Conduct Risk Assessment(s) (e.g., Security, Safety, Ethics) PMI-CPMAI Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   A company's AI chatbot starts producing offensive and discriminatory responses, and the incident is widely shared on social media. What is the most immediate reputational risk to the organization?

   • ADecreased employee morale
   • BRegulatory fines for data privacy violations
   • CIncreased operational costs from system updates
   • DLoss of customer trust and negative public perception
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   Which type of attack on an AI system aims to extract sensitive information about the training data by observing the model's outputs?

   • AData poisoning
   • BDenial of service
   • CAdversarial attack
   • DModel inversion
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   A contingency plan for an AI system that may fail during a critical business process should primarily include:

   • AA list of all potential risks without any response actions
   • BA detailed description of the AI model's architecture
   • CA guarantee that the AI system will never fail
   • DPredefined steps to maintain operations if the AI system fails
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A hospital relies on an AI system to predict patient deterioration and alert medical staff. The system is critical for patient safety, and any failure could have serious consequences. The hospital's risk manager is conducting a business continuity risk analysis. Which of the following is the most important factor to consider when assessing the risk of an AI system failure?

   • AThe technical details of the AI system's architecture and the specific algorithms used.
   • BThe number of data scientists on staff who are capable of fixing the AI system if it fails.
   • CThe cost of the AI system and the potential financial loss from a system failure.
   • DThe potential impact on patient safety and the ability of medical staff to provide care without the AI system.
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   A hospital deploys an AI system to prioritize patients for organ transplants. The system uses a scoring algorithm trained on historical data. An internal audit reveals that the model assigns significantly lower scores to patients from certain ethnic backgrounds, even when controlling for medical urgency. The hospital's leadership is concerned about the ethical and legal implications. What is the most appropriate initial action for the hospital to take?

   • ARetrain the model using a fairness-aware algorithm that explicitly optimizes for equal outcomes across all demographic groups.
   • BImmediately disable the AI system and revert to manual prioritization by a committee of doctors.
   • CConduct a formal ethical risk evaluation to determine the source of the bias, its impact on fairness and equity, and the potential legal and reputational consequences.
   • DPublish the audit findings publicly to demonstrate transparency and accountability.
   Reveal Solution Discussion Select an answer first