AAIA practice questions 41–50

1. Question 41 · 124 in bank

   When auditing a machine learning (ML) solution, false positives can BEST be assessed by examining the level of:

   • Aprecision.
   • Bcompleteness.
   • Crecall.
   • Daccuracy.

   Show answer

   Answer: A

   Next question

2. Question 42 · 124 in bank

   A healthcare organization uses an AI model to analyze patient data and provide diagnostic recommendations. Which of the following MOST effectively detects data drift related to the model’s predictions?

   • AUsing adversarial testing to simulate scenarios that stress test the model’s predictions
   • BConducting periodic model retraining to ensure alignment with updated patient data
   • CComparing incoming patient data distributions with the training data set
   • DApplying overrides to allow healthcare professionals to correct the AI model's recommendations

   Show answer

   Answer: C

   Next question

3. Question 43 · 124 in bank

   Which of the following is the MOST critical consideration for incident response in AI systems?

   • ADeveloping a process to monitor and correct biased outputs in AI models
   • BAssigning accountability for managing and resolving AI -related incidents
   • CDeploying automated incident response systems to flag unusual patterns in AI model behavior
   • DConducting regular training to enhance understanding of AI troubleshooting

   Show answer

   Answer: B

   Next question

4. Question 44 · 124 in bank

   An IS auditor is assessing an AI-powered chatbot that produces inaccurate or fabricated information for customer queries. Which of the following is the BEST recommendation when the model’s configuration emphasizes creativity over accuracy?

   • AAdd more randomness through external prompts.
   • BAllow the AI model to self-correct its hallucinated outputs.
   • CImplement stricter prompt templates to constrain the model's outputs.
   • DAdjust the model’s temperature parameter to lower randomness.

   Show answer

   Answer: D

   Next question

5. Question 45 · 124 in bank

   From a data appropriateness and bias perspective, which of the following should be of GREATEST concern when reviewing an AI model used in a credit scoring system?

   • AThe model incorporates the applicant’s loan history to assess spending habits.
   • BThe model considers the applicant’s income level as a key factor in the credit decision.
   • CThe model uses postal codes as a primary factor in determining creditworthiness.
   • DThe model utilizes historical credit data to predict future credit behavior.

   Show answer

   Answer: D

   Next question

6. Question 46 · 124 in bank

   Which of the following metrics should an IS auditor recommend as a key performance indicator (KPI) to assess the reliability of a natural language processing AI model?

   • ALandmark localization error
   • BReceiver operating characteristic curve
   • CRecall-oriented understudy for gisting evaluations
   • DMean squared error

   Show answer

   Answer: C

   Next question

7. Question 47 · 124 in bank

   An organization shares an AI model with external partners. One partner reports that sensitive data has been inadvertently exposed through the model's outputs. Which of the following is the IS auditor's BEST recommendation?

   • ARetrain the model immediately and implement privacy-preserving techniques.
   • BDisable the shared model and notify partners of the potential breach.
   • CLimit the model's outputs to anonymized results while investigating further.
   • DAudit the data pipelines of all partners to identify the source of the leak.

   Show answer

   Answer: A

   Next question

8. Question 48 · 124 in bank

   Which of the following controls helps mitigate the risk of competitors poisoning data utilized by a machine learning (ML) model performing sentiment analysis of product reviews?

   • APeer reviewing code that acquires product reviews from social media posts
   • BHiring a marketing firm to text links to customers requesting product reviews for monetary compensation
   • CRequiring customers to authenticate access to their accounts prior to writing product reviews
   • DAugmenting the unbalanced product review data set with the use of oversampling by the model developer

   Show answer

   Answer: C

   Next question

9. Question 49 · 124 in bank

   A deepfake video is presented as evidence in a court case, falsely implicating an individual in criminal activities. Which of the following is an IS auditor’s BEST recommendation to help ensure the authenticity of digital evidence?

   • AEmploying specialized tools designed to detect deepfake artifacts and inconsistencies
   • BStoring all digital evidence on secure remote cloud servers and on local servers
   • CRequiring digital watermarks for any evidence presented in legal proceedings
   • DTraining legal and forensic professionals to recognize deepfake characteristics in evidence

   Show answer

   Answer: A

   Next question

10. Question 50 · 124 in bank

    If an organization’s business objectives require an AI solution that continually learns from the results of its outputs, an IS auditor would need to confirm risk and controls around:

    • Abackpropagation.
    • Bapplying biases.
    • Cactivation functions.
    • Dapplying weights.

    Show answer

    Answer: A

    Next question

Keyboard:J Kmove between questionsA–Dselect an answerRshow or hide the answer← →change page

AAIA practice, question 41. 124 questions available.