Hiranandani mentioned that India is making progress with its conventional cloud capacity, currently at approximately 1.5 gigawatts. He emphasized that this must increase to at least three gigawatts by 2030, even before considering AI workloads. “To grow from 1.5 gigawatts to 3 gigawatts, we need around $40 billion,” he stated, which includes investments in infrastructure, power systems, and servers.
He further noted that if India aims for 5 gigawatts of capacity by the decade’s end, total capital expenditure could escalate to $70–80 billion. “Infrastructure cannot be constructed overnight,” Hiranandani remarked, highlighting that power, servers, and cloud systems require years for deployment. He anticipates sustained demand through 2030, with minimal risk of overcapacity in India due to its low starting point.
Also Read | Anant Raj to invest ₹4,500 crore in Andhra Pradesh for new data centre, IT park
Sarin pointed out that real estate is just the initial phase in the data centre value chain, with long-term returns stemming from operational models such as co-location and cloud services. “The primary requirement is land and a building, but that is merely the starting line,” he remarked.
Anant Raj currently boasts 157 megawatts of data centre buildings ready and approximately 28 megawatts of operational co-location capacity. Sarin mentioned that the company anticipates expanding capacity to 63 megawatts next year and to 117 megawatts thereafter. “The roadmap for the next three to four years is quite clear,” he added.
Regarding returns, Sarin indicated that co-location usually offers a return on investment (RoI) of 60–70%, whereas cloud infrastructure services can yield higher returns once scaled. However, Hiranandani commented that payback periods differ. “Co-location paybacks are generally around six to seven years,” he stated, while cloud computing investments may recoup costs in three to four years due to quicker depreciation cycles.
Also Read | OpenAI, NextDC plan to build $4.6 billion Sydney data centre
Both leaders asserted that a decrease in AI costs would not diminish the demand for data centres. “If AI becomes cheaper, data consumption increases,” Sarin noted. Hiranandani added that, even in the absence of global AI demand, India’s domestic cloud requirements alone support several years of capacity growth.
They also highlighted opportunities across the wider ecosystem, including power systems, mechanical, electrical, and plumbing (MEP) suppliers, as well as specialized data centre infrastructure providers, as capital expenditures accelerate.
For the full interview, watch the accompanying video
Catch all the latest updates from the stock market here
