A publication from Reuters reveals that Google and Facebook use an automated system to block extremist and violent video, similar to that used to identify copyright content.
The publication lists two anonymous sources from each business that revealed that Facebook and YouTube staff have developed a system that automatically detects extremist or violent content and removes it without interfering or supervising the user.
The system is similar to that used by both Companies to detect illegal copies of copyrighted content such as movies, music videos, or the background music of some videos. Both companies created these prototype systems after pressure from groups protectionof intellectual property, such as the RIAA and the MPAA.
Reuters reports that both Google and Facebook seem to use a modified version of this system to track ISIS videos.
The only downside is that the system can not detect the original video once it is first published on the network. Both companies are still based on users who point to the content, and one of their employees evaluates it manually.
Once added to their databases, both networks are able to recognize when another user is trying to re-upload or repost it.
Reuters claims that the two agencies (Google, Facebook) created the systems at the behest of US President Barack Obama and other US and European leaders, as they worried that terrorists had become extremely successful in promoting extremist content online, which helps them recruit new members.
The meeting was called Counter Extremism Project and took place at the end of April. In fact, Twitter and Cloudflare are said to have also participated in it.
None of the above companies wanted to give more technical details on how their system works.
