Effective model monitoring is crucial for ensuring the sustained performance and reliability of AI/ML and Generative AI (GenAI) models in production environments. Recent failures experienced by organizations, such as Zillow and a major credit bureau, underscore the importance of robust monitoring practices to detect and mitigate the risks associated with model drift, performance degradation, and misalignment with objectives. This chapter provides a comprehensive approach to model monitoring that integrates technical assessments and governance considerations. The key components of the monitoring plan include documentation of the methodology, clearly defined roles and responsibilities, appropriate monitoring frequency, and well-defined performance metrics. For GenAI models, metrics such as the Kolmogorov–Smirnov statistic, precision, recall, F1 score, BLEU, ROUGE, perplexity, and factual consistency are essential for evaluating the quality and relevance of the generated outputs. Assumption monitoring, statistical distribution monitoring using metrics such as the population stability index and Kullback–Leibler divergence, stress testing, back-testing, and sensitivity analysis are also critical for ensuring the robustness and reliability of the AI/ML and GenAI models. The monitoring plan should provide detailed results at the granular level with performance thresholds defined by the model owner based on statistical methodologies. Regular monitoring of these metrics enables early detection of anomalies, allowing timely interventions to maintain model performance, regulatory compliance, and alignment with organizational objectives. By combining rigorous technical assessments with thoughtful governance, organizations can foster AI systems that are innovative, safe, ethical, and beneficial to society while safeguarding against potential risks and failures.