A manufacturing company wants to adopt AI for predictive maintenance of its equipment. The operations team is excited, but the IT security team is concerned about the new attack surface. The company's risk appetite is low. Which approach best aligns the AI initiative with the organization's risk posture?

Let the operations team manage the AI system without security involvement

Start with a pilot project on non-critical equipment and conduct a risk assessment

Reject the AI initiative entirely

Deploy the AI system across all equipment immediately

Which mitigation is effective against adversarial examples that fool AI models?

Increasing the model's learning rate

What is an ongoing oversight activity for AI vendors?

Regularly reviewing vendor performance and security reports

Never communicating with the vendor after contract signing

Deleting all vendor documentation

Ignoring vendor updates and patches

A logistics company is decommissioning an AI model that was used for route optimization. The model was trained on proprietary data and is being replaced. What is the most important step in the decommissioning process?

Securely delete the training data and model artifacts according to a retention policy

Archive the model without any documentation

Keep the model running until the new one is fully tested

Notify all employees that the model is no longer available

A security team is responsible for an AI-based malware detection system. The system has a low false positive rate but is known to be vulnerable to adversarial examples. The team must balance the need for high detection accuracy with the need to resist attacks. Which approach is most effective?

Replace the AI model with a signature-based system

Disable the AI system during high-risk periods

Increase the model's sensitivity to catch more malware

Implement adversarial training and regularly test the model with new attack samples

GIAC Security Leadership

Domain 2Objective 5

Managing Artificial Intelligence GSLC Practice Questions (Page 4)

Part of the Technical Security Management domain, which makes up ~27% of our current practice bank. GIAC (SANS) does not publish an official question count, but from its 180-minute exam (~70–120 total, ~19–32 in this domain), expect 4–6 from this objective — we provide 61 practice questions to prepare you well beyond it. (estimate)

61questions here

13free pages

12concepts

Questions 16–20

16application · medium
A manufacturing company wants to adopt AI for predictive maintenance of its equipment. The operations team is excited, but the IT security team is concerned about the new attack surface. The company's risk appetite is low. Which approach best aligns the AI initiative with the organization's risk posture?
Select an answer first
17foundation · easy
Which mitigation is effective against adversarial examples that fool AI models?
Select an answer first
18foundation · easy
What is an ongoing oversight activity for AI vendors?
Select an answer first
19application · medium
A logistics company is decommissioning an AI model that was used for route optimization. The model was trained on proprietary data and is being replaced. What is the most important step in the decommissioning process?
Select an answer first
20expert · hard
A security team is responsible for an AI-based malware detection system. The system has a low false positive rate but is known to be vulnerable to adversarial examples. The team must balance the need for high detection accuracy with the need to resist attacks. Which approach is most effective?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to GSLC

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by GIAC (SANS). “GSLC” is a trademark of its owner, used for identification only.