In AI governance, what does 'model lifecycle management' encompass?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — you need to treat model lifecycle management as the whole journey, not just one phase. It covers a model from initial concept through development, validation, deployment, monitoring, and eventual decommissioning. Anything narrower misses the governance picture.
Full Explanation
The complete set of processes governing a model from initial concept through development, validation, deployment, monitoring, and eventual decommissioning is correct because model lifecycle management is a comprehensive governance concept that covers all stages of a model's existence: business case and concept, data preparation, model development, validation, deployment, ongoing monitoring, model changes and retraining, and eventual decommissioning. Limiting it to only the training and testing phases of model development ignores most of the lifecycle. Limiting it to the ongoing monitoring and retraining of AI models after they have been deployed to production ignores pre-deployment governance. Treating it as only the documentation and approval processes required before a model is deployed captures elements of lifecycle management but does not define it in full.