In iOS 27, Trust Insights will use AI to alert apps when a person may be a victim of fraud.
Getty Images
of Apple iOS 27 is coming soon, with a host of new features and capabilities to help boost your security and privacy. One of the most interesting and useful of these could be a new iOS 27 framework that helps apps detect when people are being scammed and notify them in real time.
That’s according to the Apple-focused publication 9 to 5 Macwhich describes how in iOS 27, Apple created what it calls the Trust Insights framework to help apps deal with social engineering attempts as they happen. This could be across multiple channels, including voice, text and email.
In iOS 27, Trust Insights will use ANotify applications when a person may be a victim of fraud. Apple says these types of scams are harder to detect automatically because the victim is often the one performing the “verified and legitimate” actions.
In one developer video unveiling the new iOS 27 framework, the iPhone maker describes how it uses “privacy-preserving machine learning to detect when someone might be led into dangerous actions.”
Developers will be able to integrate Trust Insights into their apps, interpret its signals and design thoughtful interventions that protect people while respecting their privacy, Apple said.
The new iOS 27 framework arrives as the risk of AI fraud increases
The new iOS 27 framework comes as fraud becomes increasingly common, fueled further by the ability to use artificial intelligence to create fakes and other convincing scams.
The framework primarily runs on the device, analyzing “interaction patterns, timing, context and basic sensor data.”
Among its capabilities, Trust Insights can assign what it detects as medium or high risk fraud. To the user, this will appear as in-app warnings, delays, or additional verification steps.
Most scams rely on manipulating people into making the wrong decision, so the iOS 27 framework is “an excellent and positive use of artificial intelligence,” says Jake Moore, global cybersecurity consultant at ESET. “People often don’t immediately recognize the behavioral signs of a scammer, so this iOS 27 feature could be a game-changer for how victims react to such messages.”
However, Moore warns that Apple will need to acquire it feature correct and make clear to users to be effective. “Being able to stop fraudulent payments without reading the content of your messages or emails is also extremely encouraging, but the challenge will be the accuracy of the design,” says Moore.
For example, too many false positives could cause people to ignore or bypass fraud warnings, he says.
While artificial intelligence can be used by cybercriminals to make fraud more realistic, defenders are increasingly using AI-enabled tools to fight back. The iOS 27 framework aims to do just that, and if it works, it could definitely be a game-changer.
