Welcome to Global Data Center Hub. Join investors, operators, and innovators reading to stay ahead of the latest trends in the data center sector in developed and emerging markets globally
From $30B AI megacampuses in India to sovereign cloud networks in Korea and billion-dollar GPU hubs across Southeast Asia, the first half of 2025 turned APAC into the new frontline of AI geopolitics and infrastructure strategy.
We saw:
$180B+ in AI data center commitments from India, Korea, Japan, and Southeast Asia
Sovereign compute strategies accelerating across Thailand, Malaysia, and Indonesia
Private capital flooding into powered land, GPU logistics, and modular campuses
Policy pivots in India, Korea, and ASEAN linking AI infra to national growth plans
Power and cooling constraints forcing hybrid urban-remote deployments
H1 2025 wasn’t just growth. It was a strategic realignment.
This is your APAC recap of the 15 most important shifts and what they signal for global AI infrastructure.
Here’s What’s Inside
Top 15 announcements ranked — From Reliance’s $30B AI city to SoftBank’s Stargate alignment and Alibaba’s multi-market blitz
5 key infrastructure trends — Sovereign clouds, REIT-led financing, AI-first design, energy moats, and talent arbitrage
5 emerging opportunities — Infra-as-a-Service, GPU hosting platforms, powered land banking, sovereign REITs, and green AI campuses
5 regional shifts — India inland clusters, Japan’s renewable pivot, Johor as FLAP alternative, Thailand as AI factory, Korea’s decentralized edge
5 critical constraints — Power bottlenecks, GPU scarcity, ESG pressure, policy lags, and climate volatility
Top 15 APAC Announcements (H1 2025)
Below are the most impactful strategic moves shaping the Asia-Pacific data center sector in the first half of 2025, ranked by capital scale, geopolitical significance, and long-term market implications.
1. Reliance’s $30B AI Megacampus in Mumbai
Reliance Industries announced plans to build a $30 billion AI and data center campus, positioning India as a sovereign AI powerhouse. The project will be the world’s largest single-site data center, integrating energy, telecom, and compute. It’s a direct move to localize AI infrastructure and reduce dependence on global hyperscalers. Reliance’s scale, execution capacity, and alignment with national digital goals make this a defining move. India is no longer outsourcing compute, it's building it. [Read here]
2. South Korea’s $35B AI Data Center in Jeollanam-do
Jeollanam-do secured the largest-ever AI data center investment globally, with a $35B project designed to anchor Korea’s sovereign AI infrastructure. The site is expected to house thousands of GPUs and lead green, large-scale AI compute development. This megaproject has backing from both state and private players. It demonstrates how Korea is shifting from a chip export model to hosting AI supercomputers. Execution is underway with land and permitting in progress. [Read here]
3. SoftBank Seeks $16.5B for Stargate AI Buildout
SoftBank is raising $16.5B to fund Stargate a next-generation AI data center network starting in Japan. The project is tied to the global Stargate partnership with OpenAI, xAI, Microsoft, and BlackRock. It includes sovereign-scale land banking and renewable energy. SoftBank is not just a telecom or chip player anymore, it’s positioning itself as a global compute landlord. This is Japan’s most ambitious infra raise for AI. [Read here]
4. Alibaba’s $53B AI & Cloud Infrastructure Commitment
Alibaba committed $53B over three years to expand AI and cloud infrastructure globally, with heavy investment in APAC. The push includes second data centers in Thailand and South Korea and expansions in Indonesia. Alibaba is going head-to-head with AWS and Azure across sovereign cloud zones. The strategy reflects China's pivot to compete for global AI influence via infrastructure. Execution began with new launches in Q2 2025. [Read here]
5. Amazon’s AU$20B Data Center Investment in Australia
Amazon Web Services committed AU$20 billion (~US$13.3B) to expand its Australian cloud and AI footprint. The investment will be distributed across Sydney, Melbourne, and Perth with new data centers and renewable integration. This marks the largest cloud infra investment in Australian history. It gives AWS multi-region capabilities and proximity to Southeast Asia. Amazon is also aligning the buildout with its global AI chip strategy. [Read here]
6. Thailand’s $15.4B National AI Infra Push
Thailand launched a $15.4B AI infrastructure and cloud strategy backed by the Board of Investment and foreign players. It includes Giga Data Centers, GPU zones, and tax incentives to attract Big Tech. The country is positioning itself as Southeast Asia’s new digital factory. Buildouts are accelerating in Bangkok, EEC, and northern cities. Execution is well underway, with OpenAI, AWS, and PDG reportedly engaged. [Read here]
7. CtrlS Plans 600MW AI Campus in Hyderabad
CtrlS signed an MoU with the Telangana government to develop a 600MW AI data center hub. It’s one of the largest privately planned single-site projects in India. The site will include renewable integration, sovereign compute, and GPU-ready architecture. CtrlS is positioning itself as a sovereign alternative to hyperscalers. Execution will roll out in phases through 2026. [Read here]
8. Microsoft’s Triple Data Center Launch in Malaysia
Microsoft announced three hyperscale launches in Malaysia by Q2 2025. The data centers will power Azure AI services, Microsoft Copilot, and localized cloud offerings. Microsoft also rolled out AI skilling and workforce partnerships with TAFE and local institutions. The move cements Malaysia as a rising digital infrastructure market. It’s the most coordinated hyperscaler rollout in the region this year. [Read here]
9. NTT’s 500MW AI Cluster in Navi Mumbai
NTT unveiled NAV2 a 500MW data center campus in Navi Mumbai optimized for AI workloads. It features liquid cooling, modular blocks, and hyperscaler-ready design. The launch is part of NTT’s broader $16.4B strategy to privatize its infra division and focus on AI. The project makes India a central node in NTT’s global compute map. Construction began in Q2 with a phased rollout. [Read here]
10. PDG Secures $1B+ to Scale Across APAC
Princeton Digital Group raised over $1 billion to scale AI-ready data centers in Japan, India, and Southeast Asia. The round includes green financing and refinancing of strategic campuses. PDG is one of the few Asia-native hyperscale operators with sovereign-level backing. Its portfolio aligns with national AI goals in Indonesia and India. The capital secures PDG’s spot in the regional AI arms race. [Read here]
11. AirTrunk Expands in Tokyo and Johor
AirTrunk broke ground on a 40MW Tokyo campus and advanced work on its Johor Malaysia build. These expansions signal power-secure, low-latency hubs for AI training workloads. Tokyo gives AirTrunk a northern Asia anchor; Johor taps into Singapore’s overflow. With a strong sustainability profile, the builds are also backed by institutional capital. AirTrunk continues to execute with hyperscale precision. [Read here]
12. Digital Edge & B.Grimm $1B JV in Bangkok
Digital Edge and Thai conglomerate B.Grimm launched a $1B, 100MW data center joint venture in Bangkok. The facility will serve AI and sovereign cloud workloads. This cross-border partnership signals a regional shift toward local alignment. B.Grimm brings energy, land, and local execution; Digital Edge brings design and scale. The facility will be a cornerstone in Thailand’s new AI corridor. [Read here]
13. Foxconn + Nvidia: 100MW AI Facility in Taiwan
Foxconn and Nvidia partnered to build a 100MW AI-native data center in Taiwan. It’s part of Nvidia’s move to vertically integrate compute infrastructure across Asia. Foxconn will supply power and fabrication-level systems; Nvidia will integrate GPUs and orchestration tools. This is a blueprint for OEM-AI infra collaborations. Taiwan is becoming a manufacturing + model training hybrid zone. [Read here]
14. Google & Gamuda Partner on $1B Malaysia Project
Gamuda sold land to a Google affiliate for a $1B hyperscale project in Selangor, Malaysia. It will serve Google’s Gemini and global AI clients. The project signals deeper integration of local developers into Big Tech infrastructure rollouts. Gamuda brings land and regulatory strength; Google brings capex and compute. The data center will go live in phases from 2026. [Read here]
15. Keppel Data Centre Fund III Hits $580M First Close
Keppel closed $580M for its third Asia-focused data center fund, targeting AI-native, green infrastructure. It reflects the rising institutional appetite for yield-generating AI infrastructure. The fund will focus on Singapore, Malaysia, Vietnam, and emerging markets. With prior execution across 20+ data centers, Keppel has proven sovereign alignment. This fundraise is a signal to watch for REIT and fund growth in Asia. [Read here]
Key Trends (H1 2025)
1. Compute as an Economic Imperative
Asia Pacific governments are shifting from enabling to orchestrating AI infrastructure at national scale. India’s $30B Reliance-backed AI campus, South Korea’s $35B Jeollanam-do GPU hub, and Japan’s multi-billion-dollar SoftBank Stargate program signal a structural pivot. These are not standalone facilities they are policy instruments designed to lock in digital sovereignty and GDP growth. The region now treats compute like oil: scarce, strategic, and state-protected.
2. Financial Architecture Becomes the Differentiator
Capital engineering is no longer a back-office function it’s a moat. We’re seeing REIT conversions, sovereign guarantee structures, and private credit syndicates dominate APAC builds. SoftBank’s $16.5B Stargate financing and Keppel’s $580M AI-native data center fund exemplify a playbook where debt structures unlock speed at scale. Operators fluent in both infrastructure and financial design will control the buildout curve.
3. Sovereign AI Corridors Replace Generic Hyperscale
APAC nations are not just building sites they’re designing sovereign GPU corridors. India’s three-node AI belt, Thailand’s Eastern Economic Corridor expansions, and Malaysia’s Johor mega-campus strategy show the shift toward localized, compliance-hardened compute zones. These corridors combine energy, tax, and regulatory guarantees, creating gravity wells that attract AI-native operators and cloud alliances.
4. Energy Strategy as a Competitive Weapon
AI’s appetite for megawatts is colliding with APAC’s grid constraints, and the winners will be those who integrate power generation into the development model. Japan’s Hokkaido geothermal builds, Thailand’s hydro-integrated AI parks, and Malaysia’s solar-linked hyperscale projects are early signals. In the next wave, energy-first campuses will become the standard. Compute without clean power is now a stranded asset risk.
5. Dual-Speed Geography Emerges
Inference remains urban; model training moves to the edge of the grid. Mumbai links to Lucknow, Tokyo to Hokkaido, Bangkok to Chiang Mai. This urban-remote bifurcation reflects a hybrid logic: low-latency AI services stay in cities, while GPU-intensive workloads migrate to power-rich, renewable-friendly clusters. Operators building for both topologies will capture the next trillion-dollar AI economy.
Emerging Opportunities
1. Infra-as-a-Service as a National Export
Countries like Thailand and Malaysia are packaging land, power, and compliance as a turnkey service. This approach allows other nations or private players to “rent sovereignty” for their AI workloads without building local infrastructure. By standardizing compliance, power access, and green energy integration, these nations are creating a new export category for digital infrastructure.
2. GPU Hosting Platforms Gain Traction
Specialized hosting platforms are emerging to address GPU scarcity by guaranteeing allocation, uptime, and optimized deployment. These platforms partner with chipmakers, sovereign funds, and operators to create reliable, high-speed GPU access. As AI demand surges, these platforms are becoming a critical layer of infrastructure value and profitability.
3. Powered Land Banking Becomes a Strategic Asset
Land without reliable power is dead capital, so investors are now securing pre-zoned, power-ready parcels near renewable energy corridors. This strategy effectively creates a “compute real estate” market that combines location, energy, and regulatory readiness. Operators who control these parcels can accelerate AI campus deployment while minimizing stranded asset risk.
4. Sovereign REITs Unlock Capital Velocity
Governments and institutional investors are structuring REIT-like vehicles to accelerate AI campus development across APAC. These vehicles combine stable yield with AI-driven upside, attracting global capital that seeks long-duration, inflation-resistant assets. By blending finance and infrastructure strategy, sovereign REITs are redefining how large-scale AI projects are funded and executed.
5. Green AI Campuses as the New Premium Tier
Hyperscalers and sovereign operators are racing to build AI campuses designed around renewable energy, circular cooling, and ESG-aligned operations. Sustainability is no longer just a regulatory checkbox it’s a competitive advantage that influences hyperscaler deals and investor access. Green campuses are becoming the premium standard for new AI infrastructure in APAC, signaling a shift toward energy-conscious, long-term buildouts.
Sector & Geographic Shifts
1. India Scales Beyond Metro
Mumbai remains the anchor, but the real story is moving inland. Hyderabad and Chennai are emerging as sovereign-grade AI clusters, thanks to land availability, renewable energy corridors, and state-backed infrastructure programs. Expect Tier-2 corridors like Pune and Lucknow to light up by 2026, creating a multi-polar AI landscape.
2. Japan Pushes North
Tokyo’s grid is saturated, forcing a strategic pivot to Hokkaido. The region is leveraging geothermal and offshore wind to power climate-resilient AI parks, marking Japan’s shift from dense urban hyperscale to energy-driven distributed campuses. This transition is key to Japan’s long-term energy-security strategy for AI.
3. Southeast Asia Becomes the Alt-FLAP
Johor and Cyberjaya are no longer overflow markets they’re becoming sovereign-grade compute hubs. Thailand’s Eastern Economic Corridor (EEC) is moving aggressively to position itself as an AI corridor, bundling tax breaks, land access, and GPU capacity. Southeast Asia is emerging as Asia’s counterweight to traditional FLAP markets.
4. South Korea Decentralizes
Jeollanam-do’s $35B AI cluster signals a clear shift away from Seoul. Korea’s strategy now integrates chip manufacturing, GPU cloud, and renewable-powered AI campuses to maximize efficiency and resilience. Energy-rich coastal zones will define Korea’s next wave of AI infrastructure growth.
5. Australia Doubles Down on AI Infra
AWS’s AU$20B investment cements Sydney and Melbourne as southern anchors, while new builds in Perth indicate a dual-region resiliency strategy. Australia is also piloting hydrogen and advanced storage solutions to integrate AI energy needs with its renewable grid. This positions the country as a clean-energy AI hub for the Asia-Pacific region.
Challenges & Gaps
1. Power Availability & Permitting Bottlenecks
APAC’s top markets Tokyo, Seoul, and Sydney are hitting hard power allocation ceilings, making electricity the new gating factor for AI growth. Developers face 12–24 month delays just to secure substation approvals or utility interconnections, leaving many sites “unbuildable,” according to CBRE. While India offers more flexibility, congestion around Mumbai is growing, signaling a future where energy access not land determines expansion.
2. GPU Access & Supply Chain Constraints
Even as sovereign clouds and regional players accelerate AI-ready builds, GPU scarcity remains the Achilles’ heel. Nvidia and AMD pipelines are backlogged, pushing Korea, India, and Indonesia to launch emergency procurement programs with limited success. Some projects are breaking ground without guaranteed GPU customers, risking stranded assets and ROI erosion especially outside U.S.-aligned jurisdictions.
3. Environmental Vulnerabilities: Floods, Heat, and Risk Zones
Inland and coastal expansions across APAC expose new environmental fault lines, from floods in Bangkok and Chennai to typhoons in the Philippines and seismic risk in Japan. Most new builds lack robust disaster-resilient design or adequate power/fiber redundancy, particularly in emerging edge zones. Rising insurance premiums underscore the financial drag of climate-linked vulnerabilities.
4. Policy Gaps: Sovereign AI Strategies Outpacing Regulatory Reform
Governments across APAC are aggressively announcing sovereign AI and cloud goals, but policy frameworks are lagging. India and Vietnam still lack standardized green permitting, and GPU tax incentives remain patchy across states. Korea faces local-national misalignment, forcing developers to navigate bespoke deal structures slowing execution despite political will.
5. Workforce & Technical Talent Shortages
AI data centers demand specialized skills in liquid cooling, chip-scale infrastructure, and workload orchestration capabilities that remain concentrated in a few hubs like Singapore, Tokyo, and Bangalore. Most countries lack national programs to pipeline this expertise, creating a structural talent gap that risks long-term scalability. Initiatives like Microsoft’s Datacentre Academies in Australia and Malaysia are promising but insufficient to meet projected demand.