US Big Tech Stock Divergence: AI Infrastructure ROI Defines the New Winners
The massive AI infrastructure investment phase has entered a new chapter. US Big Tech stocks are showing stark divergence as the market increasingly rewards companies that can prove tangible monetization capabilities against their heavy capital expenditures.
The Second Chapter of AI Infrastructure: From 'Spending' to 'Returns'
Entering the second half of 2026, the AI infrastructure investment boom that has driven global equity markets is entering a new phase. Moving beyond the initial blind race for capital deployment, tangible monetization capabilities are now the core variable determining corporate valuations. According to recent earnings reports, the combined capital expenditure (CapEx) of the four major US hyperscalers (Amazon, Google, Meta, and Microsoft) reached approximately $170 billion in Q2 2026, a surge of over 72% year-over-year.
Such astronomical investment costs are raising concerns among investors regarding short-term pressure on free cash flow (FCF). As the market's focus shifts drastically from 'how powerful the AI is' to 'how much return on investment (ROI) is being generated,' a stark divergence in stock price trends among Big Tech companies is intensifying.
Clear Divergence in Monetization Capabilities and Stock Prices
Recent trends in the US stock market show returns being heavily influenced by how seamlessly companies have integrated AI technology into their core business models.
- Immediate Revenue Generation Models: Companies that have maximized the efficiency of targeted advertising using AI algorithms are proving their investment outcomes in actual numbers relatively early, demonstrating positive stock performance. This suggests that AI is functioning as a core revenue driver rather than just a cost center.
Redefining Infrastructure: The War for Power and Data Centers
The nature of AI infrastructure competition is also diversifying. It is no longer just about securing advanced GPU chips; acquiring physical data center space and overwhelming power infrastructure to support AI computations has become an absolute necessity for survival. As power supply constraints are identified as the biggest bottleneck to AI expansion, Big Tech companies are rushing to sign long-term contracts with green energy and nuclear power companies. This has led to a rotation of funds broadening the scope of beneficiaries to include data center REITs and power grid infrastructure companies.
H2 2026 Investment Strategy: Separating the Wheat from the Chaff in the 'J-Curve'
The AI infrastructure industry is currently navigating the lower part of a typical 'J-curve'. It is experiencing the transitional growing pains between massive upfront investments and the full-scale harvest of returns. Therefore, rather than relying simply on the 'AI-related stock' theme, investors must strictly evaluate the efficiency of capital allocation and actual margin defense capabilities.
The new winners will be the companies that can shoulder massive CapEx without undermining the profitability of their core businesses, while driving concrete corporate productivity improvements (Agentic AI) through their AI services. Now more than ever, an investment strategy that discards blind optimism and coldly tracks company-specific monetization data is required.