Source: Wall Street News
According to industry reports, the forecast for the next five years of US electricity demand growth is double that of a year ago. By 2030, AI will consume more electricity than households.
AI is becoming the biggest topic at the world's top energy summits.
Last week, one of the world's largest energy summits, the S&P Global Energy Conference CeraWeek was held in Houston, Texas, USA. Energy executives from around the world talked endlessly about the potential of AI.
They said AI is not only expected to help improve exploration, drilling and pumping technology, but most importantly, it is creating a huge new demand for the energy industry.
At a time when AI is booming, is energy demand doubling?
The predictions are pretty amazing.
John Ketchum, CEO of US utility giant NextEra Energy, said that although electricity demand has remained flat over the past ten years, the growth rate forecast for the next five years has increased by 81%.
After a long period of stagnation in electricity demand, US utilities are raising their electricity demand forecasts to alarming levels. According to a report by Grid Strategies Consulting, the forecast for the next five years of growth in electricity demand in the US is double what it was a year ago.
Wall Street has heard this premise that generative AI technology requires a large amount of electricity. ChatGPT, a chatbot owned by OpenAI, consumes more than 500,000 kilowatt-hours of electricity every day to process about 200 million user requests, which is equivalent to more than 17,000 times the daily electricity consumption of American households.
Toby Rice, the chief executive of America's largest gas producer, quoted a forecast that by 2030, AI will consume more electricity than households.
Bill Vass, vice president of engineering at Amazon Web Services, mentioned that the world adds a new data center every three days. Microsoft co-founder Bill Gates said at the conference that electricity is a key investment that determines whether data centers can be profitable, and the amount of electricity AI consumes is amazing.
“This will be a transformation,” said ConocoPhillips CEO Ryan Lance. “The impact will be huge, and it will impact every business here.”
Is the end of AI “natural gas, coal, and nuclear power plants”?
There is consensus on the view that AI consumes a lot of electricity, but there is still controversy over which channels it should be obtained from.
Former US Secretary of Energy Ernest Moniz said that some utilities are puzzled by the size of new data centers and proposed data centers, but wind and solar farm construction is becoming more challenging, and they are concerned about how to provide enough electricity in this situation.
Moniz said utilities will have to rely more on fossil fuels such as natural gas and coal and nuclear power plants, and may also have to build new gas plants to help meet surging demand.
“We can't build 100 gigawatts of new renewable energy in a few years. Businesses are stumped by this problem.”
Rice said the technology company that built the data center is consulting on purchasing natural gas from EQT. Tech companies need reliable electricity, but renewable energy sources like wind and solar don't always provide enough electricity due to the erratic weather. However, the construction of large nuclear power plants has always been expensive and time-consuming; only one is currently under construction in the US.
“The construction of this kind of infrastructure (nuclear power plant) takes seven to ten years, and the technology industry is impatient. So all that's left is gas.”
Demand for AI power surges, pressure on power grids intensifies
The core issue is that companies are not only looking for new energy, they are also pursuing clean energy. Many tech companies and utilities have promised to drastically cut their carbon emissions.
Dominion Energy in Richmond, Virginia, has seen a sharp increase in power demand driven by data center construction in the Northern Virginia region, which has long been a hub for a large number of data centers.
To support this growth, the company has set a goal of achieving net zero carbon emissions by 2050 and plans to build at least one gas-fired power plant.
“We will achieve net zero emissions by 2050. “We are still convinced of this,” said CEO Robert Blue, “but today's growing demand complicates this goal even more.”
Notably, while AI is driving a surge in electricity demand, other factors are also increasing the pressure on the grid.
Stimulated by new tax policies under the Inflation Reduction Act, the entire US is building a series of factories, and many states are also actively promoting the use of more electricity in transportation, heating, and heavy industry.
Compared to power facilities, the rapid construction of data centers has led to tight supply. According to data from commercial real estate services company CBRE Group, the construction period of the data center has been extended by 2 to 6 years due to delays in electricity supply.
Meanwhile, the Biden administration has set a goal of eliminating carbon emissions from the US electricity industry by 2035. John Podesta, the president's key figure responsible for implementing the Inflation Reduction Act, told the media that growing demand for AI poses new challenges to achieving this goal.