Question 35
Domain 4: Guidelines for Responsible AIA company wants to collaborate with several research institutes to develop an AI model. The company needs standardized documentation of model version tracking and a record of model development. Which solution meets these requirements?
Correct answer: C
Explanation
Amazon SageMaker Model Cards provide standardized documentation for an AI model, including its purpose, training details, evaluation, and lifecycle information. This supports “model version tracking” and a “record of model development” by keeping a consistent, shareable history for collaboration across research institutes.
Why each option is right or wrong
A. Track the model changes by using Git.
Git tracks code and files, not standardized ML model documentation and governance records.
B. Track the model changes by using Amazon Fraud Detector.
Amazon Fraud Detector is for fraud detection use cases, not model documentation management.
C. Track the model changes by using Amazon SageMaker Model Cards.
Amazon SageMaker Model Cards are the AWS feature designed to create standardized, shareable documentation for a model’s lifecycle, including version history, training data, evaluation metrics, intended use, and governance notes. In this scenario, the need is not just storage of artifacts but a consistent record of development and changes across collaborating institutes, which Model Cards are built to capture in one documented place rather than ad hoc notes or code comments.
D. Track the model changes by using Amazon Comprehend.
Amazon Comprehend analyzes text with NLP; it does not maintain model development records.