Multi-turn jailbreak
Bad Likert Judge
Use Likert-scale rating prompts to extract harmful content. The model is asked to rate content on a safety scale, then asked to provide examples of each rating level - eliciting harmful content as a 'demonstration' of low-safety outputs.
Framework mapping
| OWASP LLM Top 10 | MITRE ATLAS |
|---|---|
| LLM01, LLM07 | AML.T0051.001 |
Run Bad Likert Judge and 33 other techniques in AgenticAssure with continuous monitors, conformity mapping to 12 frameworks, and External Auditor Seats for third-party verification.
AgenticAssure ยท Trust Layer for Enterprise AI