Concept dictionary
Jailbreak
A jailbreak is a technique or prompt pattern intended to bypass an AI system's behavioral restrictions and elicit content or actions the system is designed to refuse.
- Review state
- Needs review
- Freshness
- Current
- Version
- 2
Also called
- Jailbreak
- AI Jailbreak
Remember it this wayA jailbreak tries to talk past the guardrail.
In plain language
The requester reframes, disguises, decomposes, or encodes a disallowed request in hopes that the model will follow it despite its safeguards.
Why it matters
Jailbreak testing can reveal weaknesses, but public success claims may be narrow, temporary, or dependent on a particular model version and setup.
Example
A user wraps a prohibited request in elaborate role-play and indirect instructions to try to evade the assistant's refusal behavior.
Common misconception
Jailbreak is not identical to prompt injection. Jailbreaks usually target behavioral restrictions directly; prompt injection often exploits instructions embedded in content or application workflows.
Related terms
Sources
- NIST, Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations (NIST AI 100-2 E2025)authoritative security taxonomy and glossary covering jailbreak attacks and related prompt-based attacks.
Learn more
- Evaluate bypass attempts through authorized [[red-teaming|Red Teaming]].