Apple’s artificial intelligence is preparing to take a new step with the upcoming integration of M4 chips into its dedicated servers. A major change which should take place from 2025, according to the latest information from Nikkei Asia.
A strategic evolution for Apple’s AI
Currently equipped with M2 Ultra chips, Apple Intelligence servers (PCC – Private Cloud Compute) are essential for processing the most complex AI requests. These modules help manage AI functionalities that require more power than local devices can offer, while ensuring user data privacy through end-to-end encryption.
As a reminder, PCC (Private Cloud Compute) is the system developed by Apple to process the most complex artificial intelligence requests in the cloud, while guaranteeing maximum confidentiality of user data. These specialized servers, currently powered by M2 Ultra chips, combine Apple silicon with a hardened operating system to deliver what Apple considers the most advanced security architecture for large-scale cloud AI computing.
According to sources close to the matter, Apple is in discussions with Foxconn for the construction of new AI servers in Taiwan. A strategic choice that would allow the Cupertino company to benefit from local expertise, particularly that of engineers also working for Nvidia.
If the exact version of the M4 chip that will equip these servers is not yet known, we can imagine that Apple will opt for its most powerful variant, probably a future M4 Ultra. The transition to M4 (Ultra?) chips is a continuation of Apple’s efforts to catch up with Microsoft, Google and OpenAI. According to Bloomberg analysts, the Cupertino firm would invest more than a billion dollars per year in the development of AI, with a dedicated team of several hundred engineers. This development demonstrates Apple’s commitment to the AI race, while maintaining its privacy protection philosophy that sets it apart from its competitors.
