Artificial intelligence ( AI ) is creating one of the largest infrastructure investment cycles in history, transforming data centres from specialized technology assets into a mainstream institutional investment opportunity.

Across Asia-Pacific, cloud providers, technology companies and data centre operators are racing to build the computing capacity required to support AI applications. However, as facilities become larger, more power-intensive and increasingly complex, the challenge is shifting from demand creation to capital mobilization.

Global data centre capacity, according to US-listed property group JLL’s Global Data Center Outlook 2026 report, is expected to nearly double from approximately 103 gigawatts ( GW ) today to 200GW by 2030, requiring up to US$3 trillion in investment across data centre facilities, power infrastructure and technology equipment.

Nearly 100GW of new capacity, JLL estimates, will be added globally between 2026 and 2030, with AI workloads becoming a major driver of demand. Asia-Pacific is expected to remain one of the fastest-growing regions, with capacity projected to expand from approximately 32GW to 57GW by 2030.

The opportunity is significant, but so are the financing requirements.

From bank lending to capital markets

The first phase of Asia’s data centre expansion has largely been supported by traditional bank financing. Syndicated loans and project-level facilities have played a central role in funding new developments, supported by long-term customer commitments, contracted revenues and the strong credit profiles of hyperscale technology companies.

Recent transactions demonstrate the depth of bank appetite for digital infrastructure. In Singapore, AirTrunk secured a S$2.25 billion ( approximately US$1.75 billion ) syndicated green loan to support its hyperscale data centre development, while Malaysia has seen some of the region’s largest data centre financings as operators expand capacity in Johor.

DayOne Data Centers, for example, secured a major financing package comprising a 7.5-billion-ringgit Islamic term loan facility and a US$1.7 billion offshore term loan, supporting its expansion plans in Malaysia and highlighting growing lender confidence in the sector.

These transactions demonstrate that banks remain a critical source of capital for Asia’s digital infrastructure build-out. However, the scale of the artificial intelligence boom is beginning to test the limits of traditional financing models.

Modern AI data centres require significantly greater upfront investment than previous-generation facilities. Beyond land and construction, developers must fund high-performance GPUs, advanced cooling systems, grid connections and dedicated energy infrastructure.

As projects move from hundreds of megawatts ( MW ) towards GW-scale developments, the industry will increasingly require access to deeper and more diversified pools of capital, including infrastructure funds, pension funds, insurance companies, private credit investors and debt capital markets.

The United States has provided an early indication of this next phase, with data centre developers increasingly exploring bond markets, securitization structures and institutional debt solutions to complement traditional bank financing.

For Asia, the evolution of financing markets will become increasingly important. Expanding access to capital markets would allow developers to recycle capital, reduce reliance on bank balance sheets and unlock funding for the next wave of AI infrastructure investment.

Malaysia emerges as a leading Southeast Asia AI Infra hub

Singapore remains one of Asia’s most established digital infrastructure markets, but limited land availability and electricity constraints have accelerated investment across the border.

Malaysia, particularly in the state of Johor, has emerged as one of Southeast Asia’s fastest-growing data centre markets due to its proximity to Singapore, lower development costs and expanding power infrastructure.

Malaysia has one of the region’s largest data centre development pipelines, according to Cushman & Wakefield, supported by demand from cloud providers, hyperscalers and AI infrastructure operators.

Johor’s operational capacity more than doubled over the past year, increasing from approximately 401MW to 897MW, highlighting the speed of regional expansion.

However, continued growth will depend not only on attracting developers, but also on securing sufficient financing and energy infrastructure to support increasingly power-intensive AI facilities.

Malaysia’s Ministry of Energy Transition expects that data centres could consume approximately 7.7GW of electricity by 2030, reinforcing the need for coordinated investment across digital infrastructure and energy systems.

India builds domestic AI capacity

India is pursuing a different strategy by focusing on domestic AI infrastructure to support its expanding digital economy and strengthen control over critical computing resources.

India’s operational data centre capacity, according to DC Byte, is approximately 1.5GW, with projects under construction expected to push capacity beyond 1.7GW by the end of 2026.

Mumbai remains the country’s largest data centre market, accounting for approximately 768MW of operational IT capacity and attracting significant hyperscale investment.

AI-focused developments are also accelerating, including a planned 168MW AI-ready data centre in Jamnagar, Gujarat, backed by Reliance Industries and Meta.

The Indian government is also investing directly in AI computing resources through the IndiaAI Mission, which has onboarded more than 38,000 high-end GPUs and provides subsidized access to start-ups seeking to develop AI applications.

Next AI race won by capital, power

Asia-Pacific’s AI infrastructure boom represents one of the largest opportunities in global digital infrastructure. However, building the next generation of facilities will require more than technology expertise and construction capability.

The future of the sector will depend on three critical factors: access to electricity, availability of capital and the ability to develop financing structures that match the scale and complexity of demand.

Bank lending has provided the foundation for Asia’s data centre expansion. The next phase will require capital markets to play a larger role, creating new financing channels that can support the enormous investment required for AI infrastructure.

As AI reshapes the global economy, the markets that successfully combine technology, energy and capital will be best positioned to lead the next era of digital infrastructure growth.