OpenAI has introduced a specially adapted version of ChatGPT called “ChatGPT for Teens”, targeting ages 13 to 17, adding stronger safety measures and parental control tools for minors using artificial intelligence.
According to the American tech giant, the new ChatGPT for Teens is designed to offer an age-appropriate environment, with stricter limits on sensitive content such as self-harm, violence, sexual content, dangerous challenges, and eating disorders.
At the same time, OpenAI is trying to limit the risk of emotional dependence on the chatbot. The teen version is designed to avoid romantic or emotionally dependent interactions, and instead encourages users to seek support from people in their lives when needed. The move comes as OpenAI’s biggest rival, Meta, faces scrutiny in the US over practices accused of being designed in ways that could negatively affect the mental health of children and teenagers.
OpenAI announced its plan to strengthen teen safety on ChatGPT on August 18, in an effort to address growing concerns over how large language models can affect younger users.
A Study Tool, Not an Answer Machine
According to the company’s announcement, particular emphasis is placed on educational use. Through Study Mode, ChatGPT can walk students through problems step by step, helping them understand a concept and work out the solution themselves rather than simply handing them the answer.
The company has also built in reminders about responsible use when working on assignments, aiming to cut down on the chatbot being used to copy homework.
What Parents Can Do
The new experience comes paired with parental controls, letting parents link their own account to their teen’s and manage specific settings.
Among other things, parents can turn on “Quiet Hours,” during which ChatGPT won’t be available on the teen’s account, and can manage features like Voice Mode, image generation, and Study Mode.
Parents don’t get access to their teen’s conversations. However, in limited cases where the system detects serious signs of self-harm or violent behavior, a notification may be sent to the linked parent or guardian.
How the System Identifies Underage Users
OpenAI now uses an age-estimation system that draws on various signals, including how and how often an account is used, as well as the topics a user discusses.
If the system estimates a user is under 18, it can automatically switch on the teen experience, even if a different birth date was entered. If an adult is mistakenly placed into the teen experience, they can verify their age to have the restrictions lifted.
