Disinformation, AI and elections: Articles

This policy brief has three main objectives: (1) to analyse the existing components of the EU’s strategy to combat disinformation, particularly the Digital Services Act, which provides the legal framework for digital services, and the Code of Practice on Disinformation; (2) to identify the key limitations and challenges of the current regulations; and (3) to develop policy recommendations and measures to overcome these limitations and address the threat of disinformation in the light of the upcoming elections.

Even though misinformation, disinformation, and fake news are not new phenomena, they have received renewed interest since political events such as Brexit and the 2016 U.S. Presidential elections. The resulting sharp increase in scholarly publications bears the risk of lack of overview, fragmentation across disciplines, and ultimately a lack of research cumulativity. To counteract these risks, we have performed a systematic research review of 1261 journal articles published between 2010 and 2021. Results show the field is mostly data-driven, frequently investigating the prevalence, dissemination, detection or characteristics of misinformation, disinformation, and fake news.

Developing countries, many of which hold elections this year, struggle to keep up with the emerging tech absent regulations and technical know-how.

Le contexte de la guerre en Ukraine aggrave le risque d’ingérences russes dans les élections européennes, les nouvelles technologies numériques permettant une désinformation à grande échelle. Pour y répondre, l’Europe doit trouver un équilibre entre la réglementation de la désinformation et la protection de la liberté d’expression.

Politics is SUPPOSED to be about persuasion; but it has always been stalked by propaganda. Campaigners dissemble, exaggerate and fib. They transmit lies, ranging from bald-faced to white, through whatever means are available. Anti-vaccine conspiracies were once propagated through pamphlets instead of podcasts. A century before covid-19, anti-maskers in the era of Spanish flu waged a disinformation campaign. They sent fake messages from the surgeon-general via telegram (the wires, not the smartphone app). Because people are not angels, elections have never been free from falsehoods and mistaken beliefs.

Mis- and disinformation labels are increasingly weaponized and used as delegitimizing accusations targeted at mainstream media and political opponents. To better understand how such accusations can affect the credibility of real information and policy preferences, we conducted a two-wave panel experiment. This paper finds that exposure to an accusation of misinformation or disinformation lowered the perceived credibility of the video but did not affect policy preferences related to the content of the video.

The growing influence of social media platforms, and the disinformation that circulates in them, has transformed the public spheres. How to deal with disinformation is an open normative, empirical and political question in contemporary democracies. In this article, we outline an agenda on the institutional strategies pursued in the European Union (EU), the normative understandings of the public sphere that such strategies imply, and the analytical challenges to undertake this line of inquiry. We argue that there is an emerging competition in the EU field of disinformation – constructed by actors coming from different pre‐existing fields, such as journalism or foreign policy – not only to define what is ‘true’ from what is ‘fake’, but also to determine the sort of the public sphere and democracy we ought to strive for.

This article presents an overview of the characteristics of disinformation in the digital age, outlines the evolution of the EU’s efforts to tackling disinformation, and provides an overall analysis of its governance approach along policies, actors, and policy instruments. On this basis, we analyse its compatibility with the EU’s quest for digital sovereignty. Main findings are that the EU has maintained a notably stable approach by combining soft and hard instruments within a co-regulatory framework that involves private and public actors. We conclude that an interplay of the nature of disinformation, the setup of big online platforms, EU regulation, and the exposure to foreign influences creates a complex set of competing factors that facilitate and constrain digital sovereignty.

Internet and social media have revolutionised the way news is distributed and consumed. However, the constant flow of massive amounts of content has made it difficult to discern between truth and falsehood, especially in online platforms plagued with malicious actors who create and spread harmful stories. Debunking disinformation is costly, which has put artificial intelligence (AI) and, more specifically, machine learning (ML) in the spotlight as a solution to this problem. This work revises recent literature on AI and ML techniques to combat disinformation, ranging from automatic classification to feature extraction, as well as their role in creating realistic synthetic content. We conclude that ML advances have been mainly focused on automatic classification and scarcely adopted outside research labs due to their dependence on limited-scope datasets. Therefore, research efforts should be redirected towards developing AI-based systems that are reliable and trustworthy in supporting humans in early disinformation detection instead of fully automated solutions.

This article examines how governments can regulate the values of social media companies that themselves regulate disinformation spread on their own platforms. We use ‘disinformation’ to refer to motivated faking of news. We examine the effects that disinformation initiatives (many based on automated decision-making systems using Artificial Intelligence [AI] to cope with the scale of content being shared) have on freedom of expression, media pluralism and the exercise of democracy, from the wider lens of tackling illegal content online and concerns to request proactive (automated) measures of online intermediaries. We particularly focus on the responses of the member states and institutions of the European Union.

The aim of the piece is to provide a concise handbook of key terms, major actors, and policy recommendations to address current and emerging threats, including trolling campaigns, cyber-attacks, and artificial intelligence (AI). Disinformation threats are only likely to grow as both nation-state and non-state actors exploit technological advancements in hacking, AI and machine learning, and metadata.

Le projet Lakhta de l’entrepreneur d’influence Evgueni Prigojine, à l’origine de plusieurs opérations informationnelles retentissantes, est en cours de démantèlement et de reprise en main par les services de renseignement russes depuis sa mort à l’été 2023. Cette restructuration s’accompagne enfin d’une intensification d’opérations de désinformation plus souterraines, sous-traitées par les autorités russes à des contractuels de l’influence. L’article met en évidence les modalités d’adaptation de ce dispositif protéiforme à la crise politique et sécuritaire la plus importante de la période post-soviétique.

L’échec du pouvoir chinois à imposer un discours médiatique favorable autour des Jeux olympiques de 2008 a conduit le Parti communiste chinois à mettre en œuvre un programme ambitieux d’établissement de médias susceptibles de rivaliser avec CNN et Al-Jazeera dans l’espace médiatique international. CGTN, chaîne de télévision tournée vers l’étranger, est le fer de lance de cette stratégie. Si les travaux sur ce média se sont multipliés ces dernières années, les contenus informationnels et les formes discursives adoptées par CGTN demeurent peu étudiées. C’est la lacune que cet article propose de combler en se penchant sur un corpus constitué de l’ensemble des publications de l’année 2023 classées dans la rubrique « Opinion » du site internet de CGTN, auxquels ont été ajoutés les reportages de l’émission « Facts Tell », un programme qui emprunte au genre du fact-checking.

Afin d’étudier la circulation des fake news en ligne, cet article propose une analyse info-communicationnelle du traitement endogène et automatisé des contenus journalistiques sur les médias socio-numériques dans un contexte de controverse environnementale. Regarder ce qui est fait avec ces contenus pour mobiliser des publics à soutenir une cause permet de définir une typologie de pratiques visant tant à les mettre en contexte qu’en circulation. Les contenus journalistiques étant eux-mêmes standardisés pour maximiser l’audience sur les plateformes, ils deviennent le meilleur moyen d’informer, de mobiliser comme de désinformer, sans avoir nécessairement à les reproduire de manière factice. La désinformation en ligne apparait ainsi comme une conséquence de la plateformisation de l’information et de la communication qui l’autorise voire la renforce.

Depuis leur émergence en 2017, les deepfakes, vidéos générées grâce à l’intelligence artificielle, posent des risques concrets et croissants. Ces technologies, basées sur l’apprentissage profond, ont évolué pour s’appliquer non seulement à des vidéos, mais aussi à des images, des voix et du texte, ouvrant la voie à des scénarios d’usurpation et de désinformation sophistiqués et innovants. Les conséquences potentielles de ces avancées sont nombreuses et vont de l’escroquerie d’individus, d’entreprises ou d’administrations, à la manipulation des opinions publiques et des marchés. Cette capacité à créer de faux contenus de manière convaincante soulève des défis majeurs pour les organisations et les professionnels de l’intelligence économique et représente potentiellement une menace systémique pour la stabilité de la société.

This Report is aimed at supporting electoral management bodies (EMBs)and other relevant parties in developing a broad understanding of the opportunities, challenges and legal implications of the use of AI for elections. It has two main areas of focus. First, it offers a starting point to examine some of the ways in which EMBs may be able to use AI to improve the administration of elections; it presents the risks and potential mitigation strategies associated with those use cases. Second, it covers some of the ways that other, non-EMB actors may use AI to impact election processes as well as potential strategies for EMBs to respond. It describes some of the regulatory frameworks for AI starting to take shape around the world, and explains how they may impact the work of EMBs that are considering using, and responding to other actors’ use of, AI as a part of the electoral process.

The success and widespread deployment of artificial intelligence (AI) have raised awareness of the technology’s economic, social, and political consequences. Each new step in the development and application of AI is accompanied by speculations about a supposedly imminent but largely fictional artificial general intelligence (AGI) with (super-)human capacities, as seen in the unfolding discourse about capabilities and impact of large language models (LLMs) in the wake of ChatGPT. These far-reaching expectations lead to a discussion on the societal and political impact of AI that is largely dominated by unfocused fears and enthusiasms.

AI-powered tools could assist in the administering of elections, but their capability to synthesize vast amounts of data and disseminate mis- and disinformation could also pose risks to election integrity. AI has many opportunities to democratize the campaign playing field, but it could also turbocharge preexisting election interference tactics.

To protect election infrastructure and personnel from AI-generated threats, election offices and vendors must implement the best practices that experts have been urging for over a decade.

It is well documented that there has been significant enthusiasm across the globe in respect of using AI for all forms of social activity. However, the electoral process – the time, place, and manner of elections within democratic nations – is one of few sectors in which there has been limited penetration of AI. Electoral management bodies in many countries have recently started exploring and deliberating over the use of AI in the electoral process. In this paper, we consider five avenues within the core electoral process which have potential for AI usage, and map the challenges involved in using AI within them.