Summary
-
OpenAI’s blocking mode blocks live outgoing network requests to stop LLM hijacking attempts.
-
Disable or limit features that make external calls, trading some features for privacy.
-
Lockdown mode is now rolling out to self-service Free, Plus, Pro, Go, and Business ChatGPT accounts.
The world of cybersecurity changed almost completely with the introduction of LLMs. We saw hackers and security experts use new technology to develop bigger and better programs to fight each other, and the battle continues to intensify to this day. However, the rise of AI has also created a new world of attacks where hackers are hijacking LLM to do their bidding. OpenAI’s lockdown mode was a response to these attacks, but it was not available to everyone; at least, until today. Now, the company allows everyone, even free users, to be safe while using ChatGPT.
ChatGPT Block Mode arrives for all users
It’s good to have if you’re concerned about privacy.
In February, OpenAI published a blog post detailing what lock mode is. Basically, when someone sends an AI to perform a task on the web, malicious actors can sometimes “link” their web pages and materials with prompts. These attempt to ‘hijack’ the AI and make it send personal data to the attacker, all without the user knowing.
To fix this, lockdown mode was introduced. When enabled, prevents ChatGPT from making live outgoing network requests. As such, if a bad actor tries to trick your LLM into handing over your personal data, blocking mode will intervene and prevent ChatGPT from sending anything. It also disables or limits specific features that depend on outgoing requests, but it’s good if you’re concerned about privacy.
When Lockdown Mode was first released, it was only available to select users. The blog post has now been updated to read the following:
Lockdown mode is rolling out to ChatGPT personal accounts as well as ChatGPT Business self-service accounts. First introduced for ChatGPT enterprise plans, Lockdown Mode is an optional setting for individuals and teams who want a more conservative ChatGPT experience when working with sensitive information or connected features.
He Lock Mode Documentation It explicitly confirms that “free, Go, Plus, Pro, and self-service ChatGPT Business accounts” can now use it. It may take a while to fully roll out, but you can check if you have it by going to ChatGPT Settings and then Security. You should see Lockdown Mode in the Advanced Security with a Switch section.
