A cybersecurity firm uses an AI model to classify malware. During an incident, they need to understand why a particular file was classified as benign to determine if it was an adversarial example. Which technique would provide the most insight?

Check the file's hash against known malware databases.

Deploy the model in a sandbox environment to observe its behavior.

Use SHAP (SHapley Additive exPlanations) values to identify which features contributed most to the classification.

Retrain the model with more malware samples to improve its accuracy.

A retail company is developing a recommendation system. The data science team wants to ensure that the training data is not tampered with during the data collection phase. Which practice should they implement?

Encrypt the data to prevent unauthorized access.

Use cryptographic hashing to verify the integrity of data files before and after training.

Train the model on a subset of the data to reduce the attack surface.

Use a larger sample of data to reduce the impact of any tampering.

A healthcare startup is developing an AI system that triages patient symptoms. During threat modeling, the team identifies that an attacker could poison the training data by submitting fake patient records through a public web form. Which control should the team implement to mitigate this data poisoning risk?

Deploy the model behind a web application firewall to block malicious requests.

Validate and sanitize all input data before it is used for training, and restrict data collection to trusted sources.

Encrypt all patient data at rest and in transit to protect it from unauthorized access.

Use a more complex model to better distinguish between real and fake patient records.

What is the primary purpose of monitoring an AI system for adversarial activities?

To automatically retrain the model on all incoming data

To detect anomalies in model behavior that may indicate an attack

To replace the need for regular model updates

To ensure the model never makes a mistake

CompTIASecAI+

Domain 2Objective 3

Mitigate Adversarial Risks CY0-001 Practice Questions (Page 7)

Part of the Securing AI systems domain, which accounts for 40% of the CY0-001 exam. CompTIA does not publish an official question count, but from its 60-minute exam (~25–40 total, ~10–16 in this domain), expect 3–5 from this objective — we provide 34 practice questions to prepare you well beyond it. (estimate)

34questions here

7free pages

8concepts

40%of the exam

Questions 31–34

31application · medium
A cybersecurity firm uses an AI model to classify malware. During an incident, they need to understand why a particular file was classified as benign to determine if it was an adversarial example. Which technique would provide the most insight?
Select an answer first
32application · medium
A retail company is developing a recommendation system. The data science team wants to ensure that the training data is not tampered with during the data collection phase. Which practice should they implement?
Select an answer first
33application · medium
A healthcare startup is developing an AI system that triages patient symptoms. During threat modeling, the team identifies that an attacker could poison the training data by submitting fake patient records through a public web form. Which control should the team implement to mitigate this data poisoning risk?
Select an answer first
34foundation · easy
What is the primary purpose of monitoring an AI system for adversarial activities?
Select an answer first

Finished these 4 questions?

Review the revealed explanations, or continue through the curriculum.

No more pagesBack to CY0-001

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 CompTIA. “CY0-001” is a trademark of its owner, used for identification only.