Policy Design CRAGE Practice Questions (Page 9)

Questions 41–45

1. 41foundation · easy

   When incorporating ethical considerations into AI policy, what should the policy address regarding societal impact?

   • APotential unintended consequences on communities and the environment
   • BOnly the technical performance metrics
   • COnly the financial return on investment
   • DOnly the preferences of the development team
   Reveal Solution Discussion Select an answer first
2. 42application · medium

   A large e-commerce company has adopted an AI governance policy that includes fairness checks for its recommendation algorithms. The policy team now needs to ensure the policy is actually followed across multiple product teams. Which implementation and enforcement strategy is most effective?

   • AEstablish a central AI governance office that conducts regular audits, requires teams to document fairness checks, and has the authority to halt deployments that violate the policy.
   • BCreate a bug bounty program that rewards employees for reporting policy violations.
   • CProvide training on the policy to all engineers and rely on their professional judgment to apply it.
   • DPublish the policy on the company wiki and ask team leads to self-certify compliance each quarter.
   Reveal Solution Discussion Select an answer first
4. 43foundation · easy

   Which mechanism is commonly used to enforce AI governance policies within an organization?

   • APublic release of all internal algorithms
   • BEliminating all human involvement in AI oversight
   • CDisabling all AI systems after deployment
   • DRegular audits and compliance checks
   Reveal Solution Discussion Select an answer first
5. 44application · medium

   A software company develops an AI chatbot for customer service that processes personal data of EU citizens. The company wants its AI governance policy to align with GDPR and the EU AI Act. Which policy measure is most directly required by both regulations?

   • APublish the chatbot's source code on a public repository to ensure transparency.
   • BEnsure the chatbot is deployed on servers located within the EU to guarantee data residency.
   • CConduct a Data Protection Impact Assessment (DPIA) and document the AI system's purpose, data processing, and risk mitigation measures.
   • DObtain explicit consent from every user before they interact with the chatbot.
   Reveal Solution Discussion Select an answer first
6. 45application · medium

   A healthcare provider is integrating a risk assessment framework into its AI governance policy for a patient triage system. The policy must address potential harms such as misdiagnosis and unequal access. Which approach best integrates risk assessment into the policy design?

   • ARequire a risk assessment at each stage of the AI lifecycle, with mandatory mitigation actions for high-risk findings before deployment.
   • BAssign the risk assessment to the legal team and keep findings confidential to avoid liability concerns.
   • CPerform a single risk assessment at the initial design phase and document it in the policy appendix.
   • DUse a generic risk checklist from another industry without adapting it to healthcare-specific harms.
   Reveal Solution Discussion Select an answer first