India is rapidly positioning itself as a major destination for artificial intelligence infrastructure, with large technology companies investing billions of dollars in data centres. But as these facilities expand, a difficult question is gaining attention:
India is rapidly positioning itself as a major destination for artificial intelligence infrastructure, with large technology companies investing billions of dollars in data centres. But as these facilities expand, a difficult question is gaining attention: what should local communities receive in return for hosting infrastructure that consumes land, water and enormous amounts of electricity?
The debate is becoming particularly important as governments compete to attract hyperscale data centres through tax incentives, subsidised utilities, land concessions and faster regulatory approvals.
The issue is no longer simply whether India should build more data centres. The bigger question is whether the economic benefits of the AI infrastructure boom will be shared fairly with the communities that bear its environmental and social costs.
Google’s Major AI Investment in Andhra Pradesh
In October 2025, Google announced plans for an AI hub in Visakhapatnam, Andhra Pradesh. The company committed to investing $15 billion between 2026 and 2030, making it Google's largest investment in India. The project has received substantial support from the state government. Around 600 acres of land were allocated, alongside incentives estimated at nearly $2.6 billion (₹22,002 crore).
The incentive package includes a 25% discount on land value, a complete stamp-duty exemption, power tariff subsidies for 15 years, electricity bill exemptions for 15 years and a 25% water-tariff subsidy for 10 years.
However, the scale of the project has also triggered concerns among residents and activists, particularly around land acquisition, water consumption, electricity demand and the project's broader environmental footprint.
Why Governments Are Competing for Data Centre Investments
Data centres are increasingly viewed as strategic infrastructure because they provide the computing capacity required for artificial intelligence, cloud computing, digital platforms and enterprise technology.
India's data centre capacity is currently around 1.5 GW and is projected to reach approximately 5 GW by 2030. The country has also seen announcements of roughly $60–70 billion in data centre investments over the next five years.
Governments in other emerging and developed markets are following a similar approach.
Chile, for example, provides import-duty exemptions on servers and technology equipment for qualifying data centre investments above a specified threshold. Malaysia has also introduced tax incentives for eligible digital projects.
The competition is understandable. Governments want foreign investment, technology infrastructure, employment and stronger positions in the global AI economy.
But generous incentives create another question: how much public value is being exchanged for private infrastructure investment?
The Hidden Cost Behind the AI Infrastructure Race
Data centres may generate economic activity, but their benefits and costs are not necessarily distributed equally.
These facilities require substantial quantities of electricity, water and land. The communities living around them can therefore experience environmental and economic impacts even when many of the financial benefits flow to technology companies, infrastructure providers and global supply chains.
This becomes particularly complicated when data centres are owned or operated by foreign companies.
AI infrastructure can contribute to a country's computing capacity without necessarily creating genuine domestic control over the technology ecosystem. Compute sovereignty depends not only on how much infrastructure is located within a country's borders, but also on who owns the facilities and where the underlying chips come from. That distinction matters because dependence on foreign-owned infrastructure could create vulnerabilities around pricing, access and geopolitical pressure.
Electricity Is Becoming a Strategic AI Resource
Electricity is one of the most important inputs for large-scale data centres because AI computing requires reliable, uninterrupted power.
Technology companies are increasingly turning to renewable-energy agreements to meet their energy requirements. Long-term power purchase agreements between data centre operators and renewable-energy producers are becoming an important part of the infrastructure model.
The trend also highlights a growing concentration of clean-energy procurement among major technology companies.
The source notes that Microsoft, Google, Amazon and Meta accounted for 49% of global corporate clean-energy PPAs, illustrating how large technology companies are becoming increasingly influential in renewable-energy markets.
In India's case, the Google-Adani AI hub is connected to an existing renewable-energy initiative involving a power purchase agreement associated with the Khavda project in Gujarat.
The Jobs Question: Big Promises, Smaller Numbers
Employment is often one of the strongest arguments used to justify large infrastructure projects.
The Google-Adani AI hub in Andhra Pradesh is a useful example of why projected employment figures need to be examined carefully.
Of the 600 acres allocated to the project, approximately 200 acres reportedly belonged to Dalit farmers and other cash-crop cultivators. These families had occupied and cultivated the land for generations, making it an important source of long-term livelihood and financial security.
Government messaging around the project has highlighted its potential to create employment and stimulate the regional economy. Andhra Pradesh IT Minister Nara Lokesh has said the facility could generate close to 2 lakh jobs.
However, official project documents cited in the source indicate a much smaller employment figure: approximately 2,415 jobs across construction and operational phases, including permanent and temporary positions.
That difference is significant.
It does not necessarily mean the project will produce no wider economic benefits. Construction spending, supporting services and indirect business activity can create additional opportunities. But headline job estimates should not automatically be treated as equivalent to permanent local employment.
Water Consumption Adds to Environmental Concerns
Water is another major issue surrounding large data centre developments.
According to environmental clearances cited in the source, the facility's estimated water requirement is approximately 808 kilolitres per day.
That is a substantial demand, particularly in a region where activists have raised concerns about groundwater levels, uneven rainfall and existing drinking-water challenges.
The project is also located close to the catchment area of Mudasarlova reservoir, an important water source for Visakhapatnam.
Electricity demand is equally substantial. The facility is expected to require approximately 2,068 MW of power, highlighting the scale of infrastructure required to support modern AI computing.
These figures explain why communities are asking whether data centre expansion can be pursued without placing additional pressure on essential local resources.
Why Traditional Corporate Social Responsibility May Not Be Enough
Technology companies have increasingly responded to environmental concerns through initiatives such as tree planting, watershed management and other corporate social responsibility programmes.
But isolated CSR projects do not necessarily address the underlying question of resource ownership and long-term economic benefit.
For host communities, planting trees or funding local programmes may be useful, but these measures are different from receiving a sustained share of the economic value generated by infrastructure operating on their land and using local resources.
This distinction is becoming increasingly important as data centres become larger and more strategically valuable.
Could Community Benefit Agreements Offer a Solution?
One possible approach is the use of Community Benefit Agreements (CBAs).
CBAs are agreements between developers and host communities that establish specific benefits in exchange for community support for a project. Depending on the arrangement, these benefits can include local employment, skills training, infrastructure investment and funding for environmental programmes.
The concept has gained attention in the United States, where some communities have negotiated directly with data centre developers.
A cited example comes from Cedar Rapids, Iowa, where an agreement involving Google and QTS was connected to a 20-year tax exemption. The arrangement included employment requirements and financial contributions from the companies toward the local community.
Such agreements can provide greater clarity than informal promises because the benefits are defined before or during project development.
From One-Time Compensation to Long-Term Benefit Sharing
However, community agreements may not go far enough.
A more ambitious model is long-term benefit sharing, where communities continue to receive financial benefits throughout the operating life of an AI infrastructure project.
The basic idea is straightforward: if a data centre continuously uses local land, water and energy while generating economic value from AI services, part of that value could be returned to the communities hosting the infrastructure.
Instead of determining compensation entirely when land is acquired or a project begins, the financial contribution could evolve as the facility expands and its actual resource consumption and economic output become clearer.
This would create a closer connection between the value generated by AI infrastructure and the communities supporting it.
Measuring the Value of AI Infrastructure
A major challenge is determining how much should actually be shared.
The amount cannot simply be based on the size of the building or the initial investment. Policymakers may need to consider factors such as land use, water consumption, electricity consumption, computing capacity, AI chips deployed and the economic activity generated by the facility.
The source also points to emerging verification technologies that could make AI computing activity more measurable. Hardware-based verification and confidential-computing capabilities could potentially help track computing infrastructure more accurately.
Developing reliable measurement and verification mechanisms could therefore become an important part of future AI governance.
India Needs to Decide Who Benefits from the AI Boom
The debate surrounding data centres is ultimately bigger than one project or one company.
India wants to become a major global AI and digital infrastructure hub. That ambition will require massive investments in computing capacity, electricity generation, renewable energy, telecommunications and supporting infrastructure.
But rapid expansion should not mean that local communities are treated simply as the location where infrastructure happens to be built.
The challenge is to create a model where investment remains attractive while communities receive meaningful and continuing economic benefits.
That could involve a combination of community benefit agreements, local employment commitments, skills development, environmental safeguards, transparent resource accounting and long-term benefit-sharing mechanisms.
The Road Ahead for India’s Data Centre Market
India's data centre market is entering a period of rapid expansion, driven by artificial intelligence, cloud computing and increasing digital demand.
The opportunity is enormous. But so are the infrastructure requirements.
If governments provide land, tax benefits, subsidised utilities and regulatory support to attract AI infrastructure, they also need mechanisms to ensure that the resulting economic value does not remain concentrated among a small number of companies.
The goal should not be to discourage data centre investment. It should be to make the investment model more balanced.
As AI becomes increasingly dependent on physical infrastructure, the communities providing the land, water, electricity and local ecosystem should have a meaningful stake in the value that infrastructure creates.
For India, getting this balance right could determine whether its AI infrastructure boom becomes merely an investment success or a genuinely broad-based economic opportunity
References