Ethics Frameworks CRAGE Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A city government is deploying an AI system to allocate social services. The system is designed to maximize the number of people served, but it consistently deprioritizes a small group of homeless individuals who are harder to reach. A council member argues that the city has a duty to care for the most vulnerable, even if it reduces overall efficiency. Which framework does the council member's argument reflect, and what is its main limitation in this AI context?

   • AUtilitarianism; it ignores the needs of the majority.
   • BContractarianism; it assumes all citizens would agree to prioritize the vulnerable.
   • CVirtue ethics; it lacks a clear decision procedure for allocating scarce resources.
   • DDeontology; it may be inflexible when duties conflict, such as duty to serve all versus duty to prioritize the vulnerable.
   Reveal Solution Discussion Select an answer first
2. 12expert · hard

   A government is deploying an AI system to allocate public housing. The system is designed to maximize the number of families housed, but it deprioritizes families with complex needs because they require more resources. A policy advisor argues that the system should be redesigned to prioritize the most vulnerable families, even if it houses fewer families overall. The government must decide between the utilitarian system and a deontological alternative. Which framework does the policy advisor's argument reflect, and what is its primary limitation in this context?

   • AUtilitarianism; it fails to consider the needs of the most vulnerable.
   • BContractarianism; it assumes all families would agree to prioritize the vulnerable.
   • CDeontology; it may lead to inefficient use of limited housing resources.
   • DVirtue ethics; it focuses on the character of the housing officials.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A tech company is developing an AI hiring tool. The engineering team is split: one group wants to use a model that maximizes predictive accuracy but is opaque, while another wants a less accurate but fully interpretable model. The company's ethical charter states that 'we must treat every candidate with dignity and respect, regardless of the outcome.' Which framework does the charter most closely align with, and which model choice does it support?

   • AContractarianism; choose the opaque model because candidates would agree to the most accurate system.
   • BDeontology; choose the interpretable model because it respects candidates' dignity by allowing scrutiny of decisions.
   • CVirtue ethics; choose the opaque model because it demonstrates the virtue of technical excellence.
   • DUtilitarianism; choose the opaque model because it maximizes hiring accuracy.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A company is developing an AI system to recommend medical treatments. The system is highly accurate but sometimes recommends a treatment that is not FDA-approved. A deontological approach would require the system to always follow approved protocols, while a utilitarian approach would allow off-label recommendations if they improve outcomes. Which statement best describes the limitation of the deontological approach in this scenario?

   • AIt is too flexible and allows any treatment regardless of approval status.
   • BIt focuses on the character of the physicians rather than the treatment protocol.
   • CIt may prevent beneficial treatments that are not yet approved, reducing patient welfare.
   • DIt always maximizes patient welfare by following established guidelines.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A government agency is using an AI system to predict recidivism for parole decisions. A utilitarian approach would maximize public safety by detaining more high-risk individuals, but this would disproportionately affect a minority community. A deontological approach would emphasize the duty to treat every individual fairly, but might release some dangerous individuals. Which statement best describes the trade-off between these frameworks in this context?

   • ADeontology is better suited for AI because it always produces the safest outcome.
   • BUtilitarianism protects individual rights, while deontology maximizes aggregate safety.
   • CUtilitarianism prioritizes aggregate safety but may sacrifice fairness, while deontology prioritizes fairness but may reduce aggregate safety.
   • DBoth frameworks would produce identical parole decisions because they both value safety.
   Reveal Solution Discussion Select an answer first