Nvidia is moving from selling future capability to building the infrastructure for profitable "AI factories". The results are here.

The enterprise return on investment (ROI) is tangible; early data shows AI-assisted software coding output has nearly tripled globally.

As CEO Jensen Huang put it: “If you can hire a software engineer and you could generate $9 trillion worth of productive work [from $3 trillion in salaries], why wouldn't you want to hire more software engineers?”

For data center buyers, the thesis is simple: “Compute is revenues. Now compute is profit.” This fundamental utility dynamic creates massive, inelastic demand for Nvidia’s infrastructure.

Unprecedented Manufacturing Efficiency: The upcoming Vera Rubin platform represents a massive step up in production scale. While its supply chain is twice as large as Blackwell, the assembly has been radically simplified. Huang noted: “What used to take two hours to assemble one Grace Blackwell rack now only takes 5 minutes.” This slashes assembly bottlenecks and accelerates time-to-revenue for cloud providers.

The Structural Cost Advantage: Because Nvidia designs the entire hardware and networking stack vertically, they hold an immense margin advantage. The platform is engineered to deliver the “lowest cost tokens” globally. By eliminating physical cables, optimizing power distribution, and utilizing advanced hot-liquid cooling, less energy is wasted on heat, directly lowering total cost of ownership (TCO) for operators.

The Software Moat & Digital Twins: Nvidia is cementing its ecosystem lock-in through DSX, an open software platform that allows operators to build a high-fidelity "digital twin" of their data centers using Nvidia Omniverse before a single brick is laid. As Huang highlighted: “You can simulate the entire factory before you spend a dollar, validate performance before a single rack is installed.” This eliminates multi-million dollar real-world deployment risks and optimizes energy efficiency.

A Massive New Addressable Market (TAM): Nvidia unveiled the Vera CPU, explicitly designed to run automated AI software agents rather than human users. The core investing thesis rests on scale: “This market will surely be larger than the last and the reason for that is because there'll be a lot more agents than there are people.”

Defensible Performance Leads: By optimizing hardware and software together, Nvidia's new CPU runs standard database queries (SQL) 3x faster and real-time stream processing 6x faster than legacy x86 competitors.

The Paradigm Shift: This isn't just a minor hardware upgrade. It is a complete disruption of how humans interact with operating systems. “The PC 10 years from now... is going to be completely different,” moving from clicking applications to giving local agents your overall intent. They reason, plan, and use tools to complete tasks.

Ecosystem Lock-in: Nvidia is partnering with Microsoft to launch a completely overhauled consumer line—desktops, laptops, and workstations—called RTX Spark. By baking its proprietary software framework (CUDA) into 100% of this upcoming Windows AI ecosystem, Nvidia is establishing a dominant consumer lock-in reminiscent of Windows' explosive rise 40 years ago.

NVDA - Nvidia is increasing its quarterly cash dividend 2,500% from 1 cent per share to 25 cents per share.

RL - Ralph Lauren raises dividend by 9.6% to $1.00

DCI - Donaldson raises quarterly dividend by 6.7% to $0.32/share

NAT - Nordic American Tankers raises dividend by 29.4% to $0.22

MOV - Movado raises dividend by 14.3% to $0.40