Intent drift detection & taxonomy for LLMs — how multi-turn jailbreaks (gradual drift, persona anchoring, trust escalation) evade single-turn scanners, and how to detect them.
jailbreak ai-safety social-engineering red-teaming ai-security prompt-injection llm-security llm-observability intent-drift-detection goal-drift intent-drift multi-turn-jailbreak
-
Updated
Aug 30, 2026