Microsoft is planning to unveil its next-generation Maia 300 AI accelerator this fall — potentially as soon as next month — The Information reported on Monday, citing people with direct knowledge of the plans.
The company introduced its first Maia chip in November 2023, but has lagged rivals such as Alphabet and Amazon in scaling up in-house silicon as it seeks to cut its reliance on Nvidia's costly processors. Google began recognizing revenue from direct sales of its Tensor Processing Units in the quarter ended June, and Amazon has seen growing adoption of its Trainium chips.
Microsoft has been in talks with TSMC to secure manufacturing capacity for more than 300,000 Maia 300 units for delivery in 2027, and ultimately aims to secure capacity for more than 1 million chips — though component supplies and ongoing negotiations with the foundry could constrain its plans, the report said. Microsoft is also looking to ramp production and persuade major cloud customers, including Anthropic, to adopt the chip.
"Microsoft continues to invest in custom silicon as part of our long-term AI infrastructure strategy. While we don't share production volumes, the figures reported don't reflect the scale of our program," said Andrew Wall, general manager for Microsoft's Azure Maia.
The company unveiled its second-generation Maia 200 in January, built by TSMC on 3-nanometer technology and packed with SRAM memory that provides speed advantages for AI systems handling large numbers of user requests. A Maia 300 reveal would mark a quick cadence as the hyperscaler race to escape Nvidia dependence intensifies.
