AI Spending Soars: Microsoft and Uber Blow Budgets Early

Microsoft and Uber are at the forefront of a new tech financial challenge: enterprise AI spending has spiraled out of control. Despite forecasts, both giants have already exceeded their planned budgets for 2026 within the first quarter of the year. This situation presents a profound question for the industry: how sustainable is the current AI spending frenzy?

Chronology: A Rapid Unfolding

Let's rewind to the start of the year. As 2026 kicked off, both Microsoft and Uber had ambitious plans for AI integration. For Microsoft, the adoption of Claude Code, an agentic coding tool, was meant to transform its Experiences and Devices division. But by mid-May, the tech giant began winding down most internal licenses, with engineers heavily reliant on these tools up until June 30.

Uber's story is even more startling. By April, just four months into the year, the ride-hailing company announced that it had exhausted its entire AI budget for 2026. According to Uber's Chief Technology Officer, Praveen Neppalli Naga, the realization that their budgeting was wildly off the mark was a wake-up call. "I'm back to the drawing board," he admitted, after seeing AI tools like Claude Code burn through resources faster than anticipated.

Both companies were caught in the same trap: token-based pricing for AI tools, which, while initially appealing, quickly became unsustainable at scale. As Uber deployed Claude Code for 5,000 engineers, monthly costs per engineer soared to between $500 and $2,000, with AI handling 70% of the company's code.

Impact: Unforeseen Financial Strain

So, what happened when budgets blew apart? Microsoft and Uber faced a new kind of tech-induced financial strain. Token-based billing for AI tools, once seen as a cost-saving measure, rapidly outpaced any headcount savings it was supposed to create. Essentially, the cost of integration and the operational scale needed became an unexpected burden.

This financial headache isn't isolated to just these two companies. A 2025 survey by Mavvrik showed that 85% of companies missed their AI cost forecasts by more than 10%, with 84% reporting that AI spending slashed gross margins by over six percentage points. The AI cost crisis is very real and industry-wide.

For Uber, the immediate consequence was a forced re-evaluation of its AI strategy. Engineering teams, previously reliant on AI tools, had to reassess how they deployed these technologies. Microsoft's decision to scale back on Claude Code licenses similarly reflects a broader industry trend: the rush to integrate AI was moving faster than the financial checks and balances.

Outlook: A Tighter Grip on AI Budgets

So what comes next in this unfolding narrative? Companies are now reconfiguring their approach to AI to avoid further budgetary pitfalls. They're implementing quotas, internal leaderboards, and exploring cheaper model routing to navigate these financial challenges.

, the pressure is on for tech firms to manage AI spend effectively. The sudden spike in expenses has prompted a doubling of FinOps teams in the last year, from 31% to 63%, tasked with reigning in these spiraling costs. Big Tech's AI capital expenditure hit a staggering $650 billion in Q1 2026, highlighting the urgency of these financial control measures.

Anthropic, the provider of Claude Code, stands in a unique position. Despite customer complaints, the vendor is projecting its first profitable quarter with a $10.9 billion forecast for Q2. But as the industry grapples with these financial challenges, the question remains: how can companies balance the need for AI innovation with sustainable spending?

This financial friction isn't just a tech issue. It could very well influence how crypto AI infrastructure builds unfold in the coming months. As the sector grows, how companies manage these costs will be essential.