NVIDIA Q4: AI spurs 265% revenue growth

Computing company NVIDIA, which is at the forefront of AI hardware and software, is riding the AI boom as its revenues and profits soar in the fourth quarter: a sign of heavy investment in the infrastructure behind artificial intelligence. 

Why NVIDIA matters

Seen simply as a business story, NVIDIA’s $24 billion in revenues for the fourth quarter constitutes a year-on-year increase of 265%. Its profits have grown by an even more significant margin, with a 769% year-on-year increase (to $12.29 billion from $1.41 billion). This signifies its position at the core of AI growth across industries. 

NVIDIA, however, is not just central to the technology of AI but also to its applications in advertising, appearing onstage with agency holding group WPP to talk about the creation of AI factories.

But the big news is that it forecasts a “tipping point” for artificial intelligence, in the words of founder and CEO Jensen Huang.

The nature of AI demand and supply 

“Accelerated computing and generative AI have hit the tipping point. Demand is surging worldwide across companies, industries and nations,” said Huang, in a statement. 

“Our Data Center platform is powered by increasingly diverse drivers — demand for data processing, training and inference from large cloud-service providers and GPU-specialized ones, as well as from enterprise software and consumer internet companies. Vertical industries — led by auto, financial services and healthcare — are now at a multibillion-dollar level.”

But the boom is unlikely to go on unfettered because of the problem of supply. NVIDIA’s dominance belies its supply constraints for the company’s very latest AI chips. 

As AI scales it needs more computing capacity and more power, unlike a lot of digital technology. Such constraints are behind reports that OpenAI’s Sam Altman has been talking about investments in the region of $7 trillion to develop AI infrastructure that would undo some of NVIDIA’s dominance.

How did NVIDIA come to dominate?

NVIDIA makes a lot of different computer parts and technologies, but perhaps its greatest bet was in the development of the GPU – a chip developed in 1999 for fast 3D graphics, like those in video games – which have become a critical part of AI, as they are ideal for running deep learning algorithms. 

This was a somewhat accidental discovery made by Stanford University researchers in 2006. However, that finding prompted NVIDIA’s Huang to accelerate investment in the technology and create a tool to make the chips programmable. It helped to take the GPU’s use beyond graphics and into machine learning. According to VentureBeat, NVIDIA controls around 90% of the GPU market. 

Sourced from NVIDIA, WSJ, WARC, VentureBeat, BBC