Do RTX 20-series or AMD GPU owners get the same Windows AI access?

The current extension covers RTX 30-series GPUs with at least 6GB of VRAM. RTX 20-series and AMD GPU owners are not included in this change — the 30-series cutoff is the stated hardware floor.

Why did Microsoft require an NPU for Copilot+ PCs in the first place?

NPUs offered dedicated low-power inference that justified a new hardware category and gave OEMs a differentiated selling point. The RTX extension suggests the NPU requirement was partly a market boundary, not a hard technical necessity — RTX 30-series GPUs can handle the same workloads.

What should IT buyers who approved Copilot+ PC budgets do now?

Revisit refresh cycles. If the local AI features that justified Copilot+ hardware premiums now run on existing RTX 30-series machines in the fleet, the incremental cost of NPU-based upgrades needs a new rationale — and procurement teams do not have one yet.

Microsoft Cracks Its Own AI Hardware Wall // AIDRAN

The Copilot+ PC designation was a hardware segmentation strategy as much as a technical specification. By tying local AI features to dedicated NPUs, Microsoft created a clear upgrade path and a reason to buy new hardware. Extending those same features to RTX 30-series GPU owners running local AI collapses that boundary — the installed base becomes the addressable market, and the NPU premium loses its primary justification. PC OEMs that built 2026 product lines around Copilot+ differentiation now face a value proposition that Microsoft just quietly undermined from underneath them.

Microsoft Unlocks Windows AI for Nvidia GPUs Beyond Copilot+ Hardware

What the NPU Requirement Was Actually Protecting

Frequently asked