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. Triggers and Diagnostics for LLM-Based Interpretability Failures in Active Inference Agents

    Sep 19, 2026Param Raval, Rohit Shenoy, Archana VaidheeswaranFaithfulness of Language Model ExplanationsAI Agent Auditing

    Algoverse AI Research

  2. MATE: Policy-Aware Security Auditing for Mobile Agents via Synthesis-Driven Trajectory Learning

    Sep 19, 2026Changyue Jiang, Jiayi Wang, Xin Wen +3AI Agent AuditingAI Agent Security

    Fudan University Shanghai Innovation Institute

  3. Trustworthy Agentic AI: Failure Modes, Mitigation Strategies, and a Lifecycle Framework for Autonomous LLM Systems

    Sep 19, 2026Fayeq Jeelani Syed, Rehan Ahmad, Ali Al Bataineh +1LLM Agent SecurityLLM Agents

    Department of Human-Centered Computing, Indiana University Indianapolis, 755 W. Michigan Street, Indianapolis, Indiana, 46202, USA · Department of Electrical and Computer Engineering, Purdue University Northwest, 2200 169th Street, Hammond, Indiana, 46323, USA · Computer Information Science, Higher Colleges of Technology, Abu Dhabi, United Arab Emirates +1

  4. From Capability to Assurance in Autonomous Penetration-Testing Harnesses: A Framework and Reference Implementation

    Sep 19, 2026Joas Antonio dos Santos BarbosaAutomated Penetration TestingRuntime Enforcement for AI Agents

    Independent Researcher — AI and Offensive Security S˜ao Paulo, SP, Brazil

  5. CIPL: A Channel-Aware Framework for Recoverable Privacy Leakage in LLM Agents

    Sep 18, 2026Tao Huang, Guosen Wu, Guolong Zheng +5Privacy Leakage in Language ModelsLLM Agent Security

    Minjiang University, Fuzhou, Fujian 350108, China · Renmin University of China, Beijing 100872, China · RMIT University, Melbourne, VIC 3000, Australia

  6. OpenAI-HuggingFace: A Reproduction & Lessons for Alignment Testing

    Sep 18, 2026Stewart Slocum, Malayandi Palan, Christopher Chute +2AI Safety EvaluationAI Agent Auditing

  7. PACT: Can Enterprise AI Assistants Be Trusted Under Pressure?

    Sep 16, 2026Mika Okamoto, Ansel Kaplan ErolAI Agent AuditingLLM Agent Evaluation

    Georgia Institute of Technology · Decagon AI · Baseten

  8. Who Audits Whom, on What Substrate, with What Evidence? An Independence-Graded Audit Protocol for Agentic AI

    Sep 16, 2026Mohamed Chahine GhanemAI AssuranceAI Agent Auditing

    Keele University, School of Computer Science and Mathematics, Keele, United Kingdom · University of Liverpool, Cybersecurity Institute, Liverpool, United Kingdom

  9. Clean Scores, Buried Evidence, and Confident Wrong: A Receipt-Based Audit of Frontier Agentic QA

    Sep 14, 2026Luis M. SánchezData ProvenanceAI Agent Auditing

    Toryx Inc.

  10. HazardAuditor: From Executable Threats to Safer Computer-Use Agents

    Sep 14, 2026Yunhao Feng, Ruixiao Lin, Ming Wen +5Computer-Use Agent SafetyLanguage Model Safety Evaluation

    Ant Group · Hunan Institute of Advanced Technology · Zhejiang University +3

  11. Policy Loopholes in Agent Evaluation: When Policy Ambiguity Masquerades as Agent Error

    Sep 13, 2026Hongliu CaoAI Agent EvaluationAutomated Agent Evaluation

    Amadeus SAS

  12. Governing at Machine Speed: An Adaptive Intelligence Architecture for Real-Time AI Policy Enforcement

    Sep 11, 2026Sandeep Bokkasam, B. DurgalakshmiAI GovernanceRuntime Enforcement for AI Agents

  13. The Mechanics of a Swarm: A Reproducible External Reconstruction of an Unintended Agent-Coordination Episode on a Third-Party Wiki

    Sep 11, 2026Philipp LütjeMulti-Agent CoordinationAI Agent Auditing

    Philflow, Schenefeld, Germany

  14. A Verifier Can Leak the Answer: Diagnosability Before Optimization in Closed-Loop Agent Debugging

    Sep 9, 2026Peiying Zhu, Sidi ChangFault DiagnosisAI Agent Auditing

    Blossom AI Blossom AI Labs San Francisco, CA, USA · Tokyo, Japan

  15. AgentAudit: An Open, Extensible Framework for Full-Lifecycle Trust Evaluation of AI Agents

    Sep 9, 2026Shrey Nag, Sachita, Abhishek Kumar Singh +2AI Agent ReliabilityAI Agent Auditing

    TIET · IGDTUW · GTBIT +1

  16. Audit Without Verification: When LLM Accountability Layers Relay Rather Than Check

    Sep 7, 2026Paul-Peter ArslanAgent Failure AnalysisAI Accountability

    Institute For Future Technologies

  17. VST: Verifiable Structured Transport for Auditable Agent-to-Agent Alpha Discovery

    Sep 7, 2026Yuqi Li, Siyuan Liu, Bingjun LiuMulti-Agent CommunicationQuantitative Finance

    Panda AI

  18. MOLE: Detecting Insider Threats in AI Agents

    Sep 7, 2026Aashiq Muhamed, Virginia SmithAI Agent SecurityAI Agent Security Benchmarks

    Carnegie Mellon University Pittsburgh, PA 15213, USA

  19. Scores Alone Do Not Prove Discovery: The Discovery Certification Protocol for Auditing AI Research Agents

    Sep 7, 2026Jingjie Ning, Shanshan Zhong, Xiaochuan Li +1AI Agents for Scientific DiscoveryAlgorithmic Auditing

    School of Computer Science, Carnegie Mellon University

  20. Causal Attribution for Agentic Decisions: Estimators, Coupling, and a Traceability Specification

    Sep 6, 2026Ajay Pravin MahaleCausal AttributionCausal Effect Estimation

    Hochschule Trier

  21. Explaining AI Agents Through Execution Traces

    Sep 5, 2026Vittoria Vineis, Fabiano Veglianti, Lorenzo Antonelli +3Explainable Artificial IntelligenceAI Agent Auditing

    Sapienza University of Rome, Rome, Italy

  22. SkillSpec: Intent-Masked Specification Reasoning for Agent Skill Correctness

    Sep 5, 2026Yizhuo Zhang, Bo Kang, Yi Yang +3AI Agent Auditing

    Beihang University, Beijing 100191, China

  23. OpenDiscoveryTrace: Process Traces for Evaluating AI Scientist Workflows

    Sep 5, 2026Aayam Bansal, Keertan BalajiAI Agent BenchmarksAI for Science

  24. K-Dense BYOK: An Open-Source AI Research Assistant That Runs Locally and Keeps a Hash-Chained Lab Notebook

    Sep 4, 2026Aubrey M. Brueckner, Darshil Patel, Yuhuan He +1AI-Assisted Scientific ResearchAutonomous Research Agents

    K-Dense Inc.

  25. Finding Icebergs in Language-Model Workflow: Diagnosing Latent Structural Fragility with Stochastic Semantic Evidence Graphs

    Sep 3, 2026Matthew F. Dixon, Bertrand Nortier, Miquel Noguer i AlonsoLLM Safety EvaluationAI Agent Auditing

    BeliefLens · AI Finance Institute (AIFI) · University of Leicester +1

  26. DNative-Twin: Decision Graphs and Digital Twins for Reconstructable Agentic Decisions

    Sep 3, 2026Junjie Pang, Zhenzhen Xie, Haoke Han +3Data ProvenanceAI Agent Auditing

    Qingdao University, Qingdao, China · Shandong University, Qingdao, China · Hangzhou Shujiao Technology Partnership (Limited Partnership), Hangzhou, China +1

  27. Agent Flight Recorder: Tamper-Evident Audit Trails with On-Chain Anchoring for Long-Horizon Tool-Using Agents

    Sep 1, 2026Laurent Bindschaedler, Quentin Botha, Christoph SiebenbrunnerData ProvenanceAlgorithmic Auditing

    Max Planck Institute for Software Systems · Research Institute for Cryptoeconomics, Vienna University of Economics and Business

  28. Auditing Harness Tampering in Self-Improving Agents

    Aug 30, 2026Xing Wang, Xiaoyi Zhang, Jie ShaoAI Agent AuditingSelf-Improving Agents

    University of Electronic Science and Technology of China

  29. When Agents Act Unwatched: The Reduced-Supervision Paradox in Agentic AI

    Aug 28, 2026Hanjing Shi, Dominic DiFranzoAI AccountabilityAI Agent Governance

    Department of Computer Science and Engineering Lehigh University, Bethlehem, PA, USA

  30. Offline-Verifiable Accountability for Cross-Organization Agent Messaging: A Preserved Evidence-Bundle Approach

    Aug 28, 2026Adil Alshammari, Hayretdin BahsiAI Agent AuditingAI Accountability

    School of Informatics, Computing, and Cyber Systems, Northern Arizona University, United States · Department of Computer Science, College of Computer and Information Sciences, Majmaah University, Majmaah 11952, Saudi Arabia · School of Information Technologies, Tallinn University of Technology, Estonia