With the Rubin platform and orbit-based data centers, NVIDIA is rewriting the economy. Is the tech world ready for a future dominated by a single company?
If you thought NVIDIA was content with just owning the ground we stand on, Jensen Huang just proved you wrong.
At GTC 2026, he spent part of his three-hour keynote talking about the Vera Rubin Space Module. Yes, we are literally putting data centers into orbit now. It’s a wild flex, even for a company worth more than most countries. But it serves as the perfect backdrop for their new Rubin architecture.
The hardware reveal was relentless.
We got the Rubin GPU, the new 88-core Vera CPU, and the Groq 3 LPU. That last one is the most interesting part of the day. By licensing Groq technology for $ 20 billion, NVIDIA is acknowledging that general-purpose GPUs are no longer sufficient for the next phase of AI.
The chip maker needs specialized inference speed to keep their lead. This move basically turns NVIDIA into a landlord for the entire digital economy. If you want to run a model, you are likely paying rent to Jensen.
The vibes got even stranger when a robot Olaf from Disney walked onto the stage. It was a cute moment, but the message was clear.
NVIDIA is pivoting from chatbots to physical machines and autonomous agents. With their new NemoClaw platform, they want to be the operating system for every digital assistant you use in the future.
But is all this sustainable?
The power requirements for these racks are staggering. NVIDIA is building an infrastructure that requires its own mini power plants. Yet, when you look at the projection of one trillion dollars in revenue by 2027, you realize that nobody in the industry is actually trying to stop them.
We are all just watching the leather jacket show and hoping our electricity bills don’t catch fire.
