tech
OpenAI adds Lockdown Mode to ChatGPT to block data theft from prompt injection attacks
ChatGPT’s new Lockdown Mode disables live browsing, agent mode, and deep research to block data exfiltration via prompt injection. Available on all plans.
TL;DR
- OpenAI has launched Lockdown Mode for ChatGPT to combat prompt injection attacks.
- The feature disables live web browsing, agent mode, deep research, and file downloads.
- Lockdown Mode aims to prevent attackers from exfiltrating data by blocking outbound pathways.
- It does not completely solve prompt injection but reduces the risk of data exfiltration.
- Users must accept a trade-off between security and functionality.
- Lockdown Mode cannot be used simultaneously with Developer Mode.
- A separate session management feature allows users to review and log out of active sessions.