One in five accept stranger invites on Facebook
Researchers from the University of British Columbia infiltrated Facebook with a host of bot-controlled fake accounts, and sent out a large number of friend requests, resulting in a the acquisition of 250GB of personal data.
The bots managed this task in spite of Facebook’s defensive measures, which include the CAPTCHA system if an account tried to add too many friends in a short period of time, as well as the Facebook Immune System (FIS) which detects suspicious activity and blocks it.
102 bots were used in total, which sent out friend requests to 5,053 random users with public accounts, 976 of which accepted them. The bots then proceeded to sent a further 3,517 requests to friends of those users, 2,079 of which accepted. FIS blocked around 20 percent of accounts after users reported them for spamming.
The end result was a network of 3,055 profiles, with a further 1,085,785 million friends-of-friends in the extended network.
Read the full story over at: Ars Technica.
Loading ...