Mumbai: The Adani Group’s $100 billion investment commitment to an artificial intelligence (AI) ready data centre and power ecosystem—along with investment commitments of other Indian conglomerates like Reliance Industries, Tata Group and Larsen & Toubro—is taking a leaf out of the $500 billion Project Stargate playbook of the US.
The ecosystem not only gives Indian conglomerates a cost advantage but also better control over the operationalization of their multi-gigawatt data centres, experts said.
So far, investment commitments by top Indian conglomerates in the AI ecosystem have reached $125 billion, including those from Reliance Industries, the Tata Group, and Larsen & Toubro, according to Mint’s calculations based on public announcements and analyst estimates.
This rivals the $500 billion Project Stargate in the US, where top private-sector companies have committed to investing in the infrastructure needed to power the next generation of AI and secure the country’s lead over rivals like China.
The Adani Group’s investment plan, for instance, includes not just setting up data centres to handle AI demand but also the energy generation ecosystem that will power them, Gautam Adani, chairman of the Adani Group, said in a press release on Tuesday.
The Ahmedabad-based conglomerate will spend this $100 billion across its group firms by 2035, including about $7 billion already spent, as per a company executive who spoke on the condition of anonymity.
The data centres will be built in Gujarat, Maharashtra, Karnataka, Rajasthan and Andhra Pradesh, this executive said. The renewable energy infrastructure, meanwhile, will be spread across Gujarat, Rajasthan, Madhya Pradesh and Tamil Nadu.
“The world is entering an Intelligence Revolution more profound than any previous Industrial Revolution,” Adani said in the press release. “Nations that master the symmetry between energy and compute will shape the next decade.”
The contours of computing
The Adani Group has not disclosed the source of funding for this investment, which is arguably one of the largest commitments the group has made to date.
Reliance Industries is planning a similar end-to-end play, setting up renewable energy-powered data centres that are AI-ready. The Mumbai-based conglomerate has yet to outline a consolidated investment guidance for this ecosystem, but analysts at Morgan Stanley estimated in a note on 31 October 2025 that it could spend up to $15 billion per 1 GW of data centre capacity.
The Tata Group also has investments across power generation (at Tata Power Ltd) and has planned a $6.5 billion investment in AI-ready data centres at Tata Consultancy Services Ltd. However, the Mumbai-based conglomerate has not announced captive power capacity at Tata Power for TCS’s data centres, unlike the road map given by Adani and Reliance. Separately, Tata Electronics is investing $11 billion in making a semiconductor fab in Gujarat and $3 billion in an outsourced semiconductor assembly and testing (OSAT) facility in Assam.
Meanwhile, Larsen & Toubro Ltd (L&T) plans to build a moat by owning the land, physical infrastructure, and servers, and by having in-house construction capability. The company is also considering developing its own renewable energy plants to power these data centres, but a final decision has yet to be taken, Mint reported on 21 January.
“This kind of commitment is not a single bet on AI enthusiasm. It is a layered infrastructure play. Power, time, and control sit at the centre of it,” said Sanchit Gogia, the chief executive officer and chief analyst at Greyhound Research, a technology research and advisory firm.
- Indian conglomerates have committed a combined $125 billion to AI infrastructure, positioning India as a primary rival to US and Chinese tech dominance.
- This model links renewable energy generation directly to data centres, bypassing grid bottlenecks and reducing operational costs.
- These are ‘layered’ bets; the assets remain valuable even if the specific demand for AI compute fluctuates.
- The play extends beyond data centres into the silicon layer, evidenced by Tata’s $14 billion investment in semiconductor fabs and assembly.
- By controlling the ‘symmetry between energy and compute’, these firms aim to give India pricing power in the next decade of the global intelligence revolution.
Building the dedicated space
Dedicated renewable and thermal power for data centres changes the risk equation, Gogia explained. If compute demand accelerates, that capacity feeds hyperscale campuses. If utilization slows down, that same power can be sold into broader demand pools, reducing the dependency on one narrative.
These large investments also give these conglomerates greater control over the operationalisation of their data centres. Presently, many Western markets are facing a bottleneck in providing energy connectivity to data centres due to lagging grid development.
“Data halls can be built faster than they can be powered. If you can align generation, transmission coordination, storage provisioning, and campus build in one sequence, you shorten the uncertainty window. That translates into pricing power and stronger tenant negotiations,” Gogia said. In such a capacity-constrained situation, power costs also spike for data centres. Owning the generation capacity could protect these conglomerates from such cost spikes, he added.
Lastly, the scale introduces additional advantages, he said. Repeated activation of data centres reduces friction in each subsequent iteration as understanding with vendors improves, large procurement drives down costs, and commissioning becomes more predictable.
“The conviction behind a number like 100 billion does not come from assuming every AI application will succeed. It comes from building a platform where power, infrastructure, and ecosystem positioning generate layered monetisation paths,” Gogia said.
