Hello, you are using an old browser that's unsafe and no longer supported. Please consider updating your browser to a newer version, or downloading a modern browser.
Training Camp • Cybersecurity Glossary
An LLM jailbreak is a crafted prompt that bypasses a language model's safety guardrails to make it produce restricted or harmful output.
LLM Jailbreak Definition: An LLM jailbreak is a crafted prompt that bypasses a language model's safety guardrails to make it produce restricted or harmful output.
An LLM jailbreak is an adversarial prompting technique that bypasses a large language model's safety guardrails and alignment policies to elicit content the model is designed to refuse. Methods include role-play framing, prompt injection, encoding tricks, and many-shot priming, making jailbreak resistance a core concern of AI red teaming and model safety evaluation.
Turn knowledge into credentials with our instructor-led cybersecurity boot camps.
View All Courses →