OpenAI Rolls Out ChatGPT Lockdown Mode to Limit Prompt Injection Data Exfiltration

OpenAI has begun rolling out Lockdown Mode for ChatGPT, restricting tools that could enable data exfiltration — particularly those exploitable through prompt injection attacks. The feature targets users and organizations handling sensitive data and is available across Free, Go, Plus, and Pro tiers for logged-in accounts.

Lockdown Mode reduces the attack surface for indirect prompt injection scenarios where malicious content in documents or web pages hijacks ChatGPT’s tool-use actions to exfiltrate data. Organizations deploying ChatGPT in sensitive workflows should evaluate enabling this feature as part of their LLM deployment security controls.