FinOps highlights cloud + AI cost trends

FinOps teams are now treating artificial intelligence spend as a standard budget line, not an experiment, in the FinOps Foundation’s 2026 industry survey. (finops.org) The FinOps Foundation said 98% of respondents now manage artificial intelligence spend, up from 63% in 2025 and 31% in 2024. The 2026 survey drew 1,192 respondents representing more than $83 billion in annual cloud spend. (finops.org) (finout.io) The report was released in February 2026, and the foundation said artificial intelligence cost management is now the top skill set teams want to add. It also said many organizations are being asked to fund new artificial intelligence projects through savings from existing optimization work. (finops.org 1) (finops.org 2) FinOps is the practice of making engineers, finance teams, and business leaders use the same cost data before they commit to technology spending. In March 2026, the foundation updated its framework to focus on the “value of technology,” not just cloud bills. (finops.org 1) (finops.org 2) That wider scope shows up in the survey numbers. The foundation said 90% of teams now manage software as a service or plan to within a year, 64% manage software licensing, 57% manage private cloud, 48% manage data center spend, and 28% include labor costs. (finops.org) The report describes artificial intelligence, data warehouse compute, and distributed data processing as costs that increasingly land in the same budgeting conversation. The foundation’s framework update says FinOps teams are also working more closely with information technology service management, information technology asset management, information technology financial management, security, sustainability, and enterprise architecture groups. (finops.org) (finout.io) Artificial intelligence creates a different kind of cost problem than ordinary cloud infrastructure. The foundation’s guidance says teams now have to track measures such as cost per token, manage scarce graphics processing units, and watch volatile pricing across model services and supporting infrastructure. (finops.org) The survey also shows where FinOps is moving inside companies. The foundation said 78% of practices now report into the chief technology officer or chief information officer organization, and teams with executive engagement show two to four times more influence over technology selection decisions. (finops.org 1) (finops.org 2) The operating model is still mostly centralized, but not purely centralized. The data library says 60% of teams use a centralized enablement structure and 21% use a hub-and-spoke model, while the foundation says federated champions are becoming more important as spending categories multiply. (finops.org) The foundation’s conclusion is that FinOps is no longer mainly about explaining last month’s cloud invoice. In 2026, it is being positioned as the function that decides whether artificial intelligence, software, and infrastructure spending earns a place in the next budget. (finops.org)