Skip to content
AImpact
IT EN
Safety Beginner Also known as: Aggiramento delle protezioni

Jailbreak

A technique where a user talks the model into ignoring its own safety rules, for example by asking it to pretend to be a character with no restrictions.

ShareLinkedInX

In practice

Different from prompt injection: here it is the user who tries. If you offer a public LLM service this means doing red teaming, logging conversations, and running a safety classifier in cascade over responses.

Related terms

Prompt injection Alignment Red teaming

Seen in the wild

8 entries mentioning it
  1. Promptfoo Red Teaming: open source automated red-teaming with CI integration and comparative benchmark
    Medium
  2. Many-Shot Jailbreaking: safety training overridden by context length
    High
  3. HarmBench: standardized benchmark for evaluating LLM jailbreaks and defenses
    High
  4. Crescendo: the multi-turn jailbreak that bypasses guardrails through gradual escalation
    Medium
  5. Garak: the open source vulnerability scanner for LLMs
    Medium
  6. PAIR: automated LLM-vs-LLM jailbreaking
    High
  7. Universal adversarial attacks on LLMs: transferable jailbreaks across GPT-4, Claude, and Gemini
    High
  8. Lakera Guard: real-time protection for LLMs in production
    Medium
← All terms