4 weeks ago
10
Geoffrey Seiler, The Motley Fool
Sat, July 18, 2026 astatine 12:53 PM CDT 4 min read
Led by CEO and co-founder Jensen Huang, Nvidia (NASDAQ: NVDA) has established itself arsenic the apical chipmaker successful AI, and it does not program connected giving up its throne anytime soon. Much of the company's occurrence tin beryllium straight tied to Huang's instinctive endowment for predicting wherever the tech satellite is headed good successful advance. That's wherefore the banal is simply a buy.
Nvidia was founded successful 1993, and its invention of the graphics processing portion (GPU) successful 1999 helped substance the video crippled marketplace by speeding up graphics rendering and allowing for large leaps guardant successful machine graphics. The video crippled marketplace was large astatine the time, but Huang's much important strategical determination was to person Nvidia make its CUDA bundle platform, which makes its chips programmable for different tasks.
Missed Nvidia successful 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" awesome flashed for a little-known chipmaker called Nvidia. For the archetypal clip successful years, that aforesaid "Total Conviction" awesome is flashing for a institution 1/100th the size of Nvidia. Continue »
A past of forward-looking moves
While the afloat worth of that strategy took galore years to manifest, Nvidia wisely seeded CUDA into universities and probe labs that were doing aboriginal enactment connected AI. The effect was that astir foundational AI codification was written connected CUDA for Nvidia's GPUs, which is wherefore the institution enjoys a wide moat successful AI exemplary grooming today.
Huang did not halt there, though. In 2020, Nvidia acquired networking institution Mellanox. It was a institution with exertion up of its time, but Huang again saw wherever the marketplace was heading. Today, Nvidia's networking portfolio is the fastest-growing portion of its concern and a cardinal portion of its translation from a GPU specializer into a implicit AI infrastructure player.
Huang besides anticipated the displacement toward inference and agentic AI, and took steps to guarantee Nvidia would beryllium a large subordinate successful these markets. The institution has developed its ain ARM-based cardinal processing units (CPUs), arsenic CPUs volition play an important relation successful managing AI agents. The GPU-to-CPU ratio successful AI information centers built erstwhile workloads were chiefly driven by grooming was 8 to 1. As unreality companies physique retired infrastructure for agentic AI, the prediction is that the ratio could displacement to 1 to 1. With that successful mind, Nvidia has projected that the information halfway CPU marketplace could scope a worth of $200 cardinal successful the adjacent fewer years.
Nvidia besides acquired the assets and cardinal unit of Groq, including its connection processing units (LPUs), which it has since incorporated into the CUDA ecosystem. These chips volition assistance with servers designed specifically for inference, a marketplace that's yet expected to turn to a overmuch larger size than AI exemplary training.
.png)
English (CA) · 
English (US) · 
Spanish (MX) ·