Maia chip scales up
Microsoft said it will expand its Maia AI chip, used in Azure data centers, to multi-gigawatt deployments. Announced on the company's official blog, the step shows its custom silicon strategy is growing in scale.
Why it matters
The compute power needed to train and run AI models is rising fast. By meeting part of that demand with its own chips, Microsoft aims to cut both costs and dependence on external supply chains.
What the scale means
The term "multi-gigawatt" describes deployment size in terms of data center power capacity. An expansion at this scale means Maia chips will be used far more widely across Azure infrastructure.
The move stands out as a significant step in the race among major cloud providers to build custom chips for AI.
