Google's new machine learning techniques enabled it to remove 99 percent of apps with abusive content before anyone had installed them.
Google just released a security update on its Android blog in which it reveals how it fought bad apps this year, revealing it combined new machine learning models and techniques on top of actual humans monitoring app submissions. "During that process, apps are specifically reviewed for compliance against our Google Play Developer Content Policy and Developer Distribution Agreement, which prevents things like apps that are impersonating legitimate companies or deceptive behavior". Google reports that it took down more than 700,000 "bad" apps in 2017, and it did so faster than ever before. Inappropriate content, according to the company's definition, includes pornography, extreme violence, illegal activities, and hate. The company says it is 70 per cent more apps compared to previous year. Thanks to Google Play Protect, which was launched in 2017, the annual PHA installs rate was reduced by 50 per cent over year.
Google has historically had a problem with copycat apps that contain malware.
About 8 billion installs from Google Play take place every month across the world. "While small in volume, PHAs pose a threat to Android users and we invest heavily in keeping them out of the Play Store", Ahn said. It's also nice to know that most of the rejected apps were copycats that hoped to score some coin by mimicking other apps. People searching the web for Jio Coin spiked and as a result, fake websites and apps are luring netizens towards spam.
Devices running Android 6.0 Marshmallow and above can manage permissions granted for apps. We take these extremely seriously and will continue to innovate our capabilities to better detect and protect against abusive apps and the malicious actors behind them. The firm also claims that 99% of the malware was identified and removed before even being installed by anyone. At least 100,000 developers out there would rather resort to creating some sort of malicious app to take advantage of Android users rather than come up with application ideas that would generate revenue.