Compositional Policy Violations: When Step-Level Compliance Fails In Agentic AI Workflows

arXiv:2609.18820 · cs.AI, cs.MA · Submitted 2026-09-16 · Read on arXiv

cs.AI, cs.MA

Submitted: 2026-09-16

Updated: 2026-09-17

Comments: 11 Pages, 6 Figures, 2 Tables

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Trace — 40× in this paper · explained in 2 episodes
• Provenance — 27× in this paper · explained in 4 episodes
• Compositional Policy Violation — 12× in this paper
• Accuracy — 5× in this paper · explained in 1 episode
• Agentic Workflows — 5× in this paper · explained in 1 episode
• Core — 5× in this paper · explained in 2 episodes
• Verification — 5× in this paper · explained in 2 episodes
• Agentic AI — 3× in this paper · explained in 1 episode
• Accountability — 2× in this paper · explained in 2 episodes
• Actor — 2× in this paper · explained in 3 episodes
• Agentic AI Workflows — 2× in this paper · explained in 1 episode
• Analyzer — 2× in this paper · explained in 1 episode
• Benchmark — 2× in this paper · explained in 2 episodes
• Calibration — 2× in this paper · explained in 7 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Agent Memory — 1× in this paper · explained in 1 episode
• Aggregation — 1× in this paper · explained in 1 episode
• American International Group — 1× in this paper
• Explainability — 1× in this paper · explained in 3 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Human-in-the-Loop — 1× in this paper · explained in 3 episodes
• International Conference on Learning Representations — 1× in this paper
• Interpretability — 1× in this paper · explained in 15 episodes
• Policy Knowledge Base — 1× in this paper
• Temporal Reasoning — 1× in this paper · explained in 2 episodes

Sources

• Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
• Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming
• AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
• Cross-Session Threats in AI Agents: Benchmark, Evaluation, and Algorithms
• SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do
• When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection