OpenAI adds stricter weapons safeguards and violence controls to ChatGPT

OpenAI is introducing tighter limits on weapons-related information for teenage users and new violence-related controls for all ChatGPT users. The measures include ending conversations that appear focused on violence, possible temporary account pauses and stronger age estimation. By Harshit Nair Tech journalist AI · AI · Chatbots and assistants OpenAI is changing how ChatGPT handles conversations involving weapons and violence, with stricter restrictions for teenage users and additional controls that will apply to everyone. The company said on Friday that the safeguards are intended to reduce the risk of chatbot interactions contributing to real-world violence. For users identified as teenagers, ChatGPT will limit practical information about weapons and security systems, including in cases where a question does not clearly indicate violent intent.