Comprehend Artificial Intelligence (AI)/Machine Learning (ML) CCSP Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A cloud security team uses an AI/ML-based alert correlation system that processes thousands of security events daily. The system currently generates a high volume of alerts, many of which are false positives, overwhelming the incident response team. The team wants to improve the system's ability to prioritize alerts so that analysts focus on the most critical threats first. Which approach best achieves this goal?

   • AImplement a SOAR platform to automatically respond to all alerts without human intervention.
   • BIncrease the frequency of alert generation to ensure no event is missed.
   • CManually review and categorize all alerts before they reach the AI/ML system.
   • DTrain the AI/ML model to assign risk scores based on the context and severity of each alert.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   In cloud threat detection, what is the primary role of AI/ML models when analyzing telemetry data?

   • ATo encrypt telemetry data before it is stored in the cloud
   • BTo automatically patch vulnerabilities in cloud workloads
   • CTo replace all traditional signature-based detection methods
   • DTo identify patterns and anomalies that may indicate potential threats
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   What is 'model poisoning' in the context of AI/ML security?

   • AAn attack where an adversary manipulates training data to corrupt the model's behavior
   • BA method to reduce the size of a model for deployment
   • CA technique to make models more robust against adversarial examples
   • DA process of validating model accuracy on test data
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   Why is transparency important when deploying AI/ML for cloud security?

   • AIt guarantees the model is free from bias
   • BIt ensures the model runs faster
   • CIt reduces the cost of cloud storage
   • DIt allows stakeholders to understand how decisions are made
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   A security operations team deploys an AI/ML-based threat detection system that ingests cloud telemetry from VPC flow logs, identity provider logs, and API gateway logs. The system flags a sudden spike in failed login attempts from a single IP address as a potential brute-force attack. However, the team notices the same IP address is also generating legitimate traffic to a public web application. What is the most appropriate action to reduce false positives while maintaining effective detection?

   • AAdjust the AI/ML model's anomaly threshold to be more sensitive to failed login attempts.
   • BImmediately block the IP address at the network firewall to prevent further attempts.
   • CAdd context to the model by correlating the failed login attempts with successful logins and user behavior patterns.
   • DDisable the AI/ML detection for that specific IP address to avoid noise.
   Reveal Solution Discussion Select an answer first