ChatGPT cuts weapons details for under 18s

OpenAI's ChatGPT will give teens less practical information about weapons, OpenAI said on the 30th of October 2026. The limits apply even when a question seems harmless.

The rules cover users that OpenAI believes are under 18. The new safeguards start rolling out in the next few weeks, the company said.

Weapons and security systems

ChatGPT will hold back practical details about weapons and security systems that could help someone commit violence, OpenAI said. For teens this applies even with no sign of violent intent.

Teens can still use ChatGPT to help stop violence, get support and reach a trusted adult, OpenAI said.

Chats can be ended

For all users, ChatGPT will be able to end a chat when someone seems fixated on violence, Reuters reported. OpenAI may pause an account after several such chats.

OpenAI plans to tell parents with linked accounts if a teen's account is paused for this reason.

Spotting teens

OpenAI uses age prediction, which guesses a user's age from signals such as behaviour. The system now keeps checking new signals as people use ChatGPT, OpenAI said.

Adults placed in the teen version by mistake can prove their age through Persona, OpenAI's age check partner.

A report every 6 months

OpenAI will publish its first Community Safety Report in early 2027, covering the second half of 2026. Updates will follow every 6 months, it said.

Below we break down the timing

The first report covers July to December 2026, which is 6 months. With an update every 6 months, OpenAI would publish 2 reports a year, by our arithmetic.

Key facts

ChatGPT will give teens less practical information about weapons.

The limits apply even with no sign of violent intent.

ChatGPT can end chats that show a fixation on violence.

OpenAI's first Community Safety Report comes in early 2027.

The changes roll out in the next few weeks.