What India’s Forest Rights Act Can Teach Us About Data Centers - Tech Policy Press

A hyperscale data center is a large, expandable campus filled with computer servers. It supports major digital services and intensive workloads such as artificial intelligence. The article does not define the term, but it presents data centers as projects that claim land and resources. That makes their footprint a governance issue, not only an engineering issue.

The servers require large buildings and supporting infrastructure, so the campus needs substantial land. They also run continuously and consume significant electricity. Cooling systems commonly use water or other cooling methods to remove heat. These needs connect digital services to physical resources and nearby communities.

The proposed Vizag campus illustrates this connection. Its footprint grew from 480 acres to 601.4 acres, while approval papers show 2.51GW of capacity. The article links such projects to socio-ecological costs and local dissent. It argues that approval should consider cumulative impacts, land claims, and community consultation.

The proposed Google-Adani data center campus near Visakhapatnam began with a 480-acre allotment in 2025. It was described as a $15 billion project designed to build a 1GW data center campus. The project’s stated scale has since changed substantially.

The allotment now covers 601.4 acres across three locations. Tarluvada accounts for 266.6 acres, Adavivaram for 160 acres, and Rambilli for 174.8 acres. Clearance papers show approval for 2.51GW of capacity. The larger figures indicate that both land use and computing capacity expanded.

The scale matters because the sites are near sensitive local resources. The project encroaches upon the eco-sensitive zone of Kambalakonda sanctuary and lies close to Mudasarlova reservoir. Despite these concerns, environmental clearance was granted in nine days. Protests have intensified since then, making the project a major governance and land-rights dispute.

Visakhapatnam is attractive for this project partly because the state made substantial land available. The article does not provide a full list of location advantages or identify specific power infrastructure. It does show that the proposed campus occupies a large area near important ecological features, making location central to the debate.

The campus spans Tarluvada, Adavivaram, and Rambilli. It encroaches upon the eco-sensitive zone of Kambalakonda sanctuary and lies close to Mudasarlova reservoir. Data centers also require continuous electricity and cooling. The article describes data centers broadly as claims on land and resources, so these local resources are central to the project’s footprint.

The location therefore creates competing claims. The campus would use land allocated to communities while operating beside a sanctuary and reservoir. Clearance came in nine days, despite these circumstances. The article’s wider argument is that development policy must examine place-based rights and cumulative impacts, rather than treating each site as an isolated technical project.

When a data center uses land and resources at this scale, nearby communities may lose control over land and livelihoods. Ecosystems may also face pressure from construction and resource use. The article frames this as a governance problem because communities bear the real socio-ecological costs, while promised benefits from an AI future may not reach them.

In Vizag, 200 of the original 480 acres belonged to around 50 Dalit farmers in Tarluvada. The wider project now covers 601.4 acres across three sites. It also encroaches upon Kambalakonda sanctuary’s eco-sensitive zone and lies near Mudasarlova reservoir. These facts show how land, water, and ecological concerns overlap.

The article does not quantify electricity or freshwater impacts. It does report intensified protests and concerns about extraction of resources from communities. Without recorded claims and meaningful consultation, dominant groups or acquisition processes can pressure less powerful landholders. The proposed response is cumulative assessment, formal recognition of rights, and local deliberation before approval.

The Forest Rights Act, or FRA, is India’s 2006 law recognizing land and resource rights of forest-dependent communities. Its preamble identifies people who have lived in forests for generations but whose rights were never recorded. The Act matters because it addresses historical exclusion and gives communities a formal way to establish customary rights.

Under Section 3(1)(a), people may claim individual or collective forest rights over land they occupy and cultivate. Claims begin in the gram sabha, rather than the revenue department. Oral testimony and customary knowledge can be accepted as evidence. This process recognizes that official records are not the only proof of longstanding use.

The Act applies to forest land, but the article sees its method as useful beyond forests. For data centers, governments could first record individual and collective land claims. They could then negotiate with actual right-holders, instead of relying only on administrative records or dominant groups. This would make land allocation more transparent and accountable.

India’s environmental-clearance system assesses projects by categories. The article argues that separating a single large campus into three site-level applications obscured its combined scale. This matters because the project’s total land use and capacity create impacts that may not appear when each location is examined alone.

In Vizag, the three sites were processed separately under Category B2. The Environmental Impact Assessment notification of 2006 lacked a specific data-center category. As a result, officials did not assess the entire campus as one Category A project. The Notification does not require public consultation for B2 projects.

No public hearing was therefore conducted. The article calls this kind of fragmentation an obfuscation that can reduce scrutiny. It proposes cumulative assessments and an amendment to the EIA notification. Those changes would make the full project visible and create a stronger basis for public consultation before approval.

The FRA’s central lesson is procedural as well as legal. Communities should be able to identify and collectively claim land and resource rights before a major project is approved. Local institutions should lead this process because they hold knowledge about customary use and affected livelihoods.

Applied to data centers, governments could begin with the gram sabha or another local institution. Individual and collective claims would be recorded before land allotment. Oral testimony and customary knowledge could help establish those claims. Structured negotiations would then involve actual right-holders, rather than relying only on revenue records or dominant groups.

The model should also include public consultation for projects above a defined threshold. Cumulative assessments would combine all related sites and examine their shared effects. This approach could prevent Category B2 fragmentation, expose the project’s full footprint, and give communities a meaningful role in decisions affecting their land and resources.