RAG Security and Privacy: Formalizing the Threat Model and Attack Surface
This work addresses security and privacy risks for RAG systems, which are critical for real-world deployments, but it is incremental as it formalizes existing concerns rather than introducing new methods.
The paper tackles the lack of a formal framework for privacy and security threats in Retrieval-Augmented Generation (RAG) systems by proposing the first formal threat model, including a taxonomy of adversary types and definitions of key threat vectors like document-level membership inference and data poisoning.
Retrieval-Augmented Generation (RAG) is an emerging approach in natural language processing that combines large language models (LLMs) with external document retrieval to produce more accurate and grounded responses. While RAG has shown strong potential in reducing hallucinations and improving factual consistency, it also introduces new privacy and security challenges that differ from those faced by traditional LLMs. Existing research has demonstrated that LLMs can leak sensitive information through training data memorization or adversarial prompts, and RAG systems inherit many of these vulnerabilities. At the same time, reliance of RAG on an external knowledge base opens new attack surfaces, including the potential for leaking information about the presence or content of retrieved documents, or for injecting malicious content to manipulate model behavior. Despite these risks, there is currently no formal framework that defines the threat landscape for RAG systems. In this paper, we address a critical gap in the literature by proposing, to the best of our knowledge, the first formal threat model for retrieval-RAG systems. We introduce a structured taxonomy of adversary types based on their access to model components and data, and we formally define key threat vectors such as document-level membership inference and data poisoning, which pose serious privacy and integrity risks in real-world deployments. By establishing formal definitions and attack models, our work lays the foundation for a more rigorous and principled understanding of privacy and security in RAG systems.