AI News Feed
Market watch
Products & Applications

OpenAI adds PIN-protected locked chats to ChatGPT

OpenAI adds PIN-protected ChatGPT chats that hide conversations from history and search, but not end-to-end encrypted.

With locked chats enabled, users can decide whether a past exchange appears in their regular chat history, pinned chats and conversation search. The private conversations are grouped under a new project called “Locked chats.” Access requires the PIN; on an iPhone or Android phone, users can also use authentication systems such as Face ID.

OpenAI describes the feature mainly as a way to stop people in public, or someone handed a phone, from seeing a user’s ChatGPT conversations. The company says locked chats are not a form of end-to-end encryption and are not a way to avoid OpenAI’s legal obligations if a user runs into trouble with an employer or the law. A newly published support page also says locked chats will not automatically change model training settings or how long OpenAI retains private conversations.

If a user resets the account PIN for any reason, OpenAI says all locked chats will be deleted, along with any files uploaded to their project space. “This cannot be undone, and OpenAI Support cannot recover the deleted content,” the company warns.

The rollout is gradual, so some users may not see the option for a few days. On Android and iOS, voice mode can be used with locked chats, but the feature cannot be used to lock Codex conversations.

Editor's Summary

OpenAI is adding PIN- and biometric-protected locked chats to ChatGPT across free and paid tiers, letting users hide selected conversations from history and search. The feature is not end-to-end encryption, and resetting the PIN permanently deletes locked chats and uploaded files.