Research · curated 13 Jul 2026
A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots
First reported arxiv.org
Coverage timeline
Single-source research — first reported, latest, and curated coincide.
Why it matters
Indirect prompt injection via poisoned knowledge-base documents can compromise every user whose query retrieves the malicious content, so a model-agnostic layered defense offers defenders a deployable mitigation strategy for RAG systems.
A research paper by Gulshan Saleem and colleagues presents a three-layer middleware framework to defend RAG-based chatbots against both direct and indirect prompt injection, combining input screening, a provenance-based instruction hierarchy during context assembly, and output auditing. Evaluated on 5,080 samples across GPT-4o, Llama 3, and Mistral 7B, the framework reduced Attack Success Rate from 71.4% to 11.3% while maintaining a 4.8% false positive rate and 61.2 ms median latency overhead.