So, Jensen Huang, what do you have to say in your defense?
Possibly the chairman and CEO of Nvidia (NVDA) It doesn't have to say much of anything, as the ai semiconductor company left Wall Street with an impressive fourth-quarter update.
Nvidia, now the world's third-largest stock with a market value of more than $2 trillion, managed to beat already lofty sales and earnings estimates with a report that left analysts' expectations in the dust.
The chipmaker has been described by Bank of America as the “leader of the ai gold rush,” while Goldman Sachs has called Nvidia stock “the hottest stock on planet Earth.”
The company's surprising rise on February 22, which added more than $277 billion to the ai chip maker's market value (the largest market gain in U.S. history), helped boost the S&P 500 to a historical maximum of 5,087.03 points.
Nvidia earned an adjusted $5.16 per share, nearly six times more than the same period a year earlier, as revenue tripled to $21.1 billion.
Looking ahead, the company expects revenue for the first quarter of fiscal 2025 to reach $24 billion, up 9% from the fourth quarter and a gain of 233% from the first quarter of the previous year.
CEO Huang welcomes shift to 'accelerated computing'
“Many investors are shocked by how Nvidia has become a $700-plus stock, but Nvidia got to where it is thanks to extremely strong earnings and revenue,” said James Demmert, chief investment officer at Main Street Research.
“There is nothing out of the ordinary about Nvidia's valuation or its recent earnings, as the company is delivering on its earnings,” he said.
Related: Analysts Renew Nvidia Price Targets as Stock Tests $2 Trillion
He added that when a company posts 265% year-over-year revenue growth, as Nvidia did, it deserves a higher valuation.
“Many investors have been wrongly underestimating the 'E,' or earnings, part of Nvidia's price-to-earnings ratio,” Demmert said.
He noted that Nvidia hit about $330 per share in November 2021, before the bear market and when there was little demand for ai.
The stock has doubled since then and demand for artificial intelligence is now significant.
Huang, who co-founded the company in 1993, told investors that the next wave of investments in artificial intelligence “will open up a whole new world of applications that are not possible today,” while creating a market worth “hundreds of thousands of dollars.” millions of dollars”. annual.
He said the conditions for continued growth in 2024, 2025 and beyond are excellent.
“And let me tell you why,” he said during company earnings call. “We are at the beginning of two industry-wide transitions, and both are industry-wide. The first is a transition from general computing to accelerated computing.”
'A whole new type of data center'
Huang said general-purpose computing “is starting to lose steam.”
He said there's no reason to upgrade corporate and cloud networks with more CPUs (central processing units, the main component of a computer) “when you can't fundamentally and dramatically improve their performance like before. And that's why you have to speed up all”.
Huang believes that Nvidia graphics processing units or GPUs are the best solution. Coupled with purpose-built software, they are better suited to the large workloads associated with training and running ai applications.
More ai actions:
- Analyst Reveals New Broadcom Stock Price Target Linked to ai
- ai stocks Soar on New Guidelines (It's Not Nvidia!)
- Big tech stocks are doubling down on ai
“This is what Nvidia has been pioneering for some time,” Huang said. “And with accelerated computing, you can dramatically improve your energy efficiency. You can dramatically improve your cost of processing data by 20 to 1. Huge numbers. And, of course, speed.”
The company's data center division, which includes the company's H100 graphics cards used to power generative ai applications like OpenAI's ChatGPT, reported sales of $18.4 billion, a year-over-year growth of 409%.
Huang described a completely new industry where, for the first time, a data center is not just about computing data, storing it, and serving a company's employees.
“We now have a new type of data center that deals with ai generation, an ai generation factory,” Huang told analysts. “And you've heard me describe them as ai factories. But basically, you take the raw material, which is data, and transform it with these ai supercomputers that Nvidia builds and turn it into incredibly valuable tokens.”
These tokens, he said, “are what people experience in the amazing ChatGPT or Midjourney or searches these days are boosted by that.”
“Biology companies, healthcare companies, financial services companies, ai developers, big language model developers, autonomous vehicle companies, robotics companies… All these startups, big companies, healthcare, services financial, automobile and so on are working on the Nvidia platform,” Huang said. “We expect demand to continue to be stronger than what our supply offers and, throughout the year, we will do the best we can… right now, we are increasing the H200s. There is no way we can reasonably keep up Of demand. “
“We believe these two trends (accelerated computing and generative ai) will drive a doubling of the world's installed base of data center infrastructure over the next five years,” he said.
Related: A veteran fund manager picks his favorite stocks for 2024
