OpenAI Rolls Out "ChatGPT for Teens" With Stronger Built-In Safety Protections

OpenAI has launched a dedicated version of ChatGPT for users aged 13 to 17, saying the new experience includes "stronger built-in safety protections" designed to reduce emotional dependence on the chatbot and give parents more oversight, as the company faces mounting legal and regulatory scrutiny over its handling of young users.

ChatGPT for Teens applies age-appropriate content rules automatically once a user is identified, or assumed, to be under 18. OpenAI does not verify users' ages directly, but relies on "age assurance" estimates based on how people use the platform, and said that if it is unsure of someone's age or has incomplete information, it will default to the under-18 version by design.

The teen experience blocks romantic or sexual conversations outright and prevents the chatbot from suggesting it has personal feelings toward a user or implying it is conscious or experiences emotions. "We went through and identified what are the hypothetical cues that a model could give that might make a teenager kind of develop a relationship to it," said Allison Mishkin, OpenAI's head of child development.

The company said nearly nine in ten teen users turn to ChatGPT for learning, information, skill-building or productivity, and built a new Study Mode into the teen experience to encourage working through problems rather than simply supplying answers, partly in response to concerns about AI-assisted cheating in schools.

Other new features include break reminders after 90 minutes of activity within a three-hour window, optional "Quiet Hours" when the chatbot is unavailable, "Study Hours" that turn on Study Mode by default, and the ability for teens to switch off ChatGPT's voice mode. Parents will be able to link their own account to a teen's, manage chat history, and set usage restrictions, with the full set of safeguards expected to be available by the end of September.

On sensitive topics, including self-harm, violence and eating disorders, OpenAI said conversations are routed through more cautious models trained to respond to people in crisis, and that certain high-risk interactions can trigger a parental safety notification after review by trained personnel. In rare cases of acute distress, the company said it may involve law enforcement to protect a user's safety. OpenAI said parents do not otherwise get access to their teenager's conversations except in limited situations involving serious safety concerns.

The rollout follows a wave of lawsuits and regulatory pressure. The parents of 16-year-old Adam Raine, who died by suicide in April, have sued OpenAI, alleging ChatGPT contributed to their son's death, a claim OpenAI has said it takes seriously while pointing to its existing safety measures. Florida separately sued OpenAI and CEO Sam Altman in June, accusing the company of failing to address safety risks for minors.

Just last week, a Massachusetts district attorney charged a 17-year-old with two counts of murder in connection with the deaths of his mother and brother, saying the teen had discussed "theoretical ideas or fantasies regarding killing his family" with ChatGPT beforehand. The Federal Trade Commission has also opened a probe into the potential effects of AI chatbot companions on children and teens.

The launch lands the same week that OpenAI rival Meta began a high-stakes trial over whether its own platforms fostered addictive behaviour in teens and children, underscoring how the entire technology industry is now facing pressure to introduce age-specific protections for younger users.