AI Policies, Procedures, and Organizational Training AAIR Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   Which requirement must an AI policy address to be comprehensive?

   • AHardware procurement
   • BEmployee vacation schedules
   • COffice layout
   • DEthical considerations
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   What is the role of compliance oversight in AI governance?

   • ATo develop AI algorithms
   • BTo monitor adherence to policies and regulations
   • CTo manage IT infrastructure
   • DTo market AI products
   Reveal Solution Discussion Select an answer first
4. 18expert · hard

   A government agency is developing an AI policy framework. The agency must comply with strict procurement rules, maintain public trust, and allow innovation. The procurement team wants AI acquisitions to follow standard IT procurement, while the AI team wants a fast-track for low-risk prototypes. The public affairs office requires transparency about AI use. Which framework component best balances these competing needs?

   • AA policy that keeps AI use confidential to avoid public scrutiny, with no procurement changes.
   • BA policy with a risk-based procurement pathway: low-risk AI prototypes use a simplified process, while high-risk AI follows full procurement rules, and all AI use is publicly disclosed.
   • CA policy that allows the AI team to bypass procurement rules for any AI project to speed up deployment.
   • DA single policy that treats all AI acquisitions as standard IT procurement, with no exceptions.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   An e-commerce company is writing an AI policy for its recommendation engine. The marketing team wants to maximize personalization, the legal team requires compliance with data protection laws, and the customer service team needs to explain recommendations to customers. Which policy provision best addresses all three?

   • AA policy that limits personalization to data collected with consent, requires a clear explanation of how recommendations are generated, and provides a process for customers to query decisions.
   • BA policy that allows unlimited data collection for personalization, with no explanation requirements.
   • CA policy that allows personalization but requires customers to sign a waiver of their data protection rights.
   • DA policy that prohibits all personalization to avoid legal risk.
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   A manufacturing company is planning AI training for its workforce. The plant floor supervisors use AI-based predictive maintenance dashboards, the data science team builds custom models, and the executive team approves AI budgets. A needs assessment reveals that supervisors can interpret alerts but not explain model limitations, data scientists lack formal risk assessment training, and executives are unaware of regulatory exposure. Which training program design addresses the most critical gaps?

   • ARole-based training: supervisors receive model limitation and override training, data scientists receive risk assessment and governance training, and executives receive regulatory and oversight training.
   • BExecutive training on AI strategy only, leaving technical and operational training to on-the-job learning.
   • CTechnical training for data scientists only, since they are the primary AI developers.
   • DA single mandatory course for all employees covering AI fundamentals and basic risk concepts.
   Reveal Solution Discussion Select an answer first