AIToday
Large Language ModelsAI Business & IndustryAmazon AI BlogPublished: Nov 14, 2026, 04:00 JST

Amazon Bedrock adds per-user cost attribution in CUR 2.0

Send on LINE
Amazon Bedrock adds per-user cost attribution in CUR 2.0

3 Key Points

  1. What happened

    Amazon Bedrock now records the IAM identity of every inference caller and publishes it to CUR 2.0 in the line_item_iam_principal column, with no changes to API calls. Users authenticate via IAM Identity Center, and tools including Claude Code, Claude Desktop, OpenAI Codex, ChatGPT Work, and OpenCode inherit per-user cost visibility automatically.

  2. Why it matters

    The mechanism works at the endpoint layer beneath any tool-specific logic, so organizations can see the real per-user cost of AI tools without custom credential scripts or per-user IAM roles. This visibility is achieved with zero instrumentation.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

The new capability addresses a common attribution problem: when developer and productivity tools run on each user's machine and call Bedrock through a shared IAM role, billing aggregates spend to a single principal. With IAM Identity Center, the session name becomes the user's sign-in name, so each user gets their own principal, and session tags (team, cost center) can enrich the data in CUR 2.0. The post outlines a step-by-step setup with IAM Identity Center, external IdP, and CUR 2.0 configuration, and notes that the mechanism supports four access patterns: IAM users, per-application roles, federated identities, and LLM gateways proxying to Bedrock. For organizations that need enforcement beyond visibility, Bedrock-native usage controls such as AWS Budgets, CloudWatch alarms, DynamoDB-based quota tracking, and LLM gateway proxies offer different trade-offs in enforcement latency and operational complexity.

FAQ
Which AI tools are covered by Bedrock's per-user cost attribution?
The approach works identically for any AI tool that invokes Bedrock through the bedrock-runtime or bedrock-mantle endpoints, including Claude Code, Claude Desktop, OpenAI Codex, ChatGPT Work, and OpenCode.
Do I need to modify my API calls to get per-user cost attribution?
No. Amazon Bedrock records the IAM identity of every inference caller with no changes required to your API calls.
How quickly does per-user cost data appear in CUR 2.0?
Cost allocation tags can take up to 24 hours to appear and up to 48 hours to show in CUR 2.0 and Cost Explorer, and activation is not retroactive.
Amazon AI BlogRead Original Article
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleCrowdStrike to acquire Fiddler, launch Falcon Observe in FY2027 Q4