In iOS, for example, you can take advantage of the Apple Neural Engine and custom CreateML models to process the data right on the device, helping you avoid lengthy and potentially risky round trips to …

Using differential privacy, we are able to successfully identify an up-to-date set of websites for which most users prefer auto-play with sound worldwide, such as streaming services including massive …

Privacy is a more important issue than ever for your users. Learn about new features and privacy engineering techniques that can help you...

Peer group benchmark values are calculated using a technique called differential privacy, which is the gold standard for ensuring that aggregated data remains both helpful and private. Metrics are …

Tools such as on-device machine learning, Core ML, and Apple’s differential privacy frameworks allow teams to deliver adaptive app behavior without transferring sensitive data off the device.

At Apple, we believe that privacy is a fundamental human right. Learn about our four pillars of privacy, how we brought these principles together to design iCloud Private Relay, and how you can approach …

When creating peer group values, we use a technique called differential privacy, which is the gold standard for ensuring that individual values within a group remain private.

Learn about changes to the iOS & iPadOS SDK. Overview Release notes provide details on API changes, known issues, fixes, workarounds, and deprecations for recent software releases. Bug …

Starting with iOS 13 beta 4, the copy attribute of the httpBodyStream property of NSMutableURLRequest is enforced. If the body data is mutated after the property setter has been …

For AlwaysOn VPN in iOS 14.2 and later, the minimum allowed value is 14. 1, 2, and 5 are available only in iOS, macOS, and visionOS prior to iOS 26, macOS 26, and visionOS 26.