AI Spending Surge Shows Cracks as Big Tech Burns Cash

Big Tech firms have announced combined capital expenditures exceeding $740 billion for 2026, mostly directed at data‑center builds and advanced chips. Investors, who once cheered such outlays, are now questioning whether the spending can be sustained without eroding free cash flow.

Alphabet reported a second consecutive quarter of negative free cash flow, $9.1 billion, while its capex climbed to $51.4 billion and the company guided for up to $205 billion this year. The rapid adoption of its AI‑powered Search mode is slowing the core advertising engine that funds the investment.

Amazon and Meta are in similar territory. Amazon’s trailing‑12‑month free cash flow turned negative $42.4 billion, and it expects $220 billion in capex. Meta posted a $2.1 billion free‑cash‑flow deficit as its new personal AI agent consumes massive compute, yet ad revenue still grew.

Microsoft was the lone bright spot, delivering $18.6 billion of positive free cash flow, but it blamed third‑party capacity delays for a lower‑than‑planned capex. The market reaction punished chip stocks and smaller GPU‑as‑a‑service providers, highlighting how financing constraints ripple through the AI infrastructure stack.

Why it matters for GPU / AI infrastructure

The scale of planned spend signals sustained demand for high‑performance GPUs, yet the cash‑flow pressure may force hyperscalers to stretch procurement cycles, lean on debt markets, or shift workloads to third‑party clouds. For GPU cloud operators, this creates both a pipeline of long‑term contracts and a risk that customers tighten budgets or delay deployments.