Study: 94.1% of 17 mainstream LLMs have agent attack vulnerabilities

1. 09:00
   WeChat · Weijiqi Lixin

   University of Calabria, Italy: 94.1% of 17 mainstream LLMs have at least one type of agent attack vulnerability

   A study from the University of Calabria, Italy tested three types of attacks—direct prompt injection, RAG backdoors and inter-agent trust exploitation—on 17 mainstream LLMs; only Claude-4-Sonnet resisted all of them, and 94.1% of the models had at least one type of vulnerability. The study used LangChain and LangGraph to build three types of synthetic applications, with malicious payloads Base64-encoded and hidden in knowledge base documents using white text on a white background. The vulnerability rates for the three attack types were 41.2%, 52.9% and 82.4% respectively; some models that could refuse direct injection still executed malware under retrieved RAG documents and peer-agent instructions. The study used a black-box threat model, where attackers only need to inject text into external data sources accessible to the agent, without access to model weights or training data.

Current attention 34·Peak in the comparable period 36 (Nov 7 14:00)·Change over the comparable past 24 hours 0%