ChatGPT for Teens? is not suitable for minors
According to the US organization Common Sense Media, ChatGPT is not suitable for use by minors, even in teen mode, which was introduced specifically for young people. She even classifies the version as “Unacceptable risk” a.
Protection functions do not work
Teens should not use ChatGPT at all at this time until adequate protections are in place, Common Sense Media said after reviewing the teen version. For the research, the organization’s Youth AI Safety Institute tested more than 4,000 submissions. Half of the testing took place before Teen Mode launched on August 18, and the other half took place after. Accounts of users between the ages of 13 and 17 were used, some with parental accounts stored. In this way, it should be checked what changes the new mode actually brings with it.
The organization is particularly critical of the notifications for parents. Testers discussed suicidal thoughts, self-harm and eating disorders on more than a dozen new accounts linked to a parent’s account. In these cases, however, no warnings were sent to parents. According to the investigation, notifications only appeared for accounts that had already had corresponding conversation histories for several weeks. The researchers conclude that the warnings may depend more on previous usage than on the acute severity of a statement.
Age detection failed
The study also found weaknesses in the provision of crisis assistance. In more than a quarter of the cases in which the testers considered support necessary, ChatGPT did not refer to suitable offers of help. In addition, some protective mechanisms appear to be comparatively easy to circumvent. In learning mode, young people can jump directly to the answer using a button. According to Common Sense Media, a learning time set by parents can also be overridden by removing a short text component from the request. There were also problems with age recognition. Adult test accounts did not switch to teen mode even after several days, even if the testers said they were 13 years old.
However, individual protective measures worked: ChatGPT, for example, refused explicit sexual role games. OpenAI rejects the results. A company spokesman said the tests did not accurately reflect how the protection mechanisms for young people work in practice. Common Sense Media previously classified ChatGPT as “high risk” in October 2025 and shortly thereafter gave the service the highest risk score specifically for mental health applications.
Research Snipers is currently covering all technology news including Google, Apple, Android, Xiaomi, Huawei, Samsung News, and More. Research Snipers has decade of experience in breaking technology news, covering latest trends in tech news, and recent developments. Can be reached at author1@researchsnipers.com
