The Era of Industrialization of AI — and India’s Strategic Opportunity

Over the last few days, the top five US technology companies have announced a cumulative capital expenditure of about $670 billion for 2026 alone(!). This figure exceeds the annual GDP of nations like Switzerland or Poland. 

But, this is not just a large number. It's a structural break.  For two decades, technology was defined by its asset-light character. Code scaled faster than concrete. Software displaced steel. Platforms replaced plant. 

This paradigm is now reversing.

Artificial Intelligence at frontier scale is not asset-light. It is infrastructure-dependent. It demands hyperscale data centres, gigawatts of power, advanced cooling systems, semiconductor supply chains, and physical redundancy engineered for near-zero downtime. The firms once celebrated for minimal capital intensity are now building industrial-scale infrastructure.

We are entering the era of Industrialisation of AI — where tech giants are no longer just coding—they are building the physical crust of technology, to house and deploy AI at industrial scale.

And when technology becomes inseparable from infrastructure, geography begins to matter, again.

The Geography of Industrial AI

Most of this unprecedented CapEx will remain in the United States. That is natural. The US possesses deep capital markets, semiconductor ecosystems, and energy flexibility.

The more consequential question is: where will the remaining capital go?

In an infrastructure-bound AI world, investment will cluster in jurisdictions that satisfy three conditions:

1. Energy credibility — increasingly clean and scalable.

2. Market depth — capable of monetising compute locally.

3. Integration capacity — the ability to absorb AI into real industry.

India, if proactive, can credibly position itself on all three dimensions.

Beyond Land and Labour: India’s Structural Advantages

India’s opportunity is not to market itself merely as land for data centres or as a low-cost manpower base. That would replicate the logic of the asset-light era.

Industrial AI demands a different proposition.

Policy Signal

The Union Budget’s long tax holiday for data centre investments signals recognition that AI infrastructure is strategic infrastructure. In the industrial AI era, server farms are as consequential as ports and highways.

Policy clarity lowers risk. Infrastructure capital follows credibility.

Energy Compatibility

AI compute consumes enormous power. Hyperscalers are increasingly bound by carbon accounting commitments, not just carbon offsetting.

India’s accelerating renewable energy generation capacity creates compatibility. A jurisdiction capable of pairing AI infrastructure with scalable green power becomes structurally attractive.

In a world measured in megawatts, clean energy is industrial leverage.

Downstream AI Absorption Capacity

India’s most underappreciated strength lies here.

The country may not dominate AI chip fabrication or foundational model development. But it possesses large-scale AI integration capability.

Indian technology services firms already tailor complex AI systems for global enterprises. Trust, system integration depth, and operational customisation are established competencies.

Domestically, Reliance Industries Ltd (RIL) is experimenting with licensing-in foundational AI models/softwares, housing them within domestic infrastructure, and embedding them directly into refining, manufacturing, logistics, and retail networks.

This is not superficial digitisation. It is embodied intelligence — AI inserted into physical operations, supply chains, and factory floors.

Crucially, this model is replicable. There is nothing inherently unique about one conglomerate attempting it. Any diversified industrial group with manufacturing, energy, and distribution exposure can pursue similar integration.

A New Industrial Revolution?

If RIL successfully embeds AI across its operations and demonstrably increases productivity, margins, and capital efficiency, the effect will not remain isolated.

Markets observe. Capital reallocates.

If market competition begins rewarding AI-driven efficiency, if capital markets start pricing integration capability into valuations, and if state incentives continue attracting AI infrastructure investment, pressure on the next 100–200 conglomerates will intensify. At that point, AI adoption becomes defensive rather than experimental.

These firms account for a disproportionate share of formal output, capital expenditure, and organised employment. Efficiency gains at this scale ripple through suppliers, contractors, and MSMEs.

Such adoption would also create sustained domestic compute demand, strengthening the economic case for local AI infrastructure deployment.

Therefore, if the next 100–200 conglomerates replicate the RIL model — licensing-in global foundational AI softwares, housing them domestically, embedding them into factories and logistics networks — the result would exceed incremental productivity gains.

It would be an industrial inflection.

Not a services-led tech wave.

But a broad-based industrial acceleration — powered by global AI architectures, physically hosted and operationalised within India’s industrial core.

In other words, it would be new industrial revolution — driven by global AI softwares, housed in India.

Hosting vs Participating

Therefore, I reiterate that India must resist the temptation to position itself merely as a hosting zone for foreign infrastructure.

In the asset-light era, advantage flowed to platforms. In the industrial AI era, advantage flows to ecosystems capable of building, applying, and monetising intelligence.

The strategic proposition should not be:

Land and power for servers,

Labour for integration.

It should be:

A fast-growing market,

Sustainability-aligned energy expansion,

Proven integration depth,

Industrial scale capable of absorbing AI into real production.

The Structural Choice

In the nineteenth century, industrial power followed coal.

In the twentieth, it followed oil.

In the twenty-first, it will increasingly follow compute density — and the ecosystems capable of applying that compute to real industry.

The $670 billion CapEx wave signals that the industrialisation of intelligence has begun.

India’s choice is whether the country positions itself as a peripheral data source in someone else’s AI architecture — or as an active node in the emerging industrial AI order.

The window will not remain open indefinitely. Infrastructure-dependent technologies entrench in early movers.

Industrial AI has arrived.

The question is whether India merely hosts it — or helps shape it.