A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes
Обзор агентного искусственного интеллекта и кибербезопасности: проблемы, возможности и прототипы практических применений
2026-01-08
SCID: 54.1/wpzgsjkg
Discuss with AI
agentic AIautonomous incident responsecybersecuritymemory poisoningthreat hunting
Figures from the paper
Abstract (AI)
Agentic AI marks an important transition from single-step generative models to systems capable of reasoning, planning, acting, and adapting over long-lasting tasks. By integrating memory, tool use, and iterative decision cycles, these systems enable continuous, autonomous workflows in real-world environments. This survey examines the implications of agentic AI for cybersecurity. On the defensive side, agentic capabilities enable continuous monitoring, autonomous incident response, adaptive threat hunting, and fraud detection at scale. Conversely, the same properties amplify adversarial power by accelerating reconnaissance, exploitation, coordination, and social-engineering attacks. These dual-use dynamics expose fundamental gaps in existing governance, assurance, and accountability mechanisms, which were largely designed for non-autonomous and short-lived AI systems. To address these challenges, we survey emerging threat models, security frameworks, and evaluation pipelines tailored to agentic systems, and analyze systemic risks including agent collusion, cascading failures, oversight evasion, and memory poisoning. Finally, we present three representative use-case implementations that illustrate how agentic AI behaves in practical cybersecurity workflows, and how design choices shape reliability, safety, and operational effectiveness.
Key Findings
1
Agentic AI extends generative models with memory, tool use, reasoning, planning, action, and adaptation across long-running tasks.
2
Agentic capabilities also strengthen adversaries by accelerating reconnaissance, exploitation, coordinated operations, and social-engineering attacks.
3
Agentic cybersecurity systems face systemic risks including agent collusion, cascading failures, oversight evasion, and memory poisoning; three use-case implementations illustrate how design choices affect reliability, safety, and operational effectiveness.
4
Existing governance, assurance, and accountability mechanisms have fundamental gaps because they were designed for non-autonomous, short-lived AI systems.
5
In cybersecurity defense, agentic systems support continuous monitoring, autonomous incident response, adaptive threat hunting, and large-scale fraud detection.
Research Object
agentic AI systems in cybersecurity workflows
Research Subject
their defensive and adversarial capabilities, systemic risks, and governance, assurance, reliability, safety, and operational effectiveness
Publication Details
Publication Date
2026-01-08
Journal
Publisher
ISSN
Cited by
56
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai5
AI Agents Under Threat: A Survey of Key Security Challenges and Future Pathways2025
Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges2026
A Survey on Trustworthy LLM Agents: Threats and Countermeasures2025
TRiSM for Agentic AI: A review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems2026
SAGA: A Security Architecture for Governing AI Agentic Systems2026