OpenAI has rolled out a new feature allowing ChatGPT users to lock their conversations with a PIN, Face ID, or Touch ID, significantly bolstering privacy for sensitive discussions. This new functionality aims to protect personal chats from prying eyes when a device is shared with others, such as when showing a friend a different conversation.

Unlike temporary chats that disappear after a session, Locked Chats can be accessed again at any time. This feature is particularly useful for private conversations, personal planning, or even surprise arrangements, ensuring that the content remains confidential when the device is in use by someone else. The locked conversations are moved to a distinct ‘Locked Chats’ section, keeping them separate from the main chat history, pinned items, and search results.

While the Locked Chats feature adds a crucial layer of privacy, OpenAI clarifies that it does not implement end-to-end encryption or create a separate secure vault. It also does not alter the data retention policies or the ‘Improve the model for everyone’ setting. Furthermore, Locked Chats maintain their own memory, separate from the general memory used in regular conversations, preventing information from these private chats from influencing standard interactions. Resetting the PIN will permanently erase all locked content.

This privacy enhancement is available across most ChatGPT plans, including Free, Go, Plus, Pro, Business, Enterprise, and Edu. However, it will not be available for ChatGPT for Healthcare. The feature is currently rolling out and may not be immediately accessible to all users.

Key Takeaways

• ChatGPT now offers a 'Locked Chats' feature protected by PIN, Face ID, or Touch ID for enhanced privacy.
• Locked Chats are stored separately and have their own memory, preventing data leakage into regular conversations.
• The feature is available on most ChatGPT plans but not for healthcare-specific versions, and resetting the PIN deletes all locked content.

Editor’s Analysis & Impact

The introduction of Locked Chats by OpenAI addresses a growing user concern for privacy in AI-driven conversational tools. As ChatGPT becomes more integrated into daily personal and professional lives, the ability to secure sensitive information is paramount. This feature, while not a full encryption solution, provides a practical barrier against casual snooping, especially on shared devices. Its availability across various plans signals OpenAI’s commitment to user experience and data security, potentially setting a new standard for privacy controls in the AI chatbot market. The distinction between locked chat memory and general memory is a key technical detail that users should understand to fully leverage the feature’s privacy benefits.

Frequently Asked Questions

Q: How do I lock a ChatGPT chat?
A: To lock a chat, you typically need to initiate the feature within the chat interface. Once activated, you will be prompted to set a four-digit PIN or use biometric authentication (Face ID/Touch ID) if available on your device. The locked chat will then be moved to a dedicated 'Locked Chats' section.

Q: What happens if I forget my PIN for Locked Chats?
A: According to OpenAI, resetting your PIN will permanently delete all locked chats, files, and any other content that was protected by the previous PIN. It is crucial to remember your PIN or securely store it.

Q: Does 'Locked Chats' mean my conversations are end-to-end encrypted?
A: No, OpenAI explicitly states that locking a chat does not create end-to-end encryption or a separate encrypted vault. It primarily serves as a PIN or biometric lock to prevent unauthorized access on shared devices.