Finally, PC 2.0: Nvidia and Microsoft Just Reinvented the Personal Computer for the AI Era
Nvidia and Microsoft didn’t just refresh Windows PCs — they declared the birth of an entirely new category. In simultaneous announcements on May 31, 2026, the two companies unveiled the RTX Spark Superchip and a new generation of Windows PCs purpose-built for the age of personal AI agents. “The PC is being reinvented,” said Nvidia CEO Jensen Huang. “This is the new PC. The personal AI computer.”
Microsoft CEO Satya Nadella echoed the ambition: the goal is to deliver “unmetered intelligence to every home and every desk with Windows.”
The RTX Spark Superchip: Apple M-series for Windows
- A powerful Blackwell GPU with 6,144 CUDA cores and fifth-generation FP4 precision;
- A 20-core Arm-based NVIDIA Grace CPU (custom-designed in collaboration with MediaTek);
- Up to 128 GB of unified memory.
This is the closest thing the Windows world has ever seen to Apple’s M-series chips: heterogeneous architecture, massive unified memory, and exceptional power efficiency.
The result? Slim laptops as thin as 14 mm and as light as 3 pounds with all-day battery life, yet capable of running frontier-class AI workloads entirely on-device.
The real revolution: your own local AI agent
The headline feature isn’t raw performance — it’s what that performance enables. These new PCs will ship with native support for personal AI agents that run locally and securely inside Windows.
Nvidia’s new OpenShell runtime, built on fresh Windows security primitives, provides identity, containment, and policy controls so agents can act safely on your behalf.
Users stay in full control: you decide when and how the agent operates, with complete visibility into data access.
With 128 GB of unified memory, these machines can comfortably run 120-billion-parameter LLMs locally (in FP4 precision) with up to one million tokens of context. That’s enough intelligence to feel like having a real virtual Jarvis living inside your laptop.
A bumpy but inevitable takeoff
The vision is crystal clear, but the first 1–2 years may feel transitional. Deep integration across the Windows ecosystem, developer adaptation, and one or two hardware iterations will be needed. Yet once the models get smarter and the agents more capable, the shift will feel inevitable. What started as a cool demo will quickly become something most people can’t imagine living without.
Where do OpenAI and Anthropic fit in?
You’d simply select “Claude Desktop” or “GPT-5 Local” in the settings menu, wait for the (one-time) download, and suddenly your personal AI agent would be powered by the same intelligence that currently lives in the cloud — only now it’s private, instant, and runs on your own hardware.
The menu choice becomes as simple as switching between different AI assistants today.
- Delta’s Premium Cabins Just Out-Earned Economy for the First Time — And the Real Story Is More Interesting Than “The Rich Are Getting Richer”
- How Korean Underdogs Became One of the World’s Most Valuable Companies
- Analysts Name the Highest-Quality Streaming Service – And It’s Not Netflix
- The Self-Improving Tax Agent: How OpenAI and Thrive Built Tax AI with Codex
The bottom line
Nvidia and Microsoft aren’t releasing a faster laptop. They’re releasing the first true AI-native personal computer — a machine where the AI isn’t an add-on or a cloud service, but the central intelligence layer that makes the entire device feel alive.
Fall 2026 will mark the beginning of PC 2.0. The era of the passive computer is over. The age of the personal AI companion has officially begun.