A prompt crafted to get a model to ignore its own rules.
Why it mattersAssume anything user-facing will be probed. Filter what comes out, not just what goes in.
See also Prompt Injection, Guardrails
Jailbreak
A prompt crafted to get a model to ignore its own rules.
Explore more AI terms
Browse the full glossary for plain-English definitions across models, agents, data, and safety.
Back to all terms