Recent Advancements and Future Directions in AI AIE Practice Questions (Page 7)

Questions 31–35

1. 31application · medium

   A financial institution is exploring AI to detect fraudulent transactions. They want the system to flag suspicious activity but also require that human investigators can review and challenge the AI's decisions. Which design principle best aligns with this requirement?

   • ABlack-box model deployment
   • BHuman-in-the-loop oversight
   • CFully autonomous decision-making
   • DBatch processing without real-time alerts
   Reveal Solution Discussion Select an answer first
2. 32application · medium

   A university is developing an AI tutor that adapts to each student's learning pace. The tutor should provide personalized explanations and practice problems. Which future AI direction does this system exemplify?

   • AGenerative adversarial networks for content creation
   • BArtificial general intelligence
   • CFully autonomous teaching without human teachers
   • DHuman-AI collaboration in education
   Reveal Solution Discussion Select an answer first
4. 33foundation · easy

   What is the primary goal of research into Artificial General Intelligence (AGI)?

   • ATo create machines that can perform any intellectual task that a human being can.
   • BTo develop robots that can physically interact with the environment in real time.
   • CTo build systems that excel at a single, narrow task such as playing chess or recognizing faces.
   • DTo improve the interpretability of deep learning models for regulatory compliance.
   Reveal Solution Discussion Select an answer first
5. 34expert · hard

   A hospital wants to deploy an AI system that reads medical images to detect early signs of cancer. The system must achieve high accuracy, but false negatives are life-threatening. The hospital also faces strict data-privacy regulations that limit sharing patient data with external services. Which solution best balances accuracy, safety, and privacy?

   • AUse a pre-trained model with no local training and rely on automated decisions without human review.
   • BUse a cloud-based AI service that processes images externally for the highest accuracy.
   • CTrain a model on-premises using federated learning across multiple hospitals, with a high-recall threshold and human radiologist review of all positive cases.
   • DDeploy a simple rule-based system that flags obvious abnormalities only.
   Reveal Solution Discussion Select an answer first
6. 35expert · hard

   A hospital is deploying an AI system to triage emergency patients. The system must provide rapid assessments, but clinicians are concerned about liability and want to understand the reasoning behind each recommendation. The hospital also has limited IT staff and cannot maintain complex infrastructure. Which approach best balances these constraints?

   • AUse a pretrained large language model with explainability tools and a managed cloud service
   • BBuild a custom deep learning model from scratch on-premises
   • CDeploy a fully autonomous system that makes final decisions without human review
   • DUse a simple rule-based system with no explainability
   Reveal Solution Discussion Select an answer first