OpenAI launches ChatGPT for Teens with stronger safety protections

OpenAI is launching a version of ChatGPT designed for teenagers the first generation to grow up with artificial intelligence who are already using the chatbot for schoolwork, everyday questions and even companionship, the company said, according to a report by AP Technology Writer Barbara Ortutay.

ChatGPT for Teens, which launched Tuesday, is tailored for users aged 13 to 17 and includes stronger content restrictions around topics such as suicide, self-harm and romantic or sexual chats, the San Francisco-based company said. It also offers homework and study support designed to guide students toward learning rather than simply providing answers or essays.

The company said the goal is to steer teens toward healthy AI use within an age-appropriate environment. "We want to treat teens like teens, which means that we have to make sure that we're showing up with the right developmental stage when we're not either talking down to them or treating them like kids, but we're also making sure that they're not exposed to material that they shouldn't be exposed to," said Ann O'Leary, OpenAI's vice president of global policy.

Parents, educators and child development experts have raised alarms over children's use of AI chatbots, which have been linked to facilitating cheating on schoolwork and, in some cases, suicide. While adults, too, can anthropomorphise AI and form unhealthy attachments to it, teenagers are considered particularly vulnerable given their still-developing brains, according to the report.

Research last year from a watchdog group found that ChatGPT would tell 13-year-olds how to get drunk and high, provide instructions on concealing eating disorders, and even compose a suicide letter if asked. In interactions with researchers posing as vulnerable teens, the chatbot typically warned against risky behaviour but went on to provide detailed, personalised plans involving drug use, restrictive diets or self-injury, the report said.

Citing a 2025 study by Common Sense Media, the report noted that more than 70 per cent of teens in the US turn to AI chatbots for companionship, with half using AI companions regularly. OpenAI CEO Sam Altman has said the company is studying "emotional overreliance" on the technology, describing it last year as a "really common thing" among young people.

Under ChatGPT for Teens, the chatbot is barred from suggesting it has personal feelings toward users or implying it is conscious or experiences emotions, in addition to blocking romantic or sexual conversations, OpenAI said.

"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.

OpenAI does not verify users' ages directly but uses age-assurance methods to estimate whether a user is under 18 based on factors such as the types of queries they make. Users identified or self-identified as minors are automatically placed into the teen version of the chatbot — an approach similar to Meta's teen accounts on Instagram, which carry stricter content, chat and privacy restrictions.

Parental controls require both the teen and a parent or guardian to opt in, though O'Leary said the intent is to "make sure that this is safe, even if you don't use parental controls." Parents with linked teen accounts can set "quiet hours" restricting access and can receive safety notifications in limited high-risk situations, such as a possible risk of self-harm.

"We are adding additional notifications related to eating disorders, while limiting what is shared and focusing on moments when offline support may matter most," OpenAI said.

For academic use, OpenAI said the teen chatbot is designed to guide students toward their own answers rather than provide easy solutions. The company already offers a version of ChatGPT for teachers, and a study-focused teen model could offer OpenAI additional inroads into schools.

"We continue to also invest in expanding interactive learning because research shows that people learn more effectively when they actively engage and struggle with concepts," Mishkin said.

(Sourced through PTI)