AI Incident Response and Business Continuity CRAGE Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   A manufacturing company's predictive maintenance AI starts generating false alarms, causing unnecessary equipment shutdowns. The operations team is unsure if this is an AI incident or a sensor issue. What should they do to triage the incident?

   • AImmediately shut down the entire predictive maintenance system.
   • BIgnore the alarms and wait for the next scheduled maintenance.
   • CCompare the model's recent inputs and outputs to historical patterns to determine if the behavior is anomalous.
   • DReplace all sensors in the plant to rule out sensor issues.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is a common outcome of testing AI incident response plans through simulations?

   • AImmediate resolution of any ongoing incidents
   • BElimination of the need for future incident response plans
   • CGuaranteed prevention of all future AI incidents
   • DIdentification of gaps in readiness and areas for improvement
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   What is the purpose of including fallback mechanisms in a business continuity plan for AI systems?

   • ATo reduce the cost of AI training
   • BTo provide an alternative method of operation when the AI system is unavailable
   • CTo increase the speed of the AI model
   • DTo eliminate the need for human oversight
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A company has an existing IT incident management framework based on ITIL. They are now integrating AI incident response procedures. What is the best approach to align the two?

   • AUse the existing framework but ignore AI-specific response steps.
   • BReplace the existing framework with a new AI-only framework.
   • CMap AI incident types to existing incident categories and use the same escalation paths.
   • DCreate a completely separate incident response process for AI incidents.
   Reveal Solution Discussion Select an answer first
6. 30application · medium

   After an AI incident that caused a service outage, the team has restored the service. What should be done as part of the post-incident review?

   • ADelete all incident logs to free up storage space.
   • BImmediately resume normal operations without any review.
   • CAnalyze the root cause and identify improvements to the incident response plan.
   • DConduct a tabletop exercise to test the response plan again.
   Reveal Solution Discussion Select an answer first