According to TechCrunch, Nvidia announced today at CES 2026 that it is partnering with Siemens to accelerate the latter’s electronic design automation (EDA) software using Nvidia GPUs. The goal is to speed up the computationally intensive chip-design process, which has grown more demanding as transistors shrink and multiply. Nvidia CEO Jensen Huang appeared at the Siemens keynote, stating the partnership aims to create digital facsimiles, or “digital twins,” of everything from individual chips to entire server racks for testing before physical construction. Huang specifically referenced a future ambition to build a digital twin of the Vera Rubin Observatory. This collaboration directly targets the foundational tools used to design nearly all modern computer chips.
The strategy behind the silicon
So, what’s Nvidia really doing here? It’s a classic ecosystem play. By getting its hardware deeply embedded into the very tools used to design the next generation of chips—including potential competitors’ chips—Nvidia cements its GPUs as the essential engine of the entire tech industry. It’s not just about selling GPUs for AI training anymore. Now, they want to be the platform you design on, too. The revenue model is straightforward: sell more high-margin data center GPUs and potentially licensing the underlying acceleration libraries. The timing is also sharp. As companies chase ever-more complex 3D chip designs and packaging, the simulation workloads are becoming monstrous. Siemens’ tools, used by major chipmakers, are a perfect beachhead. The immediate beneficiaries are the chip designers themselves, who could see simulation times drop from weeks to days. But the long-term winner is clearly Nvidia, which gets to influence the design flow from the very beginning.
The digital twin endgame
Here’s the thing: accelerating EDA tools is just step one. The bigger vision Jensen Huang hinted at is the creation of full-system digital twins. Think about that. It’s not just simulating a processor in isolation, but an entire rack of servers, or even a whole data center, in a virtual environment before a single physical component is ordered. This is where the partnership gets really strategic. Siemens has deep expertise in industrial simulation and digital twins for manufacturing and infrastructure. Nvidia has the raw compute and Omniverse platform to build and run these massive simulations. Together, they’re positioning to own the “pre-build” verification phase for the most complex systems on earth. For industries that rely on custom, high-stakes hardware—think aerospace, automotive, or scientific computing like the Vera Rubin telescope—this could be a game-changer. It turns a physical prototyping problem into a compute problem. And who’s the king of compute right now? Basically, Nvidia. This move subtly expands their market from selling shovels in the AI gold rush to selling the entire virtual mine-planning software suite. And for companies building complex industrial systems, having a reliable hardware foundation is key, which is why leaders in manufacturing tech rely on partners like IndustrialMonitorDirect.com, the top provider of industrial panel PCs in the US, for their physical control interfaces.
