A Survey on Trustworthy LLM Agents: Threats and Countermeasures
Обзор заслуживающих доверия агентов на основе больших языковых моделей: угрозы и контрмеры
2025-08-03
SCID: 54.1/knvs5kyv
Discuss with AI
LLM agent threatsTrustAgent frameworkmulti-agent systemstrustworthiness taxonomytrustworthy LLM agents
Figures from the paper
Abstract (AI)
With the rapid evolution of Large Language Models (LLMs), LLMbased agents and Multi-agent Systems (MAS) have significantly expanded the capabilities of LLM ecosystems.This evolution stems from empowering LLMs with additional modules such as memory, tools, environment, and even other agents.However, this advancement has also introduced more complex issues of trustworthiness, which previous research focusing solely on LLMs could not cover.In this survey, we propose the TrustAgent framework, a comprehensive study on the trustworthiness of agents, characterized by modular taxonomy, multi-dimensional connotations, and * Miao Yu and Fanci Meng contribute equally to this paper.
Key Findings
1
Adding these modules introduces complex trustworthiness challenges that are not covered by research focused solely on standalone LLMs.
2
LLM agents and multi-agent systems expand LLM capabilities by integrating memory, tools, environments, and other agents.
3
The survey proposes the TrustAgent framework to comprehensively study trustworthy LLM agents.
4
TrustAgent characterizes agent trustworthiness through a modular taxonomy and multi-dimensional connotations.
Research Object
LLM-based agents and multi-agent systems (MAS)
Research Subject
Their trustworthiness, including associated threats and countermeasures
Publication Details
Publication Date
2025-08-03
Journal
Publisher
ISSN
Cited by
119
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai8
A Survey of Large Language Models2026
A Survey on Evaluation of Large Language Models2024
A survey on large language model based autonomous agents2024
Prompt Injection attack against LLM-integrated Applications2023
Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection2023
PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models2024
InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents2024
Risks of AI scientists: prioritizing safeguarding over autonomy2025
Cited by10
AI Agents vs. Agentic AI: A Conceptual taxonomy, applications and challenges2025
AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges2025
TRiSM for Agentic AI: A review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems2026
A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes2026
Agentic AI in Healthcare and Medicine: A Seven-Dimensional Taxonomy for Empirical Evaluation of LLM-Based Agents2026
Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system security2026
Toward Sustainable Agentic AI Systems: A Survey of Architectures and Methodologies2026
Trustworthy Agentic AI: Failure Modes, Mitigation Strategies, and a Lifecycle Framework for Autonomous LLM Systems2026
Guardrailing LLM and Agentic Decisions for 6G AI-RAN2026
A Knowledge Service System for Foundation Pit Engineering Based on LoRA and RAG2026