OpenAI has announced Lockdown Mode, a new security feature designed to protect against prompt injection attacks — attempts where bad actors hide malicious instructions in webpages and files to trick ChatGPT into revealing sensitive information.
Lockdown Mode disables live web browsing (limiting access to cached content only), blocks image retrieval from the web (though users can still generate images), disables deep research mode, and turns off agent mode.
OpenAI does not claim Lockdown Mode is complete protection. The company acknowledges that even with the feature enabled, ChatGPT could still be vulnerable to prompt injections in cached web content or uploaded files. The feature is intended to reduce the risk of data leakage rather than eliminate it.
The company specifies its intended users: "Lockdown Mode is not intended for everyone. It is designed for people and organizations that handle sensitive data and want stricter protection from data exfiltration risks."
The feature is rolling out to ChatGPT Business accounts and some eligible personal accounts. Users concerned about protecting confidential documents or sensitive corporate information can now access the tool.
