Human Signal™
AI Infrastructure Viability Framework · Human Signal™
LITHOGRAPHY · ENERGY · ARBITRAGE · COOLING
A framework for understanding the physical constraints AI companies must address to maintain profitability.
Premise: Transitioning from infinite software margins to an era constrained by physical realities.
The Four Constraints
Four binding physical constraints. Each one a chokepoint. All four together determine whether an AI company survives or leaks value.
Constraint One
Control of semiconductor supply chain, especially photolithography equipment, is critical. Without direct access to silicon manufacturing, you are dependent on others' capacity.
Constraint Two
Electrical grid is the limiting factor. AI training/inference require massive power. Securing gigawatt-scale power contracts is essential for operational models.
Constraint Three
Retail electricity pricing is unsustainable for large-scale AI. Success requires arbitrage: stranded energy, flare gas, off-peak power to reduce compute costs.
Constraint Four
Thermodynamics is the ultimate constraint. High-performance computing generates enormous heat. Without adequate cooling infrastructure (water/thermal management), clusters cannot run. A fundamental solvency issue — not just a facilities problem.
The Message
Developed By
Chief Sensemaking Officer · Human Signal™ | Editor in Chief, The AI Governance Record
Dr. Floyd is an independent AI governance researcher and the architect of the L.E.A.C. Protocol and Presence Signaling Architecture (PSA v1.0) — frameworks for restoring human visibility, institutional signal, and operational muscle memory in AI-disrupted environments. He developed the LEAC Protocol from forensic market analysis of AI infrastructure investment patterns, identifying the physical layer constraints that governance frameworks consistently miss.