AI Ownership, Oversight, and Accountability AAIR Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A bank deploys a machine learning model that approves or denies consumer credit applications. The model was built by the data science team, but the business unit that uses it has profit targets tied to approval rates. The bank's risk committee needs a single point of accountability for the model's outcomes. Which role should be formally designated as the accountable owner?

   • AThe chief information security officer
   • BThe model risk management team lead
   • CThe head of the consumer lending business unit
   • DThe chief data scientist who led the model development
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   An energy company has implemented an AI governance framework with defined roles and oversight committees. After two years, the company wants to verify that the framework is still effective and aligned with its evolving risk profile. What is the most appropriate action?

   • AAssume the framework is still effective because it has not been changed since implementation
   • BReplace the entire framework with a new one every year to ensure freshness
   • CWait for a regulatory audit to identify any weaknesses
   • DConduct a periodic review of the AI governance framework, including benchmarking against industry practices
   Reveal Solution Discussion Select an answer first
4. 18foundation · easy

   What is the main function of an AI oversight committee?

   • ATo review and approve AI initiatives and monitor their risks
   • BTo write code for all AI models
   • CTo market AI products to customers
   • DTo provide technical support for AI systems
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   A technology company is developing an AI system that will be embedded in a product sold to external customers. The product team, engineering, and sales all have interests in the system. The company wants to ensure that accountability for the AI system's performance and risks is clear, especially when the product is deployed at customer sites. Which allocation of accountability is most appropriate?

   • AThe customer is accountable once they deploy the product
   • BThe sales team is accountable because they sell the product to customers
   • CThe product team is accountable for the AI system's performance and risks throughout its lifecycle
   • DThe engineering team is accountable because they built the AI system
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   An insurance company has multiple AI models in production, each managed by different teams. During a recent incident, it was unclear who to contact and who had the authority to stop a malfunctioning model. The company wants to prevent this in the future. What is the most effective action?

   • ACreate a central document listing each AI model, its accountable owner, and the escalation path for incidents
   • BRequire all AI models to be approved by the IT department before deployment
   • CHold a one-time training session on AI governance for all staff
   • DAssign all AI models to the chief information officer for accountability
   Reveal Solution Discussion Select an answer first