Okuzeka
Artificial Intelligence & Machine Learning vocabulary · Bias, Safety and Responsible AI

Jailbreak

Meaning in artificial intelligence & machine learning, with examples

/ˈdʒeɪlbreɪk/nounCommon term

In artificial intelligence & machine learning, jailbreak means: A prompt or technique designed to trick a model into ignoring its safety rules and producing content it is meant to refuse.

In Turkish: jailbreak, güvenlik kısıtlarını aşma

How it is used

Also a verb: to jailbreak a model. Safety teams collect known jailbreaks, test new model versions against them and fix weaknesses. Related to, but different from, prompt injection, where hidden instructions inside data take control of an app.

Example sentences

  1. The new version resists every jailbreak in our test suite except two.

  2. Within an hour of launch, someone had found a jailbreak for the demo bot.

Common collocations

  • jailbreak attempt
  • jailbreak a model
  • known jailbreaks
  • jailbreak prompt
  • guardrailA rule or check placed around a model to stop harmful, off-topic or unsafe inputs and outputs, for example a filter that blocks personal data in answers.
  • red teamingTesting a model or AI system by deliberately trying to make it fail or behave harmfully, in order to find weaknesses before real users or attackers do.
  • system promptInstructions given to a chat model before the conversation starts, which set its role, rules, tone and limits for all the messages that follow.
  • alignmentMaking an AI system's behavior match human intentions and values, so that it is helpful and honest and avoids causing harm. Also the research field that studies this.

This word in other fields

The same English word can have a different meaning in another field:

More terms: Bias, Safety and Responsible AI

Practise with graded reading

Reading at the right level is the fastest way to make new words stick. Pick free English books by level: