Data centres are the backbone of the AI economy, but they put pressure on energy and water. Critically analyse India's data centre expansion.

Data centres are buildings full of servers that run cloud and AI services. India's capacity grew from 375 MW in 2020 to 1.57 GW (Aug 2026), and it is projected to reach about 8 GW by 2030 [1]. The expansion is strategically necessary. However, rules on its energy and water use lag behind its growth.

Strengths: why the expansion matters

• AI and cloud backbone: the IndiaAI Mission, e-governance and Digital Public Infrastructure all need compute located in India.
• Investment: about USD 70 billion of investment is underway. Infrastructure status since Budget 2022–23 makes credit easier to get [1].
• Global hub ambition: Budget 2026–27 proposes a tax holiday till 2047 for foreign cloud providers that serve global clients from MeitY-notified Indian data centres [2].
• Data sovereignty: storing data in India supports the DPDP Act, 2023 framework and protects citizens' informational privacy (Puttaswamy, Art. 21).

Concerns: energy, water and governance

• Power demand: the CEA estimates data centre demand at about 17 GW by 2031–32 [1]. This strains the grid and risks locking in coal-based power.
• Global warning: the IMF finds that if grids cannot keep up, US electricity prices could rise 8.6% and emissions 5.5% under current policies [3].
• Water blind spot: the government says it has received no concerns about water use by data centres [4]. But no complaints is not the same as measurement, and data centres cluster in water-stressed metros.
• Regulatory gap: groundwater use comes under the general Jal Shakti guidelines (2020, amended 2023), not rules written for data centres. BIS standards on PUE and WUE exist, but publishing these numbers is not compulsory [5].
• Sovereignty paradox: foreign operators drawn in by tax incentives still answer to the laws of their home countries, and the most advanced chips are still imported.

Way forward

• Make annual PUE/WUE disclosure a condition for tax incentives [2][5].
• Build data centre load into transmission planning through the Green Energy Corridor and Green Energy Open Access Rules [5].
• Set water norms by location, and require low-water cooling (direct-to-chip, immersion) in water-short areas [4][5].

India's data centre push is an economic and strategic necessity, but it is growing faster than the rules on power and water. Linking incentives to green and water-saving conditions would fit SDGs 6, 7 and 9 and the Net Zero 2070 goal. That way data centres can become a sustainable backbone of India's AI economy.