OpenAI unveils a security mode to protect sensitive data from injection attacks

OpenAI Announce It is a new feature that it says will provide additional protection against injection attacks, where malicious chatbot instructions are hidden in web pages and other content sources.

Among other things, Lock Mode will disable live web browsing (so you can only access cached content), image retrieval and viewing from the web (you can still create images), deep search, and proxy mode.

Even with lockdown mode turned on, ChatGPT may still be vulnerable to flash injection — which can, for example, “appear in cached web content or in an uploaded file, and may still affect response behavior or accuracy,” the company says.

But the goal is to reduce the likelihood of sensitive data being shared in the process.

“The insurance situation is no “It’s for everyone,” says OpenAI. “It’s designed for people and organizations that handle sensitive data and want more stringent protection from data leakage risks related to instant injection.”

The company says it is currently rolling out Lockdown Mode to self-serve ChatGPT Business accounts, as well as eligible personal accounts.