With new Macs, Apple aims to take on Microsoft, Nvidia in a rush to lower AI costs

•1mo ago

With new Macs, Apple aims to take on Microsoft, Nvidia in a rush to lower AI costs

Tap title to read story

0:00 / 1:10

Apple announced a hardware strategy aimed at reducing artificial intelligence infrastructure costs by utilizing clustered Mac Studio computers to run large language models locally. During a recent product showcase, Apple demonstrated four Mac Studios operating collectively from a standard wall outlet to execute a trillion-parameter AI model for bug detection. Apple chief hardware officer Johny Srouji stated that hardware-based on-device processing eliminates the recurring subscription and token fees associated with cloud data centers. Competitors such as Microsoft are pursuing similar on-device capabilities through Windows architectures, though they face challenges supporting fragmented multi-vendor hardware ecosystems. Meanwhile, Nvidia continues to dominate high-end data center processing for AI training workloads. Apple intends to market its unified chip architecture as a scalable solution ranging from localized mobile devices to enterprise-grade desktop workstations, potentially shifting enterprise computation expenses away from cloud providers toward localized fixed hardware investments.

Published 2026-09-22 16:55:16 UTC · Published by JQJO

Published by JQJO Editorial Desk · AI assisted

More context

Timeline

On September 1 2026 Apple showcased four Mac Studios running trillion parameter models.
On September 5 2026 industry analysts evaluated potential impacts on enterprise AI budgets.
On September 10 2026 Microsoft outlined Windows strategies for unmetered on device intelligence.
On September 12 2026 Jensen Huang discussed Nvidia strategies regarding personal computer hardware markets.
On September 15 2026 Johny Srouji detailed architectural scaling principles across Apple product lineups.
On September 18 2026 developers tested localized memory allocations for enterprise debugging environments.
On September 20 2026 corporate purchasers reviewed total cost of ownership comparisons for data centers.
On September 21 2026 hardware benchmarks confirmed power consumption metrics for stacked workstation clusters.
On September 22 2026 market researchers published findings on local versus cloud processing economics.
In coming months enterprise software vendors will release optimized local model runtimes.
By late 2027 corporate adoption of local workstation clusters is expected to grow.

Why it matters

Immediate US impact: Corporate purchasers may reduce recurring cloud token costs significantly. Possible long-term US impact: Enterprise AI infrastructure spending could shift toward fixed local hardware. Most affected groups: Enterprise developers, hardware investors, and corporate IT executives are affected. Reader priority: Monitor verified hardware benchmarks and official developer documentation updates across outlets.

Coverage notes

Left: Emphasizes corporate cost reductions and potential shifts in computing power. Center: Focuses on hardware specifications, market competition, and architectural scaling comparisons. Right: Highlights free market competition and corporate efficiency in enterprise investments.