Google Cloud announced two native cost management features: early anomalies detection on AI services and spend caps on budgets. These aim to address the unpredictability of generative AI costs, where a single prompt can trigger complex operations and significant charges.

The first feature, early anomalies on AI services, is part of the Anomalies tool in the billing console. It uses dynamic baseline modeling to analyze historical project data and build a seasonal baseline of daily service-level costs without manual thresholds. The system monitors early cost signals per service and alerts before actual costs are reported. If a daily cost signal trends abnormally, it flags the deviation and generates a Root Cause Analysis (RCA) that highlights the top three SKUs driving the surge. This allows teams to investigate and intervene before costs escalate.

The second feature, spend caps on Google Cloud Budgets, allows users to set a monthly financial cap on specific services within a project. When accumulated spend reaches the cap, the system automatically restricts further cost-incurring usage for that service. During public preview, spend caps apply to a single project and service for a fixed monthly timeframe. The enforcement is non-destructive—data and resources are not deleted, and services outside the budget scope are unaffected. Administrators and project owners receive automated email alerts at 50%, 80%, and 100% of the budget. If a spend cap is triggered, the usage block remains until manually lifted with a single click from the budgets UI. Underlying fixed commitment fees, such as Committed Use Discounts or Provisioned Throughput, continue to bill at their flat contractual rate. Spend caps for AI services trigger within minutes of hitting the threshold, offering near real-time enforcement to limit financial exposure from runaway models, infinite loops, or massive queries.

Supported services for both features include Gemini API, Agent Platform, Cloud Run, and Cloud Run Functions. The combined workflow is described as a two-step defense playbook: detect anomalies early, then enforce spend caps on identified high-velocity services. Users can access the anomalies section in the billing console and set spend caps from the Budgets & Alerts page.