An ML model deployed with Amazon SageMaker uses Model Monitor. After updating the model, the ML engineer observes data quality failures in Model Monitor checks. What action should the engineer take to address the data quality issues that Model Monitor flagged?

When Model Monitor flags data quality failures after a model update, it indicates that the current data being fed to the model deviates significantly from the baseline data it was trained and validated on. Generating a new baseline from the latest production dataset is the correct action because it recalibrates Model Monitor's expectations to reflect the current data distribution. This allows Model Monitor to accurately assess future data quality against the new, relevant standard. Tuning model parameters or adding more training data are actions to improve model performance, not directly address data quality monitoring issues. Manually starting a Model Monitor job won't resolve the underlying issue of an outdated baseline.