An enterprise is establishing an AI Trust, Risk, and Security Management (AI TRiSM) program. As part of this program, the security team needs to implement continuous monitoring of AI model outputs for concept drift and anomalous behavior changes post-deployment. Which Azure service provides model monitoring capabilities for production AI models?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — a is correct because Azure Machine Learning's model monitoring feature (preview) continuously monitors deployed models for data drift, prediction drift, and data quality issues in production. It compares production data distributions against a reference baseline and sends alerts when statistical drift is detected—a core requirement of AI TRiSM's continuous monitoring pillar.
Full explanation below image
Full Explanation
A is correct because Azure Machine Learning's model monitoring feature (preview) continuously monitors deployed models for data drift, prediction drift, and data quality issues in production. It compares production data distributions against a reference baseline and sends alerts when statistical drift is detected—a core requirement of AI TRiSM's continuous monitoring pillar. B is wrong because Defender for Cloud CSPM assesses the security posture of Azure resources (misconfigurations, compliance), not ML model output quality and drift. C is wrong because Application Insights is an APM (Application Performance Monitoring) tool for web applications; while it can monitor API latency and failures, it does not perform ML-specific statistical drift detection on model inputs and outputs. D is wrong because Sentinel analytics rules detect security threats from log data; they are not designed for statistical model performance monitoring.