Microsoft plans to unveil Maia 300 AI chip this fall, aiming for 300,000 units by 2027

Microsoft is preparing to introduce a new artificial intelligence chip, the Maia 300, this fall, according to a report that also outlines the scale of production the company is seeking.

The report says Microsoft has been in talks with TSMC to secure manufacturing capacity for more than 300,000 units, with delivery targeted for 2027. The figure points to an effort to translate AI hardware demand into a more predictable supply pipeline ahead of future cloud and enterprise deployments.

Because chip manufacturing is tightly constrained by foundry capacity, lead times, and advanced packaging and testing availability, the ability to lock production slots can become a strategic advantage for AI chip customers. For large technology companies, securing that runway is often as important as the chip design itself.

The Maia 300 effort adds to the broader industry shift toward custom AI silicon, where cloud providers and major software companies seek chips designed to better fit their workloads. In that environment, production targets can announcement both expected demand and the confidence a company has in its AI infrastructure roadmap.

Microsoft did not provide additional details in the cited report, including the chip’s specifications, power and performance characteristics, or how the Maia 300 would be allocated across customer platforms.

The report also does not state whether Microsoft expects the Maia 300 to be available to third parties through cloud partner channels, nor does it quantify margins, pricing, or internal versus external consumption of the chip.

Even with the reported unit target and timeline, several elements remain unclear. The company has not publicly laid out, at least in the cited material, the exact production partners and process nodes, the expected yield and ramp schedule, or whether the 300,000-plus unit goal is firm or contingent on demand and manufacturing outcomes.

Next to watch is whether Microsoft follows the fall unveiling with clearer guidance on deployment timing, the role of Maia 300 within its AI infrastructure, and any confirmation of the manufacturing arrangement. For the broader market, any indication of how much additional capacity Microsoft is securing through 2027 could offer a window into how aggressively it expects to scale AI compute.