CPMAI_V7 practice question 19 of 35
Your team is working on a new loan decision model that takes a number of factors and data points into consideration and then automatically approves…
Full set: 100 questionsThe disproportionate denial rate for a specific demographic, despite equivalent qualifying attributes, indicates algorithmic discrimination produced by biased training data. If historical data reflect past prejudices or sampling imbalance, the learned model internalises that bias, producing unfair outcomes—an issue well documented in credit-scoring and lending AI systems.
B. Hallucination refers to fabricated text/image content in generative models, not structured loan decisions.
C. Accuracy limits do not systematically disadvantage one protected class; bias is the core problem.
D. Privacy breaches concern data exposure, not inequitable approvals.
1. Barocas, S., Hardt, M., & Narayanan, A. “Fairness and Machine Learning”, ch. 2 “Bias in, Bias out”, pp. 23-27. (https://fairmlbook.org)
2. U.S. Federal Reserve, “Machine Learning in Credit Underwriting” (2021) §3.2 Data Bias, pp. 11-12.
3. Stanford CS221 Ethics Notes, “Algorithmic Fairness in Lending”, slides 5-7.
Keep practicing
Take the free CPMAI_V7 practice test
Ten exam-style questions with answers and explanations, plus the exam facts and study guides.
More questions
Other CPMAI_V7 practice questions
- Question 1You're working with an inexperienced team and this is all their first AI project. You're …
- Question 2Enhancing and cleaning data is an important action during which phase of CPMAI?
- Question 3Your team is working on a new facial recognition application. Since this technology has t…
- Question 4In the case that an algorithm you want to use isn't algorithmically explainable, AI syste…
- Question 5You're running an image recognition project and realize that you do not have enough data …
- Question 6Your team is working on a project and is running into some issues. You need someone on th…
