Microsoft has revealed its custom silicon project with two new server chips, Azure Maia AI Accelerator and Azure Cobalt CPU. The company shared its developments at Microsoft Ignite and admitted that the latter is ARM-based, a stark contrast to the dominance of Microsoft and Intel in the computing market. Microsoft collaborated with OpenAI to receive feedback on Azure Maia and use the company’s models for testing, aiming to train improved models and make them more affordable for customers. The custom chips will optimize Microsoft’s infrastructure instead of relying on third-party options, ensuring every element is tailored for Microsoft cloud and AI workloads. The Maia 100 AI Accelerator will power some of Microsoft Azure’s biggest internal AI workloads, improving efficiency and performance. The chips are set to be integrated into Microsoft’s data centers early next year to power services like Microsoft Copilot and Azure OpenAI Service.
Microsoft’s Innovation: Custom Chips to Enhance AI Services
