Tech Giants Commit $725 Billion in 2026 Capital Spending as AI Infrastructure Race Intensifies
Meta, Amazon, Microsoft, and Alphabet have collectively committed roughly $725 billion in capital expenditures for 2026, a 75 percent increase over the prior year. Most of the spending targets data centers, custom AI chips, and GPU clusters to support growing AI workloads.
Meta, Amazon, Microsoft, and Alphabet have collectively committed approximately $725 billion in capital expenditures for 2026, according to data compiled by TechStartups. That figure represents a more than 75 percent increase over the prior year and reflects the scale of investment required to build and run modern AI systems.
Most of the spending targets data centers, custom AI chips, and GPU clusters. Nvidia remains the dominant supplier of AI chips, though several companies are developing their own silicon to reduce dependence on a single vendor. OpenAI's effort to build custom chips with Broadcom has reportedly hit an $18 billion financing hurdle.
The spending surge is reshaping the tech workforce. Meta plans to lay off 8,000 employees in May 2026. Amazon has cut around 30,000 roles in recent months. Both companies say they are reallocating resources from payroll to compute infrastructure.
SpaceX is developing a $55 billion facility in Texas called Terafab for AI chip manufacturing. The facility is intended to support Tesla's self-driving systems, humanoid robots, and AI data centers.
Nvidia CEO Jensen Huang was notably excluded from President Trump's recent trip to China, a signal of how central the company has become to U.S.-China tensions over AI chip export controls.
China's AI sector is also attracting large investments. Moonshot AI raised $2 billion at a $20 billion valuation. DeepSeek is seeking funding at a reported $45 billion valuation, backed by China's state-funded semiconductor program.
IBM released a report in May 2026 showing that 76 percent of companies have now established a Chief AI Officer role, up from 26 percent in 2025, as organizations try to manage the pace of AI adoption.


