Google has unveiled two AI-based tools for Android devices. The tools will protect users from fraudulent phone calls and malicious apps. The company unveiled the tools last Wednesday.
Google's first tool is 'Scam Detection in Phone by Google'. It identifies which calls may be fraudulent by analyzing the conversation patterns of phone calls.
The second one is 'Google Play Protect Real-Time Alerts'. This tool helps detect malicious apps by monitoring the background activity of the app after it is installed on the device.
Google detailed how the security tools work in a blog post. Both tools are now rolling out to Google Pixel 6 and newer models. Initially, users of the Google Beta program in the US will be able to use the Scam Detection in Phone tool. Phone call based tools will primarily work on English language phone calls. The Google Play Protect real-time alerts tool is also available outside the US.
The scam detection feature is different from normal caller ID apps and services. They use phone numbers to identify which numbers may be fraudulent. Google's new tool, on the other hand, will use on-device machine learning models to identify fraudulent calls. It will process the conversation type of the call in real time. For example: In Google's Pixel 9 series, the tool will use the Gemini AI model.
Google has provided an example to illustrate how the tool will work. If someone claims to be calling from a user's bank and asking them to transfer money due to a security breach, Google will process the audio data, the company says. The AI can then use the database to identify whether such conversation patterns have been used to commit fraud before.
When it detects a call may be fraudulent, the AI model will provide audio and haptic (vibration) alerts and display a warning on the screen. The feature will be turned off by default, and users can turn it on for all calls, or for specific calls, by going to the Phone app's settings, Google noted. The company claims that no audio or transcription of the call will be stored on the device. It will not even be sent to Google servers or anywhere else and cannot be retrieved after the phone call is over.
The second feature is part of Google Play Protect. It monitors malicious and dangerous apps in play store. Using the AI-based 'Live Threat Protection' feature, Google's AI models will monitor apps installed on Android devices. If an app behaves suspiciously or interacts unnecessarily with other apps, the tool will display warning alerts on the device. Through this users can be aware.
Google claims that this AI tool can also identify apps that have been inactive for some time after installation. Also, since the tool can detect malicious behavior in real-time, it will help users take timely action. As a result, it will be possible to prevent data theft. Specifically, the tool will initially focus on identifying 'stalkerware' and 'malicious apps' that collect personal or sensitive data without the user's permission.
References: 360 gadgets
