The Linux Foundation announced plans Tuesday to launch the Tokenomics Foundation, a new industry consortium dedicated to defining open standards, benchmarks, and best practices for managing the economics of AI infrastructure as token-based pricing becomes the dominant model across hyperscalers and frontier model providers.
The Cost Problem Is Getting Worse
The timing couldn't be more urgent. Uber's CTO recently disclosed that the ride-share company burned through its entire 2026 AI coding budget by April—a stark reminder that even well-resourced tech giants are struggling to keep AI costs in check. Meanwhile, just months after Microsoft enabled its developers to use Claude Code, the company revoked those licenses due to out-of-control token expenses. These aren't edge cases; they're warning shots across the bow of every enterprise racing to deploy generative and agentic AI workloads at scale. J.R. Storment, the newly appointed executive director of the Tokenomics Foundation, put it plainly in an interview: "We're where we were with clouds in 2017-19. We need to figure out best practices and look at how inference costs, model routing, caching, and prompt engineering all fit in." While per-token prices dropped sharply between 2023 and 2025, Storment noted they have now started leveling off—and Goldman Sachs forecasts global token usage will multiply 24-fold between 2026 and 2030. In a word: ouch.
Building the Infrastructure for AI Cost Governance
The Tokenomics Foundation is positioned as a sibling to the existing FinOps Foundation, extending cloud cost governance practices into what many enterprises now identify as their fastest-growing expense category. Jim Zemlin, the Linux Foundation's CEO, emphasized that tokens have become "the new unit of technology spend" and that measuring token efficiency across models and vendors is critical for business decisions—but no neutral forum existed to develop transparent standards until now. Nishant Gupta, Salesforce's chief availability officer, highlighted why this matters structurally: "Token economics is fundamentally more abstract and more opaque than anything we've managed at this scale before. Input versus output tokens, cached versus non-cached, pricing structures that don't behave like compute or storage." He argued the industry needs to build a different operational muscle through broad experimentation, with best practices contributed back to establish durable standards.
Who Benefits and How It Works
The foundation is designed as a big-tent venue serving both sides of the AI economy. On the buyer side, it targets large enterprises already operating AI at scale that need vendor-neutral ways to compare token economics across models, clouds, and products—organizations increasingly wanting to track not just consumption volume but how token usage maps to business outcomes. On the supplier side, the foundation will engage frontier model providers, emerging NeoClouds, and what the Linux Foundation describes as the broader "token factory" supply chain: the interlocking ecosystem of model hosts, accelerator providers, and specialized AI infrastructure platforms that actually mint and process tokens on demand. A governing board will set overall direction and allocate project funding, while a technical committee develops open specifications, benchmarks, and reference frameworks. The group also plans to co-fund expansion of the FinOps Open Cost and Usage Specification (FOCUS) standard into token-based AI spending models—providing a common schema for representing and analyzing token data across vendors and internal platforms.
Early Supporters and Next Steps
A roster of major enterprise technology players has already expressed initial support, including Accenture, Booking.com, Flexera, IBM, KPMG, Oracle, Salesforce, SAP, and ServiceNow. For these organizations, shared token standards promise to make it easier to explain, govern, and ultimately optimize the AI line item on their P&L statements.
The Bottom Line
The Tokenomics Foundation won't lower your OpenAI bill—but it'll give you the vocabulary to understand why it's so high. As Goldman Sachs' 24-fold usage forecast makes clear, token costs are about to become the biggest line item in enterprise tech budgets, and without standardized measurement frameworks, companies are flying blind. This consortium is a necessary first step toward treating AI cost management as a discipline rather than a mystery.