The company says that most users do not need to use this feature.
OpenAI has begun rolling out Lockdown Mode, an optional security setting designed to provide users with advanced protection from rapid injection attacks. For those unfamiliar, prompt injection is a form of social engineering that is specific to conversational chatbots. As AI systems have become better at pulling information from the Internet, people have begun hiding malicious instructions on webpages and other places to trick those systems.
OpenAI is billing Lockdown Mode as the last line of defense against accelerated injection, building on the strong security it says it already provides through ChatGPT, its model, and backend systems. “Lockdown mode is not for everyone,” explains OpenAI. “It is designed for people and organizations that handle sensitive data and want strong protection against the risks of data intrusion related to prompt injection.”
To that end, enabling Lockdown Mode limits some features provided by OpenAI through ChatGPT and its other products. For example, you can still use image creation and upload photos to ChatGPT, but it can’t pull images from the Internet or display an image inside a response. The chatbot also can’t download files to analyze, although you can still upload documents manually if you want the information for it. Other features, such as Deep Research and Agent Mode, are disabled completely. “Lockdown mode does not change how memory, file uploads, the ability to share conversations, or whether your conversations can be used to improve models,” adds OpenAI. “Many of these settings are separately configurable by workspace administrators.”
The company also notes that lockdown mode will not prevent quick injections of content from appearing in ChatGPT processes. Instead, it is designed to prevent an attacker from extracting sensitive data from your account by limiting network requests that someone could take advantage of. Lockdown mode is available for all personal accounts, including those using ChatGPT through OpenAI’s free tier. To activate it open ChatGPT’s Settings menu and select safety and security. under advanced securityTap lockdown mode And turn on the toggle. You can temporarily disable additional protection by selecting manage Select More from the status message that appears above the chat window close this chat.
Separately, OpenAI is introducing an active session manager that allows users to view any device or browser that has been used to access their account. From there, the company offers the option to log out of individual or all sessions at once. Just note that the latter may take up to 30 minutes to complete. The company adds, “If you suspect unauthorized account activity, change your password, review your sign-in methods, and contact OpenAI Support if you use one.”
<a href