MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

1. Which type of bias occurs when an AI model’s predictions are distorted due to non-representative training data?
   a. Historical bias
   b. Selection bias
   c. Evaluation bias
   d. Amplification bias
   Answer: b. Selection biasitrexgroup+1​
2. What is algorithmic bias in the context of AI?
   a. Intentional design to favor a group
   b. Systematic errors due to skewed data or flawed algorithms
   c. Human error during system deployment
   d. Variability across environments
   Answer: b. Systematic errors due to skewed data or flawed algorithmspsbtrainingacademy+1​
3. Which fairness metric measures whether positive outcomes are evenly distributed across groups?
   a. Predictive parity
   b. Demographic parity
   c. Equalized odds
   d. Treatment equality
   Answer: b. Demographic parityverifywise​
4. What is “counterfactual fairness” in AI?
   a. Ensuring equal treatment for all inputs
   b. Modifying training data for unbiased prediction
   c. Fairness achieved by removing sensitive features
   d. Bias mitigation by data augmentation
   Answer: b. Modifying training data for unbiased predictionieeexplore.ieee+1​
5. Which bias arises when the frequency of events in a dataset doesn’t reflect reality?
   a. Reporting bias
   b. Societal bias
   c. Automation bias
   d. Cultural bias
   Answer: a. Reporting biasdevelopers.google+1​
6. Why is fairness important in AI systems?
   a. To increase efficiency
   b. To avoid unfair treatment of groups
   c. To reduce computational complexity
   d. To improve explainability
   Answer: b. To avoid unfair treatment of groupslearncbse+1​
7. What does “disparate impact” refer to in fairness evaluation?
   a. The intentional exclusion of subgroups
   b. Disproportionate effects on protected groups
   c. Reduced prediction accuracy
   d. Algorithmic transparency
   Answer: b. Disproportionate effects on protected groupspsbtrainingacademy​
8. Which technique can help mitigate AI bias in decision-making?
   a. Ignoring outliers
   b. Fairness-aware modeling
   c. Increasing dataset size
   d. Data normalization
   Answer: b. Fairness-aware modelingdevelopers.google+1​
9. What is a common challenge in ensuring fairness for AI in recruitment?
   a. Lack of data
   b. Replicating historical biases
   c. Slow candidate processing
   d. Expensive technology
   Answer: b. Replicating historical biasesresearch.aimultiple+1​
10. Which AI bias is often amplified when models learn from human prejudices?
    a. Amplification bias
    b. Confirmation bias
    c. Algorithmic bias
    d. Implicit bias
    Answer: a. Amplification biasresearch.aimultiple​
11. What is sample bias in AI?
    a. Training AI using faulty algorithms
    b. Model trained on unrepresentative samples
    c. Ignoring ethical considerations
    d. Increasing model complexity
    Answer: b. Model trained on unrepresentative sampleslumenova+1​
12. Which fairness metric evaluates whether true positive/false positive rates are similar between groups?
    a. Predictive parity
    b. Equalized odds
    c. Demographic parity
    d. Disparate impact ratio
    Answer: b. Equalized oddsverifywise​
13. What type of bias results from extrapolating individual traits to entire groups?
    a. Group attribution bias
    b. Reporting bias
    c. Measurement bias
    d. In-group bias
    Answer: a. Group attribution biasitrexgroup+1​
14. How can bias be reduced in a healthcare AI model?
    a. Use only majority demographic data
    b. Apply fairness metrics and diversify datasets
    c. Aggregate data for all demographics
    d. Increase computation power
    Answer: b. Apply fairness metrics and diversify datasetslearncbse+1​
15. What is historical bias in AI?
    a. Bias in reporting frequencies
    b. Bias reproduction from past prejudices in data
    c. Bias generated by model architectures
    d. Bias in data labeling
    Answer: b. Bias reproduction from past prejudices in datadevelopers.google+1​
16. Which is NOT a pillar of AI ethics?
    a. Explainability
    b. Fairness
    c. Efficiency
    d. Privacy
    Answer: c. Efficiencylearncbse​
17. Which of the following is NOT a type of bias in AI?
    a. Sampling bias
    b. Measurement bias
    c. Animation bias
    d. Reporting bias
    Answer: c. Animation biasgeeksforgeeks+1​
18. Which action helps ensure fairness in AI-generated resumes?
    a. Remove identity-specific info from training data
    b. Use GANs only
    c. Limit dataset size
    d. Ignore demographic data
    Answer: a. Remove identity-specific info from training datagyansetu​
19. What is meant by “automation bias” in AI?
    a. Reliance on automated outputs, ignoring human judgment
    b. Giving preference to faster algorithms
    c. Bias from human labeling errors
    d. Bias from regional preferences
    Answer: a. Reliance on automated outputs, ignoring human judgmentgeeksforgeeks+1​
20. How can case study diversity help reduce bias in AI?
    a. It doesn’t affect bias
    b. Ensures representation of all populations
    c. Increases algorithmic complexity
    d. Promotes only efficiency
    Answer: b. Ensures representation of all populationsshelf+1​
21. Which of these best describes evaluation bias?
    a. Bias caused by data labeling
    b. Testing models on non-representative test sets
    c. Only validating models locally
    d. Ignoring fairness metrics
    Answer: b. Testing models on non-representative test setsresearch.aimultiple​
22. What is the primary goal of the NIST AI Risk Management Framework?
    a. To increase profit margin
    b. Structured assessment and mitigation of AI risks
    c. Promote proprietary algorithms
    d. Increase model accuracy only
    Answer: b. Structured assessment and mitigation of AI riskslumenova​
23. What does “treatment equality” measure in AI fairness?
    a. Similar balance of false positives and negatives across groups
    b. Total accuracy of the model
    c. Data diversity
    d. Algorithmic transparency
    Answer: a. Similar balance of false positives and negatives across groupsverifywise​
24. What kind of bias is introduced by inconsistent data labeling?
    a. Label bias
    b. Societal bias
    c. Reporting bias
    d. Historical bias
    Answer: a. Label biasresearch.aimultiple​
25. What does fairness in AI ensure?
    a. Models treat everyone equitably
    b. Maximizing efficiency
    c. Reducing computational effort
    d. Increasing engagement
    Answer: a. Models treat everyone equitablylumenova+1​
26. Which law sets guidelines for fair and ethical AI practices in the EU?
    a. AI Justice Act
    b. EU AI Act
    c. General Data Protection Regulation (GDPR)
    d. Algorithmic Accountability Act
    Answer: b. EU AI Actshelf​
27. What is “ontological bias” in AI?
    a. Favoring one worldview over others
    b. Unfair treatment in hiring
    c. Measurement errors in data
    d. Mislabeling in datasets
    Answer: a. Favoring one worldview over othersresearch.aimultiple​
28. Which method is NOT effective for bias mitigation?
    a. Fairness-aware training
    b. Reweighting
    c. Ignoring attribute features
    d. Post-processing
    Answer: c. Ignoring attribute featuresdevelopers.google+1​
29. In which scenario could confirmation bias affect AI performance?
    a. Favoring predictions matching existing beliefs
    b. Preferring inaccurate data
    c. Ignoring positive outcomes
    d. Randomizing dataset selection
    Answer: a. Favoring predictions matching existing beliefsgeeksforgeeks+1​
30. What is “data governance” in ethical AI practice?
    a. Data governance ensures data quality and minimizes bias
    b. Increasing computational resources
    c. Reducing dataset size
    d. Isolating data for specific groups
    Answer: a. Data governance ensures data quality and minimizes biasshelf+1​
31. Which is an example of algorithmic bias?
    a. Hiring algorithm that unfairly favors certain demographics
    b. Oversampling majority group data
    c. Unintentional errors in code
    d. Random selection of training data
    Answer: a. Hiring algorithm that unfairly favors certain demographicslumenova+1​
32. Which fairness metric compares precision rates between demographic groups?
    a. Demographic parity
    b. Predictive parity
    c. Equalized odds
    d. Disparate ratio
    Answer: b. Predictive parityverifywise​
33. Which form of bias is introduced when performance differs across regions due to localized data?
    a. Cultural/geographic bias
    b. Measurement bias
    c. Evaluation bias
    d. Algorithmic bias
    Answer: a. Cultural/geographic biasresearch.aimultiple​
34. How can bias in facial recognition AI be mitigated?
    a. Increase dataset diversity with multiple demographics
    b. Focus on one demographic group
    c. Ignore fairness metrics
    d. Reduce data labeling complexity
    Answer: a. Increase dataset diversity with multiple demographicsitrexgroup+1​
35. What does the “disparate impact ratio” measure?
    a. Ratio of outcomes between groups
    b. Overall accuracy difference
    c. Label consistency
    d. Training time
    Answer: a. Ratio of outcomes between groupsverifywise​
36. Which standard outlines best practices for AI fairness worldwide?
    a. ISO/IEC 24029-1
    b. GDPR
    c. AI Justice Act
    d. FCRA
    Answer: a. ISO/IEC 24029-1verifywise​
37. What is measurement bias?
    a. Bias from inaccurate data acquisition tools or processes
    b. Mislabeling in datasets
    c. Random sampling errors
    d. Lack of demographic diversity
    Answer: a. Bias from inaccurate data acquisition tools or processesgeeksforgeeks​
38. What is the effect of “politeness bias” in LLMs?
    a. Models more likely to obey harmful requests if asked politely
    b. Models ignore polite instructions
    c. Improved fairness in responses
    d. Enhanced algorithmic transparency
    Answer: a. Models more likely to obey harmful requests if asked politelyresearch.aimultiple​
39. Which regulatory framework is designed for algorithmic accountability in the US?
    a. General Data Protection Regulation
    b. NIST RMF
    c. Algorithmic Accountability Act
    d. AI Act
    Answer: c. Algorithmic Accountability Actshelf​
40. What is the significance of transparency in AI fairness regulation?
    a. It allows systems to be audited and improve accountability
    b. It reduces efficiency
    c. It increases computational complexity
    d. It leads to bias amplification
    Answer: a. It allows systems to be audited and improve accountabilityshelf+1​
41. What is aggregation bias in AI?
    a. Bias hiding important differences due to data aggregation
    b. Combining data for training
    c. Ignoring attribute information
    d. Algorithmic errors
    Answer: a. Bias hiding important differences due to data aggregationresearch.aimultiple​
42. Which approach is recommended when evaluating fairness in AI?
    a. Use multiple metrics systematically
    b. Employ a single fairness metric
    c. Ignore stakeholder analysis
    d. Avoid documentation
    Answer: a. Use multiple metrics systematicallyverifywise​
43. Which is an example of bias mitigation during AI development?
    a. Evaluate fairness early and monitor in production
    b. Only test at deployment stage
    c. Ignore metric interpretation
    d. Focus solely on accuracy
    Answer: a. Evaluate fairness early and monitor in productiondevelopers.google+1​
44. Why is stakeholder analysis important in AI fairness?
    a. It clarifies who may be impacted and defines fairness
    b. It increases algorithmic complexity
    c. It promotes random data selection
    d. It is unnecessary for fairness assessment
    Answer: a. It clarifies who may be impacted and defines fairnessverifywise​
45. What does confirmation bias in AI human reviewers entail?
    a. Favoring predictions that confirm existing beliefs
    b. Rejecting all machine predictions
    c. High acceptance of random outcomes
    d. Labeling all data as unbiased
    Answer: a. Favoring predictions that confirm existing beliefsgeeksforgeeks+1​
46. What is the impact of using only local data for model testing?
    a. Model may perform poorly at broader scales due to evaluation bias
    b. Model universally generalizes
    c. Significant reduction in bias
    d. High regional accuracy
    Answer: a. Model may perform poorly at broader scales due to evaluation biasdevelopers.google+1​
47. Which AI bias can occur when designers’ cultural assumptions shape model outputs?
    a. Ontological bias
    b. Sample bias
    c. Reporting bias
    d. Automation bias
    Answer: a. Ontological biasresearch.aimultiple​
48. What is the purpose of documenting assumptions in fairness metrics?
    a. Improves transparency and accountability
    b. Reduces dataset size
    c. Increases algorithmic complexity
    d. Eliminates all bias
    Answer: a. Improves transparency and accountabilityshelf+1​
49. What is the societal bias in AI?
    a. Bias reflecting widespread social prejudices in data/algorithms
    b. Random data sampling
    c. Algorithmic bias from mathematical operations
    d. Cultural labeling errors
    Answer: a. Bias reflecting widespread social prejudices in data/algorithmslumenova+1​
50. Why is continuous monitoring necessary for AI fairness?
    a. Fairness can change over time as models are used and retrained
    b. It reduces computational efforts
    c. It increases data efficiency only
    d. It is unnecessary after deployment
    Answer: a. Fairness can change over time as models are used and retraineddevelopers.google+1​

1. https://itrexgroup.com/blog/ai-bias-definition-types-examples-debiasing-strategies/
2. https://developers.google.com/machine-learning/crash-course/fairness/types-of-bias
3. https://psbtrainingacademy.com/ethical-and-social-implications-of-ai-mcqs-solutions/
4. https://www.lumenova.ai/blog/7-common-types-of-ai-bias/
5. https://verifywise.ai/lexicon/ai-fairness-metrics
6. https://ieeexplore.ieee.org/document/10945860/
7. https://www.learncbse.in/ai-ethics-and-values-class-11-mcq/
8. https://shelf.io/blog/fairness-metrics-in-ai/
9. https://developers.google.com/machine-learning/crash-course/fairness/mitigating-bias
10. https://research.aimultiple.com/ai-bias/
11. https://www.geeksforgeeks.org/artificial-intelligence/fairness-and-bias-in-artificial-intelligence/
12. https://www.gyansetu.in/blog/70-generative-ai-mcq-questions-and-answers-gyansetu/
13. https://www.lumenova.ai/blog/fairness-bias-machine-learning/
14. https://www.sanfoundry.com/ai-bias-fairness-causes-impacts-mitigation/
15. https://www.cloudskillsboost.google/paths/1283/course_templates/985/quizzes/583542
16. https://www.scribd.com/document/809275335/30-MCQs-on-topic-Baiscs-of-AI-and-AI-ethics-1
17. https://study.com/academy/practice/quiz-worksheet-algorithmic-fairness-bias.html
18. https://www.sciencedirect.com/science/article/pii/S0167739X24000694
19. https://www.distancediary.com/iPortfolio/assets/attachments/AI_INTRO/AI%20Ethics%20Question%20Answers.pdf
20. https://testbook.com/objective-questions/mcq-on-artificial-intelligence–5eea6a0e39140f30f369e51f