Change Management Specific to AI AAIA Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A company deploys a new recommendation model. After a few hours, they notice a significant drop in user engagement. The change management plan includes a rollback strategy. What should they do first?

   • ARoll back to the previous model version
   • BRetrain the model with more data
   • CInvestigate the cause of the drop before taking any action
   • DConduct a post-change review to document the issue
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A bank's fraud detection model is updated to reduce false positives. The change is approved by the model risk committee, but the customer service team is not informed. After deployment, call volume spikes as customers question declined transactions. What should the bank have done to prevent this outcome?

   • ADocumented the change in the model inventory and version control system
   • BConfigured the model to automatically roll back if call volume increased
   • CIncluded customer service stakeholders in the change impact assessment and communication plan
   • DDeployed the model to a staging environment and ran a full regression test suite
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   Why is version control important for AI datasets?

   • ATo automatically improve data quality
   • BTo reduce the size of datasets
   • CTo track changes to data and enable reproducibility of model results
   • DTo ensure that datasets are stored in the cloud
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   What is a key benefit of applying version control to AI model code?

   • AIt automatically deploys the model to production
   • BIt guarantees that the model is free of bias
   • CIt allows multiple data scientists to work on the same code without conflicts
   • DIt eliminates the need for testing
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   What is the purpose of regression testing for AI changes?

   • ATo measure the model's training time
   • BTo ensure that the new change does not break existing functionality
   • CTo check the model's code for syntax errors
   • DTo test the model on new data for the first time
   Reveal Solution Discussion Select an answer first