Nvidia Wants to Reinvent the PC — With Microsoft. Intel and AMD Are Terrified.
In a nutshell
For the past three years, the AI story has been a data centre story. Nvidia's GPUs. Hyperscaler clusters. Hundreds of billions in infrastructure. The AI that matters runs in buildings you will never visit, on chips you will never see.
That story just changed.
The RTX Spark — Nvidia Comes for Your Laptop
During a keynote address at Taiwan's Computex conference on June 1, Nvidia CEO Jensen Huang said his company, along with Microsoft, is going to "reinvent the PC." Nvidia unveiled the RTX Spark — a powerful laptop chip for Windows machines, staking its claim in the market for next-generation consumer PCs integrated with artificial intelligence.
Nvidia's plan to build system-on-chips, or SoCs, for PCs sent shares of Advanced Micro Devices, Intel and Qualcomm downward. It's the latest sign of Nvidia moving beyond the data centre for artificial intelligence and to the so-called edge, where smaller devices like phones or computers run advanced AI models on their installed chips without tapping the cloud.
The market reaction was immediate and unambiguous. AMD, Intel and Qualcomm — the three companies that have dominated PC chip architecture for decades — saw their shares fall on the announcement. Wall Street understood immediately what Huang was saying: Nvidia is not a data centre company that also makes gaming chips. It is an AI company that intends to own every layer of the computing stack — from hyperscale clusters to the laptop on your desk.
The ARM Dimension — The Architecture War Nobody Is Discussing
Nvidia's announcement is also the latest sign of the power of ARM. For decades, CPUs have been built on the x86 instruction sets pioneered by Intel in the 1970s. ARM's alternative power-efficient architecture went mainstream when Apple adopted it for the first iPhone in 2007.
Then Amazon popularised ARM-based chips for data centres when it announced its in-house Graviton processor in 2018. Cloud rivals Google and Microsoft followed Amazon with their own custom ARM CPUs for data centres.
The RTX Spark continues this architectural shift — away from x86 toward ARM-based SoCs that integrate CPU, GPU and AI acceleration into a single chip. Apple's M-series chips proved this architecture can deliver both performance and efficiency at the device layer. Nvidia is now bringing the same approach to Windows — with Microsoft as the explicit partner.
What This Means for Every GAFAM Company
The RTX Spark announcement touches every GAFAM company simultaneously — but differently.
Microsoft is the named partner — "reinventing the PC" with Nvidia. Surface Laptop Ultra, announced at Build 2026 yesterday, almost certainly runs on or alongside RTX Spark architecture. Microsoft is positioning Windows as the AI PC operating system of the post-x86 era.
Apple has been here for four years. The M-series chips that power every Mac are the proof of concept that Nvidia is now scaling to the Windows ecosystem. Apple's architectural advantage in device-layer AI is about to face its first credible competition on Windows.
Google faces a specific challenge: Chrome OS and Android — both running on Google's AI stack — must now compete with a Windows AI PC ecosystem backed by both Microsoft and Nvidia's combined engineering resources. Aluminium OS, announced at Google I/O, is Google's answer — but it does not yet have a chip partner to match Nvidia.
Amazon and Meta are less directly affected at the device layer — but AWS and Meta's AI infrastructure both depend on Nvidia GPUs at the data centre level. A Nvidia that is confident enough to compete in the PC market is a Nvidia that has secured its data centre position completely.
The European Dimension — Chip Sovereignty Revisited
The RTX Spark announcement arrives as the European Chips Act — designed to bring semiconductor manufacturing back to Europe — faces its most significant test. European chip production capacity remains far below the stated goals of the Act. The AI PC era, driven by Nvidia and ARM architecture, will be built on chips manufactured primarily in Taiwan and South Korea.
European enterprises buying AI PCs in 2026 and 2027 will be buying devices whose most critical components are designed in California, manufactured in Asia and governed by American export control regulations. The geopolitical vulnerability of European AI infrastructure is not abstract. It is in the chip at the centre of the device on your desk.
The European Perspective
The RTX Spark announcement was made in Taipei — not in Washington, not in Brussels, not in any European city. The future of AI at the device layer is being designed in California and manufactured in Taiwan. Europe's role in this architecture is as a consumer and a regulator — not as a producer. The European Chips Act's €43 billion commitment to semiconductor manufacturing is the right response to the wrong timeline: the AI PC era is arriving now, and European chip production capacity will not be meaningful until the end of the decade at the earliest. European policymakers watching Nvidia and Microsoft announce they will "reinvent the PC" together should read that sentence carefully. The PC being reinvented runs on American AI, ARM architecture and Taiwanese silicon. European digital sovereignty is not a policy goal. It is an infrastructure deficit. gafam.ai will be watching.
We are not first. We are right.