Skip to content

Prompting

Jailbreak

A prompt technique intended to bypass an AI system's safety or behavior restrictions.

Example

A user disguises a prohibited request as a fictional exercise to bypass safeguards.

Why people use it

Teams use “Jailbreak” when they need to give models clearer instructions and diagnose weak prompts.

What you'll hear

“Can we improve this with Jailbreak before we change the model?”

Related terms