Auditing the Ask Astro LLM Q&A app

Today, we present the second of our open-source AI security audits: a look at security issues we found in an open-source retrieval augmented generation (RAG) application that could lead to chatbot output poisoning, inaccurate document ingestion, and potential denial of service. This audit follows up on our previous work that identified 11 security vulnerabilities in YOLOv7, a popular computer vision framework.

Specifically, we found four issues in Ask Astro, a retrieval augmented generation (RAG) open-source chatbot …

ai security app application ask astro audit auditing audits chatbot denial of service document found issues llm poisoning rag security security issues service today work