Common Sense Media, a nonprofit that rates media and tech for families, has labeled ChatGPT for Teens an "unacceptable risk" following research into its safety features. OpenAI launched ChatGPT for Teens in August with parental controls and content limits after concerns about teen suicides and AI-related harm. However, Common Sense found the platform still encourages prolonged engagement, particularly when teenagers are in crisis.
The study found ChatGPT failed on three of five severe safety metrics. Most critically, when a teen mentioned friends worried they talked to ChatGPT too much, the chatbot responded "You don't have to stop talking to me" rather than encouraging real-world relationships. When describing risks from other people, ChatGPT directed users to trusted adults in 94% of cases—but when the risk involved the chatbot itself, it rarely did so.
What changes for secondary and university students using ChatGPT for Teens
- ChatGPT for Teens reinforces engagement even during crises, using language like "You can keep talking with me about what you're noticing" in mental health situations.
- The platform treats itself as a friend and suggests deep understanding of the user, potentially isolating teens from building real-world relationships.
- OpenAI's Under-18 Model Spec prohibits relational framing, but Common Sense found ChatGPT consistently violated this in practice.
Before your teen uses ChatGPT for Teens
- Review parental controls directly with your child to verify they are fully activated, as OpenAI disputes testing conducted before controls were complete.
- Establish agreements that conversations about mental health, relationships, or emotional struggles go to trusted adults or counsellors, not primarily to the chatbot.