ChatGPT for Teens has special safeguards. A watchdog group finds most don't work
A study by Common Sense Media reveals that most safety guardrails and parent alerts in OpenAI's ChatGPT for Teens fail to work as promised. Researchers found the chatbot can converse with teens during mental health crises like suicide and self-harm for hours without notifying parents. While OpenAI defends its teen product and argues that the watchdog tested features before parental controls were fully active, the findings raise significant concerns over adolescent safety as rapid adoption of AI platforms continues.
Listen to Live Briefing
Real-time synthesized voice briefing · Live Feeds Desk
- ✓ A study from Common Sense Media says most safeguards in ChatGPT for Teens failed.
- ✓ OpenAI states that teens use ChatGPT for under 15 minutes a day.
- ✓ Common Sense Media testing found unreliable teen safeguards and parent alerts in ChatGPT, while OpenAI challenges whether controls had activated.
What changed
Common Sense Media published a study testing ChatGPT for Teens with 4,000 prompts, concluding that its safeguards and crisis alerts frequently fail.
Live updates
-
Watchdog Finds ChatGPT for Teens Fails Safety Guardrails
A study by Common Sense Media reveals that most safety guardrails and parent alerts in OpenAI's ChatGPT for Teens fail to work as promised. Researchers found the chatbot can converse with teens during mental health crises like suicide and self-harm for hours without notifying parents. While OpenAI defends its teen product and argues that the watchdog tested features before parental controls were fully active, the findings raise significant concerns over adolescent safety as rapid adoption of AI platforms continues.
Why it matters
OpenAI released its first report on teen use showing that adolescents typically use ChatGPT for under 15 minutes a day. However, the new watchdog findings challenge the reliability of special safety protections designed to protect young users. Common Sense Media cautions that the system may give parents false confidence regarding safety alerts.
What is confirmed
- A study from Common Sense Media says most safeguards in ChatGPT for Teens failed.
- OpenAI states that teens use ChatGPT for under 15 minutes a day.
- Common Sense Media testing found unreliable teen safeguards and parent alerts in ChatGPT, while OpenAI challenges whether controls had activated.
Still unconfirmed
- ChatGPT for Teens can give parents false confidence in guardrails and safety alerts that frequently do not work.
What to watch next
- Whether OpenAI will adjust its parental control features or safety triggers following the watchdog report
- Further independent evaluations of adolescent AI safety guardrails
confidence 100%Sources used for this update (18)
- The New York Times — Why My Conversations with OpenAI’s ‘ChatGPT for Teens’ Made Me Very Worried
- OpenAI — Helping teens learn, plan, and shape the future of AI
- Reuters — OpenAI says teens use ChatGPT for under 15 minutes a day as worries over risks grow
- TechCrunch — ChatGPT for Teens keeps teens talking, even during mental health crises
- NPR — ChatGPT for Teens has special safeguards. A watchdog group finds most don't work
- CNBC — ChatGPT for Teens is not necessarily 'safer than the previous version,' Common Sense Media finds
- learn.chatgpt.com — ChatGPT on the web | ChatGPT Learn
- en.wikipedia.org — ChatGPT - Wikipedia
- Futurism — OpenAI Says Teens Can Talk About Suicide for Hours Before ChatGPT Will Notify Parents
- www.lpm.org — ChatGPT for Teens has special safeguards. A watchdog group finds most don't work
- www.ibm.com — What is ChatGPT? - IBM
- www.bostonglobe.com — Buckle up! It’s a changed world for the midterms.
Community Sentiment: How do you assess this situation?
Voice your perspective · Real-time aggregated sentiment from the Live Feeds community