Big Tech Is Spending $725B on AI in 2026 and Still Cannot Build Fast Enough

• Amazon, Microsoft, Google, and Meta will spend $725B on AI in 2026, up 77% from $410B last year, more than the entire GDP of Switzerland
• Microsoft says it cannot build fast enough to meet AI demand and will remain capacity-constrained through all of 2026
• Amazon faces $17B in negative free cash flow this year as data centers now consume 70% of the world’s entire memory output

Amazon, Microsoft, Google, and Meta are set to spend a combined $725 billion on AI infrastructure in 2026. This was brought to light by analyst estimates compiled after the latest earnings season. The jump in AI spending comes as demand for cloud and AI services continues to rise across the industry. Even after sharply increasing their budgets, some companies say they still lack sufficient capacity to meet demand. Microsoft, in particular, warned that supply constraints are likely to continue through 2026 as it expands its AI infrastructure footprint.

Also Read: Trump’s $1T China Summit Has Every Major CEO Except Nvidia’s Jensen Huang

Why Big Tech Is Burning Cash at Historic Rates to Win the AI Race

The scale of Big Tech AI spending has moved far beyond what investors were used to just a few years ago. Amazon plans to spend about $200 billion this year, while Microsoft and Alphabet are each expected to spend close to $190 billion. Meta recently increased its own forecast to as much as $145 billion. Combined, Big Tech capex is projected to rise 77% from last year’s $410 billion total.

Goldman Sachs analysts now estimate hyperscaler capex could reach as high as $755 billion in 2026. The spending is heavily focused on data centers, AI chips, networking equipment, and memory infrastructure needed to support large-scale AI models and cloud demand. The analysts added,

“The consensus of analysts is now for the mega-cap US hyperscalers to spend $755 billion on capex in 2026, representing growth of +83% vs. 2025.“

Microsoft Says Demand Still Exceeds Sharply

Microsoft’s AI investment remains one of the largest in the sector. The company said it expects to spend around $190 billion this year, well above earlier analyst estimates.

Chief Financial Officer Amy Hood said about $25 billion of that increase is tied to higher memory and component costs. She also told investors that Microsoft expects to remain capacity-constrained through at least 2026 as it continues bringing GPU, CPU, and storage infrastructure online.

The company’s AI business is now running at a $37 billion annualized revenue rate, up 123% year over year. Microsoft added that demand for Azure AI services continues to outpace available supply.

Also Read: Anthropic Pre-IPO Value Tops $1.4T: Bigger Than SpaceX?

Rising Infrastructure Costs

It should be noted that the aggressive spending is also starting to pressure free cash flow across the sector. Goldman Sachs analysts noted,

“This capex is estimated to reach 100% of cash flows from operations this year, leaving little room to return cash to shareholders without a drawdown of cash balances or large increase in debt.”

“Accordingly, the hyperscalers cut buybacks by -64% year/year during Q1, and now allocate 20% of total spending to buybacks and dividends compared to an average of 34% from 2017-2022.”

Morgan Stanley analysts estimate Amazon could see negative free cash flow of nearly $17 billion this year due to its infrastructure expansion. Meta has also faced investor concerns after increasing its capex guidance for the second time this year.

Companies pointed to rising costs for memory chips, land, electricity, and skilled labor as major reasons behind the larger budgets. Industry estimates show data centers now consume roughly 70% of global memory output. This has helped drive gains in memory-related stocks such as Micron Technology and SK Hynix.

Despite the spending pressure, executives continue to defend the investments. This is necessary to secure long-term growth in cloud and AI services. Analysts also note that cloud revenue growth at Microsoft, Alphabet, and Amazon has remained strong. They suggest the early returns from AI infrastructure are already starting to appear.

Also Read: Elon Musk Backs Buffett’s 5-Minute Fix for Debt Bigger Than the US Economy