Building Privacy, Trust, and Safety in AI Systems CRAGE Practice Questions (Page 7)

Questions 31–35

1. 31foundation · easy

   What is the primary role of explainability in AI systems?

   • ATo ensure the AI system never makes errors
   • BTo provide understandable reasons for AI decisions to users and stakeholders
   • CTo hide the internal workings of the AI system
   • DTo increase the computational speed of AI models
   Reveal Solution Discussion Select an answer first
2. 32application · medium

   A manufacturing company is deploying an AI system that controls robotic arms on an assembly line. The system must operate safely to prevent harm to workers. The engineering team is evaluating safety mechanisms. Which combination of safety mechanisms best ensures that the AI system can be stopped or corrected if it behaves unexpectedly?

   • AIncreasing the AI system's processing speed to react faster to anomalies.
   • BImplementing a manual emergency stop button and a human supervisor who can override AI decisions in real time.
   • CTraining the AI model on more data to improve its accuracy.
   • DAdding more sensors to the robotic arms to improve data collection.
   Reveal Solution Discussion Select an answer first
4. 33foundation · easy

   How does accountability contribute to building user trust in AI systems?

   • ABy making the AI system's code open source
   • BBy allowing users to modify the AI system's algorithms
   • CBy guaranteeing the AI system never makes mistakes
   • DBy ensuring that there is a clear entity responsible for the AI system's decisions and outcomes
   Reveal Solution Discussion Select an answer first
5. 34foundation · easy

   How does transparency in AI systems foster user trust?

   • ABy openly communicating how the system works, its limitations, and its data usage
   • BBy making the system more complex to deter user questions
   • CBy ensuring the system always provides favorable outcomes to users
   • DBy keeping the system's algorithms secret to prevent misuse
   Reveal Solution Discussion Select an answer first
6. 35application · medium

   An e-commerce company uses an AI recommendation system to suggest products to customers. The company wants to build user trust by ensuring the system is reliable, transparent, and accountable. The product team is considering several initiatives. Which initiative most directly addresses the 'accountability' component of trustworthy AI?

   • AProviding users with an explanation of how recommendations are generated.
   • BConducting regular performance testing to ensure the system meets accuracy benchmarks.
   • CPublishing a public statement about the AI system's purpose and limitations.
   • DAssigning a designated human owner who is responsible for the system's decisions and outcomes.
   Reveal Solution Discussion Select an answer first