During a design review, an architect asks which choice best matches the IBM guidance. The team is focused on retirement and archive. Which recommendation is most appropriate?

Here's the deal: you would not ship a production change with no owner and no rollback plan. The same common sense applies here. In this scenario, answer D is the practical move because it keeps the implementation tied to the real IBM capability instead of chasing a shortcut. The other choices sound tempting, but they either skip governance, ignore operational reality, or solve the wrong problem.

The correct answer is D. Retire or archive AI assets when they are no longer used while preserving required evidence. That aligns with the 5.0 Define and Govern AI Use Case Inventory objective because it applies the feature or practice in the context where IBM expects a practitioner to use it. It also keeps the design reviewable, supportable, and realistic for a production environment.

Let's examine why the other options are incorrect: - Option A is incorrect because it uses an overbroad rule instead of matching the design to the actual workload and risk. - Option B is incorrect because it sounds related, but it does not solve the retirement and archive requirement described in the scenario. - Option C is incorrect because it sounds related, but it does not solve the retirement and archive requirement described in the scenario. For the exam, connect the feature to the operational outcome: the right answer is the one that preserves control, accuracy, and maintainability instead of relying on a brittle shortcut. The incorrect options — such as 'Delete all evidence when a model is retired' and 'Leave retired models active in inventory' — describe either out-of-sequence steps or unrelated configuration tasks. This concept falls under the 5.0 Define and Govern AI Use Case Inventory domain of the IBM watsonx.governance certification.