> It’s not clear how it is significantly different from allowing apps access to your contacts, calendar, photos, and so on

Those are allowed via contextual consent prompts, several of which are for specific contacts, specific photos you wish to share, and so on.

Examples of the level of access an AI agent has include:

1. To read all indexed personal data from every app installed on the device

2. To perform actions in every supporting app on the device on the user's behalf

3. To read the current displayed apps for additional context as well as sensor data like current location

If you were regulated such that you had to allow any organization this level of access, and if you were hand-tied in how much you could convey the seriousness of accepting that consent prompt to an ordinary end user, and felt that it would be you, not any legal authority, who would ultimately suffer the reputational and legal consequences for the results - what would your yes/no decision be on shipping the feature in that jurisdiction?

How is this substantially different from Safari extensions that can effectively see and act upon everything you do in the browser?

One can imagine contextual prompts for all of the examples that you give, like which data sources and which apps the AI provider is given access to — similar to how you can choose for a Safari extension which websites it has access to — and for how long.

That all seams reasonably implementable. You could even use multiple AI providers in parallel with different subsets of data and apps, which would allow you to compartementalize access by different providers in a way that isn't possible with Apple's AI.

Such integration interfaces are necessary in the long run if we don't want to lock in our whole life to a singular combination of hardware, OS, and AI provider.