AI Models, Considerations, and Requirements AAIA Practice Questions (Page 7)

Questions 31–35

1. 31application · medium

   A multinational corporation is deploying an AI model for hiring decisions. The model was developed by a third-party vendor. The company's governance board wants to ensure oversight and accountability throughout the model's lifecycle. Which governance structure is most appropriate?

   • ADelegate all oversight to the vendor since they developed the model.
   • BRely on external audits only after the model is deployed.
   • CEstablish a cross-functional AI governance committee with representatives from legal, compliance, HR, and IT.
   • DCreate a single data science team to manage all model decisions.
   Reveal Solution Discussion Select an answer first
2. 32expert · hard

   A pharmaceutical company is developing an AI model to predict drug side effects. The model is intended for use in clinical trials. The company must comply with strict regulatory requirements and ensure patient safety. The model has high accuracy but low recall for rare but severe side effects. Which action is most appropriate?

   • AAdjust the decision threshold to increase recall, even if it reduces precision.
   • BKeep the current threshold because accuracy is high.
   • CUse a more complex model to improve both precision and recall.
   • DRely on the model's high accuracy to justify its use in clinical trials.
   Reveal Solution Discussion Select an answer first
4. 33application · medium

   An insurance company is preparing for an external audit of its AI underwriting model. The auditor will assess compliance with internal policies and regulatory requirements. Which documentation is most critical to have ready?

   • AA list of the model's training data sources.
   • BThe resumes of the data science team members.
   • CA summary of the model's accuracy on the test set.
   • DA comprehensive model documentation package including development, validation, and monitoring records.
   Reveal Solution Discussion Select an answer first
5. 34foundation · easy

   Which risk management framework is specifically designed to address risks associated with artificial intelligence systems, including governance and lifecycle management?

   • AISO 27001
   • BCOBIT 2019
   • CTOGAF
   • DNIST AI Risk Management Framework
   Reveal Solution Discussion Select an answer first
6. 35application · medium

   A city government is deploying an AI system to allocate public housing resources. Community advocates are concerned about transparency and fairness. The city wants to address these ethical concerns. Which action is most aligned with ethical AI principles?

   • AKeep the model's logic confidential to prevent gaming the system.
   • BUse a proprietary model with no external oversight.
   • CLimit public access to the system's outcomes to avoid disputes.
   • DPublish an explanation of the model's decision-making process and allow appeals.
   Reveal Solution Discussion Select an answer first