On June 5, 2026, AirTrunk — the Australian data center operator backed by Blackstone and Canada’s CPP Investments — announced it will invest $30 billion in India by 2030 to develop 5 gigawatts of new capacity. TechCrunch called it one of the largest commitments ever made to India’s digital infrastructure. Worth noting: AirTrunk only entered the Indian market this year, through its acquisition of Lumina CloudInfra.
The scale is the story. India has roughly 1.5GW of data center capacity today, and research firm Bernstein projects as much as 8GW by 2030. AirTrunk alone is betting 5GW — effectively underwriting the country’s entire growth curve as a company-level wager. CEO Robin Khuda met Prime Minister Narendra Modi ahead of the announcement, and Modi posted on X that the investment would “help strengthen India’s position as a global hub for cloud computing and artificial intelligence.”
What $30B Buys: A 3GW Raigad Campus and a 600MW Pipeline
The largest single disclosed project sits in Maharashtra. Chief Minister Devendra Fadnavis posted that the state exchanged a letter of intent for land allotment at the Raigad Pen Growth Center, where AirTrunk plans a 3GW data center campus with investment of roughly ₹2 trillion (about $21 billion). One campus, in other words, absorbs around two-thirds of the headline number.
AirTrunk’s existing Indian footprint is a development pipeline of roughly 600MW across Mumbai, Chennai, and Hyderabad. The company declined to say whether Raigad accounts for most of the 5GW, or whether additional projects in other states are planned. Mint, citing Reuters, reports the pipeline will span multiple states and support India’s push to become a global hub for AI and cloud investment.
Why India: Tax Breaks to 2047 and an 8GW Forecast
Two forces make the bet look rational. The first is policy: earlier in 2026, New Delhi offered foreign cloud providers tax exemptions through 2047 on overseas-sold services — provided the workloads run from data centers inside India. That is a rule written for exactly this kind of investment. The second is demand: Bernstein expects Indian capacity to grow from about 1.5GW to as much as 8GW by 2030, with AI inference as the primary driver.
Khuda’s stated investment thesis rests on three pillars: government support, a large pool of technical talent, and access to renewable energy. Those three map neatly onto India’s biggest advantage and its biggest risk.
The Field: Every Giant Is Betting
AirTrunk is not alone in eyeing India. The roster TechCrunch assembled shows nearly every major player has placed a bet there over the past year and a half:
- Amazon: an additional $35 billion by 2030, announced December 2025, on top of $75 billion in total planned investment
- Google: a $15 billion AI infrastructure hub, announced October 2025
- Microsoft: $17.5 billion by 2029
- OpenAI: tapped Tata for 100MW in February 2026, and is reportedly eyeing 1GW
- Reliance Industries: a $110 billion AI investment plan; Adani Group: a $100 billion pledge for AI data centers
Against that field, AirTrunk’s 5GW is the largest single-operator commitment to date — but it also means fighting equally well-funded rivals for land, grid capacity, and engineers.
Power Is the Bottleneck
Data centers consume electricity, water, and land, and the executives and analysts quoted on this boom keep pointing at power as the constraint. Deloitte estimates that Asia-Pacific data center build-outs could require tens of additional terawatt-hours of electricity by the end of the decade. Read in that light, Khuda’s inclusion of renewable energy access in the investment thesis looks less like ESG boilerplate and more like risk management: whether AirTrunk can secure stable, clean power will decide how much of the 5GW blueprint actually gets built.
That is the tension in the India story. Policy and demand are in place; the grid may not expand as fast as an 8GW ambition requires.
What It Means for Developers and Product Teams
Three practical effects. First, AI products serving South Asia gain a local inference option: 5GW of in-country capacity speaks directly to both latency and data residency. Second, the 2047 tax exemption makes “India as an export base” a formal option — global teams can put the country on the shortlist for regional inference hubs. Third, the capacity market is deepening: from hyperscalers to independent operators, there are more paths to large-scale inference capacity, and more room to negotiate. We argued in our 2026 opening outlook that compute access would shape product decisions this year — India is the latest and largest footnote to that trend.
Sources
- AirTrunk commits $30B to build 5GW of AI data centers in India — TechCrunch
- Blackstone-backed AirTrunk plans $30 billion India data centre investment by 2030 — Moneycontrol
- Blackstone, CPPIB-backed AirTrunk to invest over $30 bn in India’s data centres by 2030 — Mint
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.