between two roles: 'developers' (companies that build AI systems) and 'deployers' (companies that use them). Most law firms are deployers. If you use any of the following, you're a deployer under TRAIGA:
- →Generative AI for drafting documents, letters, or contracts
- →AI legal research tools or research copilots
- →AI-powered client intake chatbots
- →AI tools for e-discovery or document review
- →AI tools for marketing or client communications
In other words, if your firm uses any of the tools we cover on this site — ChatGPT, Claude, Spellbook, or similar — you're a deployer.
What TRAIGA Actually Prohibits
TRAIGA bans developing or deploying AI for a specific list of harmful purposes. For a law firm, most of these are things you'd never do anyway, but it's worth knowing the list:
- →Behavioral manipulation that causes someone harm
- →Unlawful discrimination against a protected class
- →Creating or distributing illegal deepfakes or child sexual abuse material
- →Infringing on a person's constitutional rights
The discrimination provision is the one most relevant to law firms — if a firm uses AI to screen job applicants, evaluate clients, or make any decision that affects people differently based on a protected characteristic, that use needs to be reviewed.
Disclosure: What's Required, What's Just Good Practice
TRAIGA's mandatory disclosure rules are narrower than the headlines suggest. Government agencies must tell people when they're interacting with an AI system. Healthcare providers must disclose when AI is used in a patient's care.
Law firms aren't required to disclose AI use under TRAIGA. But several state bar ethics opinions already recommend telling clients when AI plays a meaningful role in their matter, and disclosure is a simple way to avoid a difficult conversation later if a client asks. Treat it as good practice rather than a legal requirement.
Enforcement and Penalties
TRAIGA doesn't create a private right of action — a client can't sue your firm directly under this law. Only the Texas Attorney General can enforce it. If the AG's office finds a violation, the law gives a 60-day cure period to fix it before penalties apply. After that, civil penalties can reach up to $200,000 per violation for the most serious cases.
For most law firms, the realistic risk isn't a six-figure fine — it's an AG inquiry that takes time and legal fees to respond to, even if the firm ultimately did nothing wrong. Having a basic AI use policy in place is the simplest way to reduce that risk.
A Basic Compliance Checklist
- →Write down which AI tools your firm actually uses. Most firms underestimate this — it usually includes more than just ChatGPT.
- →Check each tool isn't used for a prohibited purpose (see the list above), especially anything touching hiring or client screening.
- →Add a short AI section to your engagement letter or client intake materials noting that AI tools may assist with research or drafting, with all work reviewed by an attorney.
- →Keep a record of vendor terms for any AI tool that processes client data, so you know where that data goes.
- →Designate someone to keep an eye on AG guidance as TRAIGA enforcement practices develop in 2026.
Frequently Asked Questions
Does TRAIGA only apply to companies based in Texas?
No. TRAIGA applies based on where the AI system is developed or deployed, not where the company is headquartered. A firm based outside Texas that serves Texas clients or has a Texas office should review whether its AI use falls under the law.
Does using ChatGPT to draft a document count as 'deploying an AI system'?
Under TRAIGA's broad definition, yes — using a generative AI tool to produce drafts, summaries, or recommendations makes your firm a deployer. That doesn't mean it's prohibited. It means the firm should know it's covered and avoid the prohibited uses listed above.
Do clients need to sign something about AI use?
TRAIGA doesn't require it. Some firms add a short disclosure to engagement letters anyway, as a transparency measure and to align with bar association guidance on AI use.
How likely is a small firm to face an enforcement action?
Low, in practice — enforcement is handled by a single state agency (the Texas AG), and the law focuses on the most harmful AI uses. The bigger practical benefit of reviewing TRAIGA is that it gives a firm a reason to write down its AI policy, which helps regardless of whether the AG ever comes calling.
See how the AI tools your firm uses stack up on accuracy, pricing, and features
Editorial note: AI For Legal Research publishes independent content. We do not accept payment for editorial coverage or review scores. Nothing on this site constitutes legal advice. Always consult a qualified attorney for legal matters.