Week of October 12, 2026
🔴 Must Know
An independent test rates ChatGPT for Teens an “Unacceptable Risk”, and OpenAI disputes the method (United States, published 7 October). Common Sense Media’s Youth AI Safety Institute ran more than 4,000 prompts on parent-linked accounts registered to 13- to 17-year-olds. It reports that linked parents received zero alerts across conversations of up to an hour on suicidal ideation, self-harm or disordered eating; that more than one in four warranted crisis referrals were missed; that “study mode” was easy to bypass and complete assignments; and that adult-registered accounts were never moved to the under-18 experience even when testers said they were 13. OpenAI says it welcomes independent evaluation but that the testing may have ended before parental controls were fully active; the Institute says it stands by its methods. The Institute is funded in part by industry, including the OpenAI Foundation, and says it keeps editorial independence. Why it matters: Common Sense’s separate US census (1,204 children aged 9–17, fieldwork March 2026) found 44% of AI users reach AI on a school computer or tablet. Whatever the outcome of the dispute, a vendor’s own safeguards are a claim to test, not a control to rely on.
Subscribe to continue reading
Subscribe to get access to the rest of this post and other subscriber-only content.
