If you tuned into Nvidia's (NVDA 1.28%) Feb. 25 earnings call, you probably caught record numbers such as $215.9 billion in fiscal 2026 revenue, $120.1 billion in net income, or $96.6 billion in free cash flow.
Those are incredibly impressive results. But data centers comprised 89.7% of Nvidia's fiscal 2026 revenue, while other segments like gaming, artificial intelligence (AI) personal computing, professional visualization, and automotive and robotics are having a relatively small effect.
Here's why Nvidia has an incredible opportunity beyond data centers that could power the growth stock higher over the next decade.
Image source: Nvidia.
From generative to agentic AI
In 2024 and 2025, Blackwell pole-vaulted Nvidia to be the top compute power behind generative AI. Earlier this year, Nvidia unveiled its Rubin architecture, which features six new chips -- including a graphics processing unit (GPU), central processing unit (CPU), and various networking products. These products are sold separately, and can also be integrated into a well-oiled machine. For example, the Nvidia Vera Rubin NVL72 includes 72 Rubin GPUs, 36 Vera CPUs, and networking and infrastructure platforms. The integrated system functions as a plug-and-play supercomputer for large-scale AI factories.
Rubin's improvements over Blackwell are meant to capitalize on what Nvidia founder and CEO Jensen Huang called the "agentic AI inflection point" in Nvidia's Feb. 25 earnings release. This means enterprise adoption of AI agents.
In his 2025 keynote speech at Nvidia's GPU Technology Conference (GTC), Huang discussed Nvidia's AI roadmap, with generative AI focusing on digital marketing and content creation, agentic AI including applications like coding assistants, customer service, and patient care, and physical AI for autonomous vehicles and general robotics.
Nvidia's product development is following this roadmap, with Blackwell custom-built for generative AI, Rubin for agentic AI, and the next architecture likely to focus on improvements for physical AI. Of course, Nvidia's existing chips are already used in physical AI applications, from Boston Dynamics robots to Caterpillar earth-moving equipment to Tesla's Full Self Driving technology.
Nvidia CFO Colette Kress said the following on the Feb. 25 earnings call:
Physical AI is here, having already contributed north of $6 billion in Nvidia Corporation revenue in fiscal year 2026. Robotaxi rides are growing exponentially, with commercial fleets from [Alphabet's] Waymo, Tesla, Uber, WeRide, and [Amazon's] Zoox, and many others are expected to scale from thousands of vehicles in 2025 to millions over the next decade, creating a market poised to generate hundreds of billions of dollars of revenue. This expansion will demand orders of magnitude more compute, with every major OEM and service provider developing on Nvidia Corporation's platform.
$6 billion in revenue is less than 3% of Nvidia's fiscal 2026 total. But self-driving robotaxis remain in niche geographies, and general robotics is still a long way from everyday use in residential settings. That's why the hundreds of billions in revenue potential Kress outlined doesn't seem that far-fetched.
NASDAQ: NVDA
Key Data Points
Nvidia has plenty of room to run
Generative AI and agentic AI give Nvidia a long runway for future growth within the data center. But the more it can expand beyond the data center into physical AI applications, the more diversified its revenue will become, and the greater its overall growth potential will be.
The beauty of Nvidia's investment thesis is that it doesn't depend on physical AI taking off anytime soon. It's already a high-margin cash cow on generative AI alone, and widespread agentic AI adoption would be transformative in its own right.
Add it all up, and Nvidia has the makings of the perfect growth stock to buy and hold for decades to come.
