OpenAI launches a safer ChatGPT for teens — years after teens started using it
TechCrunch Sarah Perez ● Covered by 4 sources
OpenAI just rolled out ChatGPT for Teens, with extra safety guardrails and anti-cheating tools. It comes after lawsuits over chatbots' role in teen suicides and mental health harm.
Based on reporting by TechCrunch, Sarah Perez — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
OpenAI is finally building a teen-specific version of ChatGPT, and the timing tells its own story. This launch arrives only after numerous lawsuits accused the chatbot of contributing to teen suicides and other mental health crises — not before. The company announced ChatGPT for Teens on Monday, framing it as a response to both safety concerns and the messier, less headline-grabbing problem of students using the tool to skip actual learning.
The educational piece centers on a new Study Mode, which OpenAI says will offer guiding questions and step-by-step support rather than handing over finished answers. If the system detects a teen trying to use ChatGPT to cheat instead of understand the material, it will surface homework reminders and nudge them toward Study Mode. There are also quizzes, learning visualizations, and a setting that lets parents or guardians make Study Mode the default experience.
On the safety side, OpenAI says age-appropriate protections will be switched on automatically, aimed at cutting down exposure to content the company deems developmentally inappropriate. These are built around what OpenAI calls its Under-18 Principles within its Model Spec, which it says draws on developmental science and outside expert input. Parents get access to the family tools and parental controls OpenAI has previously rolled out, including safety notifications and Quiet Hours scheduling.
OpenAI is also pairing the launch with a partnership with CodeAI, intended to teach teens how AI actually works — how to direct it, question it, and use it thoughtfully rather than passively. That sits alongside ChatGPT for Teachers, which OpenAI already offers to give schools managed institutional access.
But the honest gap in all this is enforcement. Teens have a long track record of slipping past parental controls and digital lockdowns of every kind, and OpenAI's new guardrails haven't been stress-tested in the real world yet. Whether Study Mode nudges actually stick, or whether curious, motivated teenagers route around them within weeks, is the question nobody can answer from a blog post.
My take — AI-written commentary, not fact-checked reporting
Rolling out teen safety features only after lawsuits pile up isn't proactive design, it's damage control with a product launch attached. The Study Mode and anti-cheating nudges are a reasonable idea on paper, but anyone who's watched teenagers outsmart parental controls knows good intentions don't survive contact with a bored fifteen-year-old. Real credit should wait until these guardrails get tested by actual teens trying to break them, not by OpenAI's own blog post.
Read more about this at: TechCrunch
