A survey on AI agent security: Reasoning, Acting, and Self-Evolving
AI Plus · 2026
DOI: 10.59717/ipj.aiplus.2026.100004 ↗Kurz gefasst
Fachartikel in der Zeitschrift AI Plus, 2026. Verfasst von Daizong Liu und 23 weiteren Autor:innen. Die Arbeit nennt 251 Literaturangaben.
Abstract (Originalsprache)
Artificial intelligence (AI) agents are rapidly evolving into autonomous systems capable of reasoning, acting, and continual evolving. Their increasing autonomy enables powerful real-world applications but also introduces security risks throughout the entire agent execution cycle. However, the security challenges arising across different stages of the agent execution cycle have not yet been systematically reviewed. In this survey, we provide a comprehensive review of AI agent security from the perspective of the AI agent execution cycle. Specifically, we establish a unified taxonomy that organizes AI agent execution into three sequential stages: Reasoning, Acting, and Self-Evolving. Each stage is further divided into two interconnected components, resulting in six components in total: Thought, Plan, Skill, Tool, Adaptation, and Evolution. We then systematically review the threats, defense mechanisms, and evaluation benchmarks associated with each component, providing a comprehensive understanding of security challenges throughout the complete execution pipeline. Finally, we discuss current limitations and promising future research directions toward building secure, robust, and trustworthy AI agents. We believe that this survey provides a unified understanding of AI agent security and facilitates future research on the secure deployment of increasingly autonomous AI systems.
Zitieren
APA
Liu, D., Luo, T., Huang, S., Yang, J., Liang, S., Jia, X., Lau, C. P., Dong, J., Liu, Y., Bai, Y., Chen, X., Zhou, S., Arandjelović, O., Koffas, S., Guo, Z., Guan, R., Zhou, P., Wang, Z., Qin, Z., …, & Li, Y. (2026). A survey on AI agent security: Reasoning, Acting, and Self-Evolving. AI Plus. https://doi.org/10.59717/ipj.aiplus.2026.100004
Vancouver
Liu D, Luo T, Huang S, Yang J, Liang S, Jia X, et al. A survey on AI agent security: Reasoning, Acting, and Self-Evolving. AI Plus. 2026. doi:10.59717/ipj.aiplus.2026.100004
BibTeX
@article{Liu2026,
author = {Liu, Daizong and Luo, Tianyao and Huang, Shuwei and Yang, Junyi and Liang, Siyuan and Jia, Xiaojun and Lau, Chun Pong and Dong, Jianfeng and Liu, Yan and Bai, Yang and Chen, Xun and Zhou, Shiji and Arandjelović, Ognjen and Koffas, Stefanos and Guo, Zhongliang and Guan, Runwei and Zhou, Pan and Wang, Zhibo and Qin, Zhan and Du, Bo and Xiong, Hui and Chen, Pin-Yu and Domingo-Ferrer, Josep and Li, Yiming},
title = {A survey on AI agent security: Reasoning, Acting, and Self-Evolving},
journal = {AI Plus},
year = {2026},
doi = {10.59717/ipj.aiplus.2026.100004}
}
EUVIMED