AI Risk Management AAIA Practice Questions (Page 4)

Questions 16–20

1. 16application · easy

   A retail company uses an AI chatbot for customer service. The chatbot occasionally provides incorrect product information, leading to customer complaints and potential regulatory issues. The company decides to implement a human-in-the-loop review for all chatbot responses. This is an example of which risk response strategy?

   • AAvoidance
   • BTransfer
   • CMitigation
   • DAcceptance
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A healthcare provider uses an AI triage system that recommends patient priority levels. The risk team needs to assess the risk of misdiagnosis. They have historical data on false negatives and their clinical impact. Which approach best assesses this risk in alignment with enterprise risk management?

   • AAccept the risk because the AI system is approved by the FDA, so no further assessment is needed.
   • BQualitatively rate the risk as 'high' based on the severity of potential patient harm, without considering likelihood.
   • CCalculate the expected monetary loss by multiplying the probability of false negatives by the estimated cost of delayed treatment, and compare it to the organization's risk appetite.
   • DTransfer the risk to the AI vendor through a contractual indemnity clause, and stop further analysis.
   Reveal Solution Discussion Select an answer first
4. 18application · medium · select all that apply

   A government agency uses AI to screen visa applications. The risk team has established a risk monitoring process. Which of the following are key components of an effective AI risk monitoring and reporting process? (Select all that apply.)

   • AConduct a one-time risk assessment at deployment and document it.
   • BProvide regular risk reports to the AI governance committee.
   • CEstablish a clear escalation path for when risk thresholds are breached.
   • DAutomate alerts for significant deviations in model performance.
   • EDefine key risk indicators (KRIs) that are tracked on a regular basis.
   Check answers Discussion Select an answer first
5. 19expert · hard

   A media company uses an AI content recommendation system. The system has been criticized for promoting harmful content. The risk team must choose a response. The company wants to maintain user engagement while reducing reputational risk. Which response best achieves this balance?

   • AImplement content moderation filters and adjust the recommendation algorithm to down-rank harmful content, while monitoring engagement metrics.
   • BAccept the risk and continue current operations, arguing that the system is protected by free speech laws.
   • CRemove the recommendation system entirely and rely on user self-selection of content.
   • DTransfer the risk by hiring a third-party content moderation vendor and holding them liable for harmful content.
   Reveal Solution Discussion Select an answer first
6. 20expert · hard

   A logistics company uses AI for route optimization. The risk team has identified that the AI system may occasionally suggest routes that violate local traffic regulations. They need to establish monitoring and reporting. The company has a decentralized structure with regional operations. Which monitoring approach is most effective?

   • AAllow each region to monitor its own AI outputs and report to the regional manager, with no central oversight.
   • BRely on drivers to report any route violations they notice, and compile reports monthly.
   • CDeploy a centralized monitoring system that aggregates data from all regions, with automated alerts to regional compliance officers and monthly reports to the central AI risk committee.
   • DConduct an annual audit of AI route suggestions by an external consultant, with findings reported to the board.
   Reveal Solution Discussion Select an answer first