A production fraud detection model's AUC has dropped below the minimum acceptable threshold defined in its governance policy, triggering a retirement decision. How does watsonx.governance manage the use case status transition during retirement?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Think of it this way: in real-world AI governance, the use case status is updated to retired in the ai lifecycle and the factsheet records the retirement criteria, date, and responsible parties while preserving the complete audit trail is exactly what teams reach for when they need to handle this scenario. When a model meets retirement criteria in watsonx. On the exam, remember that this falls squarely under the 2.0 AI Lifecycle Governance domain.
Full explanation below image
Full Explanation
When a model meets retirement criteria in watsonx.governance, the correct governance action is to formally transition the use case status to Retired, which clearly marks the model as no longer active in production while preserving all historical records. The AI Factsheet is updated to document the specific retirement trigger, the retirement decision date, the stakeholders who made the decision, and any successor model that replaces the retiring model. The complete audit trail — including training history, validation results, approvals, and monitoring data — is retained in the Factsheet to support future regulatory inquiries that may arise years after decommissioning. The correct answer, "The use case status is updated to Retired in the AI lifecycle and the Factsheet records the retirement criteria, date, and responsible parties while preserving the complete audit trail", directly addresses the scenario described because it aligns with the specific governance requirement in question. The incorrect options ("The platform automatically deletes the use case and archives all Factsheet data to cold storage to free active inventory space and prevent confusion", "The model owner must file a formal retirement request with the corporate legal department before any status change can be made in watsonx.governance", "Monitoring alerts are silenced and the model remains in Deployed status indefinitely until an administrator manually archives the record months later") may seem plausible but do not satisfy the core requirement. Understanding the distinction between these concepts is critical for IBM watsonx.governance implementations and is frequently tested in the 2.0 AI Lifecycle Governance section of the certification exam.