AI Strategy and Planning CRAGE Practice Questions (Page 1)

Questions 1–5

1. 1expert · hard

   A technology company is developing an AI system for hiring. The company operates in multiple countries with varying data protection laws. The company wants to adopt a governance framework that is both internationally recognized and adaptable to local regulations. Which framework should they choose?

   • AThe EU AI Act itself, because it is the most comprehensive regulation and can be applied globally.
   • BISO/IEC 42001, because it is an international management system standard that can be certified and is designed to be adaptable to different regulatory environments.
   • CThe NIST AI Risk Management Framework, because it is mandatory for all AI systems in the US.
   • DThe IEEE Ethically Aligned Design guidelines, because they are legally binding in most countries.
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A bank is developing an AI loan approval system. The model uses demographic data, which may lead to bias. The bank must comply with fair lending laws. What should the strategy include?

   • ABias testing and fairness metrics as part of the model validation process.
   • BRely on the model's accuracy as the sole success criterion.
   • CUse a governance framework but do not test for bias.
   • DRemove all demographic data from the model.
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A social media company is deploying an AI content moderation system. The system must balance free speech with safety, and the company wants to measure success. The legal team warns about liability. What is the best approach?

   • APrioritize free speech and minimize blocking.
   • BDefine ethical principles, use a governance framework, and set KPIs for both safety and over-blocking.
   • CPrioritize safety and block aggressively.
   • DUse a governance framework but do not set KPIs.
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A university is developing an AI system to predict student dropout rates. The administration wants to use the system to allocate support resources. Faculty members are concerned about student privacy and the potential for bias. Which stakeholder engagement strategy is most effective?

   • ACreate a working group that includes faculty, student representatives, and data privacy officers to co-design the system's objectives and constraints.
   • BConduct a survey of students and faculty, but do not share the results with the administration.
   • CShare the AI system's predictions only with administrators to avoid faculty resistance.
   • DUse the AI system to automatically assign support resources without human review to save time.
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A healthcare provider is planning to deploy an AI triage system. The roadmap must include milestones, resource allocation, and risk assessment. The clinical staff are skeptical about AI accuracy, and the legal team requires transparency for patient safety. Which roadmap element is most critical to address early?

   • ASet KPIs for model accuracy and patient wait times.
   • BConduct a risk assessment that includes ethical, legal, and clinical safety risks.
   • CDefine clear milestones for model validation and clinical integration.
   • DAllocate budget for AI infrastructure and data engineering.
   Reveal Solution Discussion Select an answer first