Apple plans to enter the enterprise AI server market by 2029, leveraging its M8 Ultra chips and Nvidia's networking infrastructure to capture growing demand for high-performance inference hardware.
Apple is reportedly developing AI servers with its own chips, which could open a new market, but the product is not expected until 2029 and faces execution risks.
Apple's AI server plans would rely on Nvidia's networking technology, potentially boosting Nvidia's data center business.
Apple is reportedly developing enterprise AI servers powered by its proprietary M8 Ultra chips, marking a potential return to the server market after 2011. The project, expected to launch by 2029, would rely on Nvidia's networking technology to overcome internal chip-to-chip connectivity limitations, signaling a strategic shift toward data center compute demand.
Apple is responding to strong demand from AI labs that are currently purchasing Mac hardware in bulk to run inference tasks, prompting the company to develop a dedicated enterprise solution.
Apple's internal chip-to-chip connection technology is currently insufficient for server-scale operations, so the company plans to utilize Nvidia's networking technology to link its M8 Ultra chips.