Uber blew through its entire AI budget in four months. The company’s CTO confirmed it, as reported by multiple outlets including Forbes and TechCrunch. The budget collapse wasn’t driven by a single moonshot engineering project — widespread, heavy use of AI coding tools like Claude Code across the organization drained the annual allocation by April. That pattern is now repeating at companies across industries. The tab just arrived, and finance teams are choking on it.
The Bill Always Comes
Routine office tasks, scaled across thousands of employees, are quietly generating the biggest AI invoices.
Leaked audio from inside Accenture, reported by 404 Media, reveals executives alarmed at how fast token spend is climbing. Justice Kwak, quoted in that reporting, described “soaring token spend” and identified a surprise: the heaviest consumers aren’t engineers writing complex code. They’re office workers converting PDFs into slides, reformatting documents into markdown, automating tasks that used to cost nothing but time. According to 404 Media’s reporting, an Accenture spokesperson told CNBC the company wants to be an “AI-enabled” workplace — one requiring adoption of the latest tools. The predictable result followed.
- Non-engineers running routine tasks at scale — slide builds, format conversions — generating outsized token bills
- Uber capping employee AI tool usage after burning through its budget in four months, per TechCrunch
- Accenture developing a product called Token IQ to help clients manage token consumption, as reported by 404 Media
- According to Deloitte’s AI governance guidance, organizations should adopt real-time monitoring, model right-sizing, and FinOps-style controls
- AI pricing shifting from flat subscriptions to seat-fee-plus-pre-committed-token models, per FinOps Foundation analysis
Welcome to Token Economics
Token governance is shaping up to be the cloud cost crisis of the AI era — and the playbook for surviving it looks familiar.
This is the same movie that played during the early cloud era, just on a different streaming platform. Companies embraced unlimited usage, ignored the meter running, then spent years building FinOps disciplines to claw back control. According to FinOps Foundation frameworks, token economics requires connecting consumption directly to business outcomes — otherwise, you’re paying too much with no receipt. Deloitte’s guidance argues organizations should manage AI spend like a carefully budgeted financial resource, not an open bar with no receipt.
Expect what comes next to feel familiar: token quotas, role-based access tiers, consumption dashboards, and chargeback models arriving in your department. Vendors selling copilots and coding agents will face harder questions about usage-based pricing. The aggressive AI adoption mandate is quietly getting walked back at companies that haven’t admitted it publicly yet.
The unlimited AI buffet is closing.
The largest recurring bills aren’t coming from elite engineers pushing frontier models to their limits. They’re coming from everyday workers automating mundane tasks across thousands of seats. The real AI transformation isn’t in the model. It’s in the spreadsheet tracking who used it.
