ChatGPT Lockdown Mode: New Security Feature Explained

OpenAI has released a new security feature called ChatGPT Lockdown Mode. This mode aims to reduce the risk of data being stolen through prompt-injection attacks. It is now available for personal accounts, business users, and enterprise workspaces. Here's the thing: Prompt injection involves embedding malicious instructions into content an AI model processes. Lockdown Mode specifically targets the final step of these attacks, which is the unauthorized transfer of sensitive data. What's interesting is that this new mode does not prevent prompt injections from entering the model. It focuses only on blocking data from leaving the system. When Lockdown Mode is active, several ChatGPT capabilities are restricted. These include live web browsing, image retrieval, deep research, and agent mode. Users also cannot approve Canvas-generated code for network requests, and ChatGPT cannot download external files for data analysis. The bottom line: Administrators must manually configure access controls and app permissions to get full protection, and Lockdown Mode and Developer Mode cannot be used at the same time. This new feature helps protect sensitive information from being exfiltrated.