AI Agent Auditing

Ask about this topic

Overview

AI agent auditing studies whether autonomous systems follow their instructions and policies as they use tools, exchange information, and change external state. Researchers combine trace analysis, runtime monitoring and authorization, static and dynamic security testing, adversarial benchmarks, and tamper-evident logs to detect violations and reconstruct how they occurred. A central lesson is that correct final answers or benchmark scores do not establish safe or faithful execution; audits increasingly examine intermediate decisions, permissions, evidence provenance, and effects on the environment. The field still faces challenges in obtaining complete, independent records and reliable evaluators, while balancing stronger oversight with task utility, cost, and agent autonomy.

Latest papers

CardsList

269 papers79 papers in the last four weeks, up 39% on the four weeks before. 0.5% of all new papers.

  1. The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence

    Mar 25, 2026Biplab Pal, Santanu BhattacharyaAI Agent ReliabilityAI Agent Auditing

    CARDS (Center for Real-Time Distributed Sensing and Autonomy), University of Maryland Baltimore County, Baltimore, MD, USA · Massachusetts Institute of Technology, Cambridge, MA, USA

  2. MedOpenClaw and MedFlowBench: Auditing Medical Agents in Full-Study Workflows

    Mar 25, 2026Weixiang Shen, Chengzhi Shen, Yanzhu Hu +12Medical VLMsMedical Image Benchmarks

    Technical University of Munich (TUM) · TUM University Hospital · LMU Munich +6

  3. Willful Disobedience: Automatically Detecting Failures in Agentic Traces

    Mar 25, 2026Reshabh K Sharma, Shraddha Barke, Benjamin ZornAI Agent EvaluationAI Agent Auditing

    University of Washington Seattle, Washington, USA · Microsoft Research Redmond, Washington, USA

  4. AgentWall: A Runtime Safety Layer for Local AI Agents

    Mar 24, 2026Ashwin AravindAI Agent SafetyRuntime Enforcement for AI Agents

  5. Agent Audit: A Security Analysis System for LLM Agent Applications

    Mar 24, 2026Haiyue Zhang, Yi Nian, Yue ZhaoAI Agent AuditingLLM Agent Security

    University of Southern California Los Angeles, California, USA

  6. Before the Tool Call: Deterministic Pre-Action Authorization for Autonomous AI Agents

    Mar 21, 2026Uchi UchibekeAI Agent SecurityTool-Use Authorization

    APort Technologies Inc., Toronto, Canada

  7. When Only the Final Text Survives: Implicit Execution Tracing for Multi-Agent Auditing

    Mar 18, 2026Yi Nian, Haosen Cao, Shenzhe Zhu +4Data Provenance in Language ModelsAI Agent Auditing

    University of Southern California · University of Toronto · University of Illinois Chicago +1

  8. Ablation Study of a Fairness Auditing Agentic System for Bias Mitigation in Early-Onset Colorectal Cancer Detection

    Mar 17, 2026Amalia Ionescu, Jose Guadalupe Hernandez, Jui-Hsuan Chang +4Algorithmic FairnessAI Agent Auditing

    Cedars-Sinai Medical Center, Los Angeles, CA, USA

  9. Cryptographic Runtime Governance for Autonomous AI Systems: The Aegis Architecture for Verifiable Policy Enforcement

    Mar 15, 2026Adam Massimo MazzocchettiRuntime Enforcement for AI AgentsAI Agent Governance

    SPQR Technologies Inc.

  10. CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents

    Mar 11, 2026Marta Sumyk, Oleksandr KosovanComputer-Use Agent EvaluationAI Agent Auditing

    Ukrainian Catholic University Lviv, Ukraine

  11. VulnAgent-R2: Evidence-Calibrated Multi-Agent Auditing for Repository-Level Vulnerability Detection

    Mar 11, 2026Renwei Meng, Haoyi Wu, Jingming WangSoftware Vulnerability DetectionAI Agent Auditing

  12. Compatibility at a Cost: Systematic Discovery and Exploitation of MCP Clause-Compliance Vulnerabilities

    Mar 10, 2026Nanzi Yang, Weiheng Bai, Kangjie LuMCP SecurityAI Agent Security

    University of Minnesota Minneapolis, USA

  13. Arbiter: Detecting Interference in LLM Agent System Prompts

    Mar 9, 2026Tony MasonAI Agent AuditingAI Coding Agents

    the University of British Columbia · the Georgia Institute of Technology

  14. Give Them an Inch and They Will Take a Mile:Understanding and Measuring Caller Identity Confusion in MCP-Based AI Systems

    Mar 8, 2026Yuhang Huang, Boyang Ma, Biwei Yan +5MCP SecurityAI Agent Security

    Shandong University Qingdao, Shandong, China · City University of Hong Kong Hong Kong, Hong Kong, China

  15. Beyond Reward Suppression: Reshaping Steganographic Communication Protocols in MARL via Dynamic Representational Circuit Breaking

    Mar 7, 2026Liu Hung MingAI Agent SafetyAI Agent Auditing

    PARRAWA AI

  16. ESAA-Security: An Event-Sourced, Verifiable Architecture for Agent-Assisted Security Audits of AI-Generated Code

    Mar 6, 2026Elzo Brito dos Santos FilhoAI Agent AuditingSoftware Security

  17. EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair

    Mar 5, 2026Jiaao Chen, Jingyuan Qi, Mingye Gao +3Synthetic Data Generation for Language ModelsAI Agent Benchmarks

    Eigen AI

  18. Auditing Cascading Risks in Multi-Agent Systems via Semantic-Geometric Co-evolution

    Mar 4, 2026Zixun Luo, Yuhang Fan, Hengyu Lin +2Ollivier-Ricci CurvatureMulti-Agent LLM Systems

    Huazhong University of Science and Technology · Lingnan University · Tsinghua University +1

  19. DUCX: Decomposing Unfairness in Tool-Using Chest X-ray Agents

    Feb 28, 2026Zikang Xu, Ruinan Jin, Xiaoxiao LiAlgorithmic FairnessAlgorithmic Bias

    Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Anhui, China · The University of British Columbia, Vancouver, BC V6Z 1Z4, Canada · Vector Institute, Toronto, ON M5G 1M1, Canada

  20. The Comprehension-Gated Agent Economy: A Robustness-First Architecture for AI Economic Agency

    Feb 27, 2026Rahul BaxiAI Agent GovernanceAgentic AI

    Independent Researcher

  21. A Framework for Assessing AI Agent Decisions and Outcomes in AutoML Pipelines

    Feb 25, 2026Gaoyuan Du, Amit Ahlawat, Xiaoyang Liu +1AI Agent EvaluationAI Agent Auditing

    Amazon Web Services · Amazon Security

  22. Agile V: A Compliance-Ready Framework for AI-Augmented Engineering -- From Concept to Audit-Ready Delivery

    Feb 24, 2026Christopher Koch, Joshua Andreas WellbrockAgentic WorkflowsAutomated Software Engineering

    Agile-V.org Germany

  23. Right to History: A Sovereignty Kernel for Verifiable AI Agent Execution

    Feb 23, 2026Jing ZhangAI Agent AuditingAI Accountability

    Independent Researcher

  24. OpenPort Protocol: A Security Governance Specification for AI Agent Tool Access

    Feb 22, 2026Genliang Zhu, Chu Wang, Ziyuan Wang +2Tool Access Control for LLM AgentsAI Agent Governance

    Accentrust · Georgia Institute of Technology · University of Illinois Urbana-Champaign +2

  25. The Convergence of Schema-Guided Dialogue Systems and the Model Context Protocol

    Feb 21, 2026Andreas SchlapbachLLM Tool UseModel Context Protocol

    SBB (Swiss Federal Railways), SBB IT, Bern, Switzerland

  26. The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systems

    Feb 19, 2026Leon Staufer, Kevin Feng, Kevin Wei +6AI Agent AuditingAI Agent Safety

    University of Cambridge, United Kingdom · Massachusetts Institute of Technology, USA · Hebrew University of Jerusalem, Israel

  27. Colosseum: Auditing Collusion in Cooperative Multi-Agent Systems

    Feb 16, 2026Mason Nakamura, Abhinav Kumar, Saswat Das +5Multi-Agent LLM SystemsMulti-Agent Systems

    University of Massachusetts Amherst · University of Virginia · ELLIS Institute Tübingen +1

  28. Towards Selection as Power: Bounding Decision Authority in Autonomous Agents

    Feb 16, 2026Jose Manuel de la Chica Rodriguez, Juan Manuel Vera DíazAI Agent GovernanceRuntime Enforcement for AI Agents

    AI Lab, Grupo Santander, Madrid, Spain

  29. PAIR-SAFE: A Paired-Agent Approach for Runtime Auditing and Refining AI-Mediated Mental Health Support

    Jan 19, 2026Jiwon Kim, Violeta J. Rodriguez, Dong Whi Yoo +2Mental Health Conversational AgentsAI Agent Auditing

    University of Illinois Urbana-Champaign · Indiana University Indianapolis