Hong Kong, 15 January 2026 – The global data centre sector is set for an unprecedented period of expansion, with total capacity forecast to almost double from 103 GW to 200 GW by 2030, according to JLL’s newly released 2026 Global Data Centre Outlook. Artificial intelligence (AI) is rapidly reshaping the industry, and JLL expects AI workloads to account for half of all data centre capacity by the end of the decade.
The surge will require up to USD3 trillion in total investment over the next five years, including USD1.2 trillion in real estate asset value creation and approximately USD870 billion in new debt financing, ushering in an infrastructure investment supercycle.
“We are witnessing the most significant transformation in data centre infrastructure since the original cloud migration,” said Matt Landek, Global Division President, Data Centres and Critical Environments at JLL. “The sheer scale of demand is extraordinary. Hyperscalers alone are committing USD1 trillion to data centre spend between 2024 and 2026, while supply constraints and grid connection delays of up to four years are creating a perfect storm that is fundamentally changing how we approach development, energy sourcing and market strategy.”
AI drives transformation
AI workloads are expected to account for 50% of total data centre capacity by 2030, up from around 25% in 2025. JLL forecasts a critical inflection point in 2027, when AI inference workloads will surpass training as the dominant requirement.
“We are witnessing the emergence of a new infrastructure paradigm, where AI training facilities demand 10x the power density and command lease rate premiums of up to 60% over traditional data centres,” said Andrew Batson, Global Head of Data Centre Research at JLL. “Beyond the economics, AI has become a matter of national strategic importance, prompting countries to develop domestic capabilities through sovereign infrastructure investments, representing an estimated USD8 billion CapEx opportunity by 2030.”
AI chips are projected to increase their share of semiconductor market revenues from 20% to 50% by 2030, with custom silicon expected to capture 15% as hyperscalers design proprietary processors. Looking ahead, emerging technologies such as neuromorphic computing could deliver ultra-efficient inference capabilities, reducing infrastructure requirements and enabling data centres to achieve greater power-efficiency.
Regional growth drivers and patterns
Hong Kong is set for a sharp upswing in data centre development over the coming years. Between 2022 and 2025, approximately 1.8 million sq ft of gross floor area (GFA), supporting more than 150 MW of critical IT load, has been completed and ready for operation. Looking ahead, about a further 4.8 million sq ft of potential data centre facility is scheduled for delivery between 2026 and 2029, signalling a substantial increase in new supply starting this year. This surge marks the next major growth cycle for Hong Kong’s data centre market, reinforcing the city’s position as a regional digital infrastructure hub in Asia, underpinned by extensive submarine cable connectivity and comparatively low exposure to natural hazards.
Cathie Chung, Senior Director of Research at JLL in Hong Kong, said: “Under the government’s vision to develop Hong Kong into an international innovation and technology centre, strategic initiatives such as the tender for the Sandy Ridge Data Facility Cluster site and the development plan for the San Tin Technopole reflect the authorities’ active alignment of land supply and infrastructure planning with the needs of the data, cloud, and AI sectors. This creates a favourable development environment for the local data centre industry. With a significant pipeline of new floor space, rapid enterprise digital transformation, and supportive government policy, Hong Kong’s data centre market is well placed to capitalise on the next wave of growth opportunities.”
The Americas are expected to retain their position as the largest data centre region, accounting for around 50% of global capacity and delivering the fastest growth rate through 2030. The Asia-Pacific region is projected to expand from 32 GW to 57 GW, while Europe, the Middle East and Africa (EMEA) will add a further 13 GW of new supply.
Each region faces distinct market dynamics that will shape development strategies. In Asia Pacific, growth is being driven primarily by colocation, while on-premise capacity is forecast to decline by 6% as enterprises accelerate cloud migration. EMEA’s outlook is underpinned by strong hyperscaler demand, with expansion concentrated in established European hubs such as London, Frankfurt and Paris, alongside emerging Middle Eastern markets pursuing ambitious digital transformation strategies. In the Americas, the US continues to dominate, accounting for approximately 90% of regional capacity.
For more news, videos and research resources on JLL, please visit JLL’s newsroom.