artificial-intelligence cloud microsoft amazon google meta

Big Tech Will Invest $470B+ in AI During 2026: The Infrastructure Race

Microsoft, Meta, Amazon, and Alphabet plan to spend over $470 billion on AI infrastructure. We analyze what this means for the market and businesses.

N
Nextsoft
3 min read

This week’s Big Tech earnings revealed unprecedented figures: the four major hyperscalers plan to invest more than $470 billion in AI infrastructure during 2026, a 34% jump over 2025.

The Investment Landscape

CompanyCapex 2025Capex 2026 (est.)Growth
Amazon$118B$125-146B+6-24%
Alphabet$91-93B$115B++24%
Microsoft$70B$110-125B+57-79%
Meta$65B$95-135B+46-108%
Total~$350B$470-600B+34-71%

Where Is the Money Going?

Typical AI Capex Distribution

  • 45% - GPUs and accelerators (primarily NVIDIA)
  • 25% - Data centers (construction and expansion)
  • 15% - Networking (InfiniBand, cables, switches)
  • 10% - Storage (HBM, enterprise SSDs)
  • 5% - Cooling and power

Spending Winners

ProviderProductBenefit
NVIDIAH200, Blackwell GPUsPrimary beneficiary
ASMLLithography machinesRecord $11.5B profit 2025
SK HynixHBM3e memorySurpassed Samsung in profit
AdvantestChip testingAccelerating capacity

What Each Company Said

Amazon (Andy Jassy, CEO)

“AWS demand continues to exceed our capacity… we’re investing aggressively in AI infrastructure to meet customer needs.”

  • Guided capex: $125B for 2026
  • Focus: Custom chips (Trainium, Inferentia) to reduce NVIDIA dependency

Meta (Mark Zuckerberg, CEO)

“AI is becoming fundamental to everything we do… from recommendations to content creation to advertising.”

  • Potential capex: up to $135B
  • Focus: Llama 4, metaverse, AI for ads

Microsoft (Satya Nadella, CEO)

“We’re seeing unprecedented demand for Azure AI services, particularly Copilot adoption across enterprise.”

  • Goldman Sachs projects: $125B in 2026, $144B in 2027
  • Focus: Azure, Copilot, OpenAI partnership

Alphabet (Sundar Pichai, CEO)

“Our investments in AI are paying off… Gemini is now powering experiences across Google products.”

  • Guidance: “significant increase” over 2025
  • Focus: Gemini, TPUs, Search AI, Cloud

Investor Pressure

Despite massive investments, Wall Street is increasingly skeptical:

The Hard Questions

  1. ROI: When will we see returns on these investments?
  2. Bubble: Are we over-investing in capacity?
  3. Depreciation: $680B in accumulated depreciation coming
  4. Competition: Is the arms race worth it?

Business Implications

Opportunities

  1. Cheaper cloud: Potential oversupply = better prices
  2. AI as commodity: Democratized compute access
  3. New services: Accelerated platform innovation

Risks

  1. Dependency: Vendor lock-in with hyperscalers
  2. Variable pricing: Costs may fluctuate with demand
  3. Temporary scarcity: GPU capacity remains limited short-term

2026 Predictions

  1. Q2: Pressure to demonstrate ROI increases
  2. Q3: Possible capacity consolidation
  3. Q4: First signs of enterprise AI profitability

Need to optimize your cloud and AI strategy? Contact us for an architecture and cost evaluation.

Related Terms

Learn more about these concepts in our glossary:

View all terms →
Share article:
💬

Have a project in mind?

Let's talk about how we can help you achieve your technology goals.

Schedule a free consultation