Root cause analysis (RCA) is one of the most time-consuming and least scalable parts of the verification lifecycle. As SoCs grow exponentially in complexity and regressions span millions of cycles, engineers are left manually correlating waveforms, logs, assertions, and coverage data to answer a deceptively simple question: why did this test fail? In this session, we introduce an AI-agent–based approach to root cause analysis, developed and deployed in ChipAgents™ and already used in production verification flows at leading semiconductor companies. We will walk through how autonomous AI agents ingest simulation outputs (logs, waveforms, assertions, coverage), reason across design intent and testbench structure, and iteratively narrow failures to actionable root causes, often in minutes instead of days. Using real-world DV scenarios, we will demonstrate how AI-driven RCA: Automatically localizes failure sources across RTL, testbench, and constraints Distinguishes symptom signals from true root causes Generates explainable hypotheses engineers can validate and fix Scales across large regressions without adding human bottlenecks Attendees will leave with a clear understanding of how AI agents can be integrated into existing verification environments to dramatically reduce debug time, improve first-pass success, and free verification teams to focus on higher-value work.