A Comparative Survey of Security Risks in AI Systems: From LLMs to AI Agents and Embodied Agents
ACM Computing Surveys · 30.11.2026 · 58 (15) · 1-38
DOI: 10.1145/3837083 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Computing Surveys, online erschienen am 22.08.2026. Verfasst von Baiqi Wu und 4 weiteren Autor:innen. Die Arbeit nennt 195 Literaturangaben. Gefördert von: New Generation Artificial Intelligence-National Science and Technology Major Project, NSFC, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
Rapid AI development across industries raises pressing security and privacy risks. This work presents a unified comparison of large language models, AI agents, and embodied agents, introducing a taxonomy of risks spanning data, models, systems, content, and applications, alongside a catalog of 24 specific threats. We contrast attack surfaces and methods across the three system types to reveal common patterns and distinctive vulnerabilities. We also survey mainstream AI security assessment frameworks and evaluate how relevant laws and regulations currently address these risks. Finally, we outline concrete directions for future research and practice aimed at building robust, secure agent ecosystems.
Förderung
- New Generation Artificial Intelligence-National Science and Technology Major Project
- NSFC
- China Postdoctoral Science Foundation
- Zhejiang Key Laboratory of Decision Intelligence
- Zhejiang Provincial Natural Science Foundation Exploration of China
- State Key Laboratory of Cryptography and Digital Economy Security
- Zhejiang Provincial Natural Science Foundation
- “Pioneer and Leading Goose” R&D Program of Zhejiang
Zitieren
APA
Wu, B., Li, Q., Zhou, C., Wang, T., & Ji, S. (2026). A Comparative Survey of Security Risks in AI Systems: From LLMs to AI Agents and Embodied Agents. ACM Computing Surveys, 58(15), 1-38. https://doi.org/10.1145/3837083
Vancouver
Wu B, Li Q, Zhou C, Wang T, Ji S. A Comparative Survey of Security Risks in AI Systems: From LLMs to AI Agents and Embodied Agents. ACM Computing Surveys. 2026;58(15):1-38. doi:10.1145/3837083
BibTeX
@article{Wu2026,
author = {Wu, Baiqi and Li, Qingming and Zhou, Chunyi and Wang, Ting and Ji, Shouling},
title = {A Comparative Survey of Security Risks in AI Systems: From LLMs to AI Agents and Embodied Agents},
journal = {ACM Computing Surveys},
year = {2026},
volume = {58},
number = {15},
pages = {1--38},
doi = {10.1145/3837083}
}
EUVIMED