AAIA Practice Test

AAIA Practice Test practice questions, page 17 of 23, with answer review and explanations.

Question 81

An IS auditor is evaluating an organization’s data governance controls for its AI system. Which of the following represents the GREATEST risk in this context?

  1. Inconsistent data management practices
  2. Lack of procedures for automated data backup
  3. Limited frequency of AI system performance and data accuracy reviews
  4. Inadequate controls over data accuracy and privacy compliance

Question 82

Which of the following is the GREATEST risk of using AI to generate audit reports?

  1. The AI system uses inconsistent formatting across audit reports.
  2. The AI system misrepresents control effectiveness.
  3. The AI system cannot integrate with management dashboard tools.
  4. The AI system is not able to include historical audit findings.

Question 83

Which of the following should be an IS auditor’s GREATEST concern if class imbalance is identified in training data for an AI model?

  1. Data drift
  2. Data quality
  3. Model bias
  4. Model overfitting

Question 84

A healthcare AI tool used by a hospital recommends treatments based on patient data. The tool sometimes suggests treatments that have high success rates but significant risk. The hospital's risk management approach prioritizes patient safety over innovative solutions. Which of the following is the hospital's BEST course of action?

  1. Adjust the AI’s parameters to align with the hospital’s risk tolerance.
  2. Discontinue using the AI tool and rely solely on doctor expertise.
  3. Obtain patients' consent for the use of their data by the AI tool.
  4. Use the AI tool only for low-risk situations.

Question 85

An AI audit reveals that a loan approval model has a significantly higher rejection rate for applicants from a specific demographic group. Which of the following should be management's PRIMARY response?

  1. Accept the audit findings as within risk tolerance.
  2. Determine if audit sampling is sufficient.
  3. Conduct comprehensive bias analysis.
  4. Synthesize more data of the affected demographic group.