SOA-C02 · Question #737
A medical research company uses an Amazon Bedrock powered AI assistant with agents and knowledge bases to provide physicians quick access to medical study protocols. The company needs to generate audi
The correct answer is A. Use AWS CloudTrail to log API events from generative AI workloads. Store the events in. CloudTrail records Bedrock API calls (agents, knowledge bases, parameters) with full identity details. Storing them in CloudTrail Lake lets you query that audit data directly with SQL-like syntax - meeting the reporting need with minimal extra plumbing.
Question
A medical research company uses an Amazon Bedrock powered AI assistant with agents and knowledge bases to provide physicians quick access to medical study protocols. The company needs to generate audit reports that contain user identities, usage data for Bedrock agents, access data for knowledge bases, and interaction parameters. Which solution will meet these requirements?
Options
- AUse AWS CloudTrail to log API events from generative AI workloads. Store the events in
- BUse Amazon CloudWatch to capture generative AI application logs. Stream the logs to Amazon
- CUse Amazon CloudWatch to log API events from generative AI workloads. Send the events to an
- DUse AWS CloudTrail to capture generative AI application logs. Stream the logs to Amazon
How the community answered
(29 responses)- A79% (23)
- B14% (4)
- C3% (1)
- D3% (1)
Explanation
CloudTrail records Bedrock API calls (agents, knowledge bases, parameters) with full identity details. Storing them in CloudTrail Lake lets you query that audit data directly with SQL-like syntax - meeting the reporting need with minimal extra plumbing.
Topics
Community Discussion
No community discussion yet for this question.