Apple Intelligence gets a second shot with help from Google and Nvidia

2026-06-09 · Source: The Decoder · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, medium

Summary

Apple unveiled "Siri AI" at WWDC 2026, a significantly rebuilt virtual assistant powered by foundation models developed in collaboration with Google, building on Gemini technology. For complex queries, Siri AI leverages Nvidia GPUs within Apple's Private Cloud Compute infrastructure, which has expanded into Google Cloud. The system features a "System Orchestrator" to route queries locally or to the cloud, emphasizing privacy. New capabilities include system-wide actions, on-screen content reading, personal context integration, Visual Intelligence for tasks like bill splitting, and automatic password changes. However, the most capable on-device features require newer hardware, specifically devices with at least 12 GB of RAM, such as the iPhone 17 Pro or an M4 iPad. Crucially, Siri AI will not launch on iPhones or iPads in the EU due to the Digital Markets Act, though it will be available on macOS 27 and visionOS 27 in the region.

Key takeaway

For AI Product Managers evaluating platform integrations, Apple's "Siri AI" strategy highlights the complex interplay of hardware, cloud partnerships, and regulatory compliance. Your product's on-device AI capabilities will be gated by user hardware (e.g., 12 GB RAM minimum), and geographical market access can be severely restricted by regulations like the EU's Digital Markets Act. Consider these factors early in your development roadmap to avoid significant market exclusion and ensure your AI features can reach their intended audience across all target regions and devices.

Key insights

Apple's "Siri AI" integrates Google-derived foundation models and Nvidia cloud processing with strict privacy, but faces EU regulatory hurdles.

Principles

• Privacy-centric AI relies on local processing and secure cloud routing.
• Hardware requirements dictate on-device AI feature availability.
• Regulatory frameworks can significantly impact AI feature rollout.

Method

A "System Orchestrator" dynamically routes Siri AI queries between local device processing and Nvidia-powered Private Cloud Compute, leveraging Apple's refined foundation models and Google's Gemini technology.

In practice

• Use Visual Intelligence for bill splitting or nutritional info.
• Automate compromised password changes via Siri AI.
• Integrate custom skills using the Foundation Models Framework.

Topics

• Apple Intelligence
• Siri AI
• Foundation Models
• Google Gemini
• NVIDIA GPUs
• Digital Markets Act
• On-device AI

Best for: Product Manager, CTO, VP of Engineering/Data, AI Product Manager, Director of AI/ML, Tech Journalist

Related on AIssential

• Why Apple’s slow-and-steady AI bet is starting to look pretty smart — Apple's user-centric, integrated AI strategy, exemplified by Siri AI, prioritizes practical utility and financial prudence over a "race" for AI.
• Apple teams up with Google, Nvidia to expand private cloud capabilities — Apple's expanded Private Cloud Compute leverages partnerships to enhance AI capabilities while maintaining privacy.
• The Apple Watch needs a better Siri more than the iPhone right now — Apple's Siri and Health app need Google's Gemini AI to deliver personalized, integrated health insights and catch up in the AI race.
• Why Google Gemini Will Own AI by 2026 — Google's AI dominance stems from strategic vertical integration, vast distribution, and real-time data, not just superior models.
• OpenAI's Latest GPT 5.5 Features — Major tech companies are rapidly integrating advanced AI models while navigating significant privacy and revenue challenges.
• Apple’s new Siri AI overhaul is here (sort of) — Apple's Siri AI integrates on-device and cloud processing for enhanced context, privacy, and system-wide actions.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Decoder.