Microsoft is preparing to deploy the latest self-developed chip in data centers. The goal is to mainly rely on its own hardware to provide computing power support for its AI software. Microsoft said on Monday that it plans to deploy Maia chips with a total scale of several gigawatts in its data centers. Among them, the latest generation Maia 300 will be deployed starting next year. This type of chip is specially designed for AI inference, that is, it uses AI models to perform tasks such as summarizing conference call content or answering research questions. Microsoft, Amazon, and Alphabet's Google are all developing self-developed artificial intelligence accelerators. This move may help reduce their dependence on Nvidia hardware. Amazon and Google started chip development before Microsoft, and are considered to be ahead of the world's largest software company in this field. Microsoft is betting that by tailoring chips for its data center, AI inference can be run more efficiently while reducing some of the costs associated with purchasing Nvidia's market-leading graphics processors. Currently, most of Microsoft's AI inference work is still supported by Nvidia GPUs with computing power.

Disclaimer:Webull uses external vendor Google Translation Service for news translations where we endeavour to ensure these are correct, however, we recommend that you please double-check this information accordingly. Webull is not responsible for translation errors or issues.