The computer programmes resembling humans defeated Facebook’s fake account detection mechanisms, researchers claim
Researchers from the University of British Columbia have demonstrated that Facebook’s fake account detection mechanisms can be defeated 80% of the time with the help of automated tools such as Socialbots, computer programmes resembling humans.
A study conducted by the researchers showed that Socialbots were able to penetrate Facebook, and they harvested 250 GB of personal information belonging to many users on the social site.
Internet criminals are increasingly using Socialbots, computer programmes that mimic real Facebook profiles, which are being sold on the Internet.
In the study, the researchers used 102 Socialbots, which were capable of posting messages and sending friend requests, to send request to 5,053 randomly selected Facebook users at an average of 25 request per day in the first two weeks of the study.
About 19% of the requests were accepted by users in the first phase of the study.
In the second phase during the next six weeks, researchers sent connection request to some 3,517 Facebook friends of users who accepted in the first phase. About 59% of requests were accepted during the period.
The study found that Facebook Immune System (FIS), one of Facebook’s defense mechanisms, was ineffective in identifying and eliminating fake profiles.
Declining the claims of the new study, a Facebook representative indicated they will use the study findings as part of its process of addressing new threats.
The representative further said they have defenses in place to prevent theft of user data.