AWS launches FinOps Agent in public preview, bringing AI-driven cost investigation to cloud teams

Amazon Web Services launched the AWS FinOps Agent in public preview on June 9, 2026, an AI-powered agent designed to help engineering and finance teams investigate cloud cost anomalies, answer cost questions, and automate recurring FinOps workflows — all through the tools teams already use, including Slack and Jira.

What's new

The AWS FinOps Agent is "an agentic AI solution that investigates cost anomalies to root cause and answers cost questions for engineers across your organization, in the tools they already use," according to AWS's announcement. The agent connects to AWS Cost Explorer, Cost Anomaly Detection, Cost Optimization Hub, and Compute Optimizer, and can surface findings directly to Slack channels or create Jira tickets without manual intervention.

The core use cases:

• Cost anomaly investigation: When AWS Cost Anomaly Detection flags an unusual spend pattern, the agent automatically pulls AWS CloudTrail events, correlates them with the cost change, identifies the root cause, and posts the summary to Slack or another configured destination.
• Natural language cost queries: Engineers can ask cost questions in plain language — "What drove our EC2 spend up this week?" — and the agent responds using the organization's actual cost and usage data.
• Recurring reports: The agent generates cost reports in HTML, PDF, or PowerPoint format on a defined schedule, pulling from Cost Explorer and Compute Optimizer.
• Optimization recommendations: The agent surfaces rightsizing, idle resource, and Savings Plans recommendations and can open Jira tickets to route action items to the right teams.

The agent supports context file uploads that allow teams to customize its behavior with organization-specific knowledge — naming conventions, tagging strategies, or cost allocation rules specific to the business.

Pricing during the preview period is free, though standard charges apply for the underlying AWS services the agent calls on the user's behalf.

Context

Cloud cost management has grown significantly more complex as organizations deploy multi-service architectures across multiple AWS accounts and regions. Traditional FinOps approaches — periodic spreadsheet reviews, dashboard monitoring, and manual ticket triaging — do not scale with the pace of modern infrastructure changes. AWS is positioning the FinOps Agent as the shift toward "continuous workflows" rather than periodic reviews: "FinOps is shifting from periodic, dashboard-driven reviews toward continuous workflows."

The agent builds on AWS Bedrock's foundation model infrastructure and follows a pattern AWS has been expanding across its services: converting operational tasks (monitoring, security triage, compliance checking) into agentic workflows that run on a schedule or in response to events rather than requiring human-initiated queries.

Why it matters

Cloud cost overruns are a persistent operational problem at companies running significant AWS workloads. The window between a cost anomaly appearing and an engineer identifying its root cause is often measured in hours or days — long enough for the underlying issue to compound. An agent that can investigate anomalies in minutes, identify responsible teams, and route action items automatically shortens that window substantially.

For engineering teams, the natural language query layer removes the need to learn Cost Explorer's filtering syntax or navigate multiple dashboards. For FinOps practitioners, automated recurring reports and proactive anomaly summaries reduce the manual effort of preparing finance stakeholder updates.

The preview period also gives AWS time to refine the agent's accuracy before general availability — particularly important for anomaly root-cause attribution, where an incorrect diagnosis sent to a Slack channel could trigger unnecessary investigation or erode trust in the system.