Apple reportedly building server packed with M-series Ultra chips for AI

Apple is reportedly developing a new enterprise artificial intelligence server powered by its high-performance M-series Ultra processors, which could mark the company's return to the server hardware market for the first time in nearly two decades.

Slated for a potential debut in 2029, the system is expected to be offered in multiple configurations featuring several of Apple's upcoming M8 Ultra chips. This hardware push, which reportedly gained early backing from Apple's newly appointed chief executive officer John Ternus, aims to capitalize on the increasing trend of machine learning developers utilizing consumer Mac hardware for training AI models. Currently, major AI firms are purchasing or renting compact Apple desktop systems to run reinforcement learning tasks, highlighting a growing enterprise appetite for Apple's silicon.\n\nTo enhance the capabilities of these upcoming servers, Apple has reportedly engaged in discussions with Nvidia regarding the integration of specialized high-speed networking technologies. While the collaboration remains uncertain and the entire project could still be modified or canceled, it represents a significant strategic shift for the company following the retirement of its previous server line in 2011. However, Apple's ambitions face broader industry hurdles, particularly a global deficit of critical memory components driven by the rapid expansion of AI infrastructure. This supply shortage has already driven up manufacturing costs and retail prices for various consumer electronics, presenting a potential hurdle for Apple's hardware roadmap.