Recently, investors have begun to notice that currently booming generative AI technology requires a large amount of electricity.

Given the prospects for future growth in electricity demand, US stock investors are now looking at companies that own and operate power plants.

$Vistra Energy (VST.US)$,$Constellation Energy (CEG.US)$,$NRG Energy (NRG.US)$Due to its layout in the power sector, stock prices rose 78%, 58%, and 30% respectively during the year; companies that build transmission lines for power grids and data centers$Quanta Services (PWR.US)$The cumulative increase was over 20%.

AI bosses and Wall Street speak out collectively: There's a shortage of electricity

According to the latest report published by Digital Information World, the energy consumption generated by the data center to train AI models will be three times that of conventional cloud computing, and OpenAI training GPT-3 consumes 1.287 GWh of electricity. Maintaining the operation of ChatGPT's data center also consumes huge amounts of electricity, and the monthly electricity cost is close to 2 million US dollars. According to the International Energy Organization, the electricity consumption of global data centers will double in 2026, which is on par with the electricity consumption of all of Japan.

Not long ago, in an interview with the media, Musk predicted that the AI industry will experience a shortage of electricity next year, and “there will not be enough electricity to run all chips” at that time.

OpenAI's Sam Altman once “complained” about AI's energy needs, especially electricity. At the Davos Forum at the beginning of the year, he said,

The development of AI requires breakthroughs in energy, and AI will bring electricity demand far exceeding expectations.

Bill Gates said,

Electricity is the key to determining whether data centers can be profitable, and the amount of electricity AI consumes is astonishing. The use of AI will drive energy demand, and in the next few years, the development of AI may be constrained by chip design and power supply.

Dama's latest research report also predicts that global generative AI demand for electricity may double. The bank is expected to increase the utilization rate of GPU/custom chips from 60% to 70%, and it is expected that the percentage of electricity generated by renewable energy in data centers will decrease, and traditional energy will play a greater role.

Damo pointed out that as AI demand grows and computing power costs decrease, demand for power infrastructure and power resources will increase dramatically, and companies that can meet AI's rapidly growing energy needs, especially power solution providers that can reduce data center power supply delays, will be welcomed by the market.

At a time when demand for AI electricity is soaring, which companies are expected to usher in a new round of opportunities?

According to a recent research report released by Morgan Stanley, the cost of computational power for generative artificial intelligence is falling rapidly, but there will be a significant mismatch between the rapid increase in AI demand and the slow growth of power infrastructure.

Damo figuratively likened this mismatch to a “turtle and rabbit race,” implying that the current boom in the AI field is like a rabbit riding to the ground, yet the growth in AI demand is limited by electricity supply. In this context, more data centers will pay higher electricity price premiums because they want to secure electricity faster. As a result, Morgan Stanley is optimistic about the prospects of the power stock sector and has raised the target prices of various power companies.

US power generation and retailer$Vistra Energy (VST.US)$The increase during the year has reached 78%, but Damo still raised the target price from 62 US dollars to 78 US dollars, which means that there is still potential room for a 14% increase compared to the current latest price; in addition, major US power companies have increased by about 58% this year$Constellation Energy (CEG.US)$Damo was also raised to a target value of 193 US dollars, with a potential increase of about 4%.

America's largest developer of wind and solar farms$NextEra Energy (NEE.US)$, a diversified energy company in the United States$Public Service Enterprise Group (PEG.US)$It was worth as much as $25 and $70, respectively, and the potential increases were over 20% and 14%, respectively.

Furthermore, although solid oxide fuel cell giants$Bloom Energy (BE.US)$Poor performance during the year, with a cumulative decline of more than 25%, but Damo is still bullish as high as $22, with a potential increase of 99%; the largest independent e-commerce provider in the US$The AES Corp (AES.US)$It was estimated to be as high as $25, with a potential increase of about 45%. The company also recorded a 9% decline during the year.

Coincidentally, market strategists analyzed that the development of AI technology is beneficial to energy assets:

More and more people are beginning to realize that large-scale artificial intelligence server clusters will require a large amount of energy. This is increasing the interest of some investors, and they are beginning to expand the scope of investment to related energy fields such as electricity, oil and gas, and nuclear energy is also beginning to receive attention.

Market views, companies that build transmission lines for power grids and data centers such as$Quanta Services (PWR.US)$, and those that provide on-site fuel cell power$Bloom Energy (BE.US)$It is also expected to benefit from the AI wave.

Cow friends,

Are you optimistic about electricity stocks?

Welcome to leave your thoughts in the comments area~

Editor/Somer