Microsoft Plans Large-Scale Azure Maia AI Accelerator Deployments

Microsoft announced intentions to expand its Azure Maia AI accelerator family to multi-gigawatt deployments starting in 2027, positioning it as the primary inference system for Microsoft services. The third-generation Maia 300 chip is engineered for long-context reasoning and latency-sensitive AI tasks, aiming for 40% greater token efficiency per megawatt compared to current GPU systems. Maia 200 is already operational, supporting services like Microsoft Copilot.