Q32 — AWS AIF-C01 Ch.2

Question 32 of 100 | ← Chapter 2

A developer trained a custom model on Amazon Bedrock using a training dataset containing confidential data. The developer wants to ensure the custom model does not generate inference responses based on the confidential data. What should the developer do to prevent inference responses based on confidential data?

• A. Delete the custom model. Remove confidential data from the training dataset. Retrain the custom model. ✓
• B. Apply dynamic data masking to mask confidential data in inference responses.
• C. Use Amazon SageMaker to encrypt confidential data in inference responses.
• D. Use AWS Key Management Service (AWS KMS) to encrypt confidential data within the custom model.

Correct Answer: A. Delete the custom model. Remove confidential data from the training dataset. Retrain the custom model.

Explanation

To prevent inference responses based on confidential data, the key is ensuring the training dataset contains no confidential data. Once a model has been trained on data containing confidential information, it may have already learned features or patterns associated with that data and could potentially generate responses derived from it during inference. Therefore, the most effective approach is to delete the custom model trained on confidential data, completely remove confidential data from the training dataset, and then retrain the custom model. This ensures the new model does not encode any features of the confidential data, thereby preventing inference responses derived from it.