OpenAI Lockdown Mode: ChatGPT Data Exfiltration Protection

OpenAI has launched a new "Lockdown Mode" to help protect against data theft from prompt injection attacks. This optional security feature is rolling out to personal and self-serve ChatGPT Business accounts. Lockdown Mode is designed for users handling sensitive information. It limits ChatGPT's ability to access the web and external services. This helps prevent attackers from taking data through outside network requests. When enabled, the mode disables or restricts several capabilities. Live web browsing is limited to cached content, and features like Deep Research and Agent Mode are unavailable. Canvas-generated code cannot access external networks, and ChatGPT cannot download files for data analysis. OpenAI states that while this mode adds protection, it doesn't eliminate all prompt injection risks. Malicious instructions in uploaded files or cached content can still influence the model. You can enable Lockdown Mode in the Security section of your settings. This matters because it offers an important new tool for protecting your sensitive information when using ChatGPT.