Microsoft makes ‘biggest bet yet’ on its own silicon, unveils new AI chip to reduce reliance on Nvidia
Microsoft's Maia 300 ships in 2027. The company is betting its AI future on silicon that cuts Nvidia dependency and delivers more inference per watt. What changes for enterprise buyers when the GPU duopoly fractures.
Why it matters
Microsoft is moving from GPU-dependent AI to custom silicon at scale, targeting 2027 production of Maia 300 chips designed to handle most Copilot and enterprise AI workloads. This is a structural shift in cloud AI economics: reduced Nvidia reliance, custom efficiency claims, and a major player betting on vertical integration of the compute stack. Enterprise buyers and cloud architects now have a credible alternative pathway; chip and data-center strategies need recalibration.
The key facts
5 to knowMaia 300 rollout: 2027
Design goal: same AI work with less electricity than GPU-based systems
Scope: handle most AI work behind Copilot and other Microsoft products
Positioning: reduce reliance on Nvidia
Efficiency claim: more compute per watt (no specific benchmark disclosed)
The story so far
Earlier coverage of this storyline
- Microsoft releases new Nvidia-chip AI PCs with revamped Windows 11TechCrunch AI
- Microsoft’s new AI chip makes it less reliant on Nvidia’s GPUsThe Verge AI
- This story
Go to the source
GeekWiregeekwire.com
Publisher excerpt: Microsoft says its homegrown Maia chips will handle most of the AI work behind Copilot and its other products, with the new Maia 300 rolling out in 2027. The company says the new chip does more AI work with the same amount of electricity than the GPU-based systems that power most AI today.