Sign In

Communications of the ACM

ACM TechNews

Computer Scientists Tap AI to Identify Risky Apps

View as: Print Mobile App Share:
user views the App Danger Project website on a laptop computer

The App Danger Project is a searchable website that provides guidance on the safety of social networking apps.

Credit: Tony Luong / The New York Times

A computational model using artificial intelligence (AI) to evaluate customers' reviews of social networking applications for contextual indicators of their safety has been developed by the University of Massachusetts Amherst's Brian Levine and a dozen computer scientists.

The researchers built the App Danger Project website to provide guidance on app safety by counting user reviews about sexual predators and assessing negatively reviewed apps. The project reported finding a substantial number of reviews suggesting the Hoop app was unsafe for children, with 176 of 32,000 reviews since 2019 including reports of sexual abuse.

Levine envisions the free resource complementing Common Sense Media and other services that check app appropriateness for children by identifying those that do not police users aggressively enough.

From The New York Times
View Full Article – May Require Paid Subscription


Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account