Why Big Tech's AI Spending Is $3 Trillion Higher Than It Seems
Bubble signal: +2 · Overheating evidenceRelevance 85CapexUS
Source: NewsData: AI bubble (EN) · Published 2026-08-18
Summary
The article argues that Big Tech's actual AI spending is far higher than publicly reported, potentially underestimated by $3 trillion. This discrepancy stems from different accounting of capex, R&D, and indirect costs. The author suggests this hidden spending intensifies the risk of an AI bubble.
Bubble analysis
This news directly addresses the scale of AI investment, suggesting actual spending far exceeds reported figures, providing quantitative evidence for the bubble thesis: massive capex running far ahead of AI revenue is a core argument for a bubble.
#capex#ai-spending#big-tech
Read original ↗Related signals
linked by shared tags
- AI Spending Is Entering a More Disciplined Phase#capex#ai-spending
- Lam Research Stock Rises as Analysts Lift AI Capex Outlook#capex
- Another A-Share Company Signs Over 1.9 Billion Yuan Computing Power Order#capex
- The AI capex boom masks an uncomfortable truth about S&P 500 earnings#capex
- Nearly 400 Billion Invested, Computing Power Demand Surging: The 'Dual Engine' in Power Equipment Half-Year Reports#capex