Hong Kong must resolve acute electricity supply constraints to succeed in its goal of becoming a global artificial intelligence hub.

Addressing these energy needs is critical because AI training and inference consume vast amounts of power. Without a stable and scalable energy strategy, the city risks limiting the development of the very technologies it seeks to lead.

Currently, Hong Kong ranks third globally in the Global AI Competitiveness Index as an AI financial powerhouse [1]. This position places the city among the top tier of global markets, yet the physical infrastructure required to support high-compute workloads remains a primary concern.

The demand for power is driven by the intensive nature of AI workloads, which require significant electricity for both the initial training of large models and the subsequent inference process. These requirements place a heavy burden on the existing power grid, creating a potential bottleneck for industry growth.

Industry leaders and government officials are looking toward the Greater Bay Area region to find sustainable energy solutions. The integration of regional power resources could provide the necessary capacity to support expanding data centers, and AI research facilities.

Failure to prioritize energy infrastructure could undermine the city's competitive edge. As AI organizations transition from experimental phases to frontier success, the ability to secure reliable energy will determine whether Hong Kong can maintain its ranking and attract further investment in the sector.

Hong Kong ranks third globally in the Global AI Competitiveness Index as an AI financial powerhouse.

The tension between Hong Kong's high software and financial competitiveness and its physical energy limitations highlights a common struggle for urban AI hubs. While the city possesses the capital and talent to lead in AI, the reliance on external power sources—particularly from the Greater Bay Area—means its technological sovereignty is tied to regional energy policy and infrastructure stability.