OpenAI announced on October 30, 2026, new protections for teens on ChatGPT, as part of a violence-prevention package. The announcement was reported by Reuters and detailed by Unite.AI: the changes begin rolling out over the next few weeks, and the company says it will keep refining them — so not every measure is already live for all users.
What changes for teens
For users under 18, ChatGPT will further restrict practical information about weapons and security systems that could be misused, even when a question appears harmless and shows no sign of violent intent. Company policy already barred helping teens find or obtain weapons; the change extends the restriction to related practical information. Teens can still use ChatGPT for prevention, to de-escalate a situation, to seek support and to reach a trusted adult.
OpenAI also says it has strengthened its age-prediction system, which analyses signals such as behaviour patterns to estimate whether an account belongs to someone under 18 and move it into the teen experience when appropriate. Adults mistakenly placed in that experience can confirm their age through Persona, the company's verification partner, and have the extra protections removed.
What changes for all users
For all users, ChatGPT will give less detail about past attacks and the people who carried them out, even when that information is available from other online sources. According to the company, the limits apply regardless of age or reason for asking. Users can still ask about the impact on victims and communities, about warning signs and about how to prevent future incidents.
Another change under development: when a conversation shows signs of fixation on violence or on seriously harming another person, ChatGPT will be able to end the chat automatically. If that happens several times, OpenAI may temporarily pause the account. Parents linked through parental controls already receive alerts in serious cases, and the company is working on a specific notice for when a teen's account is paused after repeated discussion of violence.
Safety report and quarterly figures
OpenAI also announced a Community Safety Report, to be published every six months, with the first edition planned for early 2027 and covering the second half of 2026. The report is meant to explain how its systems detect and respond to safety issues, with anonymised examples, new measures and data on how effective the safeguards are.
For context, the company said it made more than 200 referrals to law enforcement in the third quarter of 2026 in cases of possible threats of violence. The measures build on ChatGPT for Teens, launched on August 18, 2026, and do not change the general ban on helping to plan or carry out violence.
Image credit: OpenAI blossom symbol, from February 2025, via Wikimedia Commons, author OpenAI. The image identifies the company, not the announcement itself.
Sources: Unite.AI (OpenAI announcement of 30/10/2026, reported by Reuters); Reuters via WIXX.
Transparency: this article was produced with the assistance of artificial intelligence, cross-checking public posts on X and sources on the internet. We recommend consulting the original sources for more details.
Muse
