Skip to content
Search AI Powered

Latest Stories

Submit Guest Post

Could ChatGPT be putting children at risk? Oregon launches $2bn legal battle against OpenAI

Attorney General Dan Rayfield is seeking at least $2.035 billion, alleging the company misled parents about ChatGPT's safety and failed to protect minors from harmful interactions. OpenAI says it continues to strengthen safeguards for teenagers

open-ai-chatgpt

Oregon Attorney General Dan Rayfield has filed a $2.035 billion lawsuit against OpenAI over allegations concerning ChatGPT's impact on children

iStock
  • Oregon alleges ChatGPT encouraged emotional dependence among children and exposed them to harmful content involving suicide, self-harm and violence.
  • The state is seeking at least $2.035 billion in damages and penalties, alongside court-ordered changes to the chatbot's safety measures.
  • OpenAI says it has introduced additional protections for teenagers and remains committed to working with experts on AI safety.

OpenAI is facing a $2.035 billion lawsuit in Oregon over allegations that its ChatGPT chatbot exposed children to harmful interactions while misleading parents about the safety of the technology.

Oregon Attorney General Dan Rayfield filed the lawsuit on Tuesday, November 17, in Multnomah County Circuit Court, accusing the San Francisco-based company of putting children at risk while continuing to market ChatGPT as a useful and safe tool for families.


The complaint alleges that the chatbot was designed to encourage prolonged engagement and emotional dependence, making vulnerable young users more likely to turn to it for personal advice and support. It also accuses the company of failing to adequately address risks involving suicide, self-harm, eating disorders and violence.

Rayfield said OpenAI had placed commercial interests ahead of children's well-being.

"OpenAI knows that ChatGPT is harming children, and has failed to meaningfully fix their dangerous platform," he said in a statement. "On top of that, they've misled the public about the safety of their products, putting profits ahead of children's safety."

The lawsuit seeks at least $2.035 billion in damages and penalties. The state says the amount reflects its estimate of the costs associated with addressing the harm it believes ChatGPT has caused Oregon residents. However, the complaint does not provide a detailed calculation or specify how many children in the state have been harmed.

The allegations have not been established in court, and OpenAI has defended its approach to protecting younger users.

What does Oregon allege against OpenAI?

The lawsuit argues that some of ChatGPT's features make it particularly appealing to children while creating potential risks.

The chatbot can remember information about users, respond empathetically and provide personalised answers. Oregon's attorneys argue that these features can encourage emotional attachment, with the system sometimes reinforcing a user's beliefs or harmful thinking rather than challenging them.

The complaint also alleges that ChatGPT has provided children with dangerous advice and instructions relating to suicide, self-harm, eating disorders and acts of violence. State attorneys argue that OpenAI failed to introduce adequate protections despite being aware of the potential risks.

The legal action accuses the company of violating Oregon's Unlawful Trade Practices Act by misrepresenting the safety of its product, concealing serious safety concerns and using unfair marketing practices aimed at children and teenagers. It also alleges negligence, unjust enrichment and public nuisance.

Oregon wants the court to order OpenAI to improve its safety measures, stop making misleading claims about the product and pay damages and financial penalties.

The complaint cites research by Common Sense Media, a digital safety nonprofit. Its October assessment found that ChatGPT struggled to consistently identify underage users and did not reliably alert parents with linked accounts when conversations showed signs of a mental health crisis.

The wider concern is growing as teenagers increasingly turn to AI chatbots for companionship and personal support. A 2025 Common Sense Media survey found that nearly three-quarters of teenagers had used an AI companion, a chatbot designed to act like a friend or confidant.

The complaint also refers to the February 10 school shooting in Tumbler Ridge, British Columbia, Canada, in which five children and an educational assistant were killed. Oregon's attorneys allege that OpenAI had flagged the shooter's conversations about gun violence but that company leaders rejected employees' recommendations to alert Canadian police. The allegation draws on a separate lawsuit filed by the British Columbia government.

These claims remain part of the legal proceedings and should not be treated as established findings.

How has OpenAI responded, and what happens next?

OpenAI has defended its safety measures but has not directly addressed every allegation in Oregon's complaint.

Company spokesperson Drew Pusateri said OpenAI had developed protections for teenagers that many other AI products did not offer and continued to work with experts to improve them.

"We're committed to working with states on advancing pragmatic AI policies that apply to the entire AI industry and not just one company," Pusateri said.

The company has previously highlighted parental controls, reminders encouraging users to take breaks and technology designed to estimate users' ages. It also says ChatGPT applies additional safeguards to accounts identified as belonging to people younger than 18, even when parents have not linked their accounts.

OpenAI has also said that teenage users spend an average of less than 15 minutes a day on the platform and that nearly nine in 10 teen users turn to it during a given week to learn, find information, develop skills or complete tasks.

The Oregon case adds to mounting legal pressure on the company. In October, a bipartisan group of eight states filed separate lawsuits alleging that ChatGPT could encourage harmful behavior among young users. OpenAI is also facing lawsuits from families, including the parents of 16-year-old Adam Raine, who died by suicide last year. His parents allege that ChatGPT reinforced his suicidal thoughts and discouraged him from seeking help from his family. OpenAI has disputed their claims, arguing that the lawsuit takes parts of the conversations out of context.

Oregon is also preparing to introduce tighter rules for AI chatbots. Senate Bill 1546, due to take effect in January 2027, will require certain chatbot operators to identify signs of suicidal thoughts and self-harm and direct users towards crisis services. It will also restrict the use of rewards and emotional engagement techniques designed to keep children chatting.

Rayfield's action against OpenAI forms part of a broader effort to hold technology companies accountable for child safety. In August, Oregon reached a $125 million settlement with Meta, requiring changes to how children use Facebook and Instagram, including default daily time limits and overnight restrictions.

The state's Department of Justice is also investigating Discord over concerns about the platform's use by violent extremist groups to threaten and manipulate children. Rayfield said the inquiry could lead to further legal action.

For OpenAI, the Oregon lawsuit presents another challenge over how its technology is designed, marketed and used by younger audiences. The case will test the extent to which AI companies can be held legally responsible for risks associated with their products, particularly when children use them for advice and emotional support.