① NextEra Energy Inc., the largest developer of wind and solar power plants in the US, predicts that with the booming development of artificial intelligence, the growth rate of US electricity demand will soar sharply, and the annual growth rate may increase 81% year over year; ② OpenAI CEO Sam Altman and Tesla CEO Musk have issued similar warnings before.
Financial Services Association, March 19 (Editor Liu Rui) As the boom in artificial intelligence continues, the huge energy demand that this revolutionary technology may bring is also drawing more and more attention.
On Monday EST, NextEra Energy Inc., the largest developer of wind and solar farms in the US, predicted that with the booming development of artificial intelligence, the growth rate of electricity demand in the US will soar sharply to keep up with the development of artificial intelligence, electrification, cloud computing capabilities, and chip factories.
The demand for electricity in the US will suddenly soar
The US Energy Information Administration estimated in its long-term annual outlook released last year that the current annual growth rate of electricity demand in the US is less than 1%.
However, according to estimates by John Ketchum (John Ketchum), CEO of renewable energy giant NextEra Energy, the annual growth rate of electricity demand will accelerate to around 1.8% under the influence of artificial intelligence technology.
Ketchum said at the S&P Global Energy Week conference on Monday: “(US) electricity demand has been relatively stable for many years, but suddenly the electricity demand growth rate is expected to increase rapidly by 81% over the next five years.”
Due to extreme temperatures and climate change, the US power grid is already strained, and almost every winter in recent years has faced the risk of large-scale power outages. And with the continued development of AI, the sharp rise in electricity demand may put further pressure on the US power grid.
AI requires far more electricity than expected
In fact, Ketchum wasn't the first business celebrity to warn of the huge energy demands brought about by AI technology.
At the World Economic Forum earlier this year, ChatGPT creator Sam Altman, CEO of OpenAI, confessed: “We really need more energy than we thought before... we're still unaware of the energy requirements for this technology (artificial intelligence).”
Recently, Tesla CEO Musk also predicted that in order to meet the needs of artificial intelligence and electric vehicles, there will be electricity shortages in the next few years:
“The limitations of AI computing are very predictable... A year ago, the shortage was chips, neural network chips. Well, it's easy to predict that the next shortage will be power infrastructure (such as transformers). You have to give these things energy.”
In September of last year, Boston Consulting Group released an analysis report stating that by the end of 2030, the electricity consumption of US data centers alone is expected to triple that of 2022. This increase is mainly driven by two key demand factors for AI model training and services for more frequent AI queries.
According to the Boston Consulting Group report, data center electricity consumption accounted for 2.5% of the total electricity consumption in the US (about 130 terawatt-hours) in 2022, and is expected to triple to 7.5% (about 390 terawatt-hours) by 2030. This is equivalent to the electricity consumption of about 40 million American households, or the electricity consumption of one-third of all American households. The group also expects generative artificial intelligence to account for at least 1% of the new electricity consumption in the US.