ChatGPT gets new security feature to fight prompt injection attacks - Help Net Security

OpenAI has launched Lockdown Mode and Elevated Risk labels for ChatGPT to combat prompt injection attacks and other advanced security threats, particularly when interacting with external systems. Lockdown Mode, an optional setting for enterprise users, limits ChatGPT’s ability to connect to external tools and data to prevent exfiltration, while Elevated Risk labels provide in-product warnings about potential security implications of certain features.