When Nvidia lost $ 600 billion in market value in one day last month, this was because some investors feared the future of the artificial intelligence chip maker. Deepseek, a Chinese start-up, said it had made its AI systems with a small fraction of the AI chips used by other companies, and at a small part of the costs.
On Wednesday, Nvidia showed that those fears were exaggerated, even when the brute pace of its growth slows down. The company, a Bellwether for AI, said that purchases from his AI chips are total turnover by 78 percent increased from a year earlier to $ 39.33 billion during the three months that ended in January. The profit increased by 80 percent to $ 22.09 billion.
In the previous quarters, Nvidia reported that sales and profit were more than doubled. But continuing to deliver that kind of profits has become more difficult as the sale and profit rise. (For a company such as Nvidia, it is common for the growth rate to slow down after a period of phenomenal increase, a phenomenon that is known as the law of large numbers.)
The quarterly results of Nvidia exceeded the expectations of Wall Street analysts for $ 38.32 billion in turnover and $ 21.08 billion in profit. The company projected that the turnover in the current quarter would rise 65 percent from a year ago to $ 43 billion, a delay compared to the previous quarter, but had predicted around $ 1 billion more than Wall Street.
Shares in Nvidia were almost close to the trade after the hours, after a profit of 3.7 percent on Wednesday. It remains the world's second most valuable listed company, behind Apple.
The NVIDIA company has been encouraged by the non -stop editions of the largest technology companies on AI data centers. After casting tens of billions of dollars in new infrastructure last year, Amazon, Microsoft, Alphabet and Meta said that this year they will each spend $ 65 billion to $ 100 billion or more.
Much of that money will flow directly to Nvidia. According to some estimates, the company controls 90 percent of the market for the graphic processing units, or GPUs, that Power AI systems. It rolls out a new, more powerful series of AI chips that is known as Blackwell and charges $ 60,000 to $ 70,000 for a characteristic chip. Blackwell contributed $ 11 billion in sales during the quarter.
How long the build-out of data centers has been a question since the start of the AI tree. Deepseek disputed a consensus of the technical industry that to build larger and better AI systems, companies should build larger and more powerful data centers. It caused the fear that companies can withdraw from their spending with Nvidia.
Since then, a new consensus has emerged that Nvidia will continue to benefit because it will be affordable for more companies to develop AI systems. An extensive area of AI companies would create more customers for the expensive chips of Nvidia, no less, as initially feared.
Nvidia said that the income from Datacenter, which includes the sale of chips, cables and powerful computing, in the quarter of a year earlier by 93 percent rose to $ 35.58 billion.
“Nobody supports his dedication to build AI data centers and the power to generate more electricity for those data centers,” said Brian Mulberry, customer portfolio manager at Zacks Investment Management, a financial company that follows Nvidia. “Everyone is still competing.”
Although Nvidia has recovered since the end of January, the concern about the expenditure for data centers will continue to cradle the shares. Shares fell nearly 6 percent at the beginning of this week after Wall Street analysts reported that Microsoft had canceled some data center contracts. Microsoft did not respond to requests for comments.
Jensen Huang, Chief Executive of Nvidia, said in a call with analysts on Wednesday that the company would benefit because companies would change the way they developed AI because the Blackwell chips were more powerful and versatile than their predecessor. It also has visibility in the amount of money spent on data centers, which he said would increase as companies introduced new tools such as AI agents who can perform autonomously tasks, such as online shopping.
“The next wave is coming,” said Mr. Huang. He added: “We are in the middle of this development and we can see great activity happening.”
“We have successfully performed the massive scale production of Blackwell AI-SUPERCOMPUTERS, with billions of dollars in sales in the first quarter. AI goes with light speed. “
Nvidia also remains confronted with geopolitical challenges. The Biden administration has pressed the sale of its chips to China and introduced a rule to limit the sale of its AI chips to more than 100 countries. Howard Lutnick, the Trump Government trade secretary, told the congress during his confirmation hearing that technology companies, including Nvidia, “should stop helping” China.
Since the US government started limiting the chip export, Nvidia's turnover in China has halved to around 15 percent of the total turnover. The company said it was not expected that the sale would return to China.
At the end of last month, Mr. Huang visited the White House to discuss AI and semiconductors with President Trump. Mr. Trump afterwards told reporters that he could not say whether he would ban China's chip sale to China. He has also threatened rates for semiconductors, including the chips of Nvidia in Taiwan.
Colette Kress, the financial chief of Nvidia, said the company was watching rates. “It is a stranger until we further understand what the US government's plan is, both the timing, the true and how much,” she said. “We are waiting for.”
