Privacy, Ethical, Trust and Safety Controls AAISM Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A city government is deploying an AI system to assist in allocating social services. Citizens are concerned about how decisions are made and who is accountable. The government wants to build trust in the system. Which measure should be prioritized?

   • AKeep the system's logic confidential to prevent gaming by citizens.
   • BPublish a plain-language explanation of how the system works and establish a clear process for citizens to challenge decisions.
   • CDelegate all decisions to the AI to remove human bias from the process.
   • DEnsure the system is highly accurate by using the most advanced AI model available.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   Which regulatory framework is specifically designed to govern the development and use of AI systems within the European Union, including requirements for high-risk AI applications?

   • AGLBA
   • BFERPA
   • CEU AI Act
   • DCCPA
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   An insurance company uses an AI system to set premiums. The system was found to produce higher premiums for a certain ethnic group, even though ethnicity was not a direct input. The company wants to mitigate this bias while maintaining the model's predictive accuracy and being able to explain decisions to regulators. What is the most appropriate action?

   • AApply a fairness constraint during model training, such as equalized odds, and use explainability tools to document the trade-off between fairness and accuracy.
   • BAdjust the premiums manually for the affected group after the model runs.
   • CRemove all demographic-related features from the model to ensure fairness.
   • DUse a more complex model that is less explainable but may reduce bias.
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   Which accountability mechanism is used to ensure that there is a designated individual or team responsible for the outcomes of an AI system?

   • AModel retraining
   • BData anonymization
   • CPerformance monitoring
   • DRole assignment
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A fintech startup is developing an AI system that analyzes customer spending to offer personalized financial advice. The startup is planning to expand to multiple countries, including the EU and the US. The startup wants to ensure compliance with GDPR and other privacy regulations while keeping development costs low. What is the most cost-effective approach?

   • AOutsource all data processing to a third-party vendor that claims to be compliant.
   • BDesign the AI system with privacy by design, such as data minimization and purpose limitation, from the outset to avoid costly retrofits.
   • CUse a single privacy policy for all countries to simplify compliance.
   • DDeploy the AI system first and address compliance after receiving customer feedback.
   Reveal Solution Discussion Select an answer first