OpenAI is putting a teen-only wrapper around ChatGPT, betting that stronger defaults and more parental oversight can make a tool already embedded in young people’s lives safer. The harder question is whether those guardrails are robust enough—and whether they arrived far too late.

The company announced ChatGPT for Teens after years in which its chatbot spread rapidly among students, alongside mounting concern about harmful content, emotional dependence and AI-assisted cheating. Users who say they are 13 to 17, or whom OpenAI’s system estimates are under 18, will be routed automatically to the new experience; children under 13 remain barred under the company’s age policy.¹

OpenAI says the version is built around learning rather than answer-hunting. Study Mode uses guiding questions and step-by-step support, while a new homework reminder is meant to detect when a teenager appears to be seeking a shortcut and steer them back toward the learning tool. “AI should support learning, not shortcut it,” the company said, adding that educators should remain in control of classroom AI experiences.²

The rollout also makes under-18 protections the default: tighter handling of self-harm, violence, eating disorders, dangerous activities and explicit content; warnings on sensitive-image uploads; break prompts; and linked-parent controls for quiet hours, selected settings and limited high-risk safety notifications. OpenAI says its updated rules also bar romantic language, encouragement of emotional dependence and suggestions that the chatbot has feelings or consciousness.

But the launch has drawn skepticism precisely because much of it is familiar. Parental controls and Study Mode were introduced roughly a year ago, age prediction arrived earlier this year, and enhanced break reminders came last month. One account of the rollout described it as largely a consolidation of existing safeguards with “a handful of smaller additions and tweaks.”³

Critics also question whether teens determined to evade the system will simply do so—and why teen-specific protections followed, rather than preceded, ChatGPT’s massive adoption. OpenAI’s answer is that access should be paired with safeguards that reflect users’ developmental stage, reinforce offline relationships and encourage healthy use over time.² The new mode is therefore both a safety upgrade and an admission: the company is now trying to govern a teenage audience it has had for years.