AI Foundations and Technology Ecosystem CRAGE Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   Which statement best distinguishes artificial intelligence (AI) from machine learning (ML) and deep learning (DL)?

   • AML is the broad field of mimicking human cognition; AI is a subset that uses statistical models; DL is a type of AI that requires no data.
   • BAI and ML are synonymous terms for any system that uses data to improve performance; DL is a separate field focused on rule-based reasoning.
   • CAI is limited to rule-based systems; ML and DL are unrelated approaches that do not fall under AI.
   • DAI is the broad field of creating systems that perform tasks requiring human-like intelligence; ML is a subset that learns from data; DL is a subset of ML using multi-layered neural networks.
   Reveal Solution Discussion Select an answer first
2. 12expert · hard

   A government agency is deploying an AI system for welfare eligibility determination. The system must be transparent and auditable. The agency has a limited budget and must choose between building an in-house model with full documentation or purchasing a commercial black-box system. Which approach best balances transparency and cost?

   • AOutsource development to a third-party consultancy and rely on their documentation.
   • BUse a hybrid approach: buy the commercial system but require the vendor to provide explanations.
   • CBuild an in-house model using interpretable algorithms and maintain thorough documentation.
   • DPurchase the commercial system because it is cheaper and faster to deploy.
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A city is implementing an AI system to allocate police resources based on historical crime data. Critics argue the system may perpetuate racial bias. The city council must decide whether to proceed. Which approach best addresses the ethical concern?

   • AUse a different algorithm that is known to be unbiased
   • BConduct a bias audit and involve community stakeholders in oversight
   • CProceed with deployment and monitor for bias after implementation
   • DLimit the system's use to non-sensitive areas
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   An AI system that predicts equipment failures in a factory has been in production for six months. The maintenance team notices that the model's accuracy has declined. Which lifecycle stage should be initiated?

   • ADeployment, because the model needs to be redeployed.
   • BDesign, because the original design was flawed.
   • CData collection, because more data is always the solution.
   • DMonitoring and maintenance, including retraining with new data.
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   Which of the following is a common AI application in the healthcare industry?

   • ARecommending movies based on viewing history.
   • BDetecting fraudulent credit card transactions.
   • CPredicting patient readmission risk using electronic health records.
   • DOptimizing delivery routes for logistics fleets.
   Reveal Solution Discussion Select an answer first