Apple’s more capable, more “personal” Siri that is currently under development will be powered by Google Gemini, reports Bloomberg. Apple will pay Google approximately $1 billion per year for an ultra powerful 1.2 trillion parameter artificial intelligence model developed by Google that would help run its long-delayed revamp of its Siri virtual assistant.
The report says Apple plans to use the Google model as an interim solution until its own models are powerful enough. The Cupertino firms is working on a 1 trillion parameter cloud-based model that it hopes to have ready for prime time as early as next year.
The current cloud-based version of Apple Intelligence uses 150 billion parameters.
While Apple models will also be used for some Siri features, Gemini will be used for functions related to summarizing and multi-step task planning and execution. The Google AI model that Apple will use will run on Apple’s Private Cloud Compute servers, meaning Google will not have access to Apple data.
Apple had tested options from OpenAI and Anthropic, it decided to go with Google’s Gemini, after deciding Anthropic wanted too much money.
Apple is planning on relying on Google’s AI short term, while it continues working on its own models and will move to an in-house solution when its LLMs are ready for prime time. Apple is already working on a 1 trillion parameter cloud-based model that could be ready as soon as 2026.
Apple had said it would debut an updated, more personal version of Siri alongside iOS 18, but that didn’t work out, as the effort to overhaul Siri’s underlying architecture didn’t work out. The smarter Siri is expected to debut in the iOS 26.4 update, which will be released sometime in spring 2026.
