A multi-month hunt for political disinformation spreading on Facebook in Europe suggests there are concerted efforts to make use of the platform to unfold bogus far proper propaganda to thousands and thousands of voters forward of a key EU vote which kicks off tomorrow.
Following the impartial investigation, Facebook has taken down a complete of 77 pages and 230 accounts from Germany, UK, France, Italy, Spain and Poland — which had been adopted by an estimated 32 million individuals and generated 67 million ‘interactions’ (i.e. feedback, likes, shares) within the final three months alone.
The bogus primarily far-right disinformation networks weren’t recognized by Facebook — however had been reported to it by marketing campaign group Avaaz — which says the pretend pages had extra Facebook followers and interactions than all the primary EU far proper and anti-EU events mixed.

“The results are overwhelming: the disinformation networks upon which Facebook acted had more interactions (13 million) in the past three months than the main party pages of the League, AfD, VOX, Brexit Party, Rassemblement National and PiS combined (9 million),” it writes in a brand new report.
“Although interactions is the figure that best illustrates the impact and reach of these networks, comparing the number of followers of the networks taken down reveals an even clearer image. The Facebook networks takedown had almost three times (5.9 million) the number of followers as AfD, VOX, Brexit Party, Rassemblement National and PiS’s main Facebook pages combined (2 million).”

Avaaz has beforehand discovered and introduced far proper disinformation networks working in Spain, Italy and Poland — and a spokesman confirmed to us it’s re-reporting a few of its findings now (such because the ~30 pages and teams in Spain that had racked up 1.7M followers and 7.4M interactions, which we lined final month) to focus on an total complete for the investigation.
“Our report contains new information for France, United Kingdom and Germany,” the spokesman added.
Examples of politically charged disinformation being unfold through Facebook by the bogus networks it discovered embody a pretend viral video seen by 10 million those who supposedly reveals migrants in Italy destroying a police automobile (however was really from a film; which Avaaz provides that this pretend had been “debunked years ago”); a narrative in Poland claiming that migrant taxi drivers rape European girls, together with a pretend picture; and faux information a few little one most cancers middle being closed down by Catalan separatists in Spain.
There’s heaps extra country-specific element in its full report.
In all, Avaaz reported greater than 500 suspicious pages and teams to Facebook associated to the three-month investigation of Facebook disinformation networks in Europe. Though Facebook solely took down a subset of the far proper muck-spreaders — round 15% of the suspicious pages reported to it.
“The networks were either spreading disinformation or using tactics to amplify their mainly anti-immigration, anti-EU, or racist content, in a way that appears to breach Facebook’s own policies,” Avaaz writes of what it discovered.
It estimates that content material posted by all of the suspicious pages it reported had been considered some 533 million instances over the pre-election interval. Albeit, there’s no approach to know whether or not or not all the things it judged suspicious really was.
In a press release responding to Avaaz’s findings, Facebook advised us:
We thank Avaaz for sharing their analysis for us to research. As we’ve mentioned, we’re targeted on defending the integrity of elections throughout the European Union and world wide. We have eliminated various pretend and duplicate accounts that have been violating our authenticity insurance policies, in addition to a number of Pages for title change and different violations. We additionally took motion towards some further Pages that repeatedly posted misinformation. We will take additional motion if we discover further violations.
The firm didn’t reply to our query asking why it didn’t unearth this political disinformation itself.
Ahead of the EU parliament vote, which begins tomorrow, Facebook invited a choose group of journalists to tour a brand new Dublin-based election safety ‘war room’ — the place it talked a few “five pillars of countering disinformation” technique to forestall cynical makes an attempt to govern voters’ views.
But as Avaaz’s investigation reveals there’s loads of political disinformation flying by solely unchecked.
One main ongoing situation the place political disinformation and Facebook’s platform is worried is that how the corporate enforces its personal guidelines stays solely opaque.
We don’t get to see all of the element — so can’t choose and assess all its choices. Yet Facebook has been identified to close down swathes of accounts deemed pretend forward of elections, whereas apparently failing solely to search out different fakes (akin to on this case).
It’s a state of affairs that doesn’t look suitable with the continued functioning of democracy given Facebook’s huge attain and energy to affect.
Nor is the corporate below an obligation to report each pretend account it confirms. Instead, Facebook will get to regulate the timing and stream of any official bulletins it chooses to make about “coordinated inauthentic behaviour” — dropping these self-selected disclosures as and when it sees match, and making them sound as routine as doable by cloaking them in its customary, dryly worded newspeak.
Back in January, Facebook COO Sheryl Sandberg admitted publicly that the corporate is obstructing greater than 1M pretend accounts day by day. If Facebook was reporting each pretend it finds it could due to this fact want to take action through a real-time dashboard — not sporadic newsroom weblog posts that inherently play down the size of what’s clearly embedded into its platform, and could also be so huge and ongoing that it’s probably not doable to know the place Facebook stops and ‘Fakebook’ begins.
The suspicious behaviours that Avaaz connected to the pages and teams it discovered that seemed to be in breach of Facebook’s said guidelines embody using pretend accounts, spamming, deceptive web page title modifications and suspected coordinated inauthentic habits.
When Avaaz beforehand reported the Spanish far proper networks Facebook subsequently advised us it had eliminated “a number” of pages violating its “authenticity policies”, together with one web page for title change violations however claimed “we aren’t removing accounts or Pages for coordinated inauthentic behavior”.
So once more, it’s value emphasizing that Facebook will get to outline what’s and isn’t acceptable on its platform — together with creating phrases that search to normalize its personal inherently dysfunctional ‘rules’ and their ‘enforcement’.
Such as by creating phrases like “coordinated inauthentic behavior”, which units a threshold of Facebook’s personal selecting for what it would and received’t choose political disinformation. It’s inherently self-serving.
Given that Facebook solely acted on a small proportion of what Avaaz discovered and reported total, we would posit that the corporate is setting a really excessive bar for performing towards suspicious exercise. And that loads of election fiddling is free flowing below its feeble radar. (When we beforehand requested Facebook whether or not it was disputing Avaaz’s discovering of coordinated inauthentic behaviour vis-a-vis the far proper disinformation networks it reported in Spain the corporate didn’t reply to the query.)
Much of the publicity round Facebook’s self-styled “election security” efforts has additionally targeted on the way it’s imposing new disclosure guidelines round political advertisements. But once more political disinformation masquerading as natural content material continues being unfold throughout its platform — the place it’s being proven to be racking up thousands and thousands of interactions with individuals’s brains and eyeballs.
Plus, as we reported yesterday, analysis carried out by the Oxford Internet Institute into pre-EU election content material sharing on Facebook has discovered that sources of disinformation-spreading ‘junk news’ generate far better engagement on its platform than skilled journalism.
So whereas Facebook’s platform can be clearly stuffed with actual individuals sharing precise information and views, the pretend BS which Avaaz’s findings suggest can be flooding the platform, will get unfold round extra, on a per unit foundation. And it’s democracy that suffers — as a result of vote manipulators are in a position to go off manipulative propaganda and hate speech as bona fide information and views as a consequence of Facebook publishing the pretend stuff alongside real opinions and professional journalism.
It doesn’t have algorithms that may completely distinguish one from the opposite, and has advised it by no means will.
The backside line is that even when Facebook dedicates way more useful resource (human and AI) to rooting out ‘election interference’ the broader drawback is that a business entity which advantages from engagement on an ad-funded platform can be the referee setting the principles.
Indeed, the entire loud Facebook publicity effort round “election security” seems like a cynical try to distract the remainder of us from how damaged its guidelines are. Or, in different phrases, a platform that accelerates propaganda can be in search of to govern and skew our views.