跳转至

The State of the AI Economy — $110B Revenue, Bottom-Up Deduplicated Model

Ch01.820 The State of the AI Economy — $110B Revenue, Bottom-Up Deduplicated Model

📊 Level ⭐⭐ | 3.3KB | entities/exponentialview-ai-economy-110b-2026.md

The State of the AI Economy — $110B Revenue, Bottom-Up Deduplicated Model

Background: Based on Exponential View's 2026-06-25 inaugural AI Economy report, using bottom-up, deduplicated financial modeling covering consumer and enterprise AI spending across the full stack.

Core Numbers

  • Past 12 months AI revenue: $110B (deduplicated)
  • Annualized run rate: $175B
  • Methodology: Bottom-up per-company P&L modeling, cross-validated, no double-counting
  • Coverage: Consumer + enterprise AI spending full stack (excluding internal AI uplift, professional services)

Methodology Innovation

Traditional AI market sizing has a severe double-counting problem: user pays Anthropic $1, Anthropic pays AWS $0.5, traditional methods report $1.5. Exponential View reports only end-user actual spend of $1.

Approach: 1. Build item-by-item financial models (P&L, balance sheet, cash flow) for largest contributing companies and business units 2. Cross-validate using high-confidence public statements, supplier data, customer feedback 3. Assign confidence weights to leaked information and self-reported data 4. All numbers are auditable — traceable to specific data points with confidence weights

Supply Side vs Demand Side

Supply side (well-understood): - Chips, memory, power transformers, cooling, data center components - Mostly public companies, trackable via disclosures, sales, forward order books

Demand side (much harder): - OpenAI, Anthropic, Cursor, ElevenLabs etc. are private — no disclosure obligations - Hyperscalers (AWS/GCP/Azure) inconsistently disclose AI segment revenue - Requires piecing together public statements, leaks, self-reports

Exclusions

  • Internal AI uplift (e.g., recommendation system improvements increasing ad revenue)
  • Efficiency savings from big tech internal tools
  • Professional services and systems integration

Differentiation from Existing Wiki Entities

Dimension This entity (Exponential View) Nadella Token Capital Dario Amodei Policy
Angle Data-driven market sizing CEO enterprise strategy AI policy/safety
Core contribution $110B deduplicated revenue model Token capital dual framework Exponential growth policy response
Methodology Bottom-up P&L modeling Strategic vision Policy analysis
Actionability Investment/market judgment Enterprise architecture decisions Regulatory/compliance

source archiveNadella Token CapitalDario Amodei Policy