Apple preps AI-focused data centers with servers running on its own chips


Data Center
Some Apple AI tasks might be handled at data centers similar to this one.
Photo: Manuel Geissinger/Pexels

Some Apple artificial intelligence services reportedly will depend on data centers with servers running on the company’s own processors. Less-demanding AI functions will run locally, on users’ own computers.

This is all apparently part of bringing more AI features to iPhone, Mac, iPad, etc., in the coming months.

Apple AI will partially depend on remote servers and data centers

AI is hot, hot, hot. And while Microsoft, Google and Samsung have rushed products to users, Apple is seemingly behind the curve. But that’s expected to change soon. CEO Tim Cook strongly hinted that new features in all the company’s operating systems will heavily focus on AI. Apple undoubtedly will unveil some at WWDC24 in June.

“Apple Inc. will deliver some of its upcoming artificial intelligence features this year via data centers equipped with its own in-house processors,” reported Bloomberg on Thursday.

This means Cupertino won’t depend on AI-focused processors from Nvidia or other chip-makers. It’s a significant advantage, as the new Nvidia “Blackwell” B200 artificial intelligence chips cost between $30,000 and $40,000. Each.

Instead, Apple reportedly will start out with the M2 Ultra, its current high-end Mac processor.

AI-focused chips are necessary

Apple apparently will utilize a two-tiered approach. “Simpler AI-related features will be processed directly on iPhones, iPads and Macs,” reports Bloomberg.

It has been known for a while that Apple wants to keep AI functionality local as much as possible. That would be more private and likely faster. But when an AI task is beyond the capabilities of a phone or even a desktop, Apple could perform the computation on a remote server.

Still, the desire to keep AI functions local emphasizes the importance of the new Apple M4 processor. It’s debuting in the 2024 iPad Pro, but is expected in Macs later this year.

“M4 has Apple’s fastest Neural Engine ever, capable of up to 38 trillion operations per second, which is faster than the neural processing unit of any AI PC today,” Apple said when unveiling the chip during Tuesday’s “Let Loose” event. “Combined with faster memory bandwidth, along with next-generation machine learning (ML) accelerators in the CPU, and a high-performance GPU, M4 makes the new iPad Pro an outrageously powerful device for artificial intelligence.”

But Macs and iPhones, whatever processor they use, apparently will be able to offload demanding AI tasks to servers running Apple chips.

Apple AI plans are … ?

Very little is known about Apple’s specific plans for AI features in iOS 18, macOS 15 and other operating systems. It’s assumed the built-in applications will be able to, for example, summarize emails, webpages and text message chats. Users probably will be able to seamlessly remove objects from images, and perhaps add them. Or they might be able to tweak their photos using spoken commands.

Some generative AI tasks reportedly might be outsourced to Google Gemini. What this means for the future of Siri remains unknown.


Daily round-ups or a weekly refresher, straight from Cult of Mac to your inbox.

  • The Weekender

    The week's best Apple news, reviews and how-tos from Cult of Mac, every Saturday morning. Our readers say: "Thank you guys for always posting cool stuff" -- Vaughn Nevins. "Very informative" -- Kenly Xavier.