Apple Intelligence is powered by Apple’s servers with the M2 Ultra chip. A Bloomberg report was the first to spoil this a month before Apple unveiled its AI strategy.
At that time, the publication said Apple was planning in-house servers with the M2 Ultra and, eventually, M4 chips for heavier AI tasks. Later, at the WWDC 2024 keynote, Apple explained that Apple Intelligence worked in two different ways.
Even though most tasks can be processed on-device with the iPhone, iPad, and Mac, Apple would also build Apple Intelligence servers that use Private Cloud Compute to send information from your device to a server while keeping the query private.
With iOS 18.1 now available with the first Apple Intelligence features and the company preparing a broader rollout for more English-speaking countries with additional features in iOS 18.2, Apple wants to expand its servers in Taiwan.
According to a Nikkei Asia report spotted by MacRumors, Apple has asked Foxconn to produce servers in Taiwan as the company focuses on its Apple Intelligence push.
These servers will use Apple’s M4 chips. However, it’s unclear if the Cupertino company will use the base model or even some of the recently announced chips. M4 Max benchmark tests reveal that this processor outperforms the M2 Ultra. Additionally, depending on when Foxconn starts making those servers, an M4 Ultra chip could be used instead.
With Apple focusing on expanding Apple Intelligence features throughout next year, including in new countries and in more languages, the company will need to expand its AI servers significantly. We also expect these features to become more complex with time, especially now that all new Macs offer at least 16GB of RAM.
BGR will let you know once we learn more about Apple’s future AI servers.
