← Back to Home Contact
Track 01

Infrastructure & Ownership

This research track studies who owns the rails of the AI economy: compute, chips, energy, cloud concentration, model infrastructure, and strategic data access. The central question is blunt: who controls the infrastructure that intelligence runs on — and who becomes dependent on it?

Track overview

Most people talk about AI as software. That is incomplete. AI is also a physical and geopolitical system shaped by semiconductor bottlenecks, energy demand, data center concentration, cloud dependence, and capital intensity. This track examines the ownership structures beneath AI and how those structures determine power, dependency, and long-run value capture.

Why this matters

Whoever controls compute, energy, cloud distribution, and the enabling infrastructure of AI will shape which firms scale, which countries remain sovereign, and which institutions become permanently downstream from someone else’s stack.

Core questions

Who controls frontier compute capacity, and how concentrated is that control becoming?
How will semiconductor supply chains shape the geography of AI power?
Will energy constraints become the real bottleneck for advanced AI deployment?
What does sovereign AI capability actually require beyond slogans and national strategy PDFs?
How much of the AI economy will be owned by hyperscalers versus independent infrastructure players?
What forms of infrastructure ownership generate durable leverage in the AI era?

Flagship briefings

Flagship briefing

The Infrastructure Layer of the AI Economy

A framing paper on why compute, chips, energy, and cloud concentration matter more than most AI commentary admits.

Flagship briefing

Compute, Energy, and the Bottlenecks of Intelligence

An analysis of how power generation, grid readiness, and data center economics constrain AI expansion.

Flagship briefing

Sovereign AI and Strategic Dependency

A study of what nations must own, access, or negotiate to avoid becoming permanently dependent consumers of foreign intelligence infrastructure.

Research notes

GPU Supply Chains and Strategic Fragility

A short memo on where control, fragility, and chokepoints really sit in the AI hardware chain.

Cloud Concentration as Economic Power

Why the hyperscaler layer may be one of the most underestimated sources of leverage in AI.

The Return of Industrial Policy Through AI

How national AI ambition is quietly pulling countries back toward industrial strategy.

Energy as the Hidden Variable in AI Scaling

Why electricity and grid access may matter as much as model quality.

Infrastructure Ownership vs Model Ownership

Who has more power: those who build the models or those who own the rails they run on?

The Geopolitics of Compute Access

A field note on export controls, alliances, and the emerging politics of compute scarcity.

Data & indexes

Global Compute Capacity Tracker

A developing index mapping where strategic AI compute is concentrated.

AI Energy Readiness Index

A framework for comparing whether regions can realistically support AI infrastructure growth.

Sovereign AI Infrastructure Map

A structured view of which countries are building real capacity versus narrative-only capability.

Want to discuss this track?

For briefings, collaborations, speaking, or strategic conversations on AI infrastructure, sovereign capability, or ownership concentration, contact the Institute directly.

Get in touch