The corridor logic: why state competition decides India’s AI geography
- No Indian corridor is strong on all six axes (compute, power, water, fibre, semiconductor adjacency, talent). Every regional bet involves a structural trade-off.
- Karnataka has the talent and policy depth but not the water. Tamil Nadu has electronics manufacturing and coastal-fibre but tightening water. Gujarat has the semiconductors but not the data centres. Andhra Pradesh is rising fastest on Google-Adani Visakhapatnam.
- Capital allocation across Indian corridors is a portfolio choice, not a single bet. The binding constraint in each corridor is different — and is the right unit of policy and execution.
- No Indian corridor is strong on all six axes (compute, power, water, fibre, semiconductor adjacency, talent). Every regional bet involves a structural trade-off.
- Karnataka has the talent and policy depth but not the water. Tamil Nadu has electronics manufacturing and coastal-fibre but tightening water. Gujarat has the semiconductors but not the data centres. Andhra Pradesh is rising fastest on Google-Adani Visakhapatnam.
- Capital allocation across Indian corridors is a portfolio choice, not a single bet. The binding constraint in each corridor is different — and is the right unit of policy and execution.
A data centre is sited where power is plentiful and grid-firm, where water is available and water rights are clear, where fibre lands or transits, where land is buildable and zoned, where the regulatory environment is predictable, and where adjacent skilled labour exists. These conditions cluster geographically — in Mumbai, Chennai, Hyderabad, Bengaluru, Pune, Visakhapatnam, Noida–Sanand–Dholera — and not nationally. The competitive question for Indian states is not whether they are doing AI; it is whether their corridor scores well across these axes against the others.
The seven corridors and their binding constraints
- Karnataka / Bengaluru — highest on talent and compute pipeline; lowest on water (Bengaluru Urban over-exploited, Bengaluru Rural at 169% of permissible extraction).
- Telangana / Hyderabad — highest on hyperscaler magnetism (AWS USD 7B, Microsoft USD 17.5B+ commitments, CtrlS 612 MW, Yotta 50 MW); water-stressed (over-exploited 2024 upgrade); Godavari Phase II/III is the structural mitigant.
- Tamil Nadu / Chennai-Coimbatore — highest electronics manufacturing depth (USD 14.65 billion FY25); coastal-fibre; Tier-2 R&D in Coimbatore; tightening on water (2019 Day Zero).
- Maharashtra / Mumbai-Pune-Navi Mumbai — largest DC mass (44% of national); eight CLS at Mumbai; cost-pressured; MERC HT cross-subsidy softening.
- Gujarat / Sanand-Dholera — only state with a major fab (Tata-PSMC ₹91,000 crore); Sanand OSAT cluster ₹1.25 lakh crore; ports and RE depth; AI-services light.
- NCR / UP / Jewar-Noida — Jewar airport cargo (250,000 → 1.8 million MT/year), HCL-Foxconn OSAT, EMC 2.0; e-commerce and fintech anchor.
- Andhra Pradesh / Visakhapatnam — rising fastest on Google-Adani 1 GW commitment, Sify Open CLS, favourable water position; transmission execution is the principal risk.
Why it matters
Every Indian corridor is strong on some axes and weak on others. There is no national winner — there are seven regional bets with structurally different binding constraints. Capital allocation across the corridors is therefore a portfolio choice, and the corridor-level outcomes will diverge sharply because the underlying endowments diverge sharply. The path-dependent geography of 2026–2030 will set the structural pattern for 2030–2035 and beyond.
For corridor-by-corridor profiles, the AI Regional Opportunity Corridors framework, and the state-by-state policy and tariff analysis, see Part V (Sections 18–25) of India’s AI Industrial Transition and Infrastructure Transformation (2026–2035).
Track the systems we watch
Signals, reports and briefings on India’s industrial transformation.