The AI Chip Fuelling Nvidia’s Meteoric Stock Surge

The AI Chip behind this year’s tech buzz isn’t a consumer gadget like a smartphone or gaming console. Instead, tech enthusiasts are captivated by the H100 processor, an obscure computer component most will never see.This chip has ushered in a new era of AI tools. It has the potential to revolutionize industries. Nvidia Corp. is now among the world’s most valuable companies. The H100 has shown that generative AI excitement translates into tangible revenue. This is evident for Nvidia and its key suppliers. Demand is so high that some customers face a six-month wait.

A Processor Named for a Pioneer

Named after computer science pioneer Grace Hopper, the H100 is an enhanced version of a graphics processing unit (GPU) typically found in PCs, boosting video game visuals. It features technology that links clusters of Nvidia chips into single units capable of processing massive data volumes and performing rapid computations. This makes it ideal for the power-hungry task of training neural networks essential to generative AI. Nvidia, founded in 1993, foresaw this market nearly two decades ago, betting that parallel processing would eventually make its chips invaluable in applications beyond gaming.

The H100’s powerful tech elevates Nvidia’s GPUs from gaming to crucial AI training, showcasing foresight, according to wsj discount.

The Role of Generative AI

Generative AI platforms learn tasks like text translation, report summarization, and image synthesis by absorbing vast amounts of existing data. The more they process, the better they become at recognizing human speech or drafting job cover letters. These systems evolve through trial and error, making billions of attempts to master tasks and consuming enormous computing power. Nvidia claims the H100 is four times faster than its predecessor, the A100, in training large language models (LLMs) and 30 times quicker in responding to user prompts. Since the H100’s 2023 release, Nvidia has introduced even faster versions—the H200 and the Blackwell B100 and B200.

Nvidia’s Strategic Dominance

Based in Santa Clara, California, Nvidia leads the world in graphics chips, the computer components that generate on-screen images. The most powerful of these are built with thousands of processing cores that handle multiple computation threads simultaneously, modeling complex 3D renderings like shadows and reflections. Nvidia’s engineers realized in the early 2000s that they could repurpose these graphics accelerators for other applications by dividing tasks into smaller units and processing them concurrently. AI researchers discovered that this type of chip could finally make their work feasible.

Snowflake Inc Faces Skepticism After Disappointing Sales Forecast

Snowflake Inc.’s recent sales forecast has failed to reassure investors about the company’s standing in the competitive artificial…

Competition Struggles to Keep Up

Nvidia now controls about 92% of the market for data center GPUs, according to market research firm IDC. Leading cloud computing providers like Amazon AWS, Google Cloud, and Microsoft Azure are developing their own chips. Nvidia’s competitors, AMD and Intel, are also involved. Despite these efforts, they haven’t made significant inroads into the AI accelerator market. Nvidia’s growing dominance has raised concerns among industry regulators.

Continued Innovation and Market Leadership

Nvidia has consistently outpaced rivals in updating its offerings, including software that supports its hardware. The company has also developed various cluster systems to help customers buy and deploy H100s in bulk. While Intel’s Xeon processors handle more complex data processing, they have fewer cores and are much slower at managing the large datasets typically used to train AI software.

The Road Ahead Blackwell Series and Beyond

The most eagerly awaited release is the Blackwell series, from which Nvidia expects substantial revenue this year. However, engineering challenges have delayed the release of some products in the lineup. Meanwhile, demand for the H series continues to surge. CEO Jensen Huang has championed the technology, encouraging governments and businesses to adopt AI early or risk falling behind. Nvidia also recognizes that once customers choose its technology for their generative AI projects, it will be much easier to sell them upgrades than it will be for competitors to lure them away.

Subscribe for 2 years to WSJ and Barron’s News on iOS, Android, PC, and Mac. Enjoy unlimited access to The Wall Street Journal and Barron’s, including Peggy Noonan’s insights, Barron’s stock picks, WSJ Live TV, audible articles, and Barron’s magazine.