#LLM Security
1 article tagged with "LLM Security".
·12 min
LAAF: Logic-Layer Attacks on Agentic LLM Systems
A case study on LAAF, a lifecycle-aware red-teaming framework for Logic-layer Prompt Control Injection (LPCI) in agentic LLM systems, and why it matters for memory, RAG, and tool-connected AI in production.
Sarah Burton