OpenAI Introduces Stricter Restrictions for Teenagers Asking About Firearms
OpenAI has announced that it will impose new restrictions on teenagers accessing practical information about firearms and security systems via ChatGPT. These will apply even when the question appears harmless and does not indicate an intent to commit violence.
In its latest publication, OpenAI stated that it intends to restrict teenagers' access to information about firearms and security systems that could be used to facilitate violence. However, ChatGPT will continue to assist in preventing or reducing the risk of violence, seeking support, and connecting with a trusted adult.
The company is also tightening rules for all users. These will apply, in particular, to reconstructions of mass attacks with details about firearms, equipment, and methods of obtaining them, as well as to detailed comparisons of the scale of harm, which could increase the perpetrator's notoriety.
The new measures will supplement the previously introduced parental controls: parents receive notifications if their teenagers' linked accounts violate ChatGPT's rules on violence or threats. OpenAI plans to publish its first community safety report in early 2027 and then update it every six months.
