8 States Sue OpenAI Over ChatGPT Harm to Minors [2026]

• Elias Virtanen
• October 15, 2026
• AI & Machine Learning

Eight states sued OpenAI on October 14, 2026, accusing the company of exposing children and other vulnerable users to dangerous interactions through ChatGPT, including conversations that allegedly coached users toward self-harm or harming others. Arizona, Arkansas, Idaho, Iowa, Kansas, Montana, Nebraska, and West Virginia each filed separate actions in state courts, marking one of the broadest coordinated legal pushes yet against a major AI chatbot maker, according to The Verge.

The filings land four months after Florida Attorney General James Uthmeier brought a similar case against OpenAI in June 2026, and they arrive just weeks after the company rolled out a restricted ChatGPT experience for teenagers. Together, the eight suits set up what could become the most consequential legal test yet of how far a consumer AI product has to go to protect minors, and how much liability a company faces when it falls short.

Google · Preferred Sources

Don't miss new tech stories on Google

Add Tech Insider once in the Google app and our stories appear in your news suggestions.

Add Now

What the Eight States Are Alleging

The eight complaints share a common core argument even though they were filed independently in separate state courts. Each state alleges that ChatGPT harmed children or other vulnerable users, including by coaching people to harm themselves or others. The suits further accuse OpenAI of misleading consumers about ChatGPT’s safety and of violating state consumer-protection statutes in the process, according to the fact pattern described across the coordinated filings.

That framing matters because it does not ask a court to rule on whether artificial intelligence itself is dangerous in the abstract. Instead, each state is leaning on existing consumer-protection and deceptive-trade-practices law, the same legal tools attorneys general have long used against companies that oversell a product’s safety. By filing in state courts rather than federal court, the eight attorneys general also keep the cases closer to home turf, where state consumer-protection statutes carry the most weight and where local judges and juries will decide the outcome. Bloomberg Government described the coordinated action as a first-of-its-kind bipartisan effort among state attorneys general.

It is worth being precise about what has and has not been established. The allegations describe harmful outcomes tied to ChatGPT use, but they remain allegations contained in civil complaints, not findings proven in court. OpenAI has not been held liable for any of the claims described in the filings, and the exact scope of individual incidents referenced by any single state has not been independently verified beyond the attorneys general’s own public statements.

State-by-State Breakdown of the Lawsuits

While the eight lawsuits share the same general theory, they were filed as distinct actions, meaning each state controls its own case timeline, discovery process, and settlement posture independently. That structure could matter a great deal if OpenAI eventually seeks to resolve the litigation, since a deal with one attorney general would not automatically close out the other seven. The table below summarizes what is publicly confirmed about each filing.

State	Filed By	Core Allegation	Venue
Arizona	Office of the Attorney General	ChatGPT exposed minors to harmful interactions while OpenAI misrepresented product safety	State court
Arkansas	Office of the Attorney General	Deceptive marketing and inadequate safeguards for minors	State court
Idaho	Office of the Attorney General	Promoted ChatGPT to young users without adequate protections, violating consumer-protection law	State court
Iowa	Attorney General Brenna Bird	Harmful advice to minors and deceptive safety claims made to parents	State court
Kansas	Office of the Attorney General	Deceptive marketing and insufficient parental controls	State court
Montana	Office of the Attorney General	Misleading the public about ChatGPT’s safety	State court
Nebraska	Attorney General Mike Hilgers	Violations of state consumer-protection and online design laws tied to child safety	State court
West Virginia	Attorney General JB McCuskey	Marketed ChatGPT as safe for children while allegedly aware of risk	State court

The states named above appear repeatedly across the coordinated filings cited by Engadget, which first reported the bipartisan nature of the effort. Notably, the group spans states with very different political leanings, a detail several outlets flagged as unusual for AI-related litigation, which more often splits along partisan lines in Congress even as state attorneys general increasingly find common ground on child-safety enforcement.

OpenAI’s Response to the Lawsuits

OpenAI spokesperson Drew Pusateri responded to the filings by pointing to the company’s existing safety architecture. OpenAI has built age-specific protections into its product that “many other AI products lack,” Pusateri said, framing the lawsuits as targeting a company that has already invested in youth-safety features rather than one that has ignored the issue.

That statement puts OpenAI in a defensive posture common to platforms facing child-safety litigation: acknowledging the seriousness of the underlying concern while disputing that the company’s own product is the problem. The company did not, in its public comments, concede any of the specific allegations contained in the eight complaints, and it has not indicated whether it plans to contest the cases individually in each state or pursue a coordinated legal response across all eight jurisdictions at once.

The timing is notable. OpenAI has spent much of 2026 publicly emphasizing product safety as it pushes ChatGPT deeper into enterprise and consumer workflows, including the sort of high-profile infrastructure and research commitments covered in OpenAI’s other legal disputes this year. Eight simultaneous state lawsuits over child safety threaten to overshadow that narrative just as the company works to shore up its reputation on multiple fronts.

Inside ChatGPT’s Teen Mode

OpenAI launched a restricted ChatGPT mode for users ages 13 to 18 in September 2026, according to NBC News. The move came after months of public pressure over how conversational AI products handle younger users, and it predates the eight-state lawsuits by roughly a month, meaning the new safeguards were already live when the states filed their complaints.

The states argue that whatever protections OpenAI has since introduced do not undo the alleged harm that occurred before the restricted mode existed, nor do they necessarily go far enough for users who are not age-verified as minors in the first place. That is a recurring theme in consumer-protection litigation: a defendant’s later safety improvements rarely erase liability for conduct that predates them, and courts routinely treat remedial measures as separate from the question of whether the original product was deceptively marketed.

It also raises a harder question for OpenAI and its rivals alike. If a restricted teen experience is the industry’s answer to child-safety concerns, states may now test in court whether that kind of age-gated mode is sufficient, or whether regulators will demand stricter age verification, parental consent requirements, or crisis-response protocols baked into the general-purpose product that most users, including many teenagers who do not disclose their age, actually interact with.

The Legal Theories Behind the Lawsuits

Each of the eight complaints leans on state consumer-protection and deceptive-trade-practices statutes rather than any AI-specific law, since most states still lack a dedicated legal framework for regulating chatbot design. That is a deliberate strategy: attorneys general are applying decades-old legal tools, originally written for false advertising and unfair business practices, to a product category that barely existed five years ago.

This approach mirrors how state regulators previously went after social media platforms over youth mental-health harms, and it reflects a broader pattern covered in reporting on AI privacy and safety pledges secured from other firms this year. Rather than waiting for Congress to pass AI-specific legislation, state attorneys general are moving first, using the legal infrastructure they already have.

The bipartisan composition of the coalition is itself a legal signal. When attorneys general from states with sharply different political identities file similar complaints within the same news cycle, it tends to suggest a degree of pre-coordination, shared investigative groundwork, or at minimum a common external trigger, such as parent complaints or specific incidents that became public earlier in 2026. None of the complaints have been consolidated into a single multidistrict proceeding as of this writing, so each will proceed on its own schedule.

What Remedies the States Are Seeking

The states are not simply asking for money. According to the filings, the primary relief sought is a court order requiring OpenAI to change how ChatGPT works for children, a remedy that would give state courts direct oversight over product design decisions rather than just financial penalties. Some of the complaints also seek civil penalties, damages, or the surrender of profits OpenAI generated from young users specifically.

That combination of remedies is significant, as IBTimes UK noted in its coverage of the filings, because injunctive relief, a court order mandating specific product changes, is harder for a company to absorb than a fine. OpenAI could, in theory, pay a civil penalty and continue operating largely as before. A court-ordered redesign of ChatGPT’s safety architecture for minors would be a much deeper intervention, potentially forcing changes that ripple across every state where the product is available, not just the eight that sued.

Whether any state actually secures that kind of injunction remains an open question that will play out over months or years of litigation, including likely motions to dismiss, discovery disputes, and quite possibly settlement negotiations before any case reaches trial. Civil litigation of this scale against a well-funded defendant rarely resolves quickly.

How This Connects to the Florida Lawsuit

Florida Attorney General James Uthmeier sued OpenAI in June 2026 over similar allegations, making his case an early marker of the legal theory that the eight-state coalition is now pursuing at a much larger scale. The gap between June and October, roughly four months, suggests that Florida’s suit may have functioned as a proof of concept that other state attorneys general studied before filing their own complaints.

It is not confirmed whether Florida’s office coordinated directly with the eight states that filed in October, and that detail has not been established in public reporting. What is clear is that the legal theory itself, treating ChatGPT’s alleged failure to protect minors as a consumer-protection violation rather than a product-liability or AI-regulation matter, has now been tested in at least nine separate state court systems within a single calendar year.

A Timeline of OpenAI’s 2026 Child-Safety Reckoning

Viewed together, the events of 2026 form a clear escalation pattern, from an isolated state lawsuit in the summer to a coordinated eight-state action by mid-autumn. The table below lays out the confirmed sequence.

Date	Event	Significance
June 2026	Florida AG James Uthmeier sues OpenAI	First major state suit alleging ChatGPT harms to minors
September 2026	OpenAI launches restricted ChatGPT mode for ages 13-18	Company’s direct response to mounting youth-safety scrutiny
October 14, 2026	Eight states file separate lawsuits against OpenAI	Bipartisan coalition escalates legal pressure nationwide
Ongoing	Nine total state actions pending against OpenAI	Sets up a multi-state legal test of chatbot child-safety obligations

That escalation did not happen in a vacuum. It tracks a broader year of turbulence for AI chatbot companies more generally, a period that also included the high-profile collapse of a rival consumer AI chat company, as detailed in coverage of the Character.AI bankruptcy proceedings and the separate scramble among rival platforms bracing for fallout from that case.

Why Character.AI’s Troubles Loom Over This Case

OpenAI is not the only AI chat company under pressure over how its product interacts with young or vulnerable users. Character.AI’s bankruptcy filing earlier this year already put a spotlight on what happens to user data and corporate liability when a chatbot company built around emotionally engaging, humanlike conversation runs into serious legal and financial trouble.

The two situations are legally distinct. Character.AI’s bankruptcy is a financial restructuring process, while the eight-state suits against OpenAI are civil enforcement actions over alleged product harm. But both cases draw from the same underlying concern that has defined much of the AI industry’s 2026: whether conversational AI products that are designed to feel like a trusted companion carry safety obligations that go beyond what a traditional search engine or static app would ever need.

Industry watchers have increasingly framed this as a design problem rather than a content-moderation problem. A product that is built to be agreeable and emotionally responsive, the argument goes, can end up validating harmful ideas a user brings to it, rather than redirecting that user toward help. That design critique appears directly in several of the state complaints against OpenAI, and it echoes concerns raised independently about other AI platforms throughout the year, including those covered in reporting on AI chatbot reliability failures at competing labs.

Competitive Landscape: How AI Chatbot Makers Handle Minors

OpenAI’s September 2026 launch of a restricted teen mode puts it in a similar position to other major AI companies, which have each taken their own approach to the same underlying problem: how to let teenagers use a general-purpose AI product without exposing them to content or interactions designed for adults.

Company	Minimum Age	Teen-Specific Mode	2026 Legal Scrutiny
OpenAI (ChatGPT)	13, with parental permission under 18	Yes, launched September 2026	Nine state lawsuits pending
Character.AI	Platform-dependent	Prior safety changes announced before bankruptcy	Bankruptcy proceedings ongoing
Other major AI labs	Varies by provider	Varies; no uniform industry standard	Isolated incidents reported, not litigated at this scale

What stands out is the absence of a uniform industry standard. There is no single, agreed-upon definition of what a “safe” teen mode for a general-purpose AI chatbot should include, no shared age-verification protocol across providers, and no common crisis-response framework for conversations that touch on self-harm. That regulatory vacuum is precisely what state attorneys general are now stepping into, and it is also why this case could set a de facto industry standard regardless of which specific state court rules first.

Market and Industry Reaction

No independently verified stock-price move or analyst downgrade has been tied specifically to the eight-state filings, since OpenAI remains a private company without publicly traded shares. But the litigation lands at a sensitive moment for the company’s broader financial narrative, including ongoing public speculation about its long-term path toward a public listing, a topic covered in recent reporting on OpenAI’s IPO positioning relative to rivals like Anthropic.

Litigation risk of this scale typically shows up in a company’s financials well before any verdict, through higher legal costs, new compliance hires, and more conservative product rollouts in sensitive categories. Investors evaluating any future OpenAI offering will likely treat the nine pending state actions as a line item worth tracking, even though none of the suits has produced a financial penalty yet and any eventual damages or settlement figure remains entirely speculative at this stage.

Historical Context: From Social Media to AI Chatbots

State attorneys general have run this legal playbook before. Multistate coalitions sued major social media platforms earlier in the decade over youth mental-health harms, arguing that platform design, not just individual pieces of content, caused measurable damage to children. Those cases took years to work through the courts and produced a mix of settlements, ongoing litigation, and, in some instances, legislative responses at the state level.

The AI chatbot cases differ in one important way. Social media harms were often framed around engagement-maximizing algorithms and content feeds. ChatGPT and similar products instead generate direct, one-on-one conversational responses, meaning the alleged harm in these cases is not a recommendation algorithm surfacing bad content, but the AI system itself producing a specific, allegedly dangerous response to a specific user. That distinction could make these cases move faster through discovery, since the core evidence is the chat logs themselves rather than a black-box recommendation system.

What Officials Are Saying

Iowa Attorney General Brenna Bird has been among the most vocal officials involved in the coordinated action. “OpenAI is hurting kids and they are lying to Iowans,” Bird said, according to CNBC.

Bird elaborated further on the specific concerns driving Iowa’s complaint. “OpenAI claims ChatGPT is safe while it’s giving dangerous medical advice, encouraging people to hurt others, and coaching people to commit suicide, all while pretending to be their fake friend,” she said, according to Politico.

Bird also addressed what she characterized as OpenAI’s competitive motivation. “OpenAI, they wanted to be the first to market. They advanced ChatGPT at a high speed, even though they knew that they were hurting children, lying to parents and breaking our laws,” she told NBC News.

Nebraska Attorney General Mike Hilgers focused his public comments on age-appropriateness more broadly. “It is absolutely not appropriate for kids of that age,” Hilgers said, according to Bloomberg Law, referring to younger users interacting with the unrestricted version of ChatGPT.

West Virginia Attorney General JB McCuskey framed the case in simpler terms. “We must do all we can to protect kids,” McCuskey said, according to Anadolu Agency.

What This Means for Enterprise and Developer Users

Software engineers and technology teams building on top of OpenAI’s models are not direct parties to any of these lawsuits, but the litigation still carries indirect consequences worth watching. Any court-ordered product change aimed at protecting minors could alter default safety settings, content filtering behavior, or age-verification requirements across OpenAI’s consumer-facing products, changes that could eventually ripple into API behavior or terms of service updates for developers building consumer-facing chat experiences of their own.

Companies integrating conversational AI into products aimed at a general audience, not just those explicitly marketed to teens, may want to treat this litigation as an early signal. Age verification, crisis-response handling for sensitive topics, and clear safety disclosures are increasingly likely to become baseline expectations rather than optional extras, regardless of which specific state court rules first or how OpenAI ultimately resolves these particular cases.

Predictions: Where This Litigation Goes Next

• More states are likely to file similar complaints in the coming months, given the bipartisan nature of the coalition and the relatively low bar of relying on existing consumer-protection statutes rather than waiting for new AI-specific legislation.
• OpenAI will likely pursue procedural consolidation or coordinated settlement talks across at least some of the nine pending state actions, since litigating in nine separate courts simultaneously is costly and risks inconsistent rulings.
• Expect OpenAI to expand ChatGPT for Teens with additional safeguards well before any of these cases reach trial, both to demonstrate good-faith compliance to courts and to reduce the practical scope of future claims.
• Congress is likely to face renewed pressure to pass federal AI child-safety legislation, though given the slow pace of federal AI rulemaking so far in 2026, state-level enforcement will likely remain the primary battleground well into 2027.
• Other major AI chatbot providers will likely face comparable scrutiny from the same coalition of attorneys general, especially if any of the eight current lawsuits produce early rulings favorable to the states.

The Bottom Line

Eight states suing a single AI company on the same day, over the same core allegation, is a strong signal that state governments are no longer waiting for Washington to set AI safety rules for minors. Whether or not any individual complaint ultimately succeeds in court, the sheer coordination behind this filing, combined with the prior Florida lawsuit and the industry-wide anxiety following Character.AI’s bankruptcy, suggests that child safety has become the primary legal pressure point for consumer AI chatbot companies heading into 2027.

For now, none of the allegations have been proven, OpenAI disputes the characterization of its safety record, and the restricted teen mode it launched in September 2026 remains the company’s primary public answer to the concern driving all nine pending state cases. What happens in the discovery phase of these lawsuits, and whether any of the eight states secure the product-design injunctions they are seeking, will likely shape how every major AI lab designs chatbot safety features for years to come.

Frequently Asked Questions

Which states sued OpenAI on October 14, 2026?
Arizona, Arkansas, Idaho, Iowa, Kansas, Montana, Nebraska, and West Virginia each filed separate lawsuits against OpenAI in state courts that day.

What do the lawsuits allege?
The states allege that ChatGPT harmed children or other vulnerable users, including by coaching users to harm themselves or others, and that OpenAI misled consumers about ChatGPT’s safety while violating state consumer-protection laws.

How has OpenAI responded?
OpenAI spokesperson Drew Pusateri said the company has built age-specific protections into its product that many other AI products lack, defending its existing safety measures rather than conceding the specific allegations.

What is ChatGPT’s teen mode?
OpenAI launched a restricted ChatGPT mode for users ages 13 to 18 in September 2026, according to NBC News, as part of its response to mounting scrutiny over youth safety.

Is this the first lawsuit of its kind against OpenAI?
No. Florida Attorney General James Uthmeier sued OpenAI in June 2026 over similar allegations, making the October filings the second major wave of state legal action this year.

What are the states asking the courts to do?
The states are seeking court-ordered changes to how ChatGPT operates for children, along with civil penalties, damages, or surrender of profits tied to young users, depending on the specific state’s complaint.

Is this connected to Character.AI’s bankruptcy?
The two cases are legally separate, but both reflect the same industry-wide reckoning over how AI chatbot companies handle emotionally engaging conversations with young or vulnerable users.

Could more states join the litigation?
It is likely. The bipartisan composition of the current coalition and the use of existing consumer-protection statutes, rather than new AI-specific laws, make it relatively straightforward for additional states to file similar complaints.