Chipmaking giant Nvidia reported revenue of $35.08 billion in the third quarter, far outstripping economists’ expected $33.2 billion. Revenue is up 97% year-on-year for the period ending Oct. 27. Net income rose to $19.3 billion compared to $9.24 billion in the same period last year. Driving the surge is its AI business, which includes chips and data centers.
Nvidia also projected $37.5 billion in sales for the current quarter, slightly above the expected $37.08 billion. The results mean the US chipmaker has now exceeded analysts’ expectations for the eighth quarter in a row, although they also mark a fourth consecutive slow-down in sales for the chipmaker overall, CNBC noted.
The world’s largest company by market cap, Nvidia is seen as a bellwether of the direction of the tech industry, and in particular, whether investor frenzy over AI could pay off.
Nvidia has more than $3 trillion market value and that size means it is “hard to overstate how big Nvidia’s third-quarter earnings are for the short-term direction of the stock market,” Bloomberg wrote, noting that Nvidia is among the world’s most valuable companies, the best example of the returns AI can bring investors, and its value has skyrocketed in the last year. The third-quarter report is expected to move the market more than other economic data or official resolutions, including the US Federal Reserve’s upcoming decision on interest rates, Barclays strategists told the outlet. Some traders predicted that the earnings could cause a swing of 8.5% to the company’s shares, or about $300 billion in gains or losses, bigger than all 25 companies in the S&P 500, after the report.
Nvidia’s rivals see innovation as a means to stay competitive, The Associated Press reported. The company’s graphics processor chips are great for building AI models from scratch, but competitors see opportunity in inference chips, which are both cheaper than Nvidia’s GPUs and better at running an already trained AI model day-to-day. “The broader the adoption of these models, the more compute will be needed for inference and the more demand there will be for inference chips,” an analyst told the AP. A future potential challenge for Nvidia will be competing with its clients who are making their own chips, CNBC noted, including its biggest customers Microsoft, Amazon, and Google: “It’s just too big of an opportunity,” one chip startup co-founder said.
While the earnings provided a positive outlook for the company’s next-generation GPU Blackwell, the chip isn’t “foolproof,” Business Insider wrote, and reports of it overheating spooked the market earlier this week. While experts dismissed the concerns, Nvidia’s customers still need to invest heavily in cooling and a huge amount of energy for its chips to function. As of March, AI had already consumed the same amount of energy as a small country, Vox reported. Another concern among investors is Nvidia’s goal to release new chips every year: Clients’ budgets could take a hit if they can’t spread their investment over several years, BI added, with some analysts reducing revenue estimates for some of the largest buyers of AI chips.