Microsoft Plans September Unveiling of Maia 300 AI Chip as It Expands In-House Silicon Push

10 August 2026

Microsoft is preparing to unveil its next generation artificial intelligence chip as early as September, marking a major expansion of the technology giant's effort to build more of the hardware powering its rapidly growing AI business.

The new processor, called Maia 300, represents Microsoft's latest attempt to reduce its dependence on expensive chips supplied by outside manufacturers such as Nvidia. According to The Information, Microsoft is planning a significantly larger production run for Maia 300 than for previous generations, signaling growing confidence in its custom silicon strategy.

Microsoft has reportedly been negotiating with Taiwan Semiconductor Manufacturing Company to secure production capacity for more than 300,000 Maia 300 chips for delivery during 2027. That would represent a dramatic increase from the tens of thousands of Maia 200 processors Microsoft has produced so far.

The company ultimately hopes to secure manufacturing capacity for more than one million Maia 300 chips, although component availability and continuing negotiations with TSMC could limit how quickly Microsoft reaches that target.

Microsoft first introduced its Maia family in November 2023 as part of a broader strategy to develop specialized processors for artificial intelligence workloads. The company followed with Maia 200 in January 2026, a chip manufactured using TSMC's advanced 3-nanometer technology.

Maia 200 was primarily designed for AI inference, the process in which trained artificial intelligence models respond to requests and generate outputs. Microsoft said the processor delivered improved performance and efficiency compared with hardware previously operating within its infrastructure. Maia 300 could represent a much more important step.

Microsoft hopes the next generation processor will eventually attract major Azure cloud customers, including artificial intelligence developer Anthropic. Winning external customers would demonstrate that Microsoft's chips can become more than internal tools for reducing infrastructure costs. The strategy reflects an increasingly competitive battle among the world's largest technology companies to control the hardware underneath artificial intelligence.

Google has spent years developing its Tensor Processing Units, while Amazon has expanded its Trainium family of AI processors. Both companies have invested heavily in custom chips to reduce their reliance on Nvidia while improving the economics of running enormous cloud computing networks. Microsoft has moved more slowly than those rivals, but the planned scale of Maia 300 production suggests it intends to close the gap.

Nvidia remains the dominant supplier of advanced artificial intelligence processors, particularly for training large AI models. Demand for its chips has surged during the generative AI boom, creating enormous costs for cloud providers racing to build increasingly powerful data centers.

Developing successful in-house alternatives could allow Microsoft to control more of those costs while ensuring it has access to enough computing capacity for Azure, Microsoft 365 Copilot and its expanding portfolio of AI products. The challenge will be convincing developers and customers that Microsoft's processors can compete not only on performance but also on software support, reliability and availability.

Securing manufacturing capacity presents another obstacle. Advanced chip production remains heavily concentrated among a small number of suppliers, particularly TSMC, and demand for leading manufacturing technology has intensified as companies race to build AI infrastructure. Still, Microsoft's reported plans for hundreds of thousands of Maia 300 chips demonstrate that its custom silicon program is moving beyond experimentation.

If Maia 300 performs as expected, Microsoft could gain greater control over one of the most expensive parts of the AI economy while placing additional competitive pressure on Nvidia and rival cloud providers.

The expected September unveiling will therefore be closely watched as Microsoft attempts to transform Maia from an internal chip project into a major component of its global artificial intelligence infrastructure.