Dynamic Contact Resistance Measurement (DCRM) is a vital diagnostic technique for assessing the health of Extra High Voltage (EHV) circuit breakers by capturing resistance variations during contact movement. This paper presents an AI-based framework that automates DCRM analysis to accurately detect abnormalities in arcing contacts, main contacts, and operating mechanisms. The system uses machine-learning models to interpret resistance signatures, classify fault patterns, and predict emerging failures before they lead to outages. By integrating historical data, real-time measurements, and intelligent pattern recognition, the proposed approach enhances reliability, reduces maintenance downtime, and enables condition-based monitoring of circuit breakers.