What you need to know
- AICore may temporarily use larger storage while updating AI models in the background on your device.
- Google keeps both old and new AI models safe for three days during updates.
- The storage used by AICore is automatically freed once the new AI model is confirmed to be stable.
If you have the AICore app installed on your Android phone and you’ve noticed that it’s taking up an unusually large amount of storage, Google has finally explained why this happens.
Over the past few years, companies like Google and Samsung have been emphasizing on-device AI in their phones. on most flagships On Android phones, a large part of that functionality is handled by the AICore app.
AICore app is basically a system service Powers the Gemini Nano on-device on Android. This enables offline and private AI features like smart replies and notification summaries on devices like the Pixel 10 Pro. But if you’ve noticed it’s using a lot of storage, you’re not alone – some users have reported this taking up 11GB.
Article continues below
Google has now updated it AICore Help Page To explain this behavior. The company says you will “occasionally find that this service uses a larger amount of storage than expected.” According to the support page, this happens while it’s updating to a new version of the AI ​​model in the background.
To make sure everything works reliably, AICore temporarily keeps both the old and new versions of the AI ​​models on your device for up to three days. This acts as a fail-safe, allowing the system to be rolled back immediately if something goes wrong during an update, saving you from having to download large files again.
Taking Android Central
I understand why Google is doing this, and it’s a smart fail-safe. But 10GB+ is still wild for a background service. At a minimum, there should be a way to limit or manage it manually.
Google also says that this extra storage is freed automatically once the update is confirmed to be stable, so you don’t need to take any action.
Honestly, this approach makes sense. If the app immediately deleted the old model and the update failed, you would have to re-download several gigabytes of data and lose access to AI features in the meantime. This method makes the entire process more reliable.
