Apple’s Secret “ACDC” Project: Building Its Own Cloud & Renting Servers with Custom Chips
Ddos 
According to a report by The Information, Apple has initiated an internal initiative codenamed “ACDC” (Apple Chips in Data Centers), which envisions deploying its proprietary chips in data centers to establish its own cloud infrastructure. The company is now weighing the possibility of offering server rentals directly to millions of iPhone and Mac developers—positioning itself as a direct competitor to the dominant players in the cloud services market.
Since launching its first in-house chip in 2010, Apple has continuously advanced the performance of its devices with the A-series and M-series processors, significantly enhancing computing power and AI inference capabilities. Previous reports suggest that Apple has already been using its custom silicon within internal data centers to support cloud-based services such as Photos and Music, while placing a strong emphasis on preserving user privacy and data security.
Sources indicate that Apple’s leadership believes its silicon can dramatically reduce costs for AI-intensive tasks. Should Apple eventually open its cloud rental services to external developers, it could offer high-performance compute resources at prices substantially lower than those of Amazon, Microsoft, and Google.
However, the future of this initiative remains uncertain following the departure of its primary champion, former cloud services chief Michael Abbott, who left the company in 2023.
Despite this, Apple continues to explore the deployment of its chips on the server side, including efforts to develop open-source frameworks such as MLX, which enable developers to run AI models directly on Apple silicon. Startups like Dragonfruit and WebAI have already begun adopting Apple hardware as an edge computing platform, powering AI applications such as real-time video analytics—demonstrating the potential of Apple chips in the cloud domain.
Historically, Apple has relied heavily on Amazon and Google for cloud services, reportedly spending around $7 billion annually, though only a small portion of that has been allocated to AI training. The company has not made large-scale purchases of NVIDIA GPUs, and has partially relied on Google’s TPU infrastructure for AI inference. Going forward, Apple appears increasingly focused on boosting AI performance through its own chip architecture.
Rather than mirroring the conventional cloud vendor strategy of building massive enterprise sales operations, Apple intends to leverage its existing developer relations teams to promote these services. This approach signals a deliberate focus on serving the developer community—rather than enterprise customers, small-to-medium businesses, or general consumers.
Related Posts:
- Proofpoint Exposes TA2900: French-Speaking BEC Actor Exploits Rental Payment System
- Apple Confirmed that All Mac and iOS Devices Are Affected by Chip Vulnerability
- Facebook plans to develop its own chip
Support Our Threat Intelligence
If you find our CVE report and cybersecurity news helpful, consider supporting our work.
Buy Me a Coffee
PayPal
