Principles of Ethical and Fair AI Use AIE Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A university is developing an AI system to predict student dropout risk. The system uses data including grades, attendance, and socioeconomic status. The university wants to ensure the system does not unfairly target students from low-income backgrounds. What is the most appropriate action?

   • AUse the model only for students with high socioeconomic status to avoid bias.
   • BEvaluate the model's false-positive and false-negative rates across different socioeconomic groups.
   • CRely on the model's overall accuracy as proof that it is fair.
   • DExclude socioeconomic status from the model to ensure fairness.
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which of the following is a core principle of AI ethics?

   • AProfitability
   • BTransparency
   • CSpeed
   • DPopularity
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A city is planning to deploy an AI system that analyzes surveillance camera footage to detect traffic violations. The system will issue automatic fines. The city wants to assess the ethical impact on the community. Which factor should be considered in the stakeholder impact assessment?

   • AThe privacy concerns of the city's IT department.
   • BThe cost of the surveillance cameras and the fines collected.
   • CThe potential for the system to disproportionately affect certain neighborhoods or demographic groups.
   • DThe technical accuracy of the system in detecting violations.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A hospital uses an AI system to predict patient readmission risk. The system is a deep learning model that achieves high accuracy but provides no explanation for its predictions. The hospital's ethics committee requires that clinicians be able to understand why a patient is flagged as high risk. The data science team proposes several options. Which option best satisfies the explainability requirement while minimizing the loss of predictive performance?

   • APublish the model's architecture and training data online.
   • BProvide clinicians with the model's confidence score for each prediction.
   • CUse a model-agnostic explanation technique such as SHAP or LIME to provide local explanations for each prediction.
   • DReplace the deep learning model with a simple logistic regression model.
   Reveal Solution Discussion Select an answer first
6. 10expert · hard

   A nonprofit organization is developing an AI system to allocate limited disaster relief resources. The system is designed to maximize the number of people helped, but it tends to allocate fewer resources to elderly populations because they are less likely to be reachable by the digital communication channels used to verify needs. The organization has limited time and budget. What is the most ethical approach?

   • ACancel the AI project and use manual allocation to avoid ethical risks.
   • BDeploy the system as-is because it maximizes overall impact, which is the primary ethical goal.
   • CModify the system to ensure a minimum allocation to elderly populations, even if it reduces overall efficiency.
   • DUse a random allocation method to avoid any bias.
   Reveal Solution Discussion Select an answer first