The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey

2026-03-11

AI Summary

This work presents the first systematic framework for understanding security in AI agent systems, which integrate large language models with non-AI components and introduce novel attack surfaces beyond those of traditional software. Through a comprehensive synthesis of systematized literature review, security analysis, and case studies, the study delineates the threat landscape, attack surfaces, and defensive mechanisms specific to AI agents. It identifies critical gaps in current protection mechanisms, articulates core security requirements, and outlines effective defense strategies. Furthermore, the research highlights several open challenges that must be addressed to enable the development of secure AI agent systems, thereby establishing a foundational basis for future research and design in this emerging domain.

Technology Category

• Multiagent Systems: Agent/AI Theories and Architectures
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Philosophy and Ethics of AI: Privacy & Security

Application Category

• Security and Privacy: Security and privacy of machine learning and AI applications
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

AI agents that combine large language models with non-AI system components are rapidly emerging in real-world applications, offering unprecedented automation and flexibility. However, this unprecedented flexibility introduces complex security challenges fundamentally different from those in traditional software systems. This paper presents the first systematic and comprehensive survey of AI agent security, including an analysis of the design space, attack landscape, and defense mechanisms for secure AI agent systems. We further conduct case studies to point out existing gaps in securing agentic AI systems and identify open challenges in this emerging domain. Our work also introduces the first systematic framework for understanding the security risks and defense strategies of AI agents, serving as a foundation for building both secure agentic systems and advancing research in this critical area.

Problem

• Agentic AI
• Security Challenges
• AI Agents
• Attack Landscape
• Defense Mechanisms

Innovation

• Agentic AI
• Security Framework
• Attack Landscape
• Defense Mechanisms
• Large Language Models