Question

A financial services company must ensure that every machine learning model is formally reviewed and approved by a compliance officer before it can be deployed to production. They need a way to catalog model versions, track their approval status, and prevent unapproved versions from being promoted. Which AWS capability best supports this governance requirement?

Accepted Answer

Use Amazon SageMaker Model Registry to register model versions and manage an approval status that gates deployment.

Answer

Store each model artifact in an Amazon S3 bucket with versioning enabled and rely on bucket policies for review.

Answer

Enable AWS CloudTrail to log all deployment API calls and review them after models go live.

Answer

Use Amazon Macie to scan model artifacts for sensitive data before deployment.