Home icon

Design secure generative AI application workflows with Amazon Verified Permissions and Amazon Bedrock Agents

Machine Learning Blog

This article discusses how to design secure generative AI application workflows using Amazon Verified Permissions and Amazon Bedrock Agents. It explains the challenges of applying fine-grained access controls to agent workflows and proposes an architecture that integrates Amazon Verified Permissions into Amazon Bedrock Agents to address this issue.

Specifically, the article covers:

  • Solution overview for an insurance claims chat assistant application
  • Fine-grained access control flows for different user roles (claims administrator, claims adjuster)
  • Entity design considerations and policy design for role-based and attribute-based access control
  • Authentication design with Amazon Cognito and passing ID tokens to Amazon Bedrock Agents
  • Integration of Amazon Verified Permissions with Amazon Bedrock Agents
  • Getting started with the provided code example
Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Oct 21
2024
Best practices for building robust generative AI applications with Amazon Bedrock Agents – Part 2
Oct 2
2024
Best practices for building robust generative AI applications with Amazon Bedrock Agents – Part 1
Sep 11
2024
Enabling complex generative AI applications with Amazon Bedrock Agents
Jun 27
2024
Build generative AI applications on Amazon Bedrock — the secure, compliant, and responsible foundation

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.