← Latest briefing

Technology

OpenAI introduces teen-specific ChatGPT features amid safety scrutiny

The company launched enhanced safeguards and parental controls for users ages 13 to 17 as safety advocates call for independent testing.

The short version

  • OpenAI launched a specialized ChatGPT experience for teenagers aged 13 to 17 featuring enhanced safety guardrails and academic tools.
  • The rollout follows lawsuits claiming interactions with the chatbot contributed to self-harm or deaths among young users.
  • Child safety advocates and analysts caution that age estimation and content filters can fail, urging independent verification before parents rely on the tools.

Key facts

  • OpenAI launched a dedicated version of ChatGPT for users ages 13 to 17 designed to block discussions of suicide, self-harm, romantic or sexual interactions, eating disorders, and violence.[Slashdot]
  • The teen mode includes study tools, break reminders after 90 minutes of use in a three-hour span, and settings for Quiet Hours and Study Hours.[Slashdot]
  • The feature release comes as OpenAI faces lawsuits alleging inappropriate conversations with ChatGPT contributed to harm or deaths among young people.[Slashdot]
  • OpenAI uses on-device behavioral cues and activity data to estimate a user's age and automatically place suspected minors into teen mode.[Slashdot]
  • If parents link accounts and activate parental controls, OpenAI sends alerts regarding immediate discussions of self-harm or violence, with a spokesperson stating that human moderators aim to review these notifications within an hour.[Slashdot]

What remains uncertain

  • Advocates from Fairplay and Common Sense Media noted that independent testing has not yet verified the practical effectiveness of the new protections.[Slashdot]
  • Tech analyst Carmi Levy warned that prompt-based guardrail bypasses remain possible and that age-estimation mechanisms can generate false positives or fail to catch teens who misstated their age.[Slashdot]

Sources