① Wall Street now thinks it's time to let go of its obsession with Nvidia; ② Damo is optimistic about power generation companies such as Bloom Energy; ③ Bank of America believes that other independent power producers, including nuclear power plant owners, will also benefit.
As the global artificial intelligence (AI) competition continues to heat up, Nvidia's stock price as the biggest beneficiary continues to soar. But Wall Street now thinks it's time to let go of it$NVIDIA (NVDA.US)$I'm obsessed.
Morgan Stanley analyst Stephen Byrd said a few days ago that artificial intelligence is changing the data center landscape and uncovering a new infrastructure requirement — power generation producers. This is a hidden investment opportunity, but it is currently underestimated by the market.
“Opportunities are around infrastructure... in everything from power generation, power electronics, data centers, etc. — the rate of growth will accelerate.” he said.
Generative artificial intelligence processing is usually done on graphics processing units (GPUs), which require more power under higher computational requirements. Demand for more electricity is driving rapid growth in data centers across the US, putting pressure on an already strained power grid.
Timothy Fox, general manager of ClearView Energy Partners, said, “Data centers not only facilitate the development of artificial intelligence, but are the backbone of our industry, commerce, transportation, and health. These are all mission-critical infrastructures. So it's important that while we promote this new industry, we don't let the existing industry turn off its lights.”
The International Energy Agency (IEA) estimates that the electricity consumption of global data centers may jump from 460 terawatt-hours in 2022 to more than 1,000 terawatt-hours in 2026, which is roughly equivalent to the electricity consumption of Japan as a whole.
Byrd said the pair$Bloom Energy (BE.US)$This is a potential boon for power generation companies. Fuel cell companies, he said, are a group of companies “capable of meeting the rapidly growing power needs of generative artificial intelligence drivers.”
“Although the current (Bloom Energy) stock is underperforming, this technology allows data center developers to get new centers up and running very quickly,” he said. “The time required to launch is absolutely critical, and it is economically valuable... Bloom Energy's fuel cells can be deployed within 50 days.”
Bloom Energy CEO KR Sridhar told analysts during last quarter's earnings call that he sees artificial intelligence data centers as the “biggest single area” for the company's growth over the next decade.
“The sales channels in this area alone are huge, not in megawatts, but in gigawatts.” he said.
In addition to fuel cell generator manufacturers, other independent power producers, including nuclear power plant owners, are also preparing to grow from future data center construction.
Bank of America research analyst Paul Cole believes$Constellation Energy (CEG.US)$, public service electricity and gas companies$Public Service Enterprise Group (PEG.US)$, energy companies$Vistra Energy (VST.US)$,$NextEra Energy (NEE.US)$und$Dominion Resources (D.US)$It is a “clear beneficiary” of the increase in electricity demand.
Although AI's energy consumption has soared over the past year, Wall Street professionals believe it has “only touched the surface” in terms of energy consumption.
“In many years to come, in terms of electricity growth, artificial intelligence will grow at an annual rate of around 100%. As computing costs drop, new use cases emerge, so the demand for computation is only getting higher. We're still in the early stages of this, and investors should now start looking at the value chain that powers these data centers,” Cole added.
Editor/Jeffrey