Siri gets a ‘new brain’: Apple and Google agree to cooperate

In that statement there is also a broader framework: the ‘next generation’ Apple Foundation Models will be based on Google’s Gemini models and cloud technology. Apple presents this as the foundation for future Apple Intelligence functions, including the ‘more personalized Siri’ that will arrive this year, that is, in 2026.

What changes in practice is also important for users. Reuters writes that ChatGPT remains in the game, but as an option for ‘opt in’ more complex queries, while Gemini becomes the main basis of the new Siri. This is a big change compared to the previous impression that Apple is “patching” Siri with external partners, because now it clearly chooses one primary model supplier.

Why this is big news for Google as well: Reuters reminds us that Apple has more than two billion active devices, so any deep integration means huge distribution. The Verge describes it as a moment where Gemini is getting its widest stage yet, and Apple is dealing with pressure from slower development of its own ‘smarter’ Siri.

In the same package, Apple emphasizes privacy and the way of processing, i.e. the combination of processing on the device and ‘Private Cloud Compute’ access when stronger processing is needed. The details of implementation and availability by language and region are still being clarified as the rollout approaches, but the direction is clear: Siri is entering a new era, and Gemini is becoming the engine that pushes it.