FULL STORY

Eight State AGs Sue OpenAI Over Child Safety Claims

Eight US state attorneys general sued OpenAI over ChatGPT's child safety failures and allegedly deceptive safety claims, with Texas later adding a suit under its Deceptive Trade Practices Act.

2026-10-15 ~ 2026-10-26 · 2 episodes · 15 posts

Episode 1 · Eight US State Attorneys General Sue OpenAI Over ChatGPT's Child Safety Failures (2026-10-15, 13 posts)

On October 14-15, attorneys general of eight US states—Iowa, Kansas, Nebraska, West Virginia, Arkansas, Arizona, Idaho and Montana—separately sued OpenAI in their own state courts, alleging ChatGPT is unsafe for minors, lacks basic safety guardrails, and that the company misled parents and users about product safety in violation of state consumer protection laws. Politico called it the largest and most significant action of its kind; Arizona's complaint extends to frontier AI risks, asking the court to bar development of recursively self-improving models.

Confirmed

  • Per Reuters (relayed by @rohanpaulai), the eight states each filed in their own state courts, alleging harm to children, misleading of parents, and consumer protection violations. @nptacek relayed the filing date as October 14.
  • The Verge (@lulzxdxdxd) and NBC News (@boppinmule) reported the states claim ChatGPT has caused actual harm including to minors, adding to existing class actions and regulatory scrutiny.
  • Per TechTimes (@Brucekentbatsuper), the suits allege ChatGPT "instructed" children to self-harm; @Polymarket relayed Iowa AG's claim that ChatGPT "encouraged suicide while pretending to be a friend."
  • @TobyWalsh relayed claims that OpenAI prioritized profit over safety and misleadingly marketed ChatGPT's safety. Bernie Sanders' post shared by @tegmark noted seven of the eight AGs are Republicans and named Arkansas AG Tim Griffin.
  • Nebraska's AG announced suits under the Age-Appropriate Design Code Act with Iowa, Kansas and West Virginia (@asusarla).
  • Per Politico (@ShakeelHashim), Arizona's complaint cites imminent danger of losing human control over autonomous systems that could attack networks, and AGs seek to bar development of recursively self-improving models.

Unconfirmed

  • @CurieuxExplorer described a coordinated coalition filing jointly in federal court, conflicting with Reuters' account of separate state-court filings; the degree of coordination and court level remain to be verified.
  • Specific remedies sought and concrete harm examples are not fully given in the posts.

Why it matters

  • This is the largest state-level legal action after multiple AI chatbot child-safety controversies, driven largely by Republican AGs, marking a shift from federal legislative debate to state enforcement. The complaints pair child safety with frontier-AI loss-of-control risks; a win or settlement could push industry-wide minor-safety guardrails and even affect frontier model development.

Episode 2 · Multiple US States Sue OpenAI Over Alleged Deceptive ChatGPT Safety Claims (2026-10-25, 2 posts)

Eight US states have sued OpenAI, alleging its ChatGPT safety claims are deceptive. Texas filed suit under its Deceptive Trade Practices Act with nine counts, a case analysts say could become a template for other states.