Apple’s rebuilt Siri arrives in beta, made with Google’s Gemini
Siri AI splits its work between the device and Apple’s Private Cloud Compute, and will not launch in the EU on iPhone at first.
Apple began rolling out Siri AI in beta on 14 September 2026, starting in English, with French, Japanese, Korean, Portuguese and Spanish due the following month. Apple describes the models behind it as custom-built in collaboration with Google and its Gemini models.
The new Siri can draw on personal context across messages, email and photos, and act inside third-party apps. It requires an Apple Intelligence device: iPhone 15 Pro or any iPhone 16 or later, iPads and Macs with M1 or later, Apple Watch Series 9 or later, and Apple Vision Pro.
Processing is split. Some features run entirely on the device, while others go to Apple’s Private Cloud Compute servers, where Apple says personal data is not stored or made accessible to Apple or anyone else. Server-dependent Siri AI features are subject to daily usage limits.
Siri AI will not be available at first in the EU on iOS, iPadOS or watchOS, which Apple says it is working to resolve. Deployment in China is waiting for regulatory approval.
Sources
This is a summary of reporting published elsewhere. Where a claim matters, follow the source — it is the record, this is not.
Which models can your PC run?
Scan your hardware and see every model that fits, with the memory each one needs.