Generative AI Leader · Free Practice Question Easy
Question 4
A company wants to train a generative AI model on a large dataset of customer support chat logs to improve its customer service responses. These chat logs contain Personally Identifiable Information (PII) such as names, email addresses, and phone numbers.
To protect customer privacy and comply with data protection regulations before using this data for training, what data processing technique should the company apply to remove or obscure this PII?
-
A
Data anonymization (or de-identification)
-
B
Data validation
-
C
Data encryption at rest
-
D
Data augmentation
Reveal correct answer
Correct answer: A
Explanation
Data anonymization (or de-identification) is the process of removing or obscuring PII to protect individual privacy.
INCORRECT: Data encryption at rest
Encryption protects data from unauthorized access if the storage is breached, but the data is still identifiable once decrypted for use. It doesn't remove PII for training purposes where the raw PII itself should not be learned by the model.INCORRECT: Data augmentation
Data augmentation is a technique to artificially increase the size of a training dataset, not to remove PII.CORRECT: Data anonymization (or de-identification)
Data anonymization involves processing data to remove or obscure personally identifiable information (PII) so that the individuals whom the data describe remain anonymous. This is a critical step when using sensitive data like customer chat logs for AI model training to protect privacy and meet regulatory requirements. Techniques can include removing, masking, or generalizing PII.INCORRECT: Data validation
Data validation checks the accuracy and quality of data but does not specifically address the removal or obscuring of PII.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
