This report is from Phase 1 of GIFCT’s research network initiative. Please note that during this time the network was known as the Global Research Network on Terrorism and Technology (GRNTT) and was delivered by the Royal United Services Institute (RUSI).
The aim of this paper is to assess whether social media platforms’ personalisation algorithms promote extremist material. The paper offers an exploratory analysis of recommender systems (for example, YouTube’s Recommended tab on its homepage, which offers video suggestions based on a number of different variables) of three social media platforms – YouTube, Reddit and Gab – when put into contact with far-right extremist content. The paper finds evidence that only one platform – YouTube – prioritises extremist material by the recommender system.
Read full report