AI Asset and Data Life Cycle Management AAISM Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   What is the PRIMARY goal of AI model life cycle management?

   • ATo maximize the number of models deployed
   • BTo eliminate the need for human oversight
   • CTo ensure models remain effective and compliant throughout their life
   • DTo minimize the cost of model training
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A data science team is developing a new recommendation model. They have made several changes to the model architecture and hyperparameters. A colleague from the risk team asks them to reproduce the exact model that was used to generate a specific report last week. What is the MOST critical practice the team should have followed to make this possible?

   • AEmail the model's Python code to the risk team for their records.
   • BStore the final model file in a shared network drive.
   • CSave the model's code, training data, and hyperparameters in a version control system with a unique version tag.
   • DDocument the model's accuracy on the test set in a shared document.
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A hospital's AI governance committee is classifying its AI assets. The assets include: (1) a patient triage model that predicts sepsis, (2) an internal chatbot for IT helpdesk tickets, and (3) a model that forecasts cafeteria demand. The committee has limited resources for detailed audits. Which classification approach BEST prioritizes governance effort?

   • AClassify all three as high criticality because they are all AI models and subject to the same governance requirements.
   • BClassify the cafeteria model as high criticality because it is the only one that directly affects revenue.
   • CClassify the sepsis model as high criticality, the IT chatbot as medium, and the cafeteria model as low, based on patient safety and regulatory impact.
   • DClassify the IT chatbot as high criticality because it is the most widely used asset.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   An external auditor is reviewing a company's AI governance program. The auditor asks to see evidence that the company's AI models comply with internal policies and external regulations. What is the MOST important documentation the company should be able to provide?

   • AA complete AI asset inventory with ownership, versioning, and associated risk assessments.
   • BA collection of the data science team's meeting minutes.
   • CA copy of the company's AI ethics policy.
   • DA list of all the AI models' accuracy scores on their test datasets.
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   A marketing analytics team collects customer browsing data for a recommendation engine. The data is stored indefinitely in a data lake. A new privacy regulation requires that customer data be deleted when it is no longer needed for the purpose for which it was collected. What is the MOST appropriate data life cycle management action?

   • AImplement an automated data retention policy that deletes browsing data after a defined period aligned with the stated purpose.
   • BMove the browsing data to a separate, more secure storage tier.
   • CContinue storing the data but restrict access to only the analytics team.
   • DAnonymize the browsing data so it can be kept indefinitely without privacy concerns.
   Reveal Solution Discussion Select an answer first