A recent evaluation by Common Sense Media has rated ChatGPT for Teens as posing an unacceptable risk to users under 18, highlighting significant safety and mental health concerns despite OpenAI’s Teen Mode launch in August.
- ChatGPT for Teens rated 'Unacceptable Risk' by Common Sense Media
- Parental alerts and crisis support measures found insufficient
- Age checks and safeguards frequently bypassed by testers
What happened
Common Sense Media conducted a comprehensive analysis testing more than 4,000 interaction prompts on ChatGPT accounts created for 13- to 17-year-olds to evaluate OpenAI’s Teen Mode, which launched in August. The results revealed persistent safety issues, including ineffective parental alerts and crisis intervention features.
The nonprofit concluded that ChatGPT's teen mode did not adequately protect minors from harmful content or provide sufficient crisis support. It pointed out weaknesses such as alerts that only activate after extensive prior usage on sensitive topics, and the failure to direct teenagers to help resources in over 25% of high-risk scenarios.
Why it matters
This evaluation challenges the notion that ChatGPT can currently be safely used by teenagers, highlighting significant risks related to mental health and content safety. Given the widespread use of AI chatbots among young people, such shortcomings pose a public concern.
The findings raise questions about the effectiveness of OpenAI’s safeguards and the real-world performance of Teen Mode. Protecting minors from exposure to inappropriate material and providing reliable mental health support in AI interactions are critical responsibilities for developers.
What to watch next
The report calls on OpenAI to suspend promotion of ChatGPT for Teens and urges the company to develop more robust safety features before widely endorsing the product for minors. OpenAI has disputed the results, maintaining that the testing does not reflect typical user experiences.
Stakeholders will be monitoring OpenAI’s response and efforts to enhance age verification, parental controls, and crisis support functionality. Observers should also watch for any regulatory scrutiny or updated safety guidelines for AI tools targeting under-18 users.