A company wants an AI tool that lets employees view open customer claims, retrieve details for a specific claim, and access supporting documents. Which solution is appropriate?

The correct solution is to use Agents for Amazon Bedrock with Amazon Bedrock knowledge bases. Agents for Amazon Bedrock can orchestrate complex tasks, breaking them down into sub-tasks and using tools to achieve goals. In this scenario, an agent can be configured to understand user requests like "view open claims" or "retrieve details for claim XYZ." Knowledge bases for Amazon Bedrock allow you to connect foundation models to your company data, such as customer claim databases and document repositories, enabling the agent to retrieve specific, up-to-date information. Using Amazon Fraud Detector is incorrect because it's designed for fraud detection, not for general information retrieval or task automation. Amazon Personalize is for building recommendation systems, which is not the requirement here. While Amazon SageMaker can build custom ML models, training one from scratch for this conversational AI task is overly complex and time-consuming compared to leveraging the pre-built capabilities of Agents for Bedrock and knowledge bases.