Released to developers a few days ago, iOS 10.3 is the third major update to the iOS 10 operating system. And looks like it packs a wallop. In a bid to personalize it’s offerings and incorporate machine learning into it’s apps and services, Apple has decided to process all iCloud data that is shared by the user with the company.
That’s right, the update includes an option for iOS users to share their iCloud data with the Cupertino-based company. Users can sync their calendar, email, notes, reminders, photos and contacts, to Safari bookmarks, providing Apple with a significantly wide array of data types to process and make it’s services (particularly the voice assistant Siri) more able.
However, as has been stated in a guideline under the tab ‘iCloud analytics and privacy,’ the company will process the data into machine learning capabilities only after following privacy prevention techniques, clearly not willing to compromise completely on it’s much acclaimed ‘privacy first’ approach when it comes to users’ data.
Users can withdraw their consent to share their iCloud data at any time.
The company has been making inroads into data mining based AI capabilities for some time now by form of a a technique called differential pricing in June last year at the WWDC.
Differential privacy is a research topic in the area of statistics and data analytics that uses hashing, sub-sampling and noise injection to enable this kind of crowdsourced learning while keeping the information of each individual user completely private.
Apple’s senior vice president of software engineering Craig Federighi had explained.
The first Apple services to come up from the AI tools were iMessage, Spotlight and Notes. iMessage now predicts emojis and words and even notes slangs as per the trends. The Spitlight search, with it’s ability to rank search results did raise privacy concerns regarding search terms would require a lot of extra information, including location data to be transmitted back to Apple.
However, when it comes to data mining, Apple is only playing catch with fierce rivals Google and Facebook, which are known to collect user’s data for leveraging purposes to create a much more personalized and unified Google and Facebook ecosystem respectively. Facebook has built its $173 billion market valuation around profiling its users and showing them targeted adverts, making it crystal clear how important data mining is to social media companies.
It has refused to allow users to subscribe with money rather than personal data. But Apple has always been shy of implementing such policies. However, it seems to be about to bow to the pressure now.