survey arXiv Mar 8, 2026 · 4w ago
Xiaolei Zhang, Lu Zhou, Xiaogang Xu et al. · Nanjing University of Aeronautics and Astronautics · Collaborative Innovation Center of Novel Software Technology and Industrialization +5 more
Surveys LLM agent security threats across three autonomy tiers: cognitive manipulation, tool misuse, and multi-agent systemic failures
Prompt Injection Insecure Plugin Design Excessive Agency nlp
Artificial Intelligence (AI) agents have evolved from passive predictive tools into active entities capable of autonomous decision-making and environmental interaction, driven by the reasoning capabilities of Large Language Models (LLMs). However, this evolution has introduced critical security vulnerabilities that existing frameworks fail to address. The Hierarchical Autonomy Evolution (HAE) framework organizes agent security into three tiers: Cognitive Autonomy (L1) targets internal reasoning integrity; Execution Autonomy (L2) covers tool-mediated environmental interaction; Collective Autonomy (L3) addresses systemic risks in multi-agent ecosystems. We present a taxonomy of threats spanning cognitive manipulation, physical environment disruption, and multi-agent systemic failures, and evaluate existing defenses while identifying key research gaps. The findings aim to guide the development of multilayered, autonomy-aware defense architectures for trustworthy AI agent systems.
llm Nanjing University of Aeronautics and Astronautics · Collaborative Innovation Center of Novel Software Technology and Industrialization · The Chinese University of Hong Kong +4 more
ML Security Papers