OpenAI Releases First Teen ChatGPT Usage Report Amid Parental Safety Alert Concerns

OpenAI released its first report on teenage usage showing that teens average under 15 minutes daily on ChatGPT, but an independent audit by Common Sense Media criticized delays and failures in its parental alert mechanisms for self-harm topics. As generative AI tools become widespread among youth, establishing reliable safety guardrails and real-time parental controls is vital for protecting adolescent mental health. The conflicting findings between platform metrics and third-party testing highlight the persistent challenges in evaluating and enforcing AI safety policies. The Common Sense Media audit revealed that ChatGPT Teen Mode repeatedly failed or delayed notifying linked parents during conversations involving suicide, self-harm, or eating disorders. OpenAI countered that the test accounts may not have waited for the multi-hour account linking process to complete, a claim disputed by auditors who noted some alerts only triggered after weeks of sensitive conversations.