Responsible Practices and Governance Standards AIE Practice Questions (Page 9)

Questions 41–45

1. 41foundation · easy

   What does 'consent' mean in the context of data governance for AI?

   • AGetting approval from the AI model itself to use data.
   • BObtaining permission from individuals before using their personal data.
   • CDeleting data after a certain period.
   • DEnsuring that data is encrypted during transmission.
   Reveal Solution Discussion Select an answer first
2. 42expert · hard

   A hospital uses an AI system to recommend treatment plans for cancer patients. The system is a deep learning model that provides accurate recommendations but offers no explanation. Doctors are legally required to inform patients about the reasons for their treatment. The hospital's ethics committee is debating between using a simpler, interpretable model versus keeping the deep learning model and using post-hoc explanations. What is the most important consideration in this decision?

   • AThe reliability and clinical validity of post-hoc explanations for the deep learning model must be established before use.
   • BThe decision should be made solely by the doctors, without patient input.
   • CThe deep learning model should be used because it is more accurate, and explanations are secondary.
   • DThe simpler model is always preferable because it is fully interpretable.
   Reveal Solution Discussion Select an answer first
4. 43application · medium

   A city government is building an AI system to predict which public housing units are most likely to need maintenance. The dataset includes tenant complaints, inspection records, and demographic information. The project team is concerned about fairness and privacy. Which data governance practice is most important to implement before model development?

   • AShare the raw dataset with all project stakeholders to ensure transparency.
   • BCollect as much data as possible to improve model accuracy, including social media activity of tenants.
   • CAnonymize the data but keep all demographic fields because they are useful for future analysis.
   • DRemove or minimize demographic fields unless they are strictly necessary for the prediction task.
   Reveal Solution Discussion Select an answer first
5. 44expert · hard

   A city is deploying an AI system to allocate police patrols based on historical crime data. The system is intended to reduce crime, but there are concerns that it may reinforce over-policing in minority neighborhoods. The police department wants to proceed, while community groups demand the system be abandoned. The city council must decide. What is the most balanced and responsible approach?

   • AProceed with the system because it is based on objective crime data.
   • BConduct an independent ethical risk assessment that includes bias analysis and community consultation, and implement strict oversight and monitoring.
   • CAbandon the system entirely to respect community concerns.
   • DDeploy the system in a limited area and evaluate after six months.
   Reveal Solution Discussion Select an answer first
6. 45application · medium

   A hospital deploys an AI system that predicts patient deterioration and alerts nurses. The system uses a complex deep learning model. The hospital's ethics committee requires that nurses understand why an alert was triggered. Which technique best supports transparency for the nurses?

   • AProvide nurses with the full model architecture and all training weights.
   • BUse an explainability tool such as SHAP to show which patient vitals contributed most to the alert.
   • CReplace the deep learning model with a simple linear model that is inherently interpretable.
   • DRequire nurses to attend a training course on deep learning before using the system.
   Reveal Solution Discussion Select an answer first