Apple reportedly doesn’t want you to know Siri will be running partly on Google’s Gemini

Last week’s reports revealed that Apple will pay Google around $1 billion a year for a custom-built Gemini AI model to power the next version of Siri, expected to debut with iOS 26.4 next spring. It’s a huge deal between two rivals, but according to a new report, Apple isn’t planning to draw attention to who’s behind the technology.

Bloomberg’s Mark Gurman says Apple has no plans to publicly acknowledge that Gemini will partly power Siri. Internally, the Gemini-based model is referred to as AFM v10, short for “Apple Foundation Model version 10,” a name chosen to obscure the partnership and avoid confusing staff and customers. Gurman adds that Apple prefers to present it as part of its own system, even though much of the core model isn’t.

The partnership helps explain how Apple has managed to speed up a project that was reportedly falling behind schedule. Earlier this year, internal data suggested the company’s AI tools weren’t gaining traction and that its major Siri overhaul could be years away. Borrowing Google’s model may be the quickest path to closing that gap, even if it means leaning on its biggest competitor in the process.

Apple will run the Gemini-based system on its own Private Cloud Compute servers, keeping user data away from Google, and still hopes to replace it with a trillion-parameter in-house model at a later time. Until that happens, Apple’s long-awaited AI comeback will partly rely on Gemini, but the company won’t be shouting this from the rooftops.