Google, Microsoft, Meta, and Amazon Capex Spending to Hit $725 Billion in 2026, up 77% From Last Year — Analyst Says Bear Thesis Is 'Garbage'

Why It Matters

The unprecedented capex wave signals that the AI economy is moving from research to massive infrastructure build‑out, reshaping competitive dynamics among the tech giants. Investors must reassess risk‑return profiles as cash‑intensive projects could pressure margins but also unlock new revenue streams.

Key Takeaways

• Combined 2026 capex reaches $725 billion, up 77% YoY
• Google Cloud revenue jumps 63% to $20 billion, boosting AI spend
• Microsoft forecasts $190 billion capex, citing $25 billion memory‑chip costs
• Meta lifts capex target to $145 billion amid data‑center cost pressures
• Analyst Jefferies calls bear thesis on AI spending 'garbage'

Pulse Analysis

The AI‑driven capital spending surge reflects a broader shift in the technology sector from software‑only models to hardware‑intensive ecosystems. As generative AI workloads demand specialized GPUs, custom chips and massive storage, the "cloud‑as‑infrastructure" paradigm forces the Big Four to pour billions into data centers, edge nodes and supply‑chain resilience. This escalation is not merely a balance‑sheet line item; it signals a strategic bet that control over compute capacity will become a core competitive moat, much like network effects have been for platforms.

At the company level, Google’s 63% cloud revenue growth validates its aggressive investment in custom AI silicon and a deepening backlog that now exceeds $460 billion. Microsoft’s $190 billion capex plan, with $25 billion earmarked for memory‑chip price inflation, underscores a capacity crunch that could delay Azure’s rollout of next‑gen services. Meta’s upward revision to $145 billion highlights the growing importance of AI‑enabled social experiences, yet investor unease persists over its historically lean cost structure. Amazon, while not detailed in the article, is expected to follow a similar trajectory, reinforcing its dominance in both retail and cloud through massive infrastructure spend.

For investors, the implications are two‑fold. On one hand, the scale of spending could compress margins in the near term, especially if component costs remain volatile. On the other, firms that successfully scale AI‑optimized hardware may capture outsized market share and command premium valuations, as evidenced by Alphabet’s post‑earnings rally toward a $4.3 trillion market cap. The prevailing narrative that AI spending is a speculative bubble is increasingly challenged by tangible revenue growth, suggesting that the bear thesis may indeed be "garbage" in a market that rewards infrastructure leadership.