Adani's Next Frontier: Building India's AI Infrastructure Supply Chain Through BHAVYA

Introduction

In his annual letter to shareholders, Mr. Gautam Adani positioned “Infrastructure + Intelligence” as the defining twin engines of the Adani Group’s future growth. He famously noted, “Before AI can think, energy must flow,” underscoring that reliable, large-scale power infrastructure is foundational to the AI era. On capacity, AdaniConneX aims to build a 2 GW data centre platform by 2030, as part of a larger $100 billion commitment to develop renewable-powered, hyperscale AI data centre capacity targeting 5 GW by 2035. This ambition is backed by the Group’s record FY26 financial performance and capital expenditure, including record renewable energy capacity expansion -- all aligned to support India’s emergence as an AI infrastructure powerhouse.

This is the right direction.

India will require enormous investments in renewable energy, transmission networks, and data centres if it hopes to emerge as a significant player in the global AI economy. Recent policy measures, including the Union Budget 2026's long-term incentives for hyperscale data-centre investments, reflect growing recognition of this reality.

However, there is a risk in viewing AI infrastructure solely through the lens of power generation and data-centre construction.

Energy is foundational. Data centres are foundational.

But between electricity generation and computing - lies an enormous industrial layer that receives comparatively little attention. This layer determines whether data centres are efficient or wasteful, reliable or unreliable, globally competitive or permanently dependent on imports.

Cooling systems.

Power-delivery infrastructure.

Optical networking equipment.

Water-management technologies.

Precision-fabricated modular systems.

Testing and validation capabilities.

Reliability engineering.

These are not peripheral industries. They are the systems that transform theoretical computing capacity into usable computing capacity.

This creates a strategic opportunity not only for India but also for the Adani Group itself.

If Adani intends to build a multi-gigawatt AI infrastructure platform, it will inevitably require vast quantities of these systems over the coming decade. Rather than sourcing them entirely from fragmented domestic suppliers or foreign vendors, the Group has an opportunity to help build the ecosystem itself.

This is where the Bharat Audyogik Vikas Yojana (BHAVYA) enters the picture.

By combining its future AI infrastructure demand with BHAVYA's industrial-park framework and partnerships with interested state governments, Adani could help create a new generation of specialised AI Infrastructure Supply Chain Parks dedicated to the components, systems, and capabilities that underpin the AI economy.

Such a strategy would do more than support Adani's own ambitions.

It would help deepen India's industrial capabilities in some of the most important sectors of the intelligence age.

Beyond Energy and Data Centres: The Missing Middle Layer

The current AI race is often described as a competition for chips, compute, and electricity.

There is truth in this characterization.

Advanced semiconductors remain essential. So do hyperscale data centres and abundant energy supplies. Yet focusing exclusively on these areas can obscure a critical reality.

Modern AI infrastructure is not simply a collection of servers connected to the electrical grid.

It is an extraordinarily complex industrial system.

The most advanced GPU in the world cannot perform optimally if cooling systems are inadequate. Cheap renewable power becomes less valuable if transmission equipment introduces reliability issues. Massive computing clusters lose effectiveness if networking systems become bottlenecks.

This is why leading hyperscale operators obsess over metrics such as Power Usage Effectiveness (PUE), Water Usage Effectiveness (WUE), uptime, redundancy, and total cost of ownership.

The competitiveness of a data centre increasingly depends not just on the chips it contains but on the ecosystem surrounding those chips.

Consider some of the critical components involved.

Advanced cooling systems must manage racks consuming 50, 100, or even more kilowatts of power. Traditional air cooling is increasingly being supplemented by liquid cooling, immersion cooling, advanced heat exchangers, and sophisticated thermal-management systems.

Power-delivery infrastructure must ensure stable operation despite fluctuations in energy supply. This includes transformers, switchgear, power-distribution units, backup systems, batteries, and intelligent power-management solutions.

Networking infrastructure requires dense optical-fibre systems, low-latency connectivity hardware, and increasingly sophisticated data-transfer technologies.

Water systems must manage treatment, recycling, cooling loops, and sustainability requirements while minimizing environmental impact.

Beyond these visible components lies an entire ecosystem of testing laboratories, calibration facilities, predictive-maintenance services, reliability engineering firms, and systems integrators.

Together, these industries form what might be described as the middle layer of AI infrastructure.

In many advanced economies, this layer is deep, specialised, and highly competitive.

In India, significant gaps remain.

Many high-performance components continue to be imported. Domestic capabilities often exist but remain fragmented. Testing and certification infrastructure is limited. Supply chains lack depth in several critical areas.

The consequence is straightforward.

India can build data centres while still depending heavily on foreign ecosystems for many of the systems that determine performance and reliability.

That dependence may become increasingly costly as AI infrastructure scales.

Why Adani Is Uniquely Positioned to Address This Gap

Most industrial-policy discussions begin with the question of how to create demand.

This proposal begins from the opposite direction.

The demand already exists.

The Adani Group's AI ambitions imply future requirements for cooling systems, electrical infrastructure, networking equipment, water technologies, modular systems, testing services, and countless supporting components.

The challenge is not generating demand.

The challenge is organizing supply around that demand.

This distinction is important because industrial ecosystems rarely emerge from infrastructure alone.

History suggests that successful industrial clusters are usually anchored by large, predictable buyers.

Automotive ecosystems grew around major vehicle manufacturers.

Aerospace ecosystems developed around aircraft producers and defence procurement programmes.

Electronics ecosystems expanded around anchor firms capable of generating long-term demand.

Supplier networks, in turn, emerged around these anchors.

The sequence is typically:

Anchor demand → Supplier ecosystem → Capability accumulation → Export competitiveness.

The Adani Group is unusually positioned to replicate aspects of this dynamic.

Few Indian firms possess comparable strengths across energy generation, transmission, logistics, ports, airports, industrial infrastructure, and now AI infrastructure.

Most companies operate within a single layer of the value chain.

Adani increasingly operates across multiple layers simultaneously.

Its renewable-energy assets provide power.

Its transmission networks move power.

Its ports and logistics systems facilitate industrial activity.

Its data-centre ambitions create demand for AI infrastructure components.

This breadth gives the Group a systems-level perspective that relatively few organisations possess.

More importantly, it creates the possibility of transforming infrastructure projects into industrial platforms.

Instead of merely constructing data centres, the Group can help build the industrial ecosystem that supports them.

BHAVYA as a Vehicle for Industrial Deepening

The announcement of the Bharat Audyogik Vikas Yojana, on 18 March, offers a potentially powerful mechanism for pursuing this objective.

The scheme's vision is ambitious: 100 plug-and-play industrial parks developed across India, supported by substantial central-government funding and designed to accelerate industrial investment.

The programme addresses a real challenge. Land acquisition, infrastructure creation, utility provision, and approval delays have long constrained industrial development.

Yet the ultimate success of BHAVYA will not be determined solely by how many parks are built.

It will depend on what kind of industrial ecosystems emerge within those parks.

In an earlier article, published on 20 March, I argued that India's next phase of industrial development requires BHAVYA to evolve beyond a generic plug-and-play model. Five upgrades are particularly important.

1. Upstream and deep value chain focus — Targeting foundational components rather than final assembly.

2. Coordinated capitalism and portfolio investment — Attracting patient capital and global JVs committed to multi-year ecosystem building.

3. Demand aggregation via procurement — Using anchor demand from large players and government projects.

4. Labor infrastructure — On-site hostels, skill intermediaries, and apprenticeship pipelines.

5. Water as the fourth utility — Integrated treatment, recycling, storage, and potentially tradable treated water systems.

Taken together, these five elements would transform BHAVYA from a land-development scheme into a platform for industrial deepening.

The AI infrastructure supply chain may be one of the best sectors through which to test this model.

The Proposal: AI Infrastructure Supply Chain Parks

The logical next step is the creation of a small number of specialised AI Infrastructure Supply Chain Parks under the broader BHAVYA framework.

Rather than attempting to spread resources thinly across numerous sectors, these parks would focus on the industrial ecosystem supporting AI infrastructure.

A practical starting point could involve four to six parks across strategically selected states.

Potential locations include Gujarat, where Adani's existing ecosystem around Mundra and Khavda offers significant advantages; Andhra Pradesh, where data-centre and digital-infrastructure investments are accelerating; and additional states such as Maharashtra or Tamil Nadu that possess strong industrial bases and supportive policy environments.

Each park could develop specialised sub-clusters.

One cluster might focus on advanced thermal-management technologies, including liquid cooling, immersion cooling, heat exchangers, and related systems.

Another could concentrate on electrical infrastructure such as transformers, switchgear, power-distribution equipment, batteries, and backup systems.

A third cluster could focus on optical networking, fibre systems, and connectivity hardware.

Additional clusters could support precision fabrication, modular infrastructure, water technologies, predictive maintenance, reliability engineering, and industrial automation.

The objective is not merely manufacturing.

It is ecosystem formation.

To support this goal, the parks should include:

1. Quality and Reliability Infrastructure: Common testing laboratories, calibration facilities, certification centres, and shared research infrastructure reduce costs for individual firms while improving quality and reliability standards across the ecosystem.

2. Capability Infrastructure: Embedded capability pipelines should form another pillar of the model. Training centres linked to local ITIs, polytechnics, colleges, and industry partners can create pipelines of technicians, supervisors, and engineers. Apprenticeships should be integrated into park operations from the outset.

3. Technology Acquisition: The parks should actively encourage partnerships with global companies possessing expertise in cooling systems, optical technologies, power infrastructure, and industrial equipment. Joint ventures can accelerate capability development while providing Indian enterprises access to advanced technologies and possibly international markets.

In effect, these parks would function not simply as industrial estates but as specialised capability ecosystems.

From Projects to Platforms

At a deeper level, this proposal reflects a broader shift in how India might think about industrial development.

Historically, much attention has focused on standalone projects. Such projects matter. But projects alone do not necessarily create industrial depth.

Industrial depth emerges when projects evolve into platforms.

Platforms create ecosystems.

Ecosystems create suppliers.

Suppliers create capabilities.

Capabilities create exports.

The Adani Group's AI ambitions create an opportunity to pursue precisely this progression.

Consider how various elements could reinforce one another.

Renewable-energy projects provide power.

Transmission infrastructure delivers reliability.

Ports and logistics networks facilitate industrial activity.

AI data centres generate demand.

Industrial parks develop suppliers.

Training centres cultivate skills.

Testing facilities improve quality.

Over time, these components begin to function as an integrated industrial platform rather than isolated projects.

This is perhaps the most interesting aspect of the proposal.

The goal is not merely to reduce imports or lower procurement costs.

The goal is to create self-reinforcing ecosystems capable of generating new capabilities.

Broader National Impact

If successful, the benefits would extend well beyond data centres.

The capabilities required for advanced cooling systems, high-performance electrical equipment, optical networking, testing, metrology, and reliability engineering are relevant across numerous sectors.

Aerospace.

Defence.

Medical devices.

Semiconductor packaging.

Electric vehicles.

Industrial automation.

Advanced manufacturing.

In this sense, AI infrastructure may serve as a catalyst for broader industrial development.

The country would not simply acquire more factories.

It would acquire deeper industrial capabilities.

This distinction matters.

Nations often succeed in attracting manufacturing activity without necessarily developing strong supporting ecosystems. The more difficult challenge is cultivating the precision, quality, reliability, and engineering capabilities that enable industries to move up the value chain.

The proposed parks could contribute to what might be called a precision-and-reliability ecosystem—an industrial environment characterised by advanced manufacturing, testing, validation, metrology, quality systems, systems integration, and reliability engineering.

Such capabilities are difficult to build but highly valuable once established.

They create resilience.

They improve competitiveness.

They support innovation.

And they generate spillovers across multiple sectors.

Conclusion

The Adani Group has already demonstrated its ability to build infrastructure at scale.

Ports, airports, transmission networks, renewable-energy projects, logistics systems, and industrial facilities have all contributed to the Group's rise as one of India's most significant infrastructure developers.

Its growing AI ambitions suggest the beginning of a new chapter.

The question is whether the Group's next strategic leap will be limited to building data centres—or whether it will extend to building the industrial ecosystem that data centres depend upon.

By leveraging its future AI infrastructure demand, partnering with state governments, and utilizing the BHAVYA framework, Adani has an opportunity to help create specialised AI Infrastructure Supply Chain Parks dedicated to cooling technologies, power systems, networking equipment, water solutions, reliability engineering, and other critical capabilities.

Such a strategy would strengthen the Group's own AI plans by improving supply-chain resilience, reducing costs, and accelerating ecosystem development.

More importantly, it could help India build the industrial depth required to compete in the intelligence age.

The most valuable outcome may not be the data centres themselves. It may be the new generation of suppliers, engineers, technicians, technologies, and capabilities that emerge around them.

In the long run, that industrial depth may prove to be the most important infrastructure of all.