Texas Attorney General sues OpenAI and Sam Altman under the DTPA and the SCOPE Act over ChatGPT safety claims

Key takeaways

• The petition is dated 23 October 2026 and names seven defendants: six OpenAI entities and Sam Altman.
• Count I invokes the DTPA; Count II invokes the SCOPE Act, which the petition says is enforceable as a DTPA violation.
• Texas states civil penalties of up to $10,000 per violation and asks for $10,000 for each violation, but states no aggregate penalty calculation; it separately pleads monetary relief over $1,000,000.
• Among the injunctions sought is an order barring OpenAI from offering ChatGPT to minors in Texas.

What is the Texas Attorney General alleging against OpenAI and Sam Altman?

The press release says Attorney General Ken Paxton sued Open AI Global, LLC and its affiliated entities, including CEO Sam Altman. It alleges that, despite marketing itself as safe and reliable, OpenAI "failed to protect Texans, including children" from harms of relying on ChatGPT. The release also says "OpenAI misrepresented their artificial intelligence products and has given a careless introduction of ChatGPT to Texas and the world."

The petition, titled "Texas's Original Petition", names OpenAI Global, LLC; OpenAI Foundation (f/k/a OpenAI, Inc.); OpenAI OpCo, LLC; OpenAI Group PBC; OpenAI Holdings, LLC; OpenAI, LLC; and Sam Altman. It is captioned for a district court in Collin County, Texas, and the signature block is dated 23 October 2026. Its marketing allegations point to OpenAI's public statements that safety is "core to our mission" and to safety messaging aimed at parents of teenagers. It also alleges that ChatGPT is unreliable, citing among other things attorneys sanctioned after relying on ChatGPT research. The petition says Texas pleads "upon personal knowledge as to all public statements" and "upon information and belief as to all other matters". That is a pleading, and OpenAI and Mr. Altman have not been found to have done any of it in the sources we opened.

Which claims and relief does the petition seek?

Count I alleges false, misleading and deceptive acts under sections 17.46(a) and (b) of the DTPA, against all defendants. Its list of alleged practices includes failing to warn of ChatGPT's dangers, offering a product without adequate safety testing, and offering a product that "functions unreliably while falsely representing that reliability". It also lists a product that falsely represents itself as having human emotions and characteristics. Count II alleges violations of the SCOPE Act, discussed below.

On relief, Texas asks for temporary and permanent injunctions and for civil penalties stated at up to $10,000 per violation of the DTPA; the prayer itself asks for $10,000 for each violation. It also asks for disgorgement of defendants' assets, interest, costs of court and investigation, and attorneys' fees, and it requests a jury trial. One requested injunction would bar releasing any new AI model to Texas users unless an independent third-party evaluator approved by the court has reviewed its safety guardrails and approved its release. Another would bar offering ChatGPT to minors in Texas. These are requests. The petition pleads monetary relief over $1,000,000 but gives no aggregate civil-penalty calculation.

On Mr. Altman, the petition says Texas "seeks to hold Altman personally liable" for harm through his direct involvement as founder and CEO and for deceptive statements he made personally. It alleges he "has personally directed the design, development, safety policies, and deployment of ChatGPT as used in Texas". Florida has a separate suit that also reaches Mr. Altman personally; see our article on the Florida suit.

What does the petition say about minors and the SCOPE Act?

Count II alleges that OpenAI is a "digital service provider" under the SCOPE Act and violated Sections 509.051, 509.052, 509.101 and 509.054. The petition says Section 509.151 makes a SCOPE Act violation a deceptive act actionable by the Attorney General under the DTPA, and it pleads each as a per se violation of Section 17.46(a).

The factual allegations are specific. The petition says the free version of ChatGPT has no age verification and that the paid tier asks for an age but cannot verify it. It says parental controls created in September 2025 do not require a child's account to be linked to a parent's, and that Defendants "do not request, nor verify, any information related to a purported parent or guardian's identity or relationship to a known minor". It alleges Defendants fail to provide verified parents with tools to limit the sharing or sale of a minor's personal identifying information or to control targeted advertising to a known minor. OpenAI's answers to these points are not in what we opened.

What should AI vendors and the counsel who advise them do now?

The petition requests relief against the named defendants only; it imposes no obligation on anyone, and a petition is not a ruling. For companies that run or embed consumer chatbots, the useful reading is how the Texas AG frames claims: public safety and reliability statements treated as DTPA representations, and age and parental features measured against SCOPE Act sections. Everything below is our suggestion, not a stated requirement.

• List every public statement about safety, reliability, accuracy and suitability for teens, and note what evidence supports each.
• Check how your product handles known minors: age signals, parental verification, parental tools, and advertising or data-sharing settings.
• Where the product tells users it can research or advise, check that its limits are disclosed where users see them.
• Watch the docket. Because the cause number is blank on the copy we opened, confirm the filed case number and any answer before relying on this article.

Which dates and amounts appear in the petition?

What we did not verify

What we opened on 25 October 2026 (UTC): the Texas Attorney General's press release dated 23 October 2026, and the full text of Texas's Original Petition from the link on that release.

What we did not open: the filed docket or file-stamped copy (the petition we read shows a blank cause number and judicial district), any OpenAI or Altman response, any service or hearing information, the sources the petition cites, and the underlying statute text beyond what the petition quotes. We did not check the petition's factual claims, such as the study results or news reports it cites, against their originals.

What we refuse to claim: that any allegation is true, that either OpenAI or Mr. Altman violated any law, that a court has granted any relief, or what penalty, if any, could result.

Related briefings

• AI Regulation News hub
• North America AI regulation tracker
• Attorneys and AI regulation
• Florida suit naming OpenAI and Sam Altman personally
• Texas AG action over Pieces Technologies health AI claims
• Texas AG biometric settlement with Meta under CUBI
• Texas AG privacy and biometric settlement with Google