{{articledata.title}}

Investing.com -- Apple Inc (NASDAQ:AAPL) is set to start analyzing data on users' devices to enhance its artificial intelligence (AI) platform. The move aims to protect user information while helping Apple keep pace with AI competitors. Apple's traditional training of AI models uses synthetic data, which is designed to replicate real-world inputs without any personal details. However, this synthetic information isn't always reflective of actual user data, which can hinder the effectiveness of its AI systems.

Apple's new method aims to solve this issue while ensuring that user data stays on the users' devices and isn't directly used to train AI models. This strategy is designed to help Apple catch up with competitors such as OpenAI and Alphabet (NASDAQ:GOOGL) Inc., which have fewer privacy restrictions.

Apple's machine learning blog post explains the company's approach to understanding usage trends while protecting privacy. Apple does not use users' private personal data or user interactions when training its foundation models. For content publicly available on the internet, Apple applies filters to remove personally identifiable information like social security and credit card numbers.

Apple is applying this work on differential privacy to Genmoji, a feature that uses differentially private methods to identify popular prompts and prompt patterns. This approach works by randomly polling participating devices for whether they’ve seen a particular fragment, and devices respond anonymously with a noisy signal. The signal Apple receives from the device is not associated with an IP address or any ID that could be linked to an Apple Account.

The tech giant currently uses differential privacy to enhance Genmoji, and in future releases, it will also use this approach, with the same privacy protections, for Image Playground, Image Wand, Memories Creation and Writing Tools in Apple Intelligence, as well as in Visual Intelligence.

For Apple Intelligence features like summarization or writing tools that operate on longer sentences or entire email messages, a new method is needed to understand trends while upholding privacy standards. To address this challenge, Apple can expand on recent research to create useful synthetic data that is representative of aggregate trends in real user data, without collecting any actual emails or text from devices.

Synthetic data mimics the format and important properties of user data, but does not contain any actual user-generated content. The creation of synthetic data aims to produce synthetic sentences or emails that are similar enough in topic or style to the real thing to help enhance models for summarization, without Apple collecting emails from the device.

Apple uses synthetic data to enhance text generation in email in beta software releases. It will soon begin using synthetic data with users who opt in to Device Analytics to improve email summaries.

In conclusion, by using techniques like differential privacy and synthetic data generation, Apple can enhance its Intelligence features while protecting user privacy for users who opt in to the device analytics program. These techniques allow Apple to understand overall trends, without learning information about any individual, like what prompts they use or the content of their emails.

This content was originally published on http://Investing.com