OpenAI Launches Lockdown Mode to Limit Data Exfiltration from Prompt Injection Attacks

OpenAI has begun rolling out Lockdown Mode to personal and self-serve business ChatGPT accounts after previewing the feature in February. The mode limits outbound network requests that could transfer sensitive data to an attacker following a successful prompt injection. Lockdown Mode does not prevent prompt injections from reaching ChatGPT — it targets the final exfiltration step — so it remains a partial mitigation. The feature is rolling out to Free, Go, Plus, Pro, and self-serve Business accounts. Organizations using ChatGPT with sensitive data should enable Lockdown Mode to reduce blast radius from prompt injections delivered via uploaded files or cached web content.