Apple partners with Google to embed the Gemini AI model into Siri, signaling a major architectural shift toward advanced conversational capabilities beyond basic voice commands.

Security and privacy remain central: personalization largely happens on-device, with encrypted cloud processing only for higher computing needs, and Apple asserts data isn’t stored or used to train external models.

Apple integrates Gemini as the primary engine for Apple Intelligence, deploying a customized licensed version tailored to Apple’s ecosystem and a hybrid on-device and cloud processing balance to keep user experience under tight control.

The system delivers longer, sustained context, improves interpretation of regional variations and accents (including Brazilian Portuguese), and offers faster responses for everyday queries.

Code-named Campos, the project accelerates Apple’s generative‑AI ambitions with a gradual rollout across 2026, culminating in updates for iOS 27 and comparable Mac and iPad releases.

Updates enable extended conversations with retained context, infer user intent, handle multi‑step tasks without reactivation, and deepen OS integration for personalized responses using local data.