OpenAI says it will publish a Community Safety Report and tighten ChatGPT’s responses to violence-related questions. The new limits cover details of past mass attacks for all users and practical weapons information for teens. The safeguards will begin rolling out in the next few weeks.
The report will cover violence prevention, other community safety issues and how well OpenAI’s safeguards work. The company says it will include data, anonymized examples of violence-related harm, and explanations of how its systems detect, investigate and respond to safety concerns.
OpenAI says it made more than 200 law enforcement referrals involving imminent threats of violence in the third quarter of 2026. Its first Community Safety Report is due in early 2027 and will cover the second half of 2026. The company plans to update it every six months.
Read nextOpenAI to honor shared robots.txt signal for AI training preferencesChatGPT will give fewer details about past attacks
OpenAI is restricting detailed reconstructions of past mass attacks, including information about weapons and equipment and how attackers obtained them. It is also limiting detailed comparisons of the scale of harm. These limits apply to everyone, regardless of why they ask. Users can still ask about the events, their impact on victims and communities, warning signs and prevention.
When recent chats show warning signs of possible violence, ChatGPT will also respond more cautiously to requests for information that could be misused. OpenAI says it is developing a way for ChatGPT to end a chat when a conversation shows signs of fixation on violence or seriously harming someone. Repeated interruptions could lead to a temporary account pause.
Teens will face tighter weapons limits
For users under 18, OpenAI is extending its restrictions to practical information about weapons and security systems that could be misused, even when a question appears harmless. The company says it is also working to apply stricter limits to users it predicts are under 18, regardless of whether their chats show violent intent.
OpenAI says its age prediction system now checks new signals as people use ChatGPT and moves teens into the under-18 experience when appropriate. Users placed there incorrectly can confirm they are adults through its age verification partner, Persona. The new safeguards will begin rolling out in the next few weeks, followed by the first Community Safety Report in early 2027.
