A study shows that the American giant has thousands of "trackers" on pornographic sites. In theory, the information collected is anonymous. But most platforms are very poorly secured. Researchers have used a software called webXray able to analyze a website and detect all the information sent to other sites, using small piece of code called tracker.

The goal is to count the audience of each web page and follow the online route of a user, in order to offer advertising according to the sites visited. According to their study, published July 15 and identified by the New York Times, pornographic sites are particularly rich in trackers of all kinds. In other words, they also collect information about the pages visited by Internet users (And therefore adult videos they have seen), before sending them back to partners.

Of the 22,484 pornographic sites analyzed by the researchers, 93% return data to third-party companies. And in the field, the main player is none other than Google, whose trackers are present on 74% of sites studied. Follow companies specializing in online advertising targeting like exoClick or JuicyAds, but also other US giants like Oracle and Facebook, present on "only" 10% of sites.

The use of the private browsing function having the sole impact of interrupting the recording of browsing history on its own machine, does not prevent this sharing of data.


USER DATA POORLY PROTECTED

In addition to the number of sites concerned, their poor security is pointed out. In total, only 17% take care to encrypt the data exchanged, leaving 83% of them vulnerable to possible hacking attempts. As the New York Times reminds us, these advertising trackers are identical to those found on millions of websites.

If Google and Facebook ensure that they do not use the information collected on porn sites to broadcast targeted advertising to users, nothing says that other actors do not benefit, or do not resell.


CAN I PERFORM AN AUDIT BY MYSELF

If this article interested you, and if you want to go further, we recently wrote a complete tutorial explaining how to use WebXray to identify domains collecting user data.