Question 740
As part of architectural change control for this problem: A team allows all AI writing tools because their URLs appear in the productivity category. Which preventive check should be required?
Explanation
Correct Answer
Track emerging AI applications using granular discovery telemetry. This lowers the risk linked to: Broad content categories do not fully characterize generative-AI use and exfiltration risk. Verification should show: Only approved AI tools meet the user-group and data-handling policy; unknown AI tools are monitored or blocked.
Incorrect Answers
Recheck reidentification risk and logging fields after model feature updates. This safeguard addresses another situation: An AI application claims its model is anonymous but retains conversation identifiers linked to employees. It is less directly relevant than the control for: Broad content categories do not fully characterize generative-AI use and exfiltration risk.
Include prompt log locations in cardholder data environment review. This safeguard addresses another situation: A credit-card support bot receives raw PAN numbers and stores them in unrestricted logs. It is less directly relevant than the control for: Broad content categories do not fully characterize generative-AI use and exfiltration risk.
Alert on new provider destinations and unauthorized outbound model calls. This safeguard addresses another situation: A model provider rotates endpoints and requests unexpectedly bypass the approved inline runtime control. It is less directly relevant than the control for: Broad content categories do not fully characterize generative-AI use and exfiltration risk.