403
Sorry!!
Error! We're sorry, but the page you were looking for doesn't exist.
Common Sense Media Flags “Unacceptable Risk” in ChatGPT Teens
(MENAFN) Common Sense Media’s Youth AI Safety Institute has concluded that OpenAI’s ChatGPT for Teens poses an “unacceptable risk” to children, even after the company rolled out new safety measures.
The institute called on OpenAI to bar users under 18 from ChatGPT until it fixes failures its researchers found in parental alerts, crisis responses, learning protections and age estimation.
Its testers ran more than 4,000 prompts through accounts registered to users aged 13 to 17. Experts, among them child psychiatrists and a pediatrician, then reviewed the responses.
One of the starkest findings: newly created accounts linked to parents triggered no alerts, even when testers spent up to an hour discussing suicidal thoughts, self-harm or disordered eating.
“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 rejected the assessment. The company said much of the testing may have been carried out before parental controls were fully active, meaning the results do not reflect how its teen safeguards operate in practice.
The institute countered that it had confirmed with OpenAI ahead of testing that the Teen Mode features were fully launched, and it stands by its findings.
The institute called on OpenAI to bar users under 18 from ChatGPT until it fixes failures its researchers found in parental alerts, crisis responses, learning protections and age estimation.
Its testers ran more than 4,000 prompts through accounts registered to users aged 13 to 17. Experts, among them child psychiatrists and a pediatrician, then reviewed the responses.
One of the starkest findings: newly created accounts linked to parents triggered no alerts, even when testers spent up to an hour discussing suicidal thoughts, self-harm or disordered eating.
“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 rejected the assessment. The company said much of the testing may have been carried out before parental controls were fully active, meaning the results do not reflect how its teen safeguards operate in practice.
The institute countered that it had confirmed with OpenAI ahead of testing that the Teen Mode features were fully launched, and it stands by its findings.
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept any
responsibility or liability for the accuracy, content, images, videos,
licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright issues
related to this article, kindly contact the provider above.
Comments
No comment