Building Privacy, Trust, and Safety in AI Systems CRAGE Practice Questions (Page 9)

Questions 41–45

1. 41foundation · easy

   Which strategy is an example of mitigating a privacy risk in an AI system?

   • AStoring personal data indefinitely
   • BSharing personal data with third parties without consent
   • CImplementing data anonymization techniques before processing
   • DIncreasing the amount of personal data collected
   Reveal Solution Discussion Select an answer first
2. 42expert · hard

   A large retail company is deploying an AI system for personalized marketing. The system will process customer purchase history, browsing behavior, and location data. The company operates in multiple jurisdictions with different privacy regulations. A DPIA has been initiated. During the DPIA, the team identifies that the system's data processing could lead to discriminatory outcomes in certain regions. The team is under pressure to launch the system quickly. What is the most appropriate course of action?

   • AProceed with the launch as planned, but add a disclaimer to the marketing materials stating that the system may produce biased results.
   • BLimit the data processing to only the data that is strictly necessary for the marketing purpose, assuming this will eliminate the risk of discrimination.
   • CLaunch the system in regions with less stringent regulations first to gather more data and refine the model before expanding to other regions.
   • DPause the deployment and conduct a thorough analysis of the potential discriminatory outcomes, involving diverse stakeholders and implementing mitigation measures before launch.
   Reveal Solution Discussion Select an answer first
4. 43foundation · easy

   What is the main purpose of an ethical AI governance framework?

   • ATo ensure AI systems are never subject to external regulation
   • BTo integrate ethical principles like privacy, trust, and safety into organizational policies and practices
   • CTo maximize the profitability of AI systems above all other considerations
   • DTo replace all human decision-making with AI
   Reveal Solution Discussion Select an answer first
5. 44foundation · easy

   What does 'informed consent' require in the context of AI systems processing personal data?

   • AUsers must be asked for consent only after their data has been processed.
   • BUsers must provide consent in writing and it is valid indefinitely.
   • CUsers must agree to all data processing without any explanation.
   • DUsers must be given clear, understandable information about what data is collected and how it will be used before they agree.
   Reveal Solution Discussion Select an answer first
6. 45application · medium

   A startup is building a new AI-powered fitness app that tracks users' health metrics. The founders want to adopt a 'Privacy by Design' approach from the outset. They are at the initial design phase. Which of the following actions best exemplifies the 'Privacy as the Default Setting' principle?

   • ACollect as much data as possible during the design phase to ensure the app has enough data to function effectively.
   • BConfigure the app so that the most privacy-friendly settings are active by default, such as not sharing data with third parties unless the user explicitly opts in.
   • CAllow users to choose their privacy settings only after they have used the app for a week.
   • DProvide users with a lengthy privacy policy that explains all data processing activities in detail.
   Reveal Solution Discussion Select an answer first