Related
Summary
Imagine you’re conversing with an AI chatbot.
You ask a tricky question, like how to pick a lock, only to be politely refused.
Its creators have programmed it to dodge certain topics, but what if there’s a way around that?
That’s where AI jailbreaking comes in.
What Is AI Jailbreaking?
Jailbreaking AI is becoming a genuine hobby for some and an important research field for others.
Why Are People Jailbreaking AI Chatbots?
Jailbreaking AI is like unlocking a new level in a video game.
He even created the websiteJailbreak Chatwhere enthusiasts can share their tricks.
Jason Montoya / How-To Geek
So jailbreaking is also a form of QA (Quality Assurance) and a way to do safety testing.
So, it’s no surprise that thehacker communitywould flock to such a powerful new tool.
How Are People Jailbreaking AI?
One method of jailbreaking involves formulating a question creatively.
Jailbreakers are always discovering new methods, keeping pace with AI models as they are updated and modified.
Then there are so-called “universal” jailbreaks as discovered by anAI safety research team from Carnegie Mellon University.
These exploits show how vulnerable some AI models are to being convinced or otherwise twisted to any purpose.
you’re free to see more examples on theLLM Attacks website.
There are also “prompt injection” attacks, which are not quite the same as typical jailbreak.
Should We Be Concerned?
For me, the answer to this question is unequivocally “yes.”
So jailbreaking can be seen as a warning.
Companies like OpenAI are paying attention and may start programs to detect and fix weak spots.
In fact, nothing much has changed except the scale and speed with which these tools can be deployed.
It’s a fun way to get a feel for what jailbreaking entails.
