Google is expanding the AI capabilities of its Google Play Protect live threat detection feature that scans apps installed on Android phones for malware, the company announced at the recently-concluded Google I/O annual developers conference.

With the expanded on-device AI capabilities, Google Play Protect live threat detection will improve fraud and abuse by catching apps that try to cloak their actions. As per Google, Google Play Protect scans 200 billion Android apps daily, helping keep more than 3 billion users safe from malware.

Live threat detection

Google explains that with live threat detection, Google Play Protect’s on-device AI will analyse additional behavioural signals related to the use of sensitive permissions and interactions with other apps and services.If the system finds suspicious behaviour, the feature will send the app to Google for additional review and warn users or disable the app if malicious behaviour is confirmed.

“The detection of suspicious behaviour is done on device in a privacy preserving way through Private Compute Core, which allows us to protect users without collecting data,” the company said.

Google Pixel, Honor, Lenovo, Nothing, OnePlus, Oppo, Sharp, Transsion, and other manufacturers will deploy live threat detection later this year.

Security tools

Apart from user-facing features, the company is also providing developers with more tools to safeguard apps from scams and fraud. “The Play Integrity API lets developers check that their apps are unmodified and running on a genuine Android device so that they can detect fraudulent or risky behaviour and take actions to prevent attacks and abuse,” Google said.

It is providing more tools to keep users safe. Developers can now check if there are other apps running that could be capturing the screen, creating overlays, or controlling the device.

Additionally, they can check if Google Play Protect is active and the user device is free of known malware before performing sensitive actions or handling sensitive data, thereby, helping them avert any dangers posed when using financial and banking apps.

Developers can also opt-in to receive recent device activity to check if a device is making too many integrity checks, which could be a sign of an attack.