Google adds new features to Pixels for detecting harmful apps, scam calls

Google Play Protect's live threat detection and the Scam Detection feature were previewed at Google I/O 2024

Google has announced the rollout of new security features on its Pixel smartphones to protect against malicious apps and scam calls. The Google Play Protect live threat detection feature alerts users in real time if it detects malware or suspicious app activity. Similarly, the Scam Detection feature uses on-device artificial intelligence to detect and notify users of potential scam calls in real time. 

 

In an update to its security blog, Google said live threat detection with real-time alerts in Google Play Protect is now available on Google Pixel 6 and newer Pixel models. The company also confirmed that the feature will be available on other Android smartphones in the coming months. The Scam Detection feature for the Pixel phone app is currently in public beta for select Pixel users in the US, with wider availability expected after the testing phase.

 

Google Play Protect live threat detection: Details

 

Google said Play Protect analyses behavioural signals related to the use of sensitive permissions and interactions with other apps and services. If a harmful app is found, the new live threat detection feature sends a real-time alert so the user can take immediate action. The company also stated that live threat detection can now find malicious apps even if they try to hide their behaviour or remain dormant for a period before engaging in suspicious activity.

For privacy, the new feature collects and analyses personal and sensitive data on devices through Private Compute Core, an isolated data processing environment within the Android operating system.

 

Scam Detection in Phone: Details

 

Google said the new Scam Detection feature uses on-device AI to notify users if it detects a potential scam call. The feature detects conversation patterns commonly associated with scams and sends audio and haptic alerts, along with visual notifications, in real time if an ongoing call appears to be a scam.

Google said the AI detection model and processing are fully on-device and no audio or transcription is stored on the device or sent to servers. The feature is disabled by default and can be enabled or disabled by the user at any time.