India's $210 Billion AI Data Centre Boom 2026: Reliance $110B, Adani $100B & the Race to Become the World's AI Infrastructure Hub
Reliance and Adani Just Committed $210 Billion to Make India the World's AI Infrastructure Hub.
$110 billion from Reliance. $100 billion from Adani. Gigawatt-scale data centres powered by their own solar farms. OpenAI-Tata and Google-Adani deals. India can't make chips — so it's building the compute the AI world will run on. Here's the complete breakdown.
At the India AI Impact Summit in New Delhi in February 2026 — the first major international AI summit hosted in the Global South, attended by Sam Altman, Sundar Pichai, and Dario Amodei — India's two largest conglomerates made an announcement that reframed the country's role in the AI age. Reliance committed $110 billion. Adani committed $100 billion. A combined $210 billion, aimed at a single goal: making India the world's AI infrastructure hub. India cannot make advanced chips. But it can build the data centres — powered by its own cheap renewable energy — where the world's AI will run. This is that story.
The $210 Billion Bet
Two of India's largest conglomerates, Reliance Group and Adani Group, are betting on AI with combined investments of around $210 billion. While Reliance Industries has committed roughly $110 billion towards building AI and digital infrastructure capabilities, Adani Enterprises has pledged $100 billion through 2035 to develop AI-enabled renewable data centers.
"Jio, together with Reliance, will invest 10,000 billion Indian rupees [$110 billion] over the next seven years starting this year. This is patient, disciplined, nation-building capital — designed to create durable economic value and strategic resilience for six decades to come," said Mukesh Ambani, chairman of Reliance Industries, speaking at the India AI Impact Summit 2026.
Reliance vs Adani: Two Strategies for the Same Goal
Both Reliance and Adani gain cost advantage by colocating data centers with their own renewable energy assets, thus reducing exposure to expensive grid power. Since India lacks presence in chip manufacturing, building data centers is the easiest way for the country to gain from the global AI growth.
Adani's vision, in Gautam Adani's words: "The world is entering an Intelligence Revolution more profound than any previous Industrial Revolution. India will not be a mere consumer in the AI age. We will be the creators, the builders and the exporters of intelligence." Adani's $100 billion is expected to create a $250 billion AI infrastructure ecosystem in India over the next decade.
Why India Can Win AI Infrastructure
Energy is the single largest operating cost for hyperscale AI data centers. In the United States, utilities in states like Virginia and Texas have warned about grid strain from AI expansion, and energy permitting timelines have slowed new builds. India's bet is that renewable integration and lower land costs can compress operating expenses relative to Western markets.
- The energy-compute symmetry. As Gautam Adani put it: "Nations that master the symmetry between energy and compute will shape the next decade. India is uniquely positioned to lead." Both Reliance and Adani are energy companies first — they can power data centres with their own captive renewable generation, sidestepping the grid strain crippling Western expansion.
- The compute cost problem. "The biggest constraint in AI is not scarcity of talent, but high cost of compute," Ambani stated. India's play is to make compute cheap — just as Jio made mobile data cheap and brought 500 million users online in the 2010s.
- Data localisation as structural demand. India's rules require financial, health, and personal data to be stored on servers physically located in India — a non-negotiable, permanent tailwind for domestic data centre capacity regardless of global conditions.
- The chip workaround. India has no advanced chip fabrication (yet). But data centres let India participate in the AI value chain now, using imported chips housed in domestically-built, domestically-powered infrastructure.
The Global Players Moving In
Global players are taking notice. OpenAI is partnering with Tata Group to develop 100 megawatts of AI capacity, with plans to scale to 1 gigawatt. Google's parent company Alphabet said it would invest $15 billion over five years to build an AI data center hub in southern India, and Adani's AI push is supported by its strategic partnership with Google, including an AI data center campus in Visakhapatnam.
| Player | India AI Data Centre Move | Scale |
|---|---|---|
| Reliance | Jamnagar gigawatt campus + Jio edge | $110B / GW-scale |
| Adani (AdaniConnex) | Mumbai, Chennai, Noida, Hyderabad, Pune, Vizag | 2GW → 5GW |
| OpenAI + Tata | 100MW AI capacity, scaling to 1GW | 100MW → 1GW |
| Google / Alphabet | $15B AI hub in southern India + Adani JV | $15B / 5 years |
| AWS, Microsoft Azure | Expanding Indian cloud regions | Ongoing scale-up |
| Nxtra (Airtel), Yotta, CtrlS | Domestic colocation operators (unlisted) | Colocation capacity |
The Capacity Numbers
India's installed data center capacity crossed 1,300 MW by the end of 2025. Industry projections put colocation capacity at 1.7 GW by the end of 2026, and some estimates have the market growing at a CAGR of 15–20% through 2030.
The demand drivers for AI workloads, cloud migration, and data localisation are not going away. AWS, Azure, and GCP are all expanding their Indian regions, which means building or leasing more data centres. The hyperscalers don't always build themselves; they often lease capacity from colocation operators, which creates a direct revenue opportunity for listed players.
The investing angle is tricky. Data center stocks in India currently offer limited pure-play listed options. The closest listed exposure is Adani Enterprises (via its AdaniConneX JV). Most major operators — Nxtra Data, Yotta, CtrlS — are unlisted. The most accessible investment angle is through IT hardware stocks in the power backup, precision cooling, and networking equipment supply chain. The capex cycle shows up in equipment order books before it shows up in operator profit margins.
The Challenges: Chips, Grid, and Water
- No domestic advanced chips. India's data centres run on imported GPUs (mostly NVIDIA). Until India's semiconductor mission delivers advanced fabrication, the highest-value component of the AI stack remains foreign-sourced — a strategic dependency.
- Grid and Right-of-Way. This scale of AI-driven data center expansion will accelerate demand for high-capacity fiber to link data center campuses. The Indian government will need to make it easier for companies to acquire Right of Way (RoW) and ensure a predictable regulatory environment.
- Water and cooling. AI data centres consume vast amounts of water and power for cooling. Sustainable cooling and water management at gigawatt scale is a genuine environmental and operational challenge in a water-stressed country.
- Execution risk at scale. $210 billion is a commitment, not a completed build. Delivering gigawatt-scale AI campuses on schedule, at cost, and with reliable uptime is an enormous execution challenge even for Reliance and Adani.
Jio made mobile data so cheap that 500 million Indians came online in a decade. Reliance and Adani are now trying to do the same for AI compute — make it so abundant and affordable that intelligence becomes infrastructure. If they succeed, India won't just use AI. It will be the place where much of the world's AI runs.
— BharatBusinessIndex Analysis, July 2026Most-Searched AI Data Centre Questions — Answered
India's $210 Billion AI Infrastructure Bet Is Its Most Ambitious Industrial Play of the Decade.
$110 billion from Reliance. $100 billion from Adani. Gigawatt-scale campuses. Captive renewable power. OpenAI-Tata and Google-Adani partnerships. A projected 1.7GW of colocation capacity by end-2026. India has identified a specific, winnable position in the global AI race — not chip fabrication (which it can't yet do), but AI infrastructure, where its energy assets and cost structure give it a genuine, defensible advantage.
The strategic logic is sound. The binding constraint on global AI is compute, compute is constrained by energy, and India's two largest companies are energy giants that can power data centres with their own solar farms while the West struggles with grid strain. That is a real edge, and the $15 billion Google commitment plus the OpenAI-Tata partnership show global players believe it too.
The honest caveats: India still depends entirely on imported chips for the highest-value layer of the stack, gigawatt-scale execution is unproven, and water and grid infrastructure at this scale are serious challenges. A $210 billion commitment is not a $210 billion completed build. But the direction is unmistakable and the ambition is credible. If India delivers even a meaningful fraction of this, it becomes one of the two or three most important places on Earth for AI infrastructure — and captures a durable position in the defining technology shift of the century. Watch Jamnagar and Visakhapatnam. That's where the future is being poured in concrete.