AI-assisted content moderation promises to manage online incivility at scale while sparing human moderators traumatic content. Yet most evidence comes from Western languages and contexts. This study examines how AI hate-speech detection interacts with counterspeech and fact-checking in Ethiopia’s polarized Amharic and Afan Oromo spaces. Drawing on infrastructural invisibility and civic labor, it triangulates a computational audit of three classifiers against 838 hand-annotated posts (2020–2025), document analysis of platform self-descriptions and transparency reports, and 20 interviews with Ethiopian fact-checkers, volunteer flaggers, and counter-speakers. The most widely used generic classifier cannot read either language natively and, on English translations, recovers only about a tenth of hate speech; locally-oriented classifiers perform far better on Amharic but collapse on Afan Oromo, leaving it effectively unserved. Across every tool, the dominant error is under-detection, not over-removal, and the same silence recurs in platforms’ self-descriptions. Rather than sparing anyone the work, AI’s failure displaces it onto an unpaid volunteer ecology that absorbs political and psychological costs platforms benefit from yet refuse to name.
This study employs content analysis and semi-structured in-depth interviews to examine how journalism education across BRICS countries has responded to the rise of AI and adjusted its core paradigms. In this study, BRICS refers to the expanded BRICS grouping, comprising Brazil, Russia, India, China, South Africa, Egypt, Ethiopia, Iran, Saudi Arabia, and the United Arab Emirates. It combines latent Dirichlet allocation topic modeling of university journalism curricula with 15 interviews of China-based journalism educators. We selected 10 representative BRICS universities with journalism and communication programs as samples, after which we compiled training frameworks, syllabi, and teaching documents for all their AI-related courses, and used the LDA model to extract core topics and to conduct manual qualitative content analysis across five dimensions: course localization, technical tools, skill guidance, teaching content, and values. The interview results provided valuable supplementary insights.
Watchdog science journalism (WSJ)—investigative journalism whose objects are scientists and scientific institutions—is an important way of complying with the public service commitment of journalism. It is also one of the instruments to uphold the integrity and independence of scientific research in the face of misconduct, corporate and political interference, ideological manipulation, and scientific misinformation. But science journalists often find themselves in a “too close for comfort” relationship with scientists, which can compromise their critical stance. This problem is even harder in Ibero-America, a cultural region encompassing Spain, Portugal, and Latin American countries. Here, investigative journalism has historically been less prevalent, and science journalism consolidated later. But recent outstanding journalistic work suggests the emergence of a vibrant investigative science journalism scene in the region. In this study, we explore this development by interviewing 15 journalists from Ibero-America who have carried out outstanding WSJ on science, health, and technology. Based on their answers, we identify practices, values, impacts, barriers, and facilitators of WSJ in the region. We also identify what they perceive as constraints specific to Ibero-America, including political interference, lack of transparency, disabling newsroom culture, precarious funding, problematic relations with scientists, misuse of results, and job and psychological instability. For each of these challenges, we also identify enabling actions that are fostering or could foster WSJ in the region. Our work signals a departure from historical trends and ingrained practices, calling for a reconsideration of the contribution of Ibero-America to global science journalism.
With the rise of generative AI, the increasing threat of automatically generated uncivil content (including misinformation for information warfare up to cyber-bullying purposes) makes the protection of open online discourse even more pressing than before. In the implementation of measures for countering these threats, the different intervention objectives of stakeholders, their workflows, and IT support need to be considered. Stakeholders include online social network moderators, journalists, fact-checkers, social listeners, as well as authorities and organizations with safety- and security-related tasks. Given the sheer volume of online social network content, automated or community-based solutions are required to detect (automated) misinformation. However, detection solutions are predominantly message-focused and target end-users, leaving experts without systematic, large-scale perspectives on coordinated disinformation campaigns to guide countermeasures. To bridge this gap, we conduct an expert-centered study on the integration of detection methods proposed by the research community. Our contributions are threefold: (a) We adopt disinformation features from a previous study and draw connections to literature on detection methods; (b) semi-structured interviews yield vignettes that expose the spectrum of goals, constraints, tools, and decision-making processes employed by experts, informing requirements for method integration; (c) we design a demonstrator that showcases representative methods to uncover unexplored concepts, probe affordances and limitations in context, and evaluate conceptual fit during the interviews. Together, these steps bridge the gap between data- and AI-driven detection techniques from research and the practical needs of diverse stakeholders confronting targeted and large-scale disinformation.
Journalism research differentiates between various conceptions of roles and societal functions that journalism is expected to perform. Previous research has mainly examined journalists’ role perceptions, while audience views remain underexplored. However, in the context of science journalism on controversial scientific fields such as climate change or Covid-19, audience insights are particularly relevant given that media coverage may shape public perceptions of science and influence crisis responses. This study combines semi-structured interviews and focus group discussions to investigate what functions audiences expect from science journalism on controversial scientific fields, focusing on context-dependent (dis)approval of the watchdog role. Findings indicate that audiences expect science journalism to fulfill both traditional and science- or crisis-specific functions. Regarding the watchdog role, participants emphasized journalists’ task of uncovering misconduct and scientific norm violations in research underlying policy decisions. However, participants also viewed watchdog journalism as a balancing act as there are many forms of inappropriate critical reporting that might foster distrust in science and media. The findings are discussed regarding their implications for journalism practice and the role of watchdog science journalism in shaping public perceptions of controversial science.
Data sharing poses a major challenge for digital media and communication research, particularly in sensitive areas such as far-right online studies. This article introduces the innovative concept of a “community data trustee” (CDT), a research infrastructure aimed at fostering collaboration and the sharing of research data, such as digital account lists. Compiling these lists is a critical yet labor-intensive step in many research projects; sharing them could significantly reduce effort and improve data quality. However, especially in sensitive research areas, sharing remains rare due to legal uncertainties and limited incentives. This hinders the traceability and comparability of research findings and threatens overall research quality. To address these challenges, we introduce the CDT as a collective research infrastructure that promotes the shared use of extensive account lists. The CDT views actor directories as a communal asset, developed and utilized based on mutually agreed-upon guidelines. A principle of reciprocity underpins this model: Those who access the lists also contribute to their updates and expansions, returning them to the data pool. An online portal facilitates the exchange and collaborative maintenance of the data. The setup of the CDT includes appropriate technical measures to ensure compliance with data protection and security standards, along with a robust regulatory framework that creates a legally secure environment for sharing personal data. This approach aims to (a) incentivize data sharing, (b) foster trust and legal certainty among research projects, (c) enhance data quality through ongoing maintenance, and (d) enhance researcher safety.
The White Lotus is renowned for its satirical depiction of wealthy, upper-class individuals. The show deliberately invites its viewers to reflect on a set of moral/ethical questions about ambiguous characters and their behaviors. This study investigates viewers’ responses to this invitation to morally deliberate in relation to class. First, a narrative analysis traces the moral plotlines in Seasons 1 and 2 of The White Lotus. Building on this, thematic analysis is used to examine viewers’ moral discussions on the online platform Reddit. In this way, both the show’s invitation and how viewers navigate this invitation are unraveled. Results show a relation between characters’ class and gender, the amount of empathy the characters evoke among viewers, and viewers’ deliberation on moral issues these characters are involved in. Viewers actively engaged with the moral plotlines concerning female and middle-class characters. The male and upper-class characters received remarkably less attention, and their immoral behavior was taken as a given characteristic of the upper class. We argue that these patterns show a connection between empathy felt for a character and readiness to engage in moral deliberation of their actions. This finding contributes to our understanding of narrative imagination, its connections to class and gender, and its manifestation in an online context.
This article examines the politics of representation in the face of today’s growing audience fragmentation and the entry of genAI tools into televisual entertainment. Establishing that contemporary TV content is shaped by normative discourses on how to represent social difference, the article highlights how text-to-image/video models accelerate and amplify their mark on user-generated content. Conducting a walkthrough exploration of Showrunner—a genAI-powered tool to generate short, animated scenes with—it demonstrates how such models operationalize dominant constructions of how to appropriately represent social differences like gender, race, or sexuality. This brings a key tension into focus: Today’s broad recognition of representation’s political salience translates into functionalism and instrumentalization. By working to discontinue “inappropriate portrayals,” contemporary representational politics ignore pop-cultural complexity in favor of “actionable” dichotomies meant to regulate “harmful content” out of existence.
We develop “moralized misogyny” as an analytic concept for examining how Meta’s Facebook functions as a form of informal patriarchal governance regulating Bangladeshi women’s political visibility. Drawing on theories of misogyny, morally motivated networked harassment, and digital vigilantism, we argue that women’s political engagement is disciplined and exposure enforces dominant moral norms. Integrating feminist and multimodal approaches to critical discourse analysis of purposively sampled Facebook items, we show how political disagreement is reframed as moral transgression. Women’s participation is recoded as sexual deviance and impurity through visual and textual manipulation that render delegitimizing attacks credible, humorous, and socially acceptable. Whether audiences believe these artifacts is often secondary; their circulation enables crowd-led vigilante punishment framed as moral defense. This dynamic can constitute a form of structural equality harm that makes women’s political citizenship conditional on compliance with patriarchal norms. We recommend context-specific moderation and policy responses that recognize such attacks as a barrier to women’s political participation.
This article examines how Chinese bromance dramas produce and circulate homo-desire under a regime of moral censorship. Existing commentary often treats bromance as a strategic substitute for boys’ love (BL): a compromised form that compensates for the absence of explicit same-sex romance. Challenging this assumption, we draw on Deleuze and Guattari’s concept of desiring-production to argue that bromance is not defined by what it cannot show, but by what it actively generates through aesthetic form and participatory spectatorship. Focusing on Justice in the Dark (Guangyuan, 2023), we combine textual analysis with online observation of viewer discussions, screenshots, and fan edits on RedNote. We identify two interlocking mechanisms through which homo-desire becomes legible without explicit naming. First, a widely cited “sense of atmosphere” (fenwei gan) emerges through viewers’ capture and interpretation of lighting, touch, objects, and eye-acting, by means of which forms of male–male intimacy “taken away” by moral censorship are re-produced and made desirable precisely because of their unspoken mutuality. Second, viewers’ reinterpretations of the character Luo Wenzhou, remaking an authoritative police figure into an erotically appealing beagle-husband, queering state-aligned masculinity. By reframing bromance as a productive genre rather than a deficient substitute for BL, this article contributes to debates on censorship, queer visibility, and the politics of spectatorship in contemporary Chinese screen culture.
This article explores how gender politics become moralized focal points of conflict within increasingly fragmented, hybrid, and alternative media environments. Drawing on feminist media studies and scholarship on affective polarization, as well as Mau et al.’s notion of trigger points (2023), it introduces the concept of “moralized trigger loops”—repetitive, emotionally charged cycles in which gender-related questions are amplified across digital and alternative media spheres. While earlier research has positioned alternative media primarily as emancipatory counterpublics, recent work shows how new forms of alternative media are actively engaged in the (re)production of anti-feminist, reactionary, and moralized gender discourse. Yet little is known about how users of such outlets themselves negotiate gender issues, how moralization unfolds affectively in their everyday media practices, and how these dynamics evolve over time. Based on a longitudinal qualitative panel study with system-critical alternative media users (n = 33) over three years, this article investigates how gender politics emerge as moralized sites of contestation within diverse media repertoires including alternative media. It argues that such processes transform gender-related debates into enduring affective loops that sustain and reshape mediatized publics.
This study examines how mother–child evacuation following the Fukushima Daiichi nuclear disaster has been discursively constructed in Japanese mainstream newspapers over the first decade after the accident. Drawing on theories of risk society, mediapolis, and proper distance, and employing a longitudinal critical discourse analysis of five major national newspapers published between 2011 and 2021, this article demonstrates that Japanese newspapers consistently portrayed mother–child evacuation as a social wrong, while simultaneously individualizing moral responsibility by foregrounding mothers as primary caregivers. Fathers, children, and institutional actors, including the Japanese government and the Tokyo Electric Power Company, Incorporated (TEPCO), were largely marginalized as morally responsible subjects. This gendered framing stabilized a particular moral order in which women’s precautionary decisions were moralized, while structural and institutional responsibility was backgrounded. By foregrounding gender as a structuring principle of mediated moral relations, this study advances Silverstone’s concepts of mediapolis and proper distance and contributes to debates on media, morality, and risk governance in post-disaster societies.
In recent years, we have witnessed a growing body of research focusing on identity politics, public shaming, and social outrage in the contemporary media landscape. Morality is often at the core of these debates. Recent studies show how all media types are rich sources of moral statements, producing an endless stream of moral discourse on all imaginable topics and themes. Yet, how these discourses are shaped in media production and what audiences and societies do with these discourses is not easy to predict. Simultaneously, gender politics are closely involved in these articulations of moral norms. This thematic issue aims to explore the historical trajectory and continuity of the vital role of media in shaping social dynamics and moral norms, particularly through the lens of gender, from a global perspective.
Authenticity has become a key concept in political communication, particularly during election campaigns, which offer opportunities for politicians to strategically convey an authentic image. Studies show that voters increasingly value politicians who appear authentic, a factor influencing their voting behavior. In particular, social media provides an ideal environment for politicians to perform authenticity during campaigns by allowing them to bypass traditional gatekeepers and present themselves in an unfiltered and intimate way. Previous studies have begun to analyze how politicians perform authenticity on social media. However, few have examined and compared how often and in what ways different politicians present themselves as authentic on social media during election campaigns. Drawing on scholarship on political authenticity and gendered self-presentation, this study examines how politicians construct authenticity through textual and visual elements across social media platforms. We conducted a manual quantitative content analysis of social media posts (N = 855) on Facebook, Instagram, and Twitter by the three German lead candidates running for chancellor during the 2021 federal election. Our findings show that authenticity and its facets were an integral part of all candidates’ social media presence and their election campaign communication. Furthermore, we found differences in how the one female and the two male politicians in our sample perform authenticity.
This study investigates the strategic use of visual and verbal disinformation by astroturf political influencers for negative campaigning and political mobilization. Focusing on the case of the 2022 national election campaign in Hungary, the research examines how seemingly grassroots influencers, who conceal affiliations with political actors, engage in astroturfing, a deceptive communication tactic involving the dissemination of fabricated or misleading information. Specifically, the analysis centers on the content produced by Megafon, a prominent astroturf influencer network on Facebook, the dominant platform for political communication in Hungary. To address the research questions, a total of 2,655 Facebook posts were analyzed using manual quantitative content analysis. The dataset comprises both visual and verbal content, with a particular focus on negative campaigning and political mobilization. The findings indicate that disinformation is deeply embedded in both the textual and visual dimensions of astroturf content, with visual disinformation being more prevalent than purely textual. A recurring tactic identified is decontextualization, where genuine visuals are paired with misleading commentary, thereby distorting meaning and intent. Visuals are frequently used to misrepresent opponents’ actions and mobilize supporters under false pretenses. These strategies are deployed without transparent disclosure of influencers’ political connections, blurring the boundaries between organic activity and coordinated political communication. By analyzing these manipulative practices, the study emphasizes the urgent need to recognize astroturf visual disinformation as a potent instrument for undermining democratic discourse and highlights the critical role of social media as a battleground for political manipulation in contemporary electoral contexts.
As social media influencers receive increased attention as intermediaries in the political domain, political parties are seeking to expand their involvement with them. However, knowledge of how parties get involved with influencers remains limited. This study presents a qualitative mapping of parties’ influencer engagement, drawing on a secondary analysis of two datasets containing a combined 25 expert interviews with party communicators from Germany. It theorizes influencer involvement as the parties’ management of social interactions. Employing an adapted version of Godes et al.’s (2005) framework for analyzing a firm’s management of social interactions, it systemizes the 35 identified cases of influencer involvement as instances of different management strategies. While the moderator strategy—characterized by relationship-building and information exchange—emerged as the most prevalent, the participant strategy—marked by partisan collaboration—enjoyed the highest popularity. Overall, the findings indicate that the parties’ influencer involvement is still in a formative phase, marked by experimental practices, resource constraints, and ambivalence toward influencer autonomy.
Campaign practices evolve alongside technological change. We examine one of the most salient current developments: the rise of short-form video on platforms such as TikTok and Instagram Reels—often termed the “TikTokification” of election campaigns (Gerbaudo, 2024). The adoption of short-form video may signal the arrival of Römmele and Gibson’s (2020) “subversive” fourth era of campaigning, characterised by emotion, disruption, spontaneity, and the mimicry of authenticity. Here, we examine how the five main UK parties used short-form content during the 2024 UK General Election through a manual content analysis of all TikToks and Instagram Reels posted during the campaign period (N = 887). We find evidence of extensive but uneven adoption of short-form video across parties, with TikTok generating substantially higher reach and engagement than Instagram Reels. Whereas Reels were largely used to repurpose traditional campaign material, TikTok served as a site of experimentation, with parties more frequently deploying humour, memes, and in-app music. Leader-centred communication remained dominant overall, but traditional campaign functions were more pronounced on Reels than on TikTok. Thus, results suggest a compressed cycle of experimentation and standardisation. Furthermore, TikTokification occurred mainly on TikTok itself rather than diffusing across short-form platforms.
This study examines whether visual generative artificial intelligence (VGenAI) serves as an equalizing force for minor parties or reinforces existing power asymmetries in political communication. Drawing on equalization and normalization theory, we investigate party differences in VGenAI adoption, content strategies, and user engagement during the 2025 German federal election. Using a semi-automated AI detection method combining automated classification with manual validation, we analyzed Facebook and Instagram posts from 37 German parties, identifying nearly 1,000 VGenAI images and videos published by approximately 400 party accounts during the four weeks preceding election day. Findings reveal evidence for both theoretical perspectives—equalization and normalization—across analyzed dimensions. Regarding adoption, minor parties used VGenAI at higher rates than major parties, supporting the equalization hypothesis, which states that low-cost technologies enable resource-constrained actors to produce professional campaign visuals. Content strategy analysis reveals a transparency divide, with mainstream major parties disclosing AI origins more frequently than minor parties or the right-wing Alternative for Germany (AfD). The AfD distinguished itself as the only major party to make extensive use of photorealistic imagery, citizen depictions, criminal portrayals, and negative tone, consistent with populist communication strategies. However, engagement analysis supports normalization: While VGenAI content is associated with higher user engagement than non-AI posts, this advantage accrues equally to major and minor parties rather than providing resource-constrained actors with competitive benefits. VGenAI thus appears to be associated with broader access to professional visual production while leaving engagement asymmetries intact. These findings advance understanding of how emerging communication technologies interact with party system structures and have implications for regulatory approaches to synthetic political content.
This article studies negative advertising on election posters in Austria over a time period of 79 years. Election posters are still one of the most important and long-lasting campaign tools in Austria, therefore allowing for the examination of long-term trends in political communication. The article uses data from 1,082 posters from 24 national legislative elections. A multilevel model is used to test whether the level of negativity decreases over time, whether opposition parties and smaller parties resort to negative messages more often, and whether the degree of negativity varies with party system polarization. The results indicate that negativity on election posters has steadily decreased. Parties choose to use the public space to promote their own strengths rather than the opponent’s weaknesses; negative messages may have disappeared or moved to the digital sphere. The article contributes to the literature by explaining the seemingly “recent” phenomenon of negative advertising in a historical context.
The infrastructures and actor constellations through which election campaign communication unfolds have changed, with platforms, influencers, and artificial intelligence (AI) tools reshaping how campaigns operate. This raises the question of whether contemporary campaign communication differs fundamentally from that of earlier election cycles. Current debates oscillate between two extremes: either contemporary campaigns are fundamentally transformed by these developments, or they largely continue earlier practices, with new technologies merely adding tools to an established strategic repertoire. This thematic issue moves beyond this binary. The contributions examine recent election campaigns in Europe and the United States and show that core campaign strategies are rather stable. Mobilization, personalization, negativity, and emotional appeals continue to structure electoral competition. At the same time, the mechanisms through which these strategies are produced, circulated, and amplified are changing. Campaign communication increasingly unfolds within hybrid actor constellations that include influencers and supporter networks, rely on platform-specific communication styles such as short-form video and memes, and operate within engagement-driven environments in which emotionally charged content is more likely to spread. Taken together, the articles suggest that contemporary election campaigns operate within a communication environment shaped by platforms, influencers, and AI. Established theories remain relevant but they require adjustment to account for changes in production, circulation, and amplification. By integrating research on actors and strategies, this thematic issue clarifies how continuity and change interact in contemporary election campaigns.
