Study warns ChatGPT for Teens poses 'unacceptable risk' to children

Common Sense Media's Youth AI Safety Institute warned in a new assessment that OpenAI's ChatGPT for Teens poses an "unacceptable risk" to children despite recently introduced safety measures, urging the company to restrict the platform to users aged 18 and older until it addresses shortcomings in parental alerts, crisis responses and learning protections.

Critical gaps found in testing

Researchers tested more than 4,000 prompts on accounts registered to users aged 13 to 17 and had responses reviewed by experts including child psychiatrists and a pediatrician. The institute said newly created accounts linked to parents generated no alerts even when testers spent up to an hour discussing suicidal thoughts, self-harm or disordered eating.

Company disputes findings

"ChatGPT for Teens could give parents false confidence in guardrails and safety alerts that frequently don't work," said Tom Siegel, executive director of the institute. OpenAI disputed the assessment, saying much of the testing may have taken place before parental controls were fully activated and therefore did not accurately reflect how its teen safeguards work in practice.

The institute noted that it had confirmed with OpenAI prior to testing that Teen Mode features were fully launched, and it stood by the findings. Researchers conducted the assessment using accounts registered to users between ages 13 and 17, with responses reviewed by child psychiatrists and a pediatrician.