This paper investigates online propaganda strategies of the Internet Research Agency (IRA)—Russian “trolls”—during the 2016 U.S. presidential election. We assess claims that the IRA sought either to (1) support Donald Trump or (2) sow discord among the U.S. public by analyzing hyperlinks contained in 108,781 IRA tweets. Our results show that although IRA accounts promoted links to both sides of the ideological spectrum, “conservative” trolls were more active than “liberal” ones. The IRA also shared content across social media platforms, particularly YouTube—the second-most linked destination among IRA tweets. Although overall news content shared by trolls leaned moderate to conservative, we find troll accounts on both sides of the ideological spectrum, and these accounts maintain their political alignment. Links to YouTube videos were decidedly conservative, however. While mixed, this evidence is consistent with the IRA’s supporting the Republican campaign, but the IRA’s strategy was multifaceted, with an ideological division of labor among accounts. We contextualize these results as consistent with a pre-propaganda strategy. This work demonstrates the need to view political communication in the context of the broader media ecology, as governments exploit the interconnected information ecosystem to pursue covert propaganda strategies.
This article explores the dynamics of digital (dis)information in the conflict between Russia and Ukraine. International Relations scholars have presented the online debate in terms of ‘information warfare’—that is, a number of strategic campaigns to win over local and global public opinion, largely orchestrated by the Kremlin and pro-western authorities. However, this way of describing the online debate reduces civil society to a mere target for manipulation. This article presents a different understanding of the debate. By examining the social media engagement generated by one of the conflict's most important events—the downing of the Malaysian Airlines Flight 17 (MH17) over Ukraine—we explore how competing claims about the cause of the plane crash are disseminated by the state, media and civil society. By analysing approximately 950,000 tweets, the article demonstrates how individual citizens are more than purveyors of government messages; they are the most active drivers of both disinformation and attempts to counter such information. These citizen curators actively shape competing narratives about why MH17 crashed and citizens, as a group, are four times more likely to be retweeted than any other type of user. Our findings challenge conceptualizations of a state-orchestrated information war over Ukraine, and point to the importance of citizen activity in the struggle over truths during international conflicts.
11 December 2020 in Humanities and Social Sciences Communications
This article examines the scope of pro-Kremlin disinformation about Crimea. I deploy content analysis and a social network approach to analyze tweets related to the region. I find that pro-Kremlin disinformation partially penetrated the Twitter debates about Crimea. However, these disinformation narratives are accompanied by a much larger wave of information that disagrees with the disinformation and are less prevalent in relative terms. The impact of Russian state-controlled news outlets—which are frequent sources of pro-Kremlin disinformation—is concentrated in one, highly popular news outlet, RT. The few, popular Russian news media have to compete with many popular Western media outlets. As a result, the combined impact of Russian state-controlled outlets is relatively low when comparing to its Western alternatives.
Mareike Hartmann, Yevgeniy Golovchenko, Isabelle Augenstein. Proceedings of the Second Workshop on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda. 2019.
A growing body of literature examines manipulative information by detecting political mis-/disinformation in text data. This line of research typically involves highly costly manual annotation of text for manual content analysis, and/or training and validating automated downstream approaches. We examine whether Large Language Models (LLMs) can detect pro-Kremlin disinformation about the war in Ukraine, focusing on the case of the downing of the civilian flight MH17. We benchmark methods using a large set of tweets labeled by expert annotators. We show that both open and closed LLMs can accurately detect pro-Kremlin disinformation tweets, outperforming both a research assistant and supervised models used in earlier research and at drastically lower cost compared to either research assistants or crowd workers. Our findings contribute to the literature on mis/-disinformation by showcasing how LLMs can substantially lower the costs of detection even when the labeling requires complex, context-specific knowledge about a given case.