Far-Right Disinformation Flourishes in Facebook Groups, Investigation Reveals
London, UK – September 28, 2025 - A new investigation by The Guardian reveals a network of highly active Facebook groups are serving as significant engines of far-right radicalisation in the UK, despite recent content moderation changes by Meta. The groups, boasting a combined membership of over 611,289 as of July 29, 2025, are disseminating disinformation and fostering anti-immigrant sentiment, often moderated by individuals over the age of 60.
the investigation uncovered deeply concerning rhetoric within the groups,including posts describing immigrants as “blood sucking lice” and accusing the government of endangering citizens by allowing “primitive minded people onto our land.” One user expressed fears of towns and cities becoming “full” of immigrants who would bring “all their crimes and third world culture” to quieter areas.
While Meta reviewed the groups analyzed by The Guardian and determined the content did not violate its hateful conduct policy, experts warn of the dangers posed by these online spaces.Dr. Julia Ebner, a radicalisation researcher at the Institute for Strategic dialog, stated that such groups “definitely play a role in the radicalisation of individuals,” acting as a “breeding ground for extremist ideologies.”
“What is new is that the online spaces amplify a lot of these dynamics,” Dr. ebner explained. “The algorithmic amplification, the speed at which people can end up in a radicalisation engine. Then there are the new technologies from fabricated videos to deepfakes to bot automation.” she also highlighted a growing trend of individuals trusting content from influencers and individual accounts over established institutions, which she described as “inherently dangerous.”
The rise of these groups marks a shift from earlier far-right online activity, which was more commonly found on platforms like 4chan, Parler, and Telegram, typically attracting younger audiences. On facebook, moderators and administrators – like a user named Tuck - are often older individuals. The Guardian identified over 40 people acting in these roles across the three groups analyzed.
This investigation comes months after Meta announced “sweeping content moderation changes” in January 2025, raising questions about the effectiveness of those policies in curbing the spread of extremist content.
The Guardian’s full methodology, including details on its use of OpenAI’s API, can be found here. The publication’s generative AI principles are outlined here.
