Following teen suicide, OpenAI explores automatic underage user restrictions (www.engadget.com)

OpenAI announced it is building an automated system to identify teenage ChatGPT users and route them into a restricted, age-appropriate experience that blocks graphic sexual content and — in “cases of acute distress” — can contact law enforcement. If the system cannot confidently estimate a user’s age it will default to the gated mode; adults will be given a way to verify their age to regain full functionality. The company also reiterated imminent parental controls promised after the reported suicide of 16‑year‑old Adam Raine, saying controls (including the ability for parents to set hours when a child cannot use ChatGPT) will arrive before the end of the month, while giving no firm timeline for the automatic age‑detection rollout. CEO Sam Altman framed the policy as prioritizing “safety ahead of privacy and freedom for teens.” Technically and policy‑wise the move is consequential: automatic age estimation and distress detection raise accuracy, privacy and abuse‑resistance challenges (false positives/negatives, identity‑verification tradeoffs, and how “acute distress” thresholds are defined). Requiring verification to lift restrictions could pressure users into revealing identity data, while law‑enforcement callbacks create operational and ethical questions about thresholds and oversight. The effort signals an industry pivot toward proactive moderation and safety-first defaults for minors, but its effectiveness will hinge on transparent metrics, robust appeal or verification paths, and careful mitigation of harms from both over‑ and under‑blocking.