CCSN POWER:AI is driving a surge in electricity demand

On May 23rd, at a symposium, the Chairman of State Power Investment Corporation put forward opinions and suggestions on deepening the reform of the power system. The reform of the power system is expected to further deepen, and the construction of a new power market system will continue to accelerate.

The surge in GPU computing power and the addition of AI large models from the training side to the inference side are driving up electricity demand. According to Alibaba Research Institute, it is expected that the total annual power consumption of global GPU computing clusters will increase by about 11 times from 2024 to 2030, with a compound annual growth rate of 52.2%. The higher demand for AI power drives the construction of infrastructure such as power grid equipment. In a recent report released by the International Energy Agency, it is predicted that global electricity demand for data centers, cryptocurrencies, and artificial intelligence will more than double over the next three years, equivalent to the entire electricity demand of a Germany. By 2026, the overall global electricity demand is expected to grow by 3.4%. According to the International Energy Agency, energy consumption in data centers and transmission networks has each accounted for 1.5% of the global total. Their annual carbon emissions are equivalent to Brazil's.

As the technological competition for AI big models enters a deeper stage, the industry is turning its attention to the energy field required for model training. Open AI founder Sam Altman previously stated at a public event that "the future of artificial intelligence depends on breakthroughs in clean energy." NVIDIA founder Huang Renxun pointed out that "the end of AI is photovoltaics and energy storage."

In the future, technological innovation and equipment upgrades are still needed to further improve power generation efficiency, enhance grid transmission capacity and stability, optimize power resource allocation, improve the flexibility of power supply, promote distributed energy systems, and reduce energy transmission losses, in order to meet the energy demand challenges brought about by the development of computing power.

CCSN Power Inc. All Rights Reserved

Posted by: Zhang Jianlong

Reviewed and translated by: Kailon Chung