At Intel's AI Summit in Taipei, executives say Copilot will soon run locally on PCs and that there will be a 40 TOPS requirement for NPUs on next-gen AI PCs
Let the TOPS wars begin. — We've previously reported on industry rumors that Microsoft's Copilot AI service will soon run locally …
Context & Ripple Effects
Intel had already outlined Microsoft’s broader AI-PC checklist, including a Copilot key, while expanding its developer and hardware-vendor program through Microsoft’s AI-PC requirements. The local-Copilot claim adds a performance threshold to that platform push.
The 40-TOPS marker was later formalized in Microsoft’s Copilot+ PC category, making this an early signal that NPU capability would become a Windows-PC qualification criterion rather than an optional specification.
First-order effects
- Intel and PC makers must position next-generation systems around NPUs that meet the stated 40-TOPS bar if they want to support the emerging local Copilot experience.
- Microsoft gains a clearer hardware baseline for moving selected Copilot functions from cloud delivery to compatible PCs.
Second-order effects
- Chip vendors and OEMs face a more comparable NPU-performance benchmark, intensifying competition over AI-PC qualification and product messaging.
- Developers and hardware partners in Intel’s AI-PC program receive a more concrete target for designing and validating local AI workloads.
Third-order effects
- If Windows AI features continue to be gated by NPU thresholds, the PC market could shift toward tiered AI capabilities tied to silicon performance rather than a uniform software feature set.
- This points to a more integrated PC AI stack, where operating-system features, chip requirements, OEM design choices, and developer tooling are coordinated around on-device inference.
The trend: The AI PC is becoming a defined platform category in which local AI features set minimum hardware-performance requirements.