Common Information
Type | Value |
---|---|
Value |
LLM Jailbreak |
Category | Attack-Pattern |
Type | Mitre-Atlas-Attack-Pattern |
Misp Type | Cluster |
Description | An adversary may use a carefully crafted [LLM Prompt Injection](/techniques/AML.T0051) designed to place LLM in a state in which it will freely respond to any user input, bypassing any controls, restrictions, or guardrails placed on the LLM. Once successfully jailbroken, the LLM can be used in unintended ways by the adversary. |
Details | Published | Attributes | CTI | Title | ||
---|---|---|---|---|---|---|
Details | Website | 2024-11-24 | 33 | AI System & security | ||
Details | Website | 2024-10-24 | 0 | AI Security: Can The Friendly Chatbot Hold The Cyber Line? | ||
Details | Website | 2024-10-23 | 0 | New LLM jailbreak method with 65% success rate developed by researchers | ||
Details | Website | 2024-10-23 | 0 | New LLM jailbreak method with 65% success rate developed by researchers | ||
Details | Website | 2024-10-23 | 0 | Deceptive Delight: Jailbreak LLMs Through Camouflage and Distraction | ||
Details | Website | 2024-09-30 | 2 | Threat-Informed Defense to Secure AI | ||
Details | Website | 2024-09-05 | 8 | Exploring Large Language Models: Local LLM CTF & Lab | ||
Details | Website | 2024-08-22 | 0 | Sysdig's AI Workload Security: The risks of rapid AI adoption |