Tech Desk 
Jun 30, 2025
While encryption and digital sleuths have always been with certain limitation when dealing with hidden apps in Android devices, Australian researchers have developed an innovative technique to uncover hidden "vault" apps on smartphones.
The invention of this remarkable method marks a milestion that might give an edge to law enforcement when carrying out digital probes.
Conducted by Edith Cowan University and the University of Southern Queensland, the study proved that machine learning (ML) can identify these vault apps with up to 98% accuracy on Android devices.
It should be noted that vault apps allow users to store files, messages, or even other applications beyond encryption. While they're often used for legitimate privacy purposes, they have increasingly been employed in illicit activities, including unauthorised surveillance.
Associate professor Mike Johnstone from ECU noted that these apps can mimic legitimate applications, which makes them even more challenging to detect.
While traditional detection tools fall short of efficiency owing to heavy reliance on prior knowledge of suspicious apps, the new ML approach can identify vault apps without requiring a pre-existing list or database of known threats.
This advancement is anticipated to prove a vital resource for police and security agencies, especially given smartphones being integral to daily life.
The research team is reportedly planning to expand their study by integrating more algorithms, a broader dataset, and testing on non-Android devices in the future.
