Ethics CT-AI Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   In the context of AI ethics, what does the principle of accountability primarily require?

   • AThat the AI system is able to explain its decisions in natural language
   • BThat the AI system is deployed only after all defects have been fixed
   • CThat there is a clear assignment of responsibility for the AI system's decisions and outcomes
   • DThat the AI system is designed to maximize accuracy on all test datasets
   Reveal Solution Discussion Select an answer first
2. 2expert · hard

   A team is testing an AI system that recommends treatment plans for cancer patients. The system is trained on data from a specific demographic group. The team is concerned about the system's performance for other groups. What is the most ethical testing strategy?

   • ATest the system only for overall accuracy, without breaking down results by group.
   • BInclude a diverse set of test cases that represent different demographic groups, and evaluate the system's performance separately for each group.
   • CUse a small sample of test data from other groups to save time.
   • DTest the system only on the demographic group it was trained on, as it is the most accurate for that group.
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A global company is testing an AI system for employee performance evaluation. The system is trained on data from one region and will be deployed in multiple countries. The test team must ensure the system is fair across different cultural contexts. What is the most critical ethical consideration?

   • AThe system may be too complex to test thoroughly.
   • BThe system may not comply with local data protection laws.
   • CThe system may reflect the cultural biases of the training data, leading to unfair evaluations in other regions.
   • DThe system may not be transparent about how it evaluates employees.
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A European company is developing an AI system for recruitment. The compliance team wants to ensure the testing process aligns with upcoming regulations. Which guideline or standard is most directly applicable to the testing of this AI system?

   • AISO/IEC 27001, which specifies information security management requirements.
   • BGDPR, which regulates the processing of personal data.
   • CEU AI Act, which classifies AI systems by risk and imposes requirements on high-risk systems like recruitment.
   • DIEEE 7000, which focuses on the ethical design of autonomous systems.
   Reveal Solution Discussion Select an answer first
6. 5expert · hard

   A team is testing an AI system that predicts recidivism for parole decisions. The system is found to have a higher false positive rate for a particular ethnic group. The team has limited time before the system goes live. What is the most ethical course of action?

   • ALaunch the system and monitor for bias in production, then fix it later.
   • BDelay the launch until the bias is investigated and mitigated, even if it means missing the deadline.
   • CReduce the false positive rate for the affected group by lowering the system's threshold, even if it increases false negatives.
   • DLaunch the system but add a disclaimer that it may be biased.
   Reveal Solution Discussion Select an answer first