Apple Inc. (NASDAQ: AAPL) is developing an artificial intelligence server that could utilize two or four planned M8 Ultra chips. The company is considering Nvidia Corp’s (NASDAQ: NVDA) NVLink Fusion technology to connect the processors.

The initiative represents a significant expansion of Apple’s in-house silicon strategy beyond consumer devices. The project reportedly began about a year ago with backing from current CEO John Ternus. If launched, it would mark Apple’s first dedicated server hardware since it discontinued Xserve in 2011.

In-House Silicon Push Expands

Apple recently refreshed its Mac lineup with M6, M5 Pro, M5 Max and M5 Ultra chips. The company stated that the 2-nanometer M6 delivers up to four times faster AI performance. This server initiative extends that computational power into data-center systems.

Apple does not expect a launch before 2029. The project could still be canceled or proceed without Nvidia technology. A partnership would potentially integrate Apple’s custom chips with Nvidia’s broader data-center infrastructure.

Broadcom Positioned For Inference Shift

Broadcom Inc. (NASDAQ: AVGO) could emerge as a key beneficiary as AI spending shifts from training models toward running large-scale inference workloads. BakerAvenue Wealth Management chief strategist King Lip said Nvidia has excelled in AI model training. However, for large, repetitive inference workloads, Broadcom is Lip’s “top pick” for the next evolution in custom AI chips.

Lip added that Broadcom does not need Nvidia to lose market share to expand. Both companies can benefit from different parts of growing AI infrastructure demand.

What the Numbers Show

The strategic divergence between Nvidia and Broadcom highlights a structural shift in AI capital expenditure. While Nvidia dominates model training, the move toward inference workloads favors Broadcom’s custom silicon approach. Apple’s entry into server hardware further validates the growing demand for specialized AI processing power beyond general-purpose GPUs.

Market Reaction

Nvidia shares were up 1.89% at $216.17 at the time of publication on Wednesday. Apple shares were up 0.40% at $332.66.

Elon Musk’s X Corp. and SpaceXAI dropped Apple Inc. from their antitrust lawsuit without publicly explaining the resolution. U.S. District Judge Mark Pittman ordered the companies to disclose any related agreements for private review by noon on Sept. 17.

The order follows an emergency motion from OpenAI, which remains a defendant in the case. OpenAI argued that it did not participate in the agreement between Musk’s entities and Apple, warning that undisclosed terms could affect its defense strategy.

Background of the Legal Dispute

X and then-xAI sued Apple and OpenAI for billions of dollars in August 2025. The plaintiffs alleged that the ChatGPT partnership represented "two monopolists joining forces to ensure their continued dominance." Musk had previously threatened Apple with antitrust action over alleged App Store favoritism toward ChatGPT.

Judge Pittman allowed the case to proceed in November 2025, noting that his decision did not address the merits of the claims. Apple argued that choosing "one partner first is not unlawful," while OpenAI described the litigation as a broader campaign against the company.

Apple’s Evolving AI Strategy

Apple and OpenAI announced their partnership in June 2024, integrating ChatGPT into Siri and other experiences. Since then, Apple has broadened its AI strategy beyond OpenAI.

In January, Apple struck a multi-year deal with Alphabet Inc.’s Google to use Gemini models for a revamped Siri. The company stated that its new Siri AI architecture uses Apple Foundation Models developed with Google’s Gemini technology.