Google’s numerous safeguards designed to prevent malicious apps from reaching Android users led to the removal of over 700,000 apps from the Google Play Store in 2017, the company said today. That’s a 70 percent increase over the total removals in 2016. “Not only did we remove more bad apps, we were able to identify and action against them earlier,” Google Play product manager Andrew Ahn wrote in a blog post. “99 percent of apps with abusive contents were identified and rejected before anyone could install them.”
Google attributes this success to its improved ability to detect abuse “through new machine learning models and techniques.”
Copycat apps designed to resemble popular mainstays remain a popular method of trying to deceive users, according to Ahn. Google removed over a quarter of a million of these impersonating apps last year. The company also says it kept “tens of thousands” of apps with inappropriate content (pornography, extreme violence, hate, and illegal activities) out of the Play Store. Machine learning plays a key role here in helping human reviewers keep an eye out for bad apps and malicious developers.
“Potentially harmful applications” (PHAs) are apps that attempt to phish users’ personal information, act as a trojan horse for malware, or commit SMS fraud by firing off texts without a user’s knowledge. “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.
Last year, Google put all of its malware scanning and detection technologies under the umbrella of Google Play Protect. The Android operating system automatically performs scans on installed applications to hunt for anything that’s out of place, and users can also manually trigger scans of their Android smartphones right in the updates section. (I’ve finally managed to stop hitting this button when checking for new versions of apps, but it took some time.)
Still, bad apps do occasionally slip through Google’s defenses. In August, Google discovered and kicked out 30 apps that were secretly using the devices they were installed on to perform DDoS attacks. Just earlier this month, the company removed 60 games from the Play Store — some of them meant for children — that were found to display pornographic ads. Google says it will continue to upgrade its methods and machine learning models against bad actors trying to trick consumers with apps that violate its policies. Those efforts indeed seem to be paying off in helping Android’s security turn a corner.