TRiSM 面向代理式 AI：基於大語言模型的代理式多智能體系統中 信任、風險與安全管理的綜述

TRiSM for Agentic AI: A Review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems

June 4, 2025

作者: Shaina Raza, Ranjan Sapkota, Manoj Karkee, Christos Emmanouilidis

cs.AI

摘要

基於大型語言模型（LLMs）構建並部署於多代理配置中的自主AI系統，正在重新定義企業和社會領域中的智能自主性、協作與決策。本綜述針對基於LLM的自主多代理系統（AMAS）中的信任、風險與安全管理（TRiSM）進行了結構化分析。我們首先探討了自主AI的概念基礎，其與傳統AI代理的架構差異，以及支持可擴展、工具使用自主性的新興系統設計。隨後，通過治理、可解釋性、模型運維（ModelOps）及隱私/安全四大支柱，詳細闡述了自主AI框架中的TRiSM，並將其置於自主LLMs的背景下進行了情境化分析。我們識別了獨特的威脅向量，並引入了一套全面的風險分類體系，輔以案例研究展示了現實世界中的脆弱性。此外，本文還調查了分佈式LLM代理系統中的信任建立機制、透明度與監督技術，以及最先進的可解釋性策略。同時，回顧了評估信任、可解釋性及以人為本性能的指標，並探討了開放基準測試的挑戰。通過加密、對抗防禦及遵守不斷演進的AI法規，本文還探討了安全與隱私問題。最後，本文提出了一條負責任自主AI的發展路線圖，建議了研究方向，以確保新興多代理系統與堅實的TRiSM原則相契合，實現安全、可問責且透明的部署。

English

Agentic AI systems, built on large language models (LLMs) and deployed in multi-agent configurations, are redefining intelligent autonomy, collaboration and decision-making across enterprise and societal domains. This review presents a structured analysis of Trust, Risk, and Security Management (TRiSM) in the context of LLM-based agentic multi-agent systems (AMAS). We begin by examining the conceptual foundations of agentic AI, its architectural differences from traditional AI agents, and the emerging system designs that enable scalable, tool-using autonomy. The TRiSM in the agentic AI framework is then detailed through four pillars governance, explainability, ModelOps, and privacy/security each contextualized for agentic LLMs. We identify unique threat vectors and introduce a comprehensive risk taxonomy for the agentic AI applications, supported by case studies illustrating real-world vulnerabilities. Furthermore, the paper also surveys trust-building mechanisms, transparency and oversight techniques, and state-of-the-art explainability strategies in distributed LLM agent systems. Additionally, metrics for evaluating trust, interpretability, and human-centered performance are reviewed alongside open benchmarking challenges. Security and privacy are addressed through encryption, adversarial defense, and compliance with evolving AI regulations. The paper concludes with a roadmap for responsible agentic AI, proposing research directions to align emerging multi-agent systems with robust TRiSM principles for safe, accountable, and transparent deployment.