AAIA · Question #50
An organization is evaluating change management practices for AI-based decision support models. Which of the following BEST demonstrates effective AI-focused change management?
The correct answer is C. Documenting model updates and retraining sessions to ensure traceability. Effective AI change management centers on maintaining a documented audit trail of all model updates and retraining sessions, which provides traceability and supports accountability for changes to AI decision-making systems. Documentation is the cornerstone of a controlled and aud
Question
An organization is evaluating change management practices for AI-based decision support models. Which of the following BEST demonstrates effective AI-focused change management?
Options
- AEngaging an independent expert to review the model's accuracy and precision on a quarterly basis
- BAssigning a single data science team member to adjust the model in order to establish
- CDocumenting model updates and retraining sessions to ensure traceability
- DDeploying two separate copies of the model after each adjustment to compare results
How the community answered
(35 responses)- A3% (1)
- B11% (4)
- C80% (28)
- D6% (2)
Why each option
Effective AI change management centers on maintaining a documented audit trail of all model updates and retraining sessions, which provides traceability and supports accountability for changes to AI decision-making systems. Documentation is the cornerstone of a controlled and auditable AI change process.
Quarterly independent expert reviews are a form of monitoring or assurance, not a change management practice, and do not ensure that changes are controlled or traceable.
Assigning a single team member to make model adjustments lacks oversight, segregation of duties, and documentation - which are all fundamental change management requirements.
Documenting model updates and retraining sessions creates a traceable record of what changed, when, why, and by whom, which is essential for understanding model behavior over time and investigating incidents or performance degradation. Traceability supports regulatory compliance, enables rollback decisions, and provides the audit evidence required to demonstrate that changes were reviewed and authorized. This directly reflects the core principles of change management - control, accountability, and reversibility.
Deploying two model copies to compare results is an A/B testing or validation technique, not a change management practice, and does not address the governance of the change process.
Concept tested: AI model change management traceability and documentation
Source: https://learn.microsoft.com/en-us/azure/machine-learning/concept-model-management-and-deployment
Topics
Community Discussion
No community discussion yet for this question.