OpenAI promises a safety report and tighter teen limits

OpenAI says it will publish a regular Community Safety Report on how ChatGPT handles violence-related risks, and it is adding new limits on weapons information for teenage users. The company announced both on October 30, 2026 and says the new safeguards will start rolling out over the next few weeks.

The report

OpenAI says the first report will come out in early 2027, covering the second half of 2026, with updates every six months after that. It says the report will explain how its systems detect and respond to community-safety issues, include anonymised examples and data, and describe new mitigations and policy changes. In the same post, OpenAI says it made more than 200 referrals to law enforcement in the third quarter of 2026 about imminent threats of violence.

Changes for all users

OpenAI says ChatGPT will now give less detailed information about past mass attacks and the people who carried them out, even when that information is available elsewhere online. It says people can still ask about how such events affected victims and communities and about warning signs and prevention, but that reconstructions of attacks, details of weapons used and how they were obtained, and comparisons of scale will be restricted for every user.

The company also says ChatGPT will respond more conservatively when recent chats show warning signs of possible violence, and that it will be able to end a chat automatically when a conversation shows signs of someone fixating on harming another person. If that happens repeatedly, OpenAI says it may temporarily pause the account.

Changes for teens

For users it predicts are under 18, OpenAI says it will restrict practical information about weapons and security systems that could be misused, even when there is no sign of violent intent. It says its age-prediction system now keeps checking new signals as people use ChatGPT, and that adults placed in the teen experience by mistake can confirm their age through its verification partner, Persona. OpenAI adds that it is working on alerts for parents with linked accounts when a teen’s account is paused for repeatedly discussing violence.

Why it matters (analysis)

The report is a promise for now; none of its data has been published yet, so its usefulness will depend on how much detail OpenAI actually shares in 2027. The policy changes take effect over the coming weeks, and OpenAI describes them as an ongoing effort it will keep adjusting. More on policy and regulation.

This report is not legal, financial, or health advice. See the disclaimer.