Southeast Asia Eyes AI Infrastructure as New Revenue Driver

Infrastructure Over Adoption

Southeast Asia's AI opportunity extends beyond deploying chatbots and productivity tools. The region is positioning itself to capture returns from the physical infrastructure that powers artificial intelligence, from data centres to semiconductor supply chains.

UOB outlined this shift during the Asean Conference 2026 in Singapore last week, emphasizing that financial institutions must act as "superconnectors" to mobilize capital across borders. Edmund Leong, head of group corporate banking and sector solutions at UOB, described AI infrastructure as an economic growth story rather than purely a technology play.

The demand stems from computing requirements. Training and running AI models requires massive server capacity, which in turn drives investment across the entire ecosystem, from chips to cooling systems to power grids.

Data Centre Build-Out Accelerates

Malaysia has emerged as the region's leading data centre hub, with Thailand, Indonesia, and Vietnam expanding capacity. But building these facilities requires more than land and construction. Power generation, cooling infrastructure, and telecommunications networks must all scale in parallel.

UOB estimates that approximately $150 billion could flow into regional energy infrastructure over the next five years, covering renewables and grid upgrades necessary to support data centre growth.

Financial institutions are stepping beyond traditional lending roles. Leong explained that banks now mobilize capital through multiple channels, including loans, bonds, and equity, depending on project structure and risk profile. Regional banks with cross-border footprints have become particularly valuable as data centre operators manage facilities across multiple markets, generating large capital flows between jurisdictions.

Not every project qualifies for institutional backing. UOB prioritizes clients with operational expertise, technical execution capability, and committed long-term shareholders. Given the capital intensity of AI infrastructure, investor commitment matters as much as technology.

Hyperscalers Drive Regional Investment

Large cloud service providers are leading the regional build-out, with investments concentrated in infrastructure that supports agentic AI systems. These systems now handle entire workflows autonomously, moving beyond simple query responses.

Amazon Web Services noted that supporting such operations requires purpose-built infrastructure, including data centres powered by renewable energy and custom chips designed for efficiency. John Kain, director of financial services and market development at AWS, emphasized that companies scaling AI successfully focus on business outcomes first, then align technology investments with those goals.

Governance frameworks built into AI deployment from the start allow organizations to experiment while ensuring compliance as regulatory standards develop. Kain described this approach as promoting innovation rather than constraining it.

Power and Chips Remain Constraints

Two bottlenecks threaten to slow the regional build-out. The first is power supply. Southeast Asia must accelerate investment in renewable energy and grid infrastructure to support data centre expansion. Leong sees potential for the region to optimize power distribution across borders, leveraging natural resource availability.

The second constraint is semiconductor supply. Demand for chips and memory components continues to outpace manufacturing capacity, driving up costs and limiting deployment speed. However, technological improvements and capacity expansion should gradually ease supply pressure, making infrastructure more affordable for enterprises.

As those constraints relax, broader AI adoption across the economy becomes feasible. The region's ability to attract and deploy capital for infrastructure will determine whether Southeast Asia captures value from the AI wave or simply imports technology built elsewhere.

Regional Coordination Required

Successful AI infrastructure projects depend on coordination among multiple parties. Local regulators, utility providers, telecommunications companies, and financial institutions must align to move projects from planning to operation.

Banks like UOB position themselves as enablers that connect these stakeholders, ensuring projects have the regulatory approvals, power commitments, and network capacity required to succeed. This coordination role becomes more critical as projects span multiple jurisdictions with different regulatory frameworks and infrastructure readiness.

The question for Southeast Asia is execution speed. If power and semiconductor constraints ease over the next several years, the region could establish itself as a key node in global AI infrastructure. If bottlenecks persist, investment may shift to markets with more reliable supply chains and energy capacity.