AIF-C01 Question 200
Select 4You are developing an AI-powered application on AWS that uses Amazon SageMaker for model training and inference. The application handles sensitive customer data during the training process. Which combination of measures should you implement to ensure the AI system is secure?
- A
Enable encryption for data at rest and in transit.
- B
Use SageMaker-provided default roles without modification.
- C
Implement VPC (Virtual Private Cloud) endpoints for SageMaker.
- D
Use multi-factor authentication (MFA) for all IAM users managing the AI system.
- E
Grant public access to the training data stored in Amazon S3 to facilitate collaboration.
- F
Monitor and audit SageMaker activities using AWS CloudTrail.
Show answer and explanation
Correct answers: A, C, D, F
Explanation
To secure AI systems, it is crucial to implement encryption, restrict access using VPC endpoints, enforce strong authentication mechanisms like MFA, and monitor activities using services like AWS CloudTrail. Avoid practices that expose data publicly or rely on overly permissive default settings. These measures collectively safeguard sensitive data and ensure the AI system operates securely.
- A. Correct.
Encryption for data at rest and in transit is critical to protect sensitive customer data from unauthorized access.
- B. Incorrect.
Using SageMaker-provided default roles without modification can lead to overly permissive access policies, which is a security risk.
- C. Correct.
Using VPC endpoints for SageMaker ensures that communication remains within a private network, reducing exposure to external threats.
- D. Correct.
MFA enhances the security of IAM accounts by requiring a second form of authentication in addition to the password.
- E. Incorrect.
Granting public access to the training data stored in Amazon S3 is a significant security risk and should be avoided.
- F. Correct.
Monitoring and auditing SageMaker activities using AWS CloudTrail helps detect and respond to unauthorized or suspicious actions.