Author: Li Xiaoyin
Source: Hard AI
The rapid development of AI technology is posing unprecedented challenges to the US power grid, while also bringing new investment themes.
Due to the poor profitability of Berkshire Hathaway Energy (BHE), a company with a generating capacity of up to 30 gigawatts, Warren Buffett stated in this year's letter to shareholders that the future demand for and corresponding capital expenditures of the US power grid will be astonishing. This view has since sparked widespread attention to the stability of the US power grid and future electricity demand.
Buffett's concerns are not unfounded. In recent years, concerns about the &#34;power grid crisis&#34; in the US have been growing. According to US electricity forecasts, the forecasted growth in electricity demand for the next five years has doubled compared to a year ago.
<h2>AI Data Centers to Become Major Power Consumers</h2>
Some media outlets have previously commented that the US is facing an imminent power grid crisis, which may be more urgent than the climate crisis. The article points out that new AI data centers, factory production subsidized by the federal government, and the government's push for the transformation of electric vehicles will be the main reasons for the surge in electricity demand.
In particular, AI data centers are expected to exert unprecedented pressure on the US power grid.
Currently, renewable energy sources such as wind and solar energy cannot provide this stability. It takes 10 to 12 years to build transmission lines connecting remote renewable energy sources to the grid. Therefore, electricity will play a major role in supplying data centers in the medium to short term.
Moreover, data centers supporting AI training consume up to 80 kilowatts, while those supporting AI inference consume half of that, and data centers without AI requirements consume only one-eighth. Therefore, without more efficient training and inference processes, further development of AI will lead to increased consumption of data centers and electricity demand.
<h2>The Next AI Trade - Energy</h2>
Against this backdrop, investors and analysts are seeking new investment opportunities related to AI.
Goldman Sachs analyst Louis Miller pointed out in a recent research report that as AI infrastructure construction advances, the electricity infrastructure and energy sector will become a major trading theme:
<blockquote>
&#34;The focus of the entire AI stack has shifted from software and AI tools to hardware, where the 'winner takes all' status may decrease, and many existing companies in the market have the potential to profit from AI. Therefore, we expect industries oriented towards hardware and electricity to catch up with the broader AI theme.&#34;
</blockquote>
Specifically, Miller believes that attention can be focused on four major areas: smart grid infrastructure (weighting 30%), non-regulated utilities (weighting 25%), power generation materials (weighting 25%), and other utilities (weighting 20%). Companies in these areas may gain new orders and sources of revenue due to the growth in electricity demand
In addition, the report states that with the growth in electricity demand, the demand for copper and uranium and other power generation raw materials is also expected to surge.
It is reported that these materials are crucial for electricity transmission and the construction of new nuclear power plants. Therefore, producers of related materials may also become new favorites for investors.
The report concludes:
<blockquote>
&#34;In the short term, whether software companies successfully launch their AI products or not, hardware and power companies will emerge victorious in this competition.&#34;
</blockquote>

Goldman Sachs pointed out that with the advancement of AI infrastructure construction, the electricity infrastructure and energy sector will become a major trading theme

The next AI trading - Energy