India’s data center boom is leaving the people it displaces with nothing
Communities are being left behind as AI companies secure tax breaks, cheap land from willing governments.
India's data center boom is displacing communities, leaving them without options. Over the past two decades, internet usage has skyrocketed from 1.2% to more than 70% of the population. Yet, the digital revolution may be an extractive one, with profits and control concentrated among a few. Protests have erupted globally, from India to Malaysia to Chile, questioning the placement and purpose of data centers.
These facilities are projected to consume vast amounts of water and electricity, equal to that of 1.3 billion people and 650 million respectively, by 2030. The issue has been edited for brevity and clarity. In the U.S., calls have been made for governments to ban or restrict data centers, but in India, the government is viewing them as a developmental activity.
The government is partnering with data center companies without consulting local communities, a concerning practice given the significant land and water resources required. The conversation also highlights the potential job displacement due to AI, as even data creators could be replaced. Local communities, including landowners, farmers, and women's groups, have not been adequately consulted.
In Telangana, where Microsoft is building a data center, some acquired land is under dispute. The question arises: why are we allowing private entities to build sovereign infrastructure, and what are we gaining in return? Could we not develop hyperlocal, multi-distributed data centers serving local needs? While companies may claim they won't use groundwater, the real question is what they are offering in return for the environmental impact.
The promises of job creation and community benefits lack concrete guarantees. The existence of the data center itself is not even up for discussion.
Written by urgent.news from Rest of World Tech's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
