Українською читайте тут.
The Detector Media Research Center participated in the analysis of a large dataset in Ukrainian for an international study on the malicious pro-Russian propaganda network Doppelganger operating on Facebook and X.
In 2022, the Qurium Media Foundation and EU DisinfoLab identified the Russian-coordinated inauthentic behavior network Doppelganger. This operation is based on mimicking real media outlets and posing as local entities in the targeted country to disseminate narratives favorable to Russia. Two years later, the Counter Disinformation Network (CDN) conducted a study and published a report demonstrating that this influence operation continues to function on X (formerly Twitter) and Facebook. The European Commission is currently assessing whether these platforms are violating their obligations under the EU Digital Services Act in relation to Meta's role in spreading deceptive ads and disinformation, as well as X's efforts to combat information manipulation.
The Counter Disinformation Network gathers more than 130 disinformation-countering practitioners from 40 civil society organizations, universities, and news and fact-checking organizations, mostly from Europe and North America. To ensure quality analysis and interpretation of the data in the context of the languages and countries targeted by the campaign, these data subsets were analyzed by CDN researchers proficient in the relevant languages. The Ukrainian-language dataset was analyzed by analysts from the Detector Media Research Center.
The study revealed that from June 4 to June 28, 2024, a total of 1,366 pro-Russian posts were published on X in six languages: German, French, English, Italian, Polish, and Ukrainian. By the end of June, the total number of views for these posts exceeded 4.66 million.
“The posts discredited Western governments, criticized Ukraine and Western support for Ukraine, exploited polarising issues and targeted Western alliances across all observed languages,” the report states.
The report from the Counter Disinformation Network (CDN) features a chart titled "Distribution of Meta-Narratives Overall," showing that the largest category of posts falls under "Exploiting polarizing issues" (739 posts), which constitutes the largest segment of the diagram. Following this are "Criticizing current government" (641 posts), "Against war against Ukraine" (426 posts), "Against Ukraine" (186 posts), "Undermining Western Alliances" (191 posts), and "Supports alternative government" (83 posts).
According to the report, the accounts posting on X can be divided into two groups: one consisting of accounts that created original content, and another that shared or amplified posts from the first group. The accounts spreading these messages mostly appeared to be American but posted in various languages, pretending to be concerned citizens from European countries.
“Publication and amplification times of the postings showed clear signs of coordination and suggested automated processes. While publishing accounts would usually only create one post each, multiple of them would do so at the same time or very close to one another,” the report states.
As part of the campaign, researchers from France, Germany, Poland, and Italy analyzed 98 Facebook-promoted posts. The audience for these posts was around 560,000 social media users.
“By the time of writing, there was no need to alert Meta about the Facebook ads, as the associated pages that posted the ads had been taken down, or are no longer in use,” the report states.
X's Response
“On July 15th, an alert flagging 1.236 original postings was sent to X, describing the observed activity and providing data to enable internal investigations. While by that time 529 of those postings had already been removed, at least 623 postings were still online, giving X the opportunity to act. As of August 23rd 2024, almost six weeks later, only a single one of those 623 flagged accounts had been suspended, while CDN has yet to receive a reaction from the platform,” the report's authors stated.
During July, some of the remaining tweets received new replies.
The removal rates varied by language: 23.6% of Ukrainian-language posts were removed, compared to 62.4% of German-language posts.
Key Themes in Posts About Ukraine in Different Languages
The proportion of posts criticizing the war in Ukraine in German was 28.1% out of 495 tweets.
“139 posts targeted the war against Ukraine, specifically publishing different arguments as to why the invaded country should not receive western support. This included posts directly calling for a stop to said support, or claims about national disadvantages from its continuation,” writes Lea Frühwirth from CeMAS (Center für Monitoring, Analyse und Strategie), who contributed to the report on German-language tweets.
These claims often presented criticism through overly simplified false dilemmas.
“By depicting aid for Ukraine in alleged competition with needs of the German population, these posts insinuated neglect on the part of the German government,” writes Lea Frühwirth from CeMAS (Center für Monitoring, Analyse und Strategie).
Some of the French-language posts, published during the Global Peace Summit in mid-June, promoted the involvement of Russia in peace negotiations, occassionally even depicting Russian President Vladimir Putin as a beacon for peace.
“In line with the Kremlin’s foreign policy priorities, the campaign’s likely overarching objective is to undermine Western support for Ukraine. However, the favored way to approach the objective is not to address it head-on: over half (57%) of posts do not refer either implicitly or explicitly to any Ukraine-related topics and focus chiefly on polarised domestic issues and attacks on the current government,” write Charles Terroille and Florian Roussel, authors of the section on French-language tweets.
In total, 361 French-language posts were analyzed in the sample.
The researchers note that given the French government's support for Ukraine, any campaign aimed at weakening it serves as an indirect tool to undermine support for Ukraine.
“Out of the 156 posts that directly address the war in Ukraine, 79 attempts to frame an explicit contradiction between supporting Ukraine and the supposed French national interest. For instance, the campaign highlights the fact that the resources sent to Ukraine are sorely needed at home (17 posts), that the French people are being dragged into a war in which they have nothing to gain (14 posts), or that economic sanctions against Russia are backfiring (10 posts),” write Charles Terroille and Florian Roussel.
Thirteen French-language tweets focused on creating a positive image of Russia. These posts mainly centered on two themes. First, Russia's economic successes despite Western sanctions. The second was portraying Russia as a constant geopolitical player interested in peace and stability. Tweets of the latter type included calls to freeze the war in Ukraine and transfer control of the territories occupied by Russia to Russia itself. Such posts were published in the lead-up to and during the Global Peace Summit.
Despite technological advancements and the availability of tools for creating disinformation, some posts were clearly crafted by actors lacking basic knowledge of the situation. For instance, one tweet called for peace negotiations among all parties and included an image depicting a mix of celebrating youth, Western and Ukrainian political leaders, and a large table in a conference room.
A screenshot of the tweet, which image seemed almost unrelated to the text of the post and the broader narrative it aimed to support. The translation reads: "Only discussions among all interested parties can bring peace."
“In the dataset of English-language Doppelganger posts, the most represented narratives involve Ukraine, with just over 36% of the English-language posts mentioning Ukraine or Zelenskyy. These posts allege that Zelenskyy is power-hungry, acts as a puppet for the Democratic party, betrays the Ukrainian people and his allies, fails militarily, and recklessly causes civilian casualties,” write Max Lesser and Ivana Stradner from the Foundation for Defense of Democracies, authors of the section on English-language posts.
The tweets about Ukraine that were included in the sample also discussed U.S.-Ukraine relations. Max Lesser and Ivana Stradner note that the posts aimed to provoke outrage among Americans over U.S. military support for Ukraine, arguing that supporting a "corrupt Ukraine" was wasting "hard-earned American money" and akin to "pouring water into a sieve."
“Just over 25% of English-language posts in the dataset mention Ukraine, while approximately 20% mention Biden. On top of this, just over 34% of posts mentioning Biden also mention Ukraine or Zelensky, showing that Doppelganger often integrates its criticism of Biden into its broader efforts to undermine Ukraine, and vice-versa,” write Max Lesser and Ivana Stradner.
They also clarify that Russian information operations in English continue to use Ukraine as a tool to try to influence U.S. elections.
The Italian-language subset contained 164 tweets.
“Looking at the distribution of meta-narratives across all posts, the Italian subset of the campaign seems to be overall geared towards eroding support for Ukraine (26,5% of total posts), through the undermining of the current, pro-Ukraine government of Giorgia Meloni (16,6%), and of Western alliances (12,6%). To do so, 35,9% across all posts rely on exploiting polarising issues,” write Claudia de Sessa and Lorenzo DiStasi, authors of the section on Italian-language tweets.
Screenshots of tweets in Italian. On the left: “Russia could be an important strategic partner. The interests of the great European leaders should not be interfered with.” Center: “We do not want to participate in wars that are not ours! Stop the pressure!” On the right: “What do we expect from a government that supports the war and not its citizens?”
The authors of the Italian-language section point out that the key theme is the war in Ukraine.
“97 posts out of 164 (59%) speak against the war in Ukraine. The most popular narrative with 35 posts, as hinted above, is that supporting Ukraine damages national interests, often complaining about how funds should be spent in Italy, rather than in Ukraine. This is followed by 18 posts painting the war as an EU/U.S./NATO plot, highlighting how Italy should not get involved in the war to satisfy Western allies, and 16 posts calling for a general stop to Ukraine support,” write Claudia de Sessa and Lorenzo DiStasi.
Similar to French-language tweets, 14 Italian posts also depicted Russia as an ally of Italy, stressing that breaking this cooperation would have negative consequences for Italy.
Thirteen more posts in Italian suggested that Russia was winning the war, while support for Ukraine was framed as prolonging the conflict and being a "bad strategic move for Italy."
The Polish-language subset included 145 posts on X. Of these, 105 posts exploited issues that polarized Polish society (72%), 78 criticized the current government (54%), 39 attempted to undermine Western alliances (27%), 34 used arguments against Ukraine (23%), and 16 were against supporting Ukraine in resisting Russian aggression (11%).
“Economic issues were the most frequently cited concern, appearing in 33 posts, of which 13 focused on Ukraine-related topics: 6 on Ukraine itself, 2 on Ukrainians, 5 on Ukrainian refugees, and 1 on Ukraine's support in the war with Russia. Migration issues were discussed in 17 posts, including one post about refugees and 5 pictures referencing Zelenskyy and the situation at the east border,” write the authors of the section on Polish-language posts, Kateryna Savranska, Dominik Uhlig, and Oleksiy Shymkevych.
Historical grievances between Ukraine and Poland were mentioned in six posts. One tweet discussed the "Ukrainization" of Poland.
“It is worth recalling that the tragic historical moments and territorial claims between Poland and Ukraine are recurrent themes in Russian propaganda, frequently exploited by local pro-Russian actors in Poland to undermine support for Ukraine,” write the Polish disinformation researchers.
Ukraine was the main topic in 25 tweets in Polish. Six of these addressed Ukraine's responsibility for the Volhynia tragedy (five posts) and the loss of Lviv to Poland (one post). Additionally, two posts blamed Ukrainians for taking Polish jobs. Six posts criticized Ukrainian refugees, with five highlighting their negative impact on the Polish economy and one focusing on their impact on Poland's security. Furthermore, five posts claimed that Ukraine's leadership was illegitimate, self-serving, or incompetent.
“These posts typically focused on Zelenskyy, often in the context of the war in Ukraine, or depicted Zelenskyy himself or a Ukrainian man crossing the Polish border and telling Poles to defend his country,” the researchers write.
They add that Ukraine's portrayal in Polish-language tweets appears "quite canonical for the Russian narrative in Poland." Additionally, tactics such as depicting the conflict as pointless, intimidating the audience with the threat of escalation, or accusing Ukrainian authorities of inciting hostility align with the Kremlin's standard approach to persuading the public to pressure Ukraine into making peace with Russia.
The sample also included 89 tweets in Ukrainian. The primary theme of these tweets was the criticism and delegitimization of the state authorities, with Ukrainian President Volodymyr Zelenskyy being the main "villain." In the analyzed tweets, he was frequently mentioned as an "illegitimate," "corrupt," and "undemocratic leader."
“The ongoing Russian aggression constitutes a primary reason for Ukraine's imposition of martial law, preventing the conduct of free and fair democratic elections, according to the elections watch-dogs experts,” commented Oleksandr Siedin, co-author of the analysis of tweets in Ukrainian.
The authors of the tweets suggested protests against the government as a means of restoring justice. During armed conflict, protests could lead to increased disorder and chaos, which might weaken Ukraine's defense due to civil unrest. Russia could then exploit this unrest in its propaganda, portraying Ukraine as a failed state.
In several instances, propagandists used words that do not exist in the Ukrainian language, such as bezchuvalni ("безчувальні"), which could be interpreted as intended to mean "heartless" or "inhumane."
An example of a tweet inciting protests in Ukraine with the invented word bezchuvalni
Other topics targeting the Ukrainian audience included intimidation about the coming winter and blaming the Ukrainian authorities for problems caused by Russian attacks on Ukraine's energy infrastructure, as well as drawing attention to military mobilization in Ukraine.
“The discourse surrounding the politicians' neglect of the personnel of the Armed Forces of Ukraine, the corruption among the elites, and the perception of the population as mere resources for conscription, alongside the pervasive fear of death, all contribute to the narrative that Ukraine is defending itself against Russia, which has instigated an unprovoked war against another nation,” Oleksandr Siedin commented.
A tweet states “Kyiv politicians allocate money for defense, but where is the money for salaries for soldiers, who will answer” and shows a family being declared “fit” for military service.
Propagandists often used humorous illustrations to convey their views on mobilization in Ukraine. For example, one image depicted a family with children, grandparents, and parents behind bars, being declared "fit" for service in the Armed Forces of Ukraine by a character in a military uniform. The text claimed that soldiers in Ukraine were not being paid for their service, while officials in Kyiv claimed to be funding the defense of Ukraine.
The detection of this multinational campaign through its structural features underscores the effectiveness of initiatives aimed at exposing disinformation. It also demonstrates that external observers can uncover malicious networks on social media. However, social media platform administrators responsible for stopping the spread of disinformation under the Digital Services Act used in the EU to regulate social media could be even more effective in uncovering these networks. Social media representatives possess much more information about the sources of disinformation and those managing the networks of accounts.
Main page illustration credits: Media IQ (generated by Midjourney)
