Capex Divergence: The $725B AI Buildout vs. Apple's Capital-Light Buybacks
The seven largest US tech companies are experiencing an unprecedented divergence in how they allocate capital. On one side of the ledger, the "hyperscaler" cohort—comprising Microsoft, Alphabet, Amazon, and Meta—is engaged in the largest capital spending commitment in corporate history, pouring hundreds of billions of dollars into artificial intelligence infrastructure. On the other side, Apple remains highly capital-light, keeping its quarterly capital expenditures (capex) to a bare minimum while returning massive amounts of cash to shareholders through buybacks and dividends.
According to recent analysis, the combined 2026 AI infrastructure capex across Amazon, Microsoft, Alphabet, and Meta is now estimated to be near $725 billion, up from prior expectations of $650 billion. This massive scale has dramatically raised the cost of entry for top-tier AI platforms. For example, during its Q1 2026 earnings release, Meta raised its full-year 2026 capex guidance to an astronomical $125–$145 billion, up from its prior range of $115–$135 billion, citing rising component pricing and additional data center costs to support future capacity.
In stark contrast, Apple’s capital allocation playbook is entirely focused on returning cash to shareholders. Apple's capex for the quarter ending March 31, 2026, was just $1.97 billion, despite generating a record $28.70 billion in operating cash flow. Instead of building massive data centers, Apple authorized a fresh $100 billion stock repurchase program—following its historic $110 billion program in 2024—and raised its cash dividend by 4% to $0.27 per share.
This creates a fundamental structural divergence. While the hyperscalers are sacrificing near-term free cash flow to secure physical infrastructure control, Apple is leveraging its dominant ecosystem and brand to maintain high capital efficiency, relying on partnerships (such as its satellite deal with Amazon's Project Kuiper) rather than building its own capital-intensive networks.