AI RESEARCH
Near-Miss: Latent Policy Failure Detection in Agentic Workflows
arXiv CS.CL
•
ArXi:2603.29665v1 Announce Type: new Agentic systems for business process automation often require compliance with policies governing conditional updates to the system state. Evaluation of policy adherence in LLM-based agentic workflows is typically performed by comparing the final system state against a predefined ground truth. While this approach detects explicit policy violations, it may overlook a subtle class of issues in which agents bypass required policy checks, yet reach a correct outcome due to favorable circumstances.