Hypercloud Players Reveal Their Latest AI FinOps Products

SAN DIEGO — At FinOps X, the FinOps Foundation‘s annual conference last week, Amazon Web Services (AWS), Google Cloud, and Microsoft all had things to say about their AI FinOps offerings.

AWS

John Phillips, Director of AWS Insights, emphasized the integration of AI-driven cost optimization features into its developer tools, notably through Amazon Q for developers. The cloud power announced that Amazon Q now includes advanced cost optimization functions, helping teams identify savings opportunities and generate actionable execution plans. This is part of a broader push to embed FinOps best practices into the developer workflow, ensuring that cost management is a seamless part of the software lifecycle.

Additionally, AWS introduced I/O optimizations for Aurora, its MySQL and PostgreSQL compatible Relational Database Manager System (RDBMS). Here, the name of the game is to automate reducing the number of input/output operations per second (IOPS) required by your AI from its database. On top of that, AWS Cost Explorer will now include new AI cost comparison capabilities. These enhancements are designed to make it easier for organizations to monitor, analyze, and optimize their cloud spending, especially as AI workloads, with their unique consumption patterns and GPU-intensive requirements, become more prevalent.

Google

Sarah McMullin, Head of Google Cloud FinOps Product, has said earlier that FinOps number one jobs were workload optimization and waste reduction. After all, It’s extremely difficult to optimize workloads or applications if customers cannot fully understand how much is even being used. Why purchase a committed use discount for compute cores that you might not even be fully using?”

At FinOps X, she spelled out what that meant in practice at Google Cloud. The centerpiece of Google’s new FinOps Hub 2.0, which is now in preview, is the deep integration of Gemini Cloud Assist, a generative AI tool now embedded directly into the FinOps Hub. Gemini enables users to quickly generate personalized cost reports, analyze spending trends, and receive actionable recommendations for cost optimization. McMullin emphasized that you could do all this without additional agents or complex setup. This AI-driven approach allows FinOps practitioners to automate time-consuming tasks, such as identifying waste and summarizing optimization opportunities, then seamlessly communicate these insights to engineering and business teams for rapid remediation.

She added that Google Cloud has enhanced the granularity of its cost data, now supporting detailed breakdowns for compute nodes and disks. The platform’s AI-powered anomaly detection operates at an hourly rate and delivers root cause analysis, allowing organizations to quickly pinpoint and resolve unexpected cost spikes. Since its rollout, more than a million cost anomaly alerts have been delivered, helping customers avoid unnecessary spend.

FinOps Hub 2.0 consolidates Google Cloud’s cost management tools into a single dashboard. It provides a unified view of savings, optimization opportunities, and spend forecasts. The platform supports full allocation of cloud spending, accurate forecasting, and now offers enhanced support for multicloud environments via its BigQuery AI data warehouse. Its materialized views and Looker dashboards enable organizations to analyze and optimize costs across diverse cloud landscapes.

Microsoft

In the meantime, Microsoft was showing off its cloud cost management strategy by deeply integrating FinOps principles and AI-driven agents into its core products. In particular, they’ll be adding this functionality into GitHub Copilot,

According to Cyril Belikoff, Microsoft VP of Cloud, these “new AI agents within GitHub Copilot will pick up the load for developers modernizing apps. These capabilities will allow organizations to modernize apps onto Azure faster and more cost-effectively.” The AI agents assist with code assessments, remediation, and configuration, streamlining the transition to the cloud and reducing manual workloads for developers. Support for Java is available now, with .NET coming soon, reflecting Microsoft’s ongoing commitment to open source technologies.

Belikoff also praised the capabilities of Azure AI Foundry. Belikoff is “an amazing service that really brings together a set of capabilities and allows you to manage an AI infrastructure in an integrated way so that you can evaluate models, decide and pick the best one based on price performance and many other factors. In addition, a new agent service within AI Foundry will enable you to monitor, optimize, and manage the growing number of AI agents operating within their environments.

For FinOps pros, Microsoft is introducing comprehensive modernization guidance, helping ensure that cloud migrations are not only efficient but also cost-effective. One new feature is Azure AI Foundry Provisioned Throughput Reservations. This allows organizations to reserve compute capacity across multiple AI models. This includes those from OpenAI, Microsoft, Llama, Grok, Mistral, and Black Forest under a single cost structure. This flexibility enables IT teams to optimize deployments and align cloud spending with budget goals.

Additionally, FinOps workers can benefit from continued investments with Copilot for Cost Management which now delivers more sophisticated recommendations for resource optimization. Additionally, Microsoft Fabric is now “focus aware.” This enables organizations to ingest multicloud cost data and leverage AI-generated analytics and conversational insights directly within the platform. In short, Microsoft is responding to strong business demand for tools that maximize value from AI and cloud investments.

The Future of AI-Driven FinOps Tools

Really, all the hyperclouds have that goal in mind. It’s all about giving people control of their cloud and AI spend, a la FinOps Open Cost and Usage Specification (FOCUS) version 1.2. The methodologies vary; Microsoft is big into AI agents, while Google avoids the agentic approach, but they’re all going to the same destination. Which one will work best for you will depend on your AI and cloud needs. These tools won’t determine which services you’ll invest in. That said, whichever way you go, you should look into using these tools so you’ll have control over the money you’ll pour into whatever services you use. You’ll be glad of the savings. Only a fool wouldn’t use FinOps.