Nvidia, the king of chips that power artificial intelligence, released quarterly financial results on Wednesday that reinforced how the company has become one of the biggest winners of the artificial intelligence boom, and said demand for its products will fuel continued sales growth.
The Silicon Valley chipmaker has seen extraordinary growth over the past 18 months, driven by demand for its specialized and expensive semiconductors, which are used to train popular artificial intelligence services such as OpenAI’s ChatGPT chatbot. Nvidia has become known as one of the “magnificent seven” technology stocks that, along with others such as Amazon, Apple and Microsoft, have helped fuel the stock market.
Nvidia’s valuation has risen more than 40% to $1.7 trillion since the start of the year, making it one of the most valuable public companies in the world. Last week, the company briefly eclipsed the market values of Amazon and Alphabet before falling to fifth place among the most valuable tech companies. Its stock market gains are largely the result of it repeatedly beating analysts’ growth expectations, a feat that is becoming increasingly difficult as they continue to improve their forecasts.
On Wednesday, Nvidia reported that revenue in its fiscal fourth quarter more than tripled from a year earlier to $22.1 billion, while profits increased nearly ninefold to $12.3 billion. Revenue was well above the $20 billion the company forecast in November and above Wall Street estimates of $20.4 billion.
Nvidia forecast revenue in the current quarter would be about $24 billion, also more than triple the same period a year earlier and higher than analysts’ average forecast of $22 billion.
Jensen Huang, co-founder and CEO of Nvidia, says a sea change to upgrade data centers with chips needed to train powerful AI models is still in its early stages. That will require spending about $2 trillion to equip all buildings and computers to use chips like Nvidia’s, he predicts.
“Accelerated computing and generative artificial intelligence have reached the tipping point,” Huang said in a press release. “Demand is increasing around the world among businesses, industries and nations.”
In an interview later Wednesday, Mr. Huang said there will be much more growth for the company. “We’re a year into generative AI,” he said. “My guess is that we are literally in the first year of a decade-long cycle of diffusion of this technology into every single industry.”
Some analysts had predicted a sell-off after Nvidia’s announcement, a reaction to its rising stock price. But shares rose more than 8% in after-hours trading.
“Despite concerns about its lofty valuation, Nvidia’s unrivaled AI-related intellectual property, rooted in decades of visionary investments, sets it in a league of its own,” Hans Mosesmann, an analyst at Rosenblatt Securities, wrote in a research report before the press conference. he reported the company.
One of the factors behind Nvidia’s latest revenue growth is the ability of the company’s manufacturing partners, led by Taiwan Semiconductor Manufacturing Company, to increase supplies of Nvidia’s flagship AI chip, which has prices ranging from $ 15,000 to $40,000.
On a conference call with analysts on Wednesday, Huang said the availability of such chips has improved significantly, but noted that the company will soon introduce new products that will again be in short supply.
“Every time we have new products, the number increases from zero to a very large number, and you can’t do that overnight,” he said on the call.
But giant cloud computing companies like Amazon, Google and Microsoft are designing their own AI chips to use in addition to Nvidia’s, and rival chipmakers continue to introduce their own AI products.
Intel, which has long dominated the industry in standard microprocessor chips but is a laggard in artificial intelligence, gathered a number of partners and potential customers in Silicon Valley on Wednesday to discuss its plans to offer manufacturing services, which could increase the industry’s ability to build AI chips. Among the attendees was Sam Altman, who relies heavily on Nvidia chips as CEO of OpenAI.
“Intel was once the evil Borg of the industry,” said Daniel Newman, chief executive of Futurum Research, which tracks the semiconductor industry. Now, he said, “companies are banding together to make sure Nvidia doesn’t get too powerful.”
The Biden administration has thrown up another set of hurdles for Nvidia and other U.S. chipmakers, placing restrictions on their chip sales in China. Nvidia responded by selling less powerful versions of some products on the market.
However, the company said Wednesday that its sales in China fell to a mid-single-digit percentage of its data center chip revenue, from 19% in fiscal 2023.
Meanwhile, some experts worry that the global rollout of the company’s expensive, power-hungry chips could overburden countries’ power grids and budgets.
Mr. Huang addressed some of these concerns in February at the world governments summit in Dubai. He said Nvidia’s chips were cheap and efficient compared to using slower standard microprocessors to do the same job—and that much faster chips were on the way, some of which the company is expected to unveil in March.
“If you assume that computers never get faster, you might come to the conclusion that we need 14 different planets, three different galaxies and four more suns to power all of this,” Huang said. “But of course computer architecture continues to advance.”