Cast AI, the leading Kubernetes automation platform for cloud-native and AI infrastructure, today announced it has joined the Tokenomics Foundation as a founding Premier Member. Laurent Gil, Co-founder and President of Cast AI, has been named to the Foundation’s Governing Board. The Tokenomics Foundation, hosted by the Linux Foundation, was established to define open industry standards, benchmarks, and best practices in AI economics.
Cast AI joins at a moment when enterprises are under mounting pressure to demonstrate ROI on AI investments, yet lack the frameworks to connect their infrastructure spending to the value AI actually delivers. Token consumption is forecast to increase 24-fold by 2030, according to Goldman Sachs, but most organizations are still measuring costs only at the model invoice layer – missing the largest and least visible portion of their AI spend.
Cast AI’s position is grounded in data. Across thousands of production deployments, Cast AI research puts enterprise GPU utilization at an average of 5%. The implication is significant: most of what a token costs to produce is idle hardware, not active compute. The industry has been standardizing the meter in the middle, the token itself, while both ends of the economics remain ungoverned.
Governance and AI sovereignty is another gap of the token factories. Enterprises moving AI workloads into production still have no standard way to prove where inference runs, which jurisdictions their data crosses, or who can audit the path from prompt to output. Regulated industries are forced to invent their own controls and take each vendor’s guarantees on faith. The Foundation’s open standards work gives the industry a vendor neutral place to address these kinds of challenges, and they belong in the same conversation as ROI: a company that cannot trace where its tokens are produced has no way to certify compliance on them.
“Enterprises are under pressure to show ROI on tokens, but almost none can say what a token costs to produce,” said Laurent Gil, Co-founder and President of Cast AI and Governing Board Member of the Tokenomics Foundation. “Our research puts enterprise GPU utilization at 5%, so most of a token’s cost is idle hardware. The return shows up at the business outcome, and almost no one measures it. The industry is standardizing the meter in the middle and ignoring both ends. Cast AI joined the Tokenomics Foundation to help measure both.”
“We’re excited to welcome Cast AI to the Tokenomics Foundation. Their data on infrastructure efficiency and GPU utilization highlights a critical gap in how the industry understands the true cost of AI,” said J.R. Storment, Executive Director of the Tokenomics Foundation. “By bringing visibility to both the production and performance aspects of AI economics, Cast AI helps the community move beyond just tracking token usage alone to better understand the larger AI picture in supporting the development of open best practices.”
The Tokenomics Foundation’s roadmap includes the Big-T Framework for classifying token cost complexity ahead of model routing decisions, Token Cost Telemetry improvements to the FOCUS billing specification, and AI Value Frameworks that connect total cost of ownership to business outcomes. Cast AI’s involvement spans both the production layer, where infrastructure decisions determine the floor cost of every token, and the consumption layer, where model routing, caching, and prompt architecture determine the ultimate cost of using those tokens.
