Nvidia: Profit for the year jumps to 117 billion, fourth quarter beats estimates

Wall Street was looking to Nvidia's accounts for yet another indication of the hundreds of billions of dollars earmarked for data centre infrastructure. It should not be forgotten, in fact, that big companies such as Alphabet, Microsoft, Amazon, and Meta are expected to spend a total of at least $630 billion in 2026, with the bulk of the resources going into data centres and processors.

In all this, however, signs of increased competition are now emerging for Nvidia, which remains the industry leader, but with some growing insidiousness. Rival AMD will unveil a new flagship AI server later this year and has secured contracts with long-standing Nvidia customers such as Meta. Google, on the other hand, has entered into an agreement to supply its TPU chips to Anthropic, developer of the chatbot Claude (ChatGPT's most accredited antagonist), and is in talks to supply Meta as well, according to press rumours. The large technology groups are also strengthening their in-house development of semiconductors for their data centres.

For Nvidia, in short, the road may no longer be as smooth as up to this point in history, where its chips have made the difference. But the numbers are still on its side.

The Californian company pointed out that its forecast for the current quarter does not include expected revenues from the sale of data centre chips in China. The company did, however, report that it received licences from the US government this month to ship 'small quantities' of H200 chips to Chinese customers. Analysts had considered the possible resumption of sales in China after the export restrictions imposed by Washington. And AMD, for instance, re-instated sales of AI chips to China in its estimates after obtaining licences for modified versions of its processors.

Finally, the company stated that it has ensured sufficient inventory and production capacity to meet demand beyond the next few quarters, and that it will include stock-based remuneration in its non-GAAP financial metrics, amid strong competition to attract engineers and researchers specialising in AI.

"Business adoption of agents is soaring. Our customers are racing to invest in AI processing, factories fuelling the AI industrial revolution and their future growth,' Huang said. The big word, in short, remains optimism.