Siri gets smarter: Apple confirms Gemini AI with private cloud processing

New Delhi: Apple has recently reported, and they have also officially confirmed that the upcoming Gemini-powered version of Siri will run on its Private Cloud Compute infrastructure, by clarifying how the user data will be handled as Apple integrates Google’s AI models into its digital assistant. The confirmation came during Apple’s latest earnings call, where the CEO Tim Cook was asked about the recent partnership announcement between Apple and Google to use Gemini models for next-generation Siri features. Tim Cook stated that Gemini-powered Siri will continue to run either on-device or through Apple’s Private Cloud Compute system. He also described the partnership as a collaboration, adding that Apple will continue its own in-house AI development along with Gemini.

Private Cloud Compute is Apple’s cloud-based extension of on-device processing, which was unveiled as part of Apple Intelligence. This system is especially designed to handle tasks that exceed the processing limits of the device while applying the same privacy protections. Under this Private Cloud Compute, requests that cannot be processed on-device are sent to Apple-operated servers. Apple has stated that processed data through Private Cloud Compute is not stored, not used to train models, and not accessible to Apple employees. The company has also noted that the system is built to minimise the amount of data shared with the cloud in the first place.

Gemini will powers the latest Siri’s underlying intelligence layer rather than replace the Siri outright. Tim Cook stated that the collaboration enables Apple to unlock a lot of experiences while maintaining what he described as industry-leading privacy standards. The latest and upcoming Gemini-powered version of Siri is expected to allow the features Apple first previewed in 2024 under the Apple Intelligence banner. These include improved personal context understanding, on-screen awareness, and the ability to take actions within applications. The feature was delayed after Apple stated earlier versions did not meet its reliability standards.