Security Considerations for Multi-agent Systems

• URL: http://arxiv.org/abs/2603.09002v1
• Date: Mon, 09 Mar 2026 22:46:27 GMT
• Title: Security Considerations for Multi-agent Systems
• Abstract summary: Multi-agent artificial intelligence systems or MAS are systems of autonomous agents that exercise delegated tool authority, share persistent memory, and coordinate via inter-agent communication.<n>This study systematically characterizes the threat landscape of MAS and quantitatively evaluates 16 security frameworks for AI against it.
• Score: 4.004329206052807
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent artificial intelligence systems or MAS are systems of autonomous agents that exercise delegated tool authority, share persistent memory, and coordinate via inter-agent communication. MAS introduces qualitatively distinct security vulnerabilities from those documented for singular AI models. Existing security and governance frameworks were not designed for these emerging attack surfaces. This study systematically characterizes the threat landscape of MAS and quantitatively evaluates 16 security frameworks for AI against it. A four-phase methodology is proposed: constructing a deep technical knowledge base of production multi-agent architectures; conducting generative AI-assisted threat modeling scoped to MAS cybersecurity risks and validated by domain experts; structuring survey plans at individual-threat granularity; and scoring each framework on a three-point scale against the cybersecurity risks. The risks were organized into 193 distinct main threat items across nine risk categories. The expected minimal average score is 2. No reviewed framework achieves majority coverage of any single category. Non-Determinism (mean score 1.231 across all 16 frameworks) and Data Leakage (1.340) are the most under-addressed domains. The OWASP Agentic Security Initiative leads overall at 65.3\% coverage and in the design phase; the CDAO Generative AI Responsible AI Toolkit leads in development and operational coverage. These results provide the first empirical cross-framework comparison for MAS security and offer evidence-based guidance for framework selection.

Related papers

• Adversarial Prompting Framework for AI Safety Assessment [0.7619404259039283]
  This paper presents the implementation of an Adversarial Prompting Framework (APF) for a comprehensive assessment of AI safety.<n>The framework systematically evaluates the resilience of the AI model through the generation of structured adversarial prompts.<n>Results indicate significant variations in the model vulnerabilities across different attack vectors.
  arXiv  Detail & Related papers  (2026-07-15T05:26:43Z)
• Architecture Matters for Multi-Agent Security [1.699014226195857]
  We study how multi-agent systems shape the tradeoff between task performance and attack resistance.<n>We find that multi-agent architectures are more vulnerable than standalone agents in the majority of configurations.
  arXiv  Detail & Related papers  (2026-04-25T22:25:58Z)
• A Formal Security Framework for MCP-Based AI Agents: Threat Taxonomy, Verification Models, and Defense Mechanisms [0.0]
  The Model Context Protocol (MCP) has become the de facto standard for connecting large language model (LLM)-based agents to external tools and data sources.<n>This paper presents MCPSHIELD, a comprehensive formal security framework for MCP-based AI agents.
  arXiv  Detail & Related papers  (2026-04-07T15:02:47Z)
• OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage [59.3826294523924]
  We investigate the security vulnerabilities of a popular multi-agent pattern known as the orchestrator setup.<n>We report the susceptibility of frontier models to different categories of attacks, finding that both reasoning and non-reasoning models are vulnerable.
  arXiv  Detail & Related papers  (2026-02-13T21:32:32Z)
• Securing Agentic AI Systems -- A Multilayer Security Framework [0.0]
  Securing Agentic Artificial Intelligence (AI) systems requires addressing the complex cyber risks introduced by autonomous, decision-making, and adaptive behaviors.<n>Existing AI security frameworks do not adequately address these challenges or the unique nuances of agentic AI.<n>This research develops a lifecycle-aware security framework specifically designed for agentic AI systems.
  arXiv  Detail & Related papers  (2025-12-19T20:22:25Z)
• Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse [50.87630846876635]
  We develop nine detailed cyber risk models.<n>Each model decomposes attacks into steps using the MITRE ATT&CK framework.<n>Individual estimates are aggregated through Monte Carlo simulation.
  arXiv  Detail & Related papers  (2025-12-09T17:54:17Z)
• The Evolution of Agentic AI in Cybersecurity: From Single LLM Reasoners to Multi-Agent Systems and Autonomous Pipelines [0.0]
  Cybersecurity has become one of the earliest adopters of agentic AI.<n>This survey presents a five-generation taxonomy of agentic AI in cybersecurity.
  arXiv  Detail & Related papers  (2025-12-07T05:10:16Z)
• ANNIE: Be Careful of Your Robots [48.89876809734855]
  We present the first systematic study of adversarial safety attacks on embodied AI systems.<n>We show attack success rates exceeding 50% across all safety categories.<n>Results expose a previously underexplored but highly consequential attack surface in embodied AI systems.
  arXiv  Detail & Related papers  (2025-09-03T15:00:28Z)
• SoK: Cybersecurity Assessment of Humanoid Ecosystem [25.852577434268273]
  We introduce a seven-layer security model for humanoid robots, organizing 39 known attacks and 35 defenses across the humanoid ecosystem.<n>We demonstrate our method by evaluating three real-world robots: Pepper, G1 EDU, and Digit.
  arXiv  Detail & Related papers  (2025-08-24T18:13:33Z)
• Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms [1.03121181235382]
  Large Language Model (LLM) agents face security vulnerabilities spanning AI-specific and traditional software domains.<n>This study bridges this gap through comparative evaluation of Function Calling architecture and Model Context Protocol (MCP) deployment paradigms.<n>We tested 3,250 attack scenarios across seven language models, evaluating simple, composed, and chained attacks targeting both AI-specific threats and software vulnerabilities.
  arXiv  Detail & Related papers  (2025-07-08T18:24:28Z)
• AILuminate: Introducing v1.0 of the AI Risk and Reliability Benchmark from MLCommons [62.374792825813394]
  This paper introduces AILuminate v1.0, the first comprehensive industry-standard benchmark for assessing AI-product risk and reliability.<n>The benchmark evaluates an AI system's resistance to prompts designed to elicit dangerous, illegal, or undesirable behavior in 12 hazard categories.
  arXiv  Detail & Related papers  (2025-02-19T05:58:52Z)
• EARBench: Towards Evaluating Physical Risk Awareness for Task Planning of Foundation Model-based Embodied AI Agents [53.717918131568936]
  Embodied artificial intelligence (EAI) integrates advanced AI models into physical entities for real-world interaction.<n>Foundation models as the "brain" of EAI agents for high-level task planning have shown promising results.<n>However, the deployment of these agents in physical environments presents significant safety challenges.<n>This study introduces EARBench, a novel framework for automated physical risk assessment in EAI scenarios.
  arXiv  Detail & Related papers  (2024-08-08T13:19:37Z)
• ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models [65.79770974145983]
  ASSERT, Automated Safety Scenario Red Teaming, consists of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. We find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings.
  arXiv  Detail & Related papers  (2023-10-14T17:10:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.