A bank is developing an AI system to detect fraudulent transactions. The project is in the deployment phase. The compliance team requires an audit trail of model decisions and the ability to roll back to a previous model version if issues arise. Which practice should the program manager ensure is in place BEFORE going live?

A/B testing to compare the new model with the old model.

Model versioning and logging of predictions with timestamps and input data.

A data retention policy for training data only.

Data pipeline monitoring to ensure data freshness.

EC-CouncilCertified AI Program Manager

Domain 1Objective 5

AI Project Life Cycle, MLOps, and DataOps CAIPM Practice Questions (Page 10)

Part of the AI Fundamentals for Business Adoption domain, which makes up ~19% of our current practice bank.

46questions here

10free pages

7concepts

Questions 46–46

46application · medium
A bank is developing an AI system to detect fraudulent transactions. The project is in the deployment phase. The compliance team requires an audit trail of model decisions and the ability to roll back to a previous model version if issues arise. Which practice should the program manager ensure is in place BEFORE going live?
Select an answer first

Finished these 1 questions?

Review the revealed explanations, or continue through the curriculum.

No more pagesBack to CAIPM

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by EC-Council. “CAIPM” is a trademark of its owner, used for identification only.

AI Project Life Cycle, MLOps, and DataOps CAIPM Practice Questions (Page 10)

Questions 46–46

Keep studying — AI Fundamentals for Business Adoption