Big Tech AI Spending Sparks Cash-Flow Reckoning

Big Tech AI spending pushed Alphabet (GOOG, GOOGL) on July 22, 2026, to raise its 2026 capital-expenditure guidance, a move that coincided with record quarterly revenue and the company’s first negative free cash flow since 2004. This shift intensified investor scrutiny across the sector over the cash-flow and margin impact of AI infrastructure investments.

Alphabet’s Quarter and Capital-Expenditure Guidance

Alphabet reported about $118 billion in revenue for Q2 2026, its largest quarterly top line on record. The company’s free cash flow turned negative for the first time since its 2004 IPO after capital spending of $44.9 billion exceeded operating cash inflows. Management raised 2026 capital-expenditure guidance to $195–$205 billion, up from a prior $180–$190 billion range, and signaled a significant further increase for 2027. Spending is focused on AI data centers, specialized processors including TPUs, and infrastructure supporting Gemini models and AI-enhanced search and cloud services. Over the year, the firm’s guidance rose sharply from an initial plan near $115 billion to roughly $185 billion after Q1, reflecting the rapid escalation of AI investments.

Hyperscaler Spending and Market Reaction

Analysts estimate that Alphabet, Microsoft, Amazon, and Meta will commit roughly $724–$730 billion in capital expenditures in 2026, with projections approaching $950 billion in 2027. These outlays concentrate on data centers, high-performance chips, and power infrastructure needed for AI training and inference. A broader group of hyperscalers, sometimes including Apple and Tesla, plans about $700–$725 billion in 2026 capex, highlighting the concentration of spending among a few platform operators.

Microsoft and Amazon emerged as relative winners in the earnings cycle. Microsoft signaled further acceleration in cloud revenue growth and reclassified some AI spending from capital expenditure to operating expense while keeping full-year investment broadly unchanged. Some trackers estimate Microsoft’s 2026 capex near $190 billion. Amazon’s 2026 capex is roughly $200 billion, largely tied to AWS and custom-chip programs. Investors rewarded signs that AI spending is translating into visible cloud monetization and profit gains.

The cash-flow picture was more uneven at other peers. Amazon and Tesla reported negative free cash flow in the latest quarter, while Meta’s cash generation plunged about 91%. Market participants linked this deterioration to aggressive AI infrastructure buildouts and sharply higher high-bandwidth memory costs, which have inflated training and inference expenses. Tesla committed roughly $25 billion in capex for 2026, about three times its historical annual level, as it ramps up Optimus and robotaxi programs.

Meta raised the lower end of its 2026 capex outlook to about $130–$145 billion to fund AI training infrastructure, Llama-class models, and data-center expansion. This update coincided with an 8–9% drop in the company’s shares after hours. The combination of heavier spending and elevated component prices has shifted investor focus from raw investment scale to whether AI will lift margins and restore cash generation.

Soaring high-bandwidth memory costs are inflating AI buildout expenses. A UBS Chief Investment Office note warned that near-term AI capex remains strong and monetization is becoming more visible but said pressure on free cash flow could constrain AI spending beyond 2027. Investor scrutiny has shifted from applauding scale to demanding evidence of payback. Cloud growth, margins, and free cash flow are now the metrics determining whether the current wave of multiyear AI investments produces durable returns or forces a pullback in capital intensity.