Objective 2.4 Identify Ethical Challenges That May Conflict or Require Compromise with Regulatory And/or Business Constraints or Demands CERTIFIED-ETHICAL-EMERGING-TECHNOLOGIST Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A logistics company optimizes its delivery routes to minimize fuel consumption. The optimization algorithm reduces fuel costs by 12% by consolidating deliveries in urban areas and reducing service frequency in rural areas. Rural customers now experience delivery delays of up to three days. The company's leadership is focused on the cost savings. What is the most ethical response?

   • AModify the optimization to include a service-level constraint that guarantees a maximum delivery delay for rural customers, then re-optimize.
   • BAccept the cost savings because the company has a fiduciary duty to shareholders to maximize efficiency.
   • CKeep the optimized routes but offer rural customers a discount on future deliveries.
   • DRevert to the previous routing algorithm to restore rural service levels.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   Which of the following is a common compromise strategy to balance data minimization with the need for data utility?

   • ACollect all available data and rely on strict access controls to protect it
   • BCollect only the minimum data required by law, regardless of the model's needs
   • CDelay data collection until a regulatory framework is fully established
   • DUse data anonymization and aggregation techniques to reduce the risk of re-identification while retaining analytical value
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   In a high-stakes context such as healthcare diagnostics, why might an organization choose a less accurate but more interpretable model over a highly accurate 'black-box' model?

   • ABecause regulators and clinicians need to understand and trust the reasoning behind decisions that affect patient care
   • BBecause interpretable models are always faster to train and deploy
   • CBecause interpretable models are less expensive to maintain in the long run
   • DBecause interpretable models are less likely to be biased
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   A global social media platform rapidly deploys a new AI feature that automatically translates user posts. Due to the fast rollout, the translation model was not tested on regional dialects and slang, leading to offensive mistranslations in certain cultures. What is the core ethical issue in this scenario?

   • AThe AI feature reduces user autonomy
   • BThe AI feature is not compliant with data protection laws
   • CThe development speed led to a lack of cultural sensitivity, causing unintended harm to users
   • DThe AI feature is too expensive to maintain
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   A global e-commerce company is rolling out a new AI-powered customer service chatbot. The development team, based in the US, used English-language training data and Western conversational norms. The chatbot is being deployed in Japan and India, where customer service expectations differ significantly. The company has a tight launch deadline. What should the team do?

   • ADelay the launch in Japan and India until the chatbot is trained on local language data and tested with local users.
   • BDeploy the chatbot as-is to meet the deadline, and plan to localize it in a future update.
   • CLaunch the chatbot globally but add a disclaimer that it may not understand local customs.
   • DHire local translators to translate the English responses into Japanese and Hindi without changing the underlying logic.
   Reveal Solution Discussion Select an answer first