QUICK TAKE:
Contents
| Company: | Adani Group — India’s largest industrial conglomerate; Gautam Adani, Chairman. AdaniConnex (50:50 JV with EdgeConnex) is the data centre arm. |
| Commitment: | $100 billion by 2035 — renewable-energy-powered, hyperscale AI-ready data centres. India’s single largest private digital infrastructure commitment in history. |
| Capacity Target: | 5 GW total data centre capacity by 2035 (up from existing 2 GW AdaniConnex platform) — targeting world’s largest integrated data centre platform |
| Ecosystem Effect: | $150B in catalysed investments across server manufacturing, sovereign cloud, advanced electrical infrastructure = $250B total AI ecosystem by 2035 |
| Partners: | Google (Visakhapatnam GW-scale campus + Noida) · Microsoft (Hyderabad + Pune) · EdgeConnex (JV). Discussions ongoing with additional global tech firms. |
| Energy Source: | 100% renewable — Adani Green Energy’s 30 GW Khavda project, Gujarat (10 GW already operational). World’s largest hybrid solar+wind park. |
| Clean Energy Add-On: | Separate $55B commitment to expand renewable energy portfolio including world-scale battery energy storage systems (BESS) |
| Manufacturing: | Co-investment in domestic production of transformers, power electronics, thermal management systems to reduce supply chain exposure |
| Skills: | AI Infrastructure Engineering curricula with leading universities · Applied AI research labs · National fellowship programme for skills gap |
| Announced: | February 17, 2026 — Day 2, India AI Impact Summit, Bharat Mandapam, New Delhi. Stock reaction: Adani Enterprises +2.3%, Adani Green Energy +1.8% |
THE STORY
On the second day of the India AI Impact Summit in New Delhi, Adani Group Chairman Gautam Adani announced what may be the single most audacious private infrastructure commitment in India’s history: $100 billion to build renewable-energy-powered, hyperscale AI-ready data centres across India by 2035. The plan targets a 5-gigawatt data centre platform — growing AdaniConnex’s existing 2 GW base to the world’s largest integrated data centre footprint — backed by strategic partnerships with Google (Visakhapatnam, Noida) and Microsoft (Hyderabad, Pune), and powered entirely by Adani’s own renewable energy assets. Adani simultaneously committed a separate $55 billion to expand its renewable energy portfolio, including what would be one of the world’s largest battery energy storage systems. Together, the two commitments total $155 billion in direct investment — with a further $150 billion in catalysed investments projected across server manufacturing, sovereign cloud, and related industries. The total projected AI infrastructure ecosystem: $250 billion by 2035.
“Nations that master the symmetry between energy and compute will shape the next decade. India will not be a mere consumer in the AI age. We will be the creators, the builders and exporters of intelligence.” — Gautam Adani, Chairman, Adani Group
“Renewable energy. Grid resilience. Hyperscale AI. A $250B intelligence ecosystem in motion. India will not follow the AI century. India will shape it.” — Gautam Adani
The $250 Billion Equation — Decoded
| $100B
Direct AI Data Centre Investment |
Building 5 GW of hyperscale AI-ready data centre infrastructure across 4+ campuses (Visakhapatnam, Noida, Hyderabad, Pune + more) by 2035. 100% renewable-powered. Optimised for large high-density compute clusters and next-gen AI workloads with advanced liquid cooling. |
| $55B
Renewable Energy + Battery Storage |
Separate commitment to expand Adani Green Energy’s renewable portfolio alongside one of the world’s largest battery energy storage systems (BESS). Anchored by the 30 GW Khavda project in Gujarat — already the world’s largest hybrid solar+wind park with 10 GW operational. |
| $150B
Catalysed Ecosystem Investments |
Adani-projected spillover investments from partner companies in server manufacturing, sovereign cloud platforms, advanced electrical infrastructure, transformers, power electronics, and thermal management. This is India’s AI industrial multiplier effect. |
| $250B
Total AI Ecosystem by 2035 |
The sum: $100B direct + $150B catalysed = Adani’s projection of India’s total AI infrastructure ecosystem value created by this single initiative over the decade. For context: India’s current GDP is ~$3.7 trillion. This is 6.7% of current GDP being committed to one infrastructure vertical. |
| 5 GW
Data Centre Capacity Target |
From 2 GW (AdaniConnex today) to 5 GW by 2035. For scale: the entire US data centre market currently consumes ~17 GW. A 5 GW Adani platform would represent roughly 29% of current total US data centre power — concentrated in one country, owned by one company, powered 100% renewably. |
| 30 GW
Khavda Renewable Energy Park |
Gujarat’s Kutch district: 30 GW hybrid solar + wind across 72,600 hectares of wasteland. Already 10 GW operational. Full 30 GW expected by 2030. Generates enough electricity to power 18 million Indian homes. This is the energy backbone of the entire AI data centre plan. |
The 4-City Campus Map — Where Adani Is Building India’s AI Backbone
| Campus | City / State | Partner | Details & Strategic Significance |
|---|---|---|---|
| Campus 1 | Visakhapatnam, Andhra Pradesh | Gigawatt-scale AI data centre campus. Part of Google’s $15B India investment over 5 years. Adani invested $5B in Google’s AI infrastructure hub here in November 2025. Andhra Pradesh’s proximity to the Bay of Bengal enables subsea cable connectivity to Southeast Asia and the Middle East. | |
| Campus 2 | Noida, Delhi-NCR | Second Google-partnered campus in the National Capital Region. Proximity to India’s largest enterprise IT corridor, ISRO’s data infrastructure, and the government’s Digital India backbone. Key for citizen services AI and government enterprise deployments. | |
| Campus 3 | Hyderabad, Telangana | Microsoft | Microsoft has committed $17.5B to India after its December meeting with PM Modi. Hyderabad is India’s de-facto second tech capital — home to Microsoft India HQ, Google India HQ, Amazon’s largest campus outside the US. A data centre here integrates into the deepest tech talent and enterprise client base in the country. |
| Campus 4 | Pune, Maharashtra | Microsoft | Maharashtra is India’s largest state economy. Pune is the engineering and deep-tech anchor of the Mumbai-Pune corridor. Microsoft’s second Adani campus here serves western India’s manufacturing, pharma, and BFSI AI workloads. |
| Campuses 5+ | TBD — Discussions Underway | Multiple Global Tech Firms | Adani has stated discussions are ongoing with additional global technology companies for further campuses. Chennai, Bengaluru, and Mumbai are the highest-probability locations given existing data centre density, renewable energy access, and subsea cable landing infrastructure. |
The Energy + Compute Integration — Why Adani’s Moat Is Unique Globally
Every major AI infrastructure company in the world faces the same bottleneck right now: energy. Nvidia’s Blackwell GPUs can be manufactured. Land can be acquired. Engineers can be hired. But clean, reliable, cheap electricity at gigawatt scale — that takes decades to build. Microsoft, Google, and Amazon are all scrambling to sign power purchase agreements (PPAs) with renewable energy providers, nuclear operators, and even decommissioned coal plants, just to power their AI data centres. They are buyers of energy, completely dependent on others to supply it.
Adani’s structural advantage is that it is both the largest renewable energy producer in India AND the data centre operator. It does not need to sign a PPA. It owns the power plant. The 30 GW Khavda project is Adani’s energy, supplying Adani’s data centres, serving Adani’s hyperscaler partners. This vertical integration — renewable generation, grid infrastructure, battery storage, and compute — is what Gautam Adani means by ‘the symmetry between energy and compute.’ No other data centre operator in the world controls this complete stack at this scale.
How Adani Compares to Global AI Infrastructure Players on Energy:
|
India AI Infrastructure Week — How Adani’s Pledge Stacks Up
| Company | India Investment | Focus | Nature of Commitment |
|---|---|---|---|
| Adani Group | $100B + $55B RE | AI Data Centres | Domestic private conglomerate owning full stack: energy + data centres + manufacturing. Partnerships with Google & Microsoft. |
| Microsoft | $50B (Global South by decade-end) | AI Infrastructure + Skills | Global South-wide commitment; $17.5B India-specific Dec 2025. Data centres in Hyderabad + Pune with Adani. 20M Indians to be skilled by 2030. |
| $15B (India, 5-year) | AI Infrastructure + Connectivity | Largest AI data centre hub outside US in Visakhapatnam. America-India Connect subsea cables. Reliance Jio cloud clusters. 50MW solar in Rajasthan. | |
| L&T + NVIDIA | Not disclosed | AI Factory Infrastructure | Gigawatt-scale AI factories: 30 MW Chennai + 40 MW Mumbai. Joint venture for AI-ready data centre infrastructure and ecosystem support. |
| Blackstone / Neysa | $600M (equity) | Sovereign AI Cloud | 20,000+ GPUs for AI training and inference. India’s largest independent AI cloud platform. Series A equivalent for AI infrastructure. |
| AMD + TCS | Not disclosed | AI Infrastructure | 200 MW of AI infrastructure capacity deployment for sovereign AI and enterprise workloads across India. |
| India Government | $1.1B (VC Fund) | AI Startups + Manufacturing | State-backed VC fund for AI and advanced manufacturing startups under IndiaAI Mission. 58,000 GPUs at ₹65/hour subsidised rate. |
THE HONEST CAVEATS — WHAT WE DON’T KNOW YET
StartupFeed publishes every announcement with the same rigour we’d apply to a funding round: we report what was said, we note what wasn’t answered.
|
STARTUPFEED INSIGHT
| The Geopolitical Dimension: Adani’s announcement is not just a business investment — it is a sovereignty statement. Saudi Arabia has NEOM. France has its national AI strategy. The UAE has the Emirati AI initiative. Every major non-Western nation is building its own AI infrastructure rather than renting it from AWS, Azure, or Google Cloud. India, through Adani, is making the same move — but with a structural advantage none of those countries have: a domestic conglomerate that already owns the renewable energy assets to power the compute. The energy-compute integration is India’s unique answer to the global AI sovereignty race, and Adani is the private sector vehicle executing it. | |
| For Startups: | If Adani’s timeline holds, by 2027-28 India will have the cheapest clean AI compute in the Asia-Pacific region — at rates potentially below AWS Mumbai, Google Cloud Hyderabad, or Azure India. For compute-hungry AI startups (especially foundation model builders, inference-heavy applications, and data platforms), Adani’s infrastructure could become the default infrastructure layer. Watch for AdaniConnex to announce an ‘AI Startup Cloud’ programme with subsidised access to Blackwell/next-gen GPU clusters in 2027. |
| For Investors: | The $150B catalysed investments across server manufacturing, sovereign cloud, and electrical infrastructure is the real investment thesis. Indian companies building transformers, power electronics, thermal management systems, and AI server components are the picks-and-shovels play on Adani’s data centre boom. Listed names to watch: Hitachi Energy India, BHEL (power infrastructure), Kaynes Technology (electronics manufacturing), Netweb Technologies (AI server manufacturing), Dixon Technologies (electronics). |
| For Policy Makers: | Adani’s commitment to AI Infrastructure Engineering curricula and a national fellowship programme for skills addresses India’s most critical AI bottleneck: not compute, not capital, but engineers who can design, deploy, and maintain gigawatt-scale AI infrastructure. The government should co-invest here — matching Adani’s fellowship programme with AICTE mandates, IIT partnerships, and National Skill Development Corporation certification for AI Infrastructure Engineering as a formal trade qualification. |
| Our 2035 Prediction: By 2030 (not 2035), Adani will have deployed the first 2.5 GW of AI data centre capacity across Visakhapatnam and Hyderabad, and will announce its IPO of AdaniConnex as a standalone listed entity on NSE/BSE — potentially India’s largest technology infrastructure IPO ever, targeting a $30-40 billion valuation. The clean energy + compute integration moat will attract long-term strategic investors from sovereign wealth funds (GIC, ADIA, Mubadala) and infrastructure-focused PE (Brookfield, Macquarie). By 2035, Adani’s AI infrastructure platform will be India’s second-largest foreign exchange earner after IT services — as global AI companies pay to train and serve models on Indian sovereign compute at renewable energy rates unavailable anywhere else in the world. Gautam Adani’s bet is not just about India. It is about becoming the world’s preferred home for AI compute — and winning on energy economics. | |
