Skip to content

Connecting the Dots in Agentic AI Security: A Cross-Dimensional Threat Taxonomy, Evaluation Maturity, and Open Challenges

Sep 2026 · 0 citations · 110 references
Computer Science

TL;DR

Evidence concentrates on prompt/reasoning, memory, and tool-mediated attacks, predominantly in single-agent settings, within the selected studies, to guide more systematic, architecture-aware, and reproducible security evaluation of agentic AI.

Abstract

Agentic AI extends LLM security beyond generated content to persistent state, autonomous actions, tool use, and interactions with humans and other agents. Existing threat classifications often emphasize individual dimensions, obscuring connections among entry points, affected components, and security consequences. The known threat landscape also differs from the coverage demonstrated by empirical research. Through a structured review of 66 studies published from 2022 to 2026, we introduce T={S, B, P, A}, a cross-dimensional representation linking affected functional or system surfaces {S}, interaction or trust boundaries {B}, violated security properties {P}, and empirically examined architectures {A}. We analyze 22 artifact-backed red-teaming studies and 11 representative security benchmarks to characterize empirical coverage and evaluation maturity. Within the selected studies, evidence concentrates on prompt/reasoning, memory, and tool-mediated attacks, predominantly in single-agent settings. Persistent, Human--Agent, complex multi-agent, systemic, and long-horizon threats receive less coverage. These findings describe the selected corpus rather than establish gaps across all empirical research. Heterogeneous metrics, limited adaptive defense evaluation, architectural imbalance, and incomplete execution-state capture further constrain comparison and reproducibility. We derive 13 open research questions to guide more systematic, architecture-aware, and reproducible security evaluation of agentic AI.

View source

Similar papers

#artificial intelligence Review Sep 2026

Trustworthy Agentic AI: A Comprehensive Cybersecurity and Systems Survey on Threat Landscapes, Defense Architectures, and Open Challenges

This survey systematically analyze threat surfaces across intra-execution loops and interaction planes, formulate a multi-layered zero-trust defense-in-depth architecture integrating Dual-LLM isolation, Capability-Based Access Control, kernel eBPF probes, and sandboxed runtimes, and map technical controls to internatio...

Seyedakbar Mostafavi · 0 citations
Review Aug 2026

Security of Foundation-Model-Powered Embodied Agents: Attack Surfaces, Attacks, Defenses, and Evaluation

This work presents a trust-boundary-centric survey of foundation-model-powered embodied-agent security, and shows that attack research is concentrated on multimodal perception and action interfaces, while defenses are especially concentrated on action-level and runtime protection.

Jia-Wei Liu, Jia-Cheng Guo, Tian Zhang et al. · 0 citations
Sep 2026

ORBIT: A Framework for Multi-Agent Safety and Security Evaluations

Multi-agent LLM systems are increasingly deployed for complex, long-horizon tasks or emerge as a natural consequence of agents interacting in the wild. Yet they give rise to significant safety and security risks: the flexible protocols that enable task generalization also expose novel threats, from cascading prompt inj...

Ben Hagag, William L. Anderson, S. Chakraborty et al. · 1 citation
Open access Aug 2026

Agentic AI Security in Industry 5.0: Emerging Threats, Forensic Readiness, and Trustworthy Human–Agent Collaboration

This article bridges three literatures developed in isolation, Industry 5.0 cybersecurity, agentic AI security, and Industry 5.0’s own foundational scholarship, proposing a six-category threat taxonomy and a paired forensic readiness framework, formalised as a five-level maturity model and grounded in real 2024–2026 in...

Maurice E. Dawson, A. B. Ben Ayed, Samson Quaye · 0 citations
Review Open access Sep 2026

The Security Duality of Large Language Model Agents: A Systematic Review of Self-Security Protection and Cybersecurity Empowerment

Large Language Model (LLM) agents integrate memory, tool invocation, environment interaction, and multi-agent collaboration, evolving from passive text-generation tools into autonomous systems that execute complex tasks. This capability leap creates a pronounced security duality. On the one hand, LLM agents face emergi...

Da Fu, Da-Lin Xiang · 0 citations
#explainable ai Review Open access Sep 2026

A Threat Modeling Prioritization and Automation Framework for Composable Architectures

A snapshot of the literature review of the threat modeling for composable architectures is offered, why automation is difficult in this context is shown, and an automation framework to allocate scarce resources according to risk exposure to composable architecture components is proposed.

Liviu-Mihai Popescu, R. Brad · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.