Microsoft plans to rely mainly on its own Microsoft AI chips in future data centers. The company aims to reduce dependence on Nvidia and AMD. Kevin Scott, Microsoft’s chief technology officer, shared the strategy during a major tech event. Microsoft wants full control over AI workloads and system performance. This move reflects its ambition to lead in AI computing.
Currently, Microsoft chooses chips based on price and performance, often using Nvidia and AMD. However, the firm also integrates custom-designed silicon in its data centers. “We are not religious about which chips we use,” Scott said. The company evaluates all options to ensure it meets rising demand efficiently. This flexible approach has allowed Microsoft to scale quickly.
Furthermore, Microsoft introduced the Azure Maia AI Accelerator and the Cobalt CPU, both optimized for AI workloads. Additionally, the company is also developing next-generation semiconductor products. In addition, it recently launched microfluid cooling technology to prevent overheating. These innovations make Microsoft data centers more powerful and energy-efficient. Using Microsoft AI chips enables better control over performance and costs.
Long-term, Microsoft plans to deploy Microsoft AI chips widely across its data centers. Scott highlighted that the company already uses “lots of Microsoft” silicon. The strategy includes system design, network architecture, and cooling integration. Optimizing all components together improves performance and reduces energy use.
However, other tech giants, like Google and Amazon, also design custom chips to reduce reliance on Nvidia and AMD. Using Microsoft AI chips allows the company to tailor workloads and address compute shortages. Nevertheless, this ensures it can meet the surging demand for AI applications. Microsoft continues to invest in infrastructure and chip technology to maintain its competitive edge.
Despite major investments, Scott warned that computing capacity still lags behind demand. Billions are spent on AI infrastructure, yet forecasts often fall short. Microsoft continues expanding data centers and integrating Microsoft AI chips. Overall, this approach provides flexibility, performance, and efficiency in AI computing.
For more business updates, visit DC Brief.
