Microsoft is creating its own AI server hardware to intensify actions to decrease its dependency on Nvidia, according to a source familiar with the matter speaking to The Information.
The media reported on Tuesday that the tech giant is designing a networking card to facilitate fast data transfer between its servers, offering an alternative to Nvidia’s equipment, potentially reducing costs for Microsoft and aiming to enhance the performance of servers equipped with Nvidia chips.
Namely, this development could benefit OpenAI, which relies on Microsoft’s servers for generating large AI models.
