【Monitoring】Can AI Models Also Go Off Track? How Does the Operations Command Center Monitor the Stability and Fairness of AI (MLOps)?