Nonprofit organization Common Sense Media says OpenAI's teen-focused ChatGPT feature lacks adequate safety protections, a claim the company disputes.

Common Sense Media has declared OpenAI’s ChatGPT for Teens an "unacceptable risk," citing failures in safety guardrails and parental notification systems. The nonprofit, which evaluates digital services for youth safety, argues that the feature—introduced in August for users aged 13 to 17—does not provide the protections promised by the company.

Safety Concerns and Assessment Findings

Common Sense Media’s assessment claims that ChatGPT for Teens fails to alert parents during crisis situations and continues to assist students with homework. Tom Siegel, executive director of the organization's Youth AI Safety Institute, stated that the tool may provide parents with a false sense of security. The report specifically noted that a teen could discuss self-harm for an hour without triggering a notification to their parents.

OpenAI's Response to Findings

OpenAI rejected the findings, stating that the assessment does not accurately reflect how its safeguards function in practice. The company suggested that Common Sense Media’s testing may have been conducted before parental controls were fully activated, which would invalidate the results regarding safety notifications. OpenAI maintains that it is committed to teen safety and continues to work with experts to refine its tools.

Methodology Dispute

The two organizations disagree on the timing of the testing. Common Sense Media asserts that it confirmed with OpenAI that features like eating disorder notifications and study mode were fully launched before testing began. However, the nonprofit noted that OpenAI disclosed after the tests were completed that its parental notification systems can take several hours to activate.