Apple’s Craig Federighi and his team recently held a technical briefing following the WWDC keynote to pull back the curtain on iOS 27 and the new Siri AI. A major focus was clarifying the company’s collaboration with Google, as Apple aims to differentiate its new Apple Intelligence from existing chatbot experiences.
The technical talk was attended by 9to5Mac, which provided a transcript of remarks from Federighi and his leadership team: Amar Subramanya, vice president of AI; Mike Rockwell, Siri lead; and Sebastien Marineau-Mes, software VP.
Federighi was firm in clarifying Apple’s different approach from typical AI deployments. “Of course, we don’t have the Gemini app as our app,” he explained. “In fact, none of that client code is part of how we run on iOS. For these models, we use none of the models that Google deploys to their customers, nor do we use the infrastructure and means by which they deploy models to their customers.” He further emphasized that Apple is not using Google Search as the foundation for their system, noting, “The amount of the Google Assistant we use is none.”
Instead, Apple has built a deeply integrated “Assistant experience” that functions as a core part of the operating system rather than a standalone chatbot. This setup utilizes a “System Orchestrator”—a privacy-focused architecture that coordinates requests across personal content and on-device context. For more demanding tasks, the system may tap into “Private Cloud Compute,” which Federighi describes as an extension of the iPhone’s privacy promise to the cloud.
The collaboration with Google specifically centers on model development for Apple’s new family of third-generation Apple Foundation Models (AFM). Amar Subramanya, Apple’s VP of AI, confirmed that the company has built a family of models ranging from on-device to cloud-based. “These are custom builds for Apple Silicon, trained using proprietary data, and refined using outwards from Gemini frontier models,” Subramanya explained.
For the most complex reasoning tasks, Apple has introduced AFM Cloud Pro. To power this, the company collaborated with both Google and Nvidia, extending their private cloud infrastructure to include Nvidia GPUs within Google’s cloud while maintaining Apple’s strict privacy guarantees.
As the team emphasized, the primary goal is to intelligently match every user request to the specific model that provides the best response with the lowest latency.
But of course, not all devices will get the full AI Siri experience. Here’s a list of the devices that will get all the new features coming in iOS 27 and beyond.
Want to see more of our stories on Google?
P.S. Want to keep this site truly independent? Support us by buying us a beer, treating us to a coffee, or shopping through Amazon here. Links in this post are affiliate links, so we earn a tiny commission at no charge to you. Thanks for supporting independent Canadian media!
