3.4 Explain AI Governance Considerations (policy, Risk Management, Compliance) AI-TECHNICAL-PRACTITIONER Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A multinational bank is developing an AI system to approve personal loans. The policy team must create an AI policy that satisfies both the EU's GDPR and the bank's internal operational needs. The policy must clearly define how the AI's decisions are reviewed and how customers can contest them. Which element is most critical to include in the policy?

   • AA clause that allows the bank to override customer objections if the model is deemed accurate.
   • BA policy that limits human review to only high-value loan applications.
   • CA mandate to store all customer data indefinitely to support future audits.
   • DA requirement for a human review process for all automated loan decisions that are contested by customers.
   Reveal Solution Discussion Select an answer first
2. 22expert · hard

   A company is developing an AI policy. The policy must cover the entire lifecycle of AI systems, from development to deployment and decommissioning. The policy team is debating whether to include a section on model monitoring and maintenance. Which argument best supports including this section?

   • AMonitoring is unnecessary if the model was accurate at the time of deployment.
   • BMonitoring should be the responsibility of the data science team only.
   • CMonitoring is only needed for models that are deployed in regulated industries.
   • DModels can degrade over time as data distributions change, so ongoing monitoring is essential to maintain performance and safety.
   Reveal Solution Discussion Select an answer first
4. 23expert · hard

   A multinational corporation is implementing an AI governance framework. The company has operations in regions with different cultural norms regarding data privacy. The governance team must create a framework that is globally consistent but also respects local norms. Which approach best achieves this?

   • ACreate entirely separate frameworks for each region.
   • BCreate a global framework with core principles and allow regional adaptations for specific local requirements.
   • CCreate a single global framework that ignores local differences.
   • DAdopt the framework of the region with the most lenient regulations.
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   What is the primary purpose of data protection regulations like GDPR in the context of AI?

   • ATo standardize the algorithms used in AI models across all industries
   • BTo require all AI systems to be open-source
   • CTo protect individuals' personal data and ensure its processing is lawful and transparent
   • DTo ensure AI systems are deployed on cloud infrastructure
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   Which of the following should be included in an AI policy to address operational requirements?

   • AThe brand of servers used to host AI models
   • BA list of approved programming languages for AI development
   • CThe mathematical formulas used to calculate model accuracy
   • DProcedures for monitoring AI system performance and handling failures
   Reveal Solution Discussion Select an answer first