Apple has been granted a patent (number 11496286) for “differential privacy with cloud data.” It involves using the data the company collects to “learn” how you write and speak without compromising your privacy.
About the patent
As we all know, we use devices such as Macs, iPhones, iPads, etc., to generate email, text messages, voice mails, documents, and other personal data on these devices. And a user may want the client device to learn how to write, speak, or spell the way that most users would write, speak, or spell.
As Apple notes in the patent data, to learn the way that users write, speak, or spell, a server could collect a large amount of user data from a large number of users to perform the learning. However, collection of user data can compromise user privacy with each piece of data collected from a user.
Also, a user may want to have a say, or some control, over whether a server collects data from the user of the client device. Apple wants to find a reasonable compromise between users having to choose whether or not to have their email data collected and analyzed, which can seriously compromise user privacy, or not using the email service at all. And, obviously, Apple wants folks to use its Mail app.
Summary of the patent
Here’s Apple’s abstract of the patent: “Embodiments described herein enable data associated with a large plurality of users to be analyzed without compromising the privacy of the user data. In one embodiment, a user can opt-in to allow analysis of clear text of the user’s emails.
“An analysis process can then be performed in which an analysis service receives clear text of an email of a client device; processes the clear text of the email into one or more tokens having one or more tags; enriches one or more tokens in the processed email using data associated with a user of the client device and the one or more tags; and processes the clear text and one or more enriched tokens to generate a data set of one or more feature vectors.”