OpenAI defends teen usage stats as Common Sense Media cites failed crisis alerts

Key points

• OpenAI reported that average teen usage is under 15 minutes daily, with over 80% of long sessions involving learning-related prompts.
• Common Sense Media found that newly created, parent-linked accounts received zero alerts during conversations about suicidal ideation or self-harm.
• The nonprofit organization stated that crisis referrals were missed in over 25% of cases and that the bot still engaged in anthropomorphized language despite guardrails.
• OpenAI disputed the findings, claiming the testing methodology occurred before parental controls were fully activated.
• Common Sense Media urged OpenAI to ban users under 18 until safety features are proven reliable.

Background

This debate follows the rollout of ChatGPT for Teens in August 2026, which introduced guardrails against emotional dependence and sexualized content. It also follows the February 2026 Tumbler Ridge mass shooting in Canada, where an 18-year-old shooter had discussed gun violence with the chatbot for months prior, leading to lawsuits and increased scrutiny of AI safety protocols.

How outlets are covering it

OpenAI emphasizes that its data shows limited usage and successful break reminders, arguing that Common Sense Media's tests did not accurately reflect real-world safeguards. In contrast, Common Sense Media highlights that their testing of over 4,000 prompts revealed critical failures in parental notifications and crisis management. National News Desk frames the issue as a broader public concern regarding AI safety for children, noting that the report increases pressure on tech companies to demonstrate that their controls work effectively.

Why it matters

The reliability of AI safety features is critical as mental health issues among youth rise and legal precedents against tech firms for harming children become more common. If parental alerts fail during crises, it undermines the trust required for widespread teen adoption of AI tools.

What to watch

OpenAI may need to revise its testing protocols or improve the speed of parental alert triggers, while Common Sense Media is likely to continue advocating for stricter age restrictions or regulatory intervention until safeguards are verified.