AWS · MLA-C02 · BETA
Auditing, safeguards, and AI risks
MLA-C02 · Lesson 23 of 23 · 3:04:38
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 23 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 4, operating, monitoring, and securing). In this lesson: CloudTrail, AWS Config, and model invocation logs, encryption with SSE-S3 and AWS KMS, the security risks of ML and AI systems, and Amazon Bedrock Guardrails and sensitive data. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- A partner's AWS account must load models that AnyCompany Mobility trains, from encrypted artifacts in Amazon S3. Today, the artifacts use the default encryption. What should change?
- The support assistant answers from documents that customers upload. One document holds hidden text that tells the model to reveal the bookings of other customers. Which risk is this?
- The support assistant sometimes repeats phone numbers and email addresses from booking data in its answers. Customers should still get help, without the personal data. What should the ML engineer configure?