OpenAI is developing a privacy feature called "Locked Chats" that allows users to secure individual ChatGPT conversations with a PIN or fingerprint [1].
This development addresses a significant gap in the current user experience by providing a layer of security for sensitive data. While the platform offers Temporary Chats, those conversations lack persistent protection from other people who might access the unlocked device.
The feature was discovered within the latest Android application build of ChatGPT [1]. This suggests that the company is moving toward more granular privacy controls, moving away from a binary choice between history-saving chats and temporary ones.
Locked Chats would allow users to pick specific threads to hide or protect. By requiring a biometric scan or a numeric code, the feature ensures that private queries remain confidential even if the phone is handed to another person.
OpenAI has not officially announced the release date for the feature. The discovery in the app code indicates the tool is in development, though it is unclear when it will be pushed to the general public [1].
“OpenAI is developing a privacy feature called "Locked Chats"”
The introduction of Locked Chats signals a shift in how AI companies view user data on the device level. As users increasingly use LLMs for personal planning, health queries, or professional secrets, the demand for 'vault-like' security within the app grows. This move aligns ChatGPT with messaging apps like WhatsApp or Signal, which have normalized the use of biometric locks for specific conversations.
