NVIDIA Confidential Computing Powers Apple Intelligence Expansion on Google Cloud
At its Worldwide Developers Conference (WWDC), Apple announced a significant expansion of its Private Cloud Compute (PCC) infrastructure, bringing Google Cloud into its secure AI ecosystem. As part of this move, NVIDIA’s Confidential Computing technology is now being used to support confidential AI inference workloads for Apple Intelligence.
The collaboration marks a major step in combining advanced AI capabilities with strong privacy protections. NVIDIA GPUs equipped with Confidential Computing are being integrated into Apple’s Private Cloud Compute architecture running on Google Cloud, helping power next-generation Apple Intelligence features.
Apple’s server-side AI workloads will leverage Apple Foundation Models, developed by Apple and Google using technologies related to Google’s Gemini family of AI models. NVIDIA’s latest Blackwell GPUs will provide the computational power needed for these large-scale inference tasks while maintaining strict security standards.
Privacy-Centric AI Infrastructure
As AI applications increasingly rely on a combination of on-device and cloud-based processing, protecting user data has become a critical challenge. NVIDIA Confidential Computing addresses this issue by creating secure, hardware-protected environments where AI workloads can process sensitive information without exposing it to cloud operators or system administrators.
The technology isolates workloads within trusted execution environments and uses cryptographic verification methods to ensure that infrastructure has not been altered before any user data is transmitted.
For users, this means personal data, conversations, and AI interactions remain protected throughout the processing cycle, even when handled in cloud environments.
How NVIDIA Confidential Computing Works
NVIDIA’s Confidential Computing platform includes several key security features designed for enterprise and AI workloads:
- Hardware-rooted trust that verifies systems are running on authentic, untampered NVIDIA GPUs.
- Encrypted communication channels that secure data moving between servers and components.
- Remote attestation capabilities that allow software to verify platform security before releasing sensitive information.
- Support for AI training and inference workloads without sacrificing GPU performance.
These protections are becoming increasingly important as organizations deploy AI systems that handle sensitive personal, financial, healthcare, and business data.
Growing Demand for Secure AI
The adoption of NVIDIA Confidential Computing by Apple and Google highlights a broader industry trend toward privacy-focused AI infrastructure. As AI services become more integrated into everyday digital experiences, technology companies are investing heavily in security architectures that can deliver powerful AI capabilities while preserving user trust.
The partnership between Apple, Google, and NVIDIA demonstrates how leading technology companies are working together to build secure AI platforms capable of supporting the next generation of intelligent services without compromising privacy.