In an Azure Machine Learning project, you need to manage and package a model using MLflow. The model should be evaluated following responsible AI guidelines. What approach should you take to ensure optimal model management and adherence to ethical AI principles?