Which control is primarily used to ensure that only authorized users and services can access an AI/ML data set stored in a cloud object store?

Identity and access management (IAM)

A security company has developed a model to detect malware. They want to protect the model from being reverse-engineered by attackers who have access to the model's API. Which technique would best prevent attackers from learning the model's decision boundaries?

Limit the number of API requests per user

Watermark the model to track its usage

Add noise to the model's outputs using differential privacy

Encrypt the model's parameters and store them in a hardware security module (HSM)

A marketing company is building a customer segmentation model using purchase data. They want to protect customer privacy while still being able to analyze purchasing patterns. Which technique would best allow them to do this?

Use differential privacy when training the model

Store the data in a secure data warehouse with access controls

Encrypt the purchase data with a strong encryption algorithm

Delete all customer names from the dataset

Certified Cloud Security Professional

Domain 2Objective 9

Comprehend Data Protection of Artificial Intelligence (AI) and Machine Learning (ML) Data CCSP Practice Questions (Page 2)

Part of the Cloud Data Security domain, which accounts for 20% of the CCSP exam. ISC2 does not publish an official question count, but from its 180-minute exam (~70–120 total, ~14–24 in this domain), expect 2–3 from this objective — we provide 24 practice questions to prepare you well beyond it. (estimate)

24questions here

5free pages

4concepts

20%of the exam

Questions 6–10

6foundation · easy
Which control is primarily used to ensure that only authorized users and services can access an AI/ML data set stored in a cloud object store?
Select an answer first
7foundation · easy
Which process is used to confirm that a model's predictions remain accurate and consistent after deployment, as part of model integrity validation?
Select an answer first
8foundation · easy
A data scientist wants to publish aggregate statistics from a training data set while limiting the risk of identifying any individual record. Which technique is designed for this purpose?
Select an answer first
9application · medium
A security company has developed a model to detect malware. They want to protect the model from being reverse-engineered by attackers who have access to the model's API. Which technique would best prevent attackers from learning the model's decision boundaries?
Select an answer first
10application · medium
A marketing company is building a customer segmentation model using purchase data. They want to protect customer privacy while still being able to analyze purchasing patterns. Which technique would best allow them to do this?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CCSP

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 ISC2. “CCSP” is a trademark of its owner, used for identification only.

Comprehend Data Protection of Artificial Intelligence (AI) and Machine Learning (ML) Data CCSP Practice Questions (Page 2)

Questions 6–10

Which control is primarily used to ensure that only authorized users and services can access an AI/ML data set stored in a cloud object store?

Which process is used to confirm that a model's predictions remain accurate and consistent after deployment, as part of model integrity validation?

A data scientist wants to publish aggregate statistics from a training data set while limiting the risk of identifying any individual record. Which technique is designed for this purpose?

A security company has developed a model to detect malware. They want to protect the model from being reverse-engineered by attackers who have access to the model's API. Which technique would best prevent attackers from learning the model's decision boundaries?

A marketing company is building a customer segmentation model using purchase data. They want to protect customer privacy while still being able to analyze purchasing patterns. Which technique would best allow them to do this?

Keep studying — Cloud Data Security