FinOps Foundation Launches New FinOps for AI Certification
SAN DIEGO — It’s AI boom time: Gartner predicts an estimated $644 billion will be spent on generative AI in 2025.
However, few, if anyone, know how to manage the cost of AI, much less understand its value.
At FinOps X this week, the FinOps Foundation introduced a certification for AI. The new credential is for practitioners who seek to understand AI spending and the value it offers.
The certification is built into three parts and will evolve, Stacy Case, the FinOps Foundation’s vice president for professional development, told The New Stack.
The certificate will be updated over the coming months with additional levels. New learning materials and badges will be introduced, culminating in full certification next year.
The certificate costs $500; the introduction is available immediately and will be followed by three levels:
- Level 1, available in September
- Level 2, available in November.
- Level 3, available in January 2026.
The rolling nature of the FinOps for AI certificate curriculum reflects the rapid pace of market development.
If the FinOps Foundation had prepared the AI certificate upfront, then it would be obsolete.
“It will typically take us three to six months to create a certification,” said Case.
Worse, practitioners may not have a clear direction if they follow a curriculum that is outdated, due to the rapid evolution of the market.
Instead, the certificate will enable FinOps professionals to initiate training and take action pretty quickly.
How Do You Gather the Data?
The certificate covers AI-specific cost allocation, chargeback models, workload optimization, unit economics and sustainability, according to the FinOps Foundation. Practitioners will also learn how to govern AI investments, forecast usage and align financial strategies with AI innovation.
The first section focuses on the fundamentals for understanding FinOps in the context of AI. How do you gather the data?
“Because until you have a clear understanding of what that data is and understanding what everything is, there’s no point in doing anything else.” Case said.
She likened the curriculum’s structure to acquiring building blocks: “So you’re building with the foundation and the first one, and with the second one, you’re gonna be building onto that. You can now develop more strategic oversight using the available data. What are some of the things that you can do? How can you plan? How can you forecast? How can you estimate your spending in your usage?”
The certification is scheduled to culminate in March 2026, with a focus on advanced practices in AI. At that time, participants will begin to learn how to optimize AI workloads. They’ll also get a LinkedIn badge and a pin awarded to them at FinOpsX 2026.
“Instead of waiting for everything to be completed, we’re taking people along with us on this journey,” Case said.
J.R. Storment, executive director of the FinOPs Foundation, told The New Stack that enterprise managers are trying to get ahead of the whiplash. Company managers say the amount of spending on AI is now starting to raise questions: “Hey, you’re using this AI service and spending too much. Do you know what it’s for?”
So far, the focus is primarily on how AI aligns with the costs and the value it provides.
Adopting FinOps Practices for AI
The FinOps community emerged in the cloud era. The scope of AI is vast, and the pace of market adoption is unprecedented. But what people learned in the cloud era makes a difference. Still, the change is constant, and that’s an adjustment that Storment said requires people in FinOps roles to adopt FinOps practices for AI, particularly what the organization refers to as “scopes.”
A presentation at FinOpsX illustrates the concept of “scopes” for AI usage. (Photo by Alex Williams)
Data collection is crucial at this point to understand what to optimize. When the FinOps Foundation did its State of FinOps survey earlier this year, AI optimization showed up seventh or eighth on the list.
Practitioners, instead, wanted to know how to allocate spending, anticipate anomalies and address unexpected expenditures. For example, how does it tie back to unit economics, a term that essentially means how your revenues and costs add up on a per-customer or single unit of a product?
It used to be that engineers could spin up resources, and they were the only ones who did, Storment said. They had the keys to the kingdom.
Now, Storment said, “you’ve got a bunch of other personas in the org masquerading as engineers, deploying resources via SaaS or other areas, like token-based charging.”
