Malaysia is rapidly transforming into a regional hub for data centers, particularly those focused on artificial intelligence. However, this technological boom carries serious infrastructure risks. According to my analysis of data from the ASEAN+3 Macroeconomic Research Office, the strain on electricity supply, water resources, and the labor market has already reached critical levels.
Water Footprint of AI Infrastructure
A single 100 MW data center consumes approximately 4.2 million liters of water daily. For comparison, this is equivalent to the daily water consumption of a small town. Given that about 80% of all operational capacity in the country is concentrated in the state of Johor, the region faces a threat of freshwater scarcity. This is especially acute during dry seasons when cooling systems operate at full capacity.
Energy Collapse by 2029
Current projections indicate that by 2029, the total capacity of Malaysian data centers could grow to 3–4 GW. This is comparable to the output of four large nuclear power plants. The national power grid, already under strain from aging coal-fired plants, may not cope with such a surge in demand. Without large-scale investments in renewable energy and grid modernization, the country faces rolling blackouts.
Labor Market: Competition for Engineers
Beyond resources, the competition for skilled personnel is intensifying. Specialists in cooling, electrical systems, and AI infrastructure management are becoming a scarce commodity. Salaries in this sector are already 30–40% higher than the market average, draining talent from other sectors of the economy, including traditional manufacturing and agriculture.
My expert assessment: Malaysia is following the path of Singapore, which previously faced similar problems and imposed a moratorium on the construction of new data centers. If Kuala Lumpur does not take preventive measures—such as introducing strict water consumption standards and incentivizing the transition to renewable sources—the country risks losing its status as an attractive jurisdiction for AI investments. In the long term, this could lead to a slowdown in technology adoption in the region.