Anthropic $30T TAM vs Global Economy Decomposer

Scenario Presets

"Eclipses China ($18.5T), equals entire US ($28.7T), and represents a quarter of world GDP ($120T)." — Fortune

TAM Decomposition Levers

Labor Substitution Rate 38%

% of $54T global white-collar payroll automated

Software Budget Share 18%

% of enterprise IT spending captured by AI agents

New Autonomous Value $7.5T

Brand new AI-native industries & cognitive services

Productivity Multiplier 1.25x

GDP expansion feedback from cognitive leverage

Calculated AI TAM
$30.0T
25.0% of Global $120T GDP

vs US GDP ($28.7T)

1.05x

104.5% of US Economy

vs China GDP ($18.5T)

1.62x

162.2% of China Economy

Cost / Knowledge Worker

$27,273

Across 1.1B Global Workers

Macroeconomic Scale Comparison ($ Trillions) D3 Dynamic Proportional Slices

Economic Sanity & Feasibility Diagnostics

White-Collar Labor Captured: $20.52T / yr Based on $54T baseline knowledge payroll

Enterprise Software Absorbed: $1.98T / yr Global enterprise tech allocation

Global Corporate Profit Share: ~187% of S&P 500 Equivalent to absorbing 2.2x global enterprise margins

Required Compute CapEx: $3.75T / yr Estimated at 12.5% revenue compute buildout

Trace the fixed labor buckets and normalized multiplier

Your browser does not support video.

Read the explanation

This explainer traces a saved hypothetical market model, not current economic statistics or an endorsement of its headline. The source fixes global payroll at fifty-four trillion dollars. Its default thirty-eight percent substitution gives twenty point five two trillion. A hypothetical twenty percent setting gives ten point eight trillion. Both bars use twenty pixels per trillion dollars. This captures the slider arithmetic without proving that substitution is achievable. The default labor term twenty point five two, software term one point nine eight, and new-demand term seven point five sum to thirty trillion dollars. The software term is eleven trillion multiplied by eighteen percent. Raising new demand from seven point five to ten gives a thirty-two point five trillion base with the other controls unchanged. Both bars use fifteen pixels per trillion dollars. The model adds these buckets without a separate overlap correction. The multiplier slider is divided by one point two five before scaling the bucket sum. Consequently a displayed one point two five gives an effective factor of one, leaving thirty trillion unchanged. A slider value two point five gives an effective factor of two and sixty trillion. Both bars use eight pixels per trillion dollars. This is the precise source convention, not direct multiplication by the displayed slider value. The simulator uses fixed GDP and worker benchmarks; they are not verified current facts.