Apple’s AI server dreams come with a Nvidia cameo

Apple wants a bigger seat at the AI table

Apple is reportedly developing an AI server that could run on two or four planned M8 Ultra chips, which is a pretty dramatic way of saying: the company wants its own silicon doing more than powering iPhones and Macs.

Nvidia might get a guest spot

The twist is that Apple is considering Nvidia’s NVLink Fusion technology to connect the processors. That means this could become one of those awkward-but-profitable tech relationships where rivals still need each other to make the plumbing work.

• Apple says the project could extend its chip strategy beyond consumer devices.
• The server work reportedly started about a year ago, with backing from CEO John Ternus.
• Apple doesn’t expect a launch before 2029, and the whole thing could still get shelved.

Why investors should care

If Apple actually ships this, it would be a sign the company wants a bigger slice of the AI infrastructure pie — not just shiny gadgets on the edge, but the boring, powerful stuff in the data center.

Broadcom gets a shoutout too because the market is already thinking about the next wave of AI spend: less model training fireworks, more inference muscle. Big picture: Apple may be late to the server party, but it’s clearly eyeing the buffet.