OpenAI blocks data exfiltration in Lockdown Mode

Summary

If you're processing untrusted content or sensitive data in ChatGPT, this mitigates the exfiltration vector of prompt injection attacks—the easiest attack surface to actually close. Without it, default ChatGPT lacks robust protection against determined data theft attempts.

Why it matters

If you're processing untrusted content or sensitive data in ChatGPT, this mitigates the exfiltration vector of prompt injection attacks—the easiest attack surface to actually close. Without it, default ChatGPT lacks robust protection against determined data theft attempts.

Sources

1. 1.rolling out to eligible personal accounts, including Free, Go, Plus, and Pro, and self-serve ChatGPT Business accounts
2. 2.Lockdown Mode is designed to help prevent the final stage of data exfiltration from a prompt injection attack by limiting outbound network requests
3. 3.The only way to solve the trifecta is to cut off one of the three legs, and by far the easiest leg to restrict without making your LLM systems far less useful is the exfiltration vectors to steal data
4. 4.using mechanisms that are deterministic and, crucially, are not evaluated by AI systems that themselves can be subverted