Key Takeaways:
India's data-centre market is set to expand more than fivefold over the next five years as AI inferencing demand drives a wave of infrastructure investment.
Brookfield Asset Management expects about 6.5 gigawatts of data-centre capacity for artificial intelligence to come online in India over the next five years, a bet that the country's AI inferencing market will rival established hubs in the US and Europe.
"AI inference demand in India will grow as enterprises deploy trained models at scale," Arpit Agrawal, managing partner and head of India and the Middle East for Brookfield's infrastructure group, said.
India currently has about 1.5 GW of installed data-centre capacity, almost all for non-AI workloads, Agrawal said. Brookfield expects non-AI capacity to rise to about 3 GW, with the 6.5 GW of AI-specific capacity representing incremental demand. The firm's Digital Connexion venture — a partnership with US-listed Digital Realty and Reliance Industries — has about 160 MW of capacity in India, of which 60 MW is operational and fully leased.
The forecast is part of Brookfield's broader India push. The New York-based alternative asset manager oversees more than $32 billion in Indian assets and aims to more than triple that to $100 billion within five years. Its energy platforms expect to add 4 GW to 4.5 GW of renewable capacity this year alone — power that will be critical to running energy-hungry AI data centres.
India's AI Infrastructure Gap
India's current data-centre stock of 1.5 GW is tiny relative to major markets. The US alone had more than 20 GW of hyperscale capacity as of 2025, according to Synergy Research Group. The 6.5 GW of new AI-specific capacity Brookfield forecasts would bring India closer to parity with markets like Singapore and Japan, which have been competing for AI data-centre investment.
The 6.5 GW estimate refers to India's wider data-centre market, not Brookfield's own planned capacity. AI inference — the process by which trained models generate answers, content or predictions — is expected to be the primary driver, as Indian enterprises across banking, e-commerce and logistics deploy AI applications at scale.
Power and Partnerships
Running 6.5 GW of AI data centres requires enormous electricity. Brookfield's energy platforms are positioned to supply some of that demand: the firm expects to add 4 GW to 4.5 GW of renewable capacity in India this year, Nawal Saini, managing partner in Brookfield's Energy Group, said. That would help data-centre operators meet power requirements while managing carbon targets.
Digital Connexion, Brookfield's Indian data-centre venture, is building out the remaining 100 MW of its 160 MW portfolio after the first 60 MW was fully leased. The partnership structure — combining Brookfield's capital, Digital Realty's operating expertise and Reliance's local infrastructure — provides a template for scaling.
The buildout represents a long-term capital commitment to India's digital infrastructure. Companies exposed to the trend include data-centre REITs, power utilities and AI hardware suppliers. Brookfield's $100 billion India target implies the firm sees a decade-plus runway for data-centre investment in the country.
This article is for informational purposes only and does not constitute investment advice.
