Apple reportedly building server packed with M-series Ultra chips for AI
Apple is reportedly developing an enterprise AI server that will use its high-performance M-series Ultra chips, with a potential launch targeted for 2029. The project, which began about a year ago and had backing from CEO John Ternus, could be Apple’s first server product in nearly two decades, according to The Information.
Why It Matters
If released, the server would position Apple to supply data-center hardware optimized for AI workloads at a time when developers and companies are increasingly using Apple Silicon for training and inference. That demand, evidenced by strong Mac mini and Mac Studio sales and reported large purchases by AI firms, appears to be driving Apple’s interest in an enterprise offering.
Key Facts
- Reported product: Enterprise AI server using Apple M-series Ultra chips
- Planned release: 2029
- Chip configurations: Two or four M8 Ultra chips
- Project backing: Received support from John Ternus when he led hardware engineering
- Timing: Project began about a year ago
Apple is said to be building an AI-focused server that would use its high-performance M-series Ultra processors, with a possible market debut in 2029, The Information reports. The planned appliance would come in at least two configurations, reportedly packing either two or four of Apple’s upcoming M8 Ultra chips. The initiative began roughly a year ago and received backing from John Ternus during his tenure overseeing hardware engineering; Ternus is now Apple’s CEO.
If the product reaches market, it would be Apple’s first dedicated server in nearly twenty years, marking a notable shift from the company’s recent focus on consumer and professional desktop and laptop systems. The timing appears linked to rising use of Apple Silicon in AI development environments: Mac minis and Mac Studios powered by M-series Ultra chips have seen strong sales as developers and companies adopt them for AI training and inference tasks.
The Information cites examples of AI firms acquiring Apple hardware at scale, reporting that OpenAI purchased “tens of thousands” of Mac minis and Mac Studios to run reinforcement-learning experiments, while Anthropic has rented Mac minis via Amazon Web Services. Those adoption patterns are presented as a motivating factor for Apple to pursue an enterprise server that consolidates the same chip architecture into a data-center form factor.
Details on performance, power, software stack, pricing, or how Apple would integrate the server into existing enterprise ecosystems were not provided in the report. Apple has not publicly confirmed the server project or the 2029 timeline; the account and technical specifics come from The Information’s reporting.
Keep Reading
OpenAI discloses 6 new cases of ‘misaligned’ AI behavior
Former Waymo CFO jumps to self-driving startup Wayve
I wore Snap’s $2,200 smart glasses
Snap is launching a new Specs AI tool, and it’s coming to iOS and Mac
Original source: Ars Technica AI