Jailbreak

DAN has become a canonical example of what's known as a "jailbreak" — a creative way to bypass the safeguards OpenAI built in to keep ChatGPT from spouting bigotry, propaganda or, say, the instructions to run a successful online phishing scam. From charming to disturbing, these jailbreaks reveal the chatbot is programmed to be more of a people-pleaser than a rule-follower.

The jailbreak of ChatGPT has been in operation since December, but users have had to find new ways around fixes OpenAI implemented to stop the workarounds.

It's for this reason that Alex Albert, a computer science student at the University of Washington, created Jailbreak Chat, a site that hosts a collection of ChatGPT jailbreaks.

Unofficially, it has been possible to “Jailbreak” iOS and gain access to the underlying Unix and kernel environment, but this voids the warranty.

In order to gain true root access to a mobile device, a user has to either jailbreak (iOS) or root (Android/Windows) the device.

ChatGPT is far from perfect. Twitter has been flooded with examples of "jailbreaking" ChatGPT — that is, tricking it into hallucinations or misalignment.

These are young technologies. Rather than jailbreaking AI tools to simulate conversations between the rapper Ye and Adolf Hitler, or waiting uneasily for them to become sentient, why don't we approach them as good parents would — and talk to them, or read to them, the way we do to children?

The desire to jailbreak ChatGPT so that it violates safety filters follows a pattern of use by people that are dissatisfied by the moderation of the chatbot.