LIVE

Amazon Introduces Built-In Cost Tracking and Controls for AI Developer Tools on Bedrock

🏢 Big Tech 2 min read 👁 7

Amazon Web Services has detailed built-in cost attribution capabilities for Amazon Bedrock, allowing organizations to track AI spending down to individual users and teams across various developer productivity tools.

Amazon Introduces Built-In Cost Tracking and Controls for AI Developer Tools on Bedrock🏢Big Tech

Photo: AWS Machine Learning Blog

Key points

  • Amazon Bedrock offers built-in cost attribution that tracks inference calls to the IAM principal without changing API code.
  • The tracking works across multiple developer tools like Claude Code, ChatGPT Work, and OpenAI Codex via standard endpoints.
  • Data flows directly into AWS Cost and Usage Reports (CUR 2.0) and Cost Explorer for detailed team and user budgeting.

Amazon Web Services has highlighted native cost attribution and control features within Amazon Bedrock, designed to help finance and platform teams monitor expenditures on AI coding assistants and productivity tools. As enterprise adoption of generative AI scales, organizations face challenges in mapping application spending to specific users, departments, and business value.

According to the AWS Machine Learning Blog, Bedrock provides zero-instrumentation tracking that records the IAM principal behind every inference call. This mechanism operates at the endpoint layer—specifically through the bedrock-runtime and bedrock-mantle endpoints—meaning it works automatically regardless of whether developers use tools like Claude Code, Claude Desktop, OpenAI Codex, ChatGPT Work, or OpenCode. By integrating AWS IAM Identity Center with an external identity provider and passing session tags, enterprises can capture precise per-user metrics alongside organizational dimensions.

The resulting billing data flows directly into AWS Cost and Usage Reports (CUR 2.0) and AWS Cost Explorer without requiring modifications to standard API calls. Organizations can further combine these IAM principal tracking features with Application Inference Profiles, Projects, Workspaces, and metadata tagging to enforce spending limits and achieve deeper, prompt-level visibility into their AI operations.

What to watch next: Monitor how enterprise adoption of these native IAM attribution features influences FinOps strategies and budget management for large-scale AI deployments.

Source AWS Machine Learning Blog — aws.amazon.com/blogs/machine-learning/cost-attribution-and-controls-for… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.