Das ZeMKI auf der AoIR 2025
13. Oktober 2025
Die 26. Jahreskonferenz der Association of Internet Researchers (AoIR) wird vom 15. bis 18. Oktober 2025 in Rio de Janeiro, Brasilien stattfinden. Das diesjährige Thema „Ruptures“ widmet sich alternativer Geschichte des Internets und technologischen Perspektiven jenseits westlicher Dominanz.
Das ZeMKI wird auf der AoIR 2025 mit zahlreichen Beiträgen vertreten sein.
Eine Übersicht der Beiträge der ZeMKI-Mitglieder:
Donnerstag, 16. Oktober 2025
9:00-10:30 Uhr, Room 7a – Groundfloor
Tariq Choucair, Laura Vodden, Ahrabhi Kathirgamalingam, Paul Pressmann, Fabio Giglietto, Katharina Esau, Cornelius Puschmann, Axel Bruns, Giada Marino, Bruna Paroni: „Humans and machines in the loop: Rethinking LLMS for conflict and disagreement in content analysis of social complex phenomena“
The study of polarization, conflict, and ideological divergence has long challenged scholars across media, communication, and political science. Understanding these phenomena requires engaging with fundamental questions about how opinions are formed, reinforced, and contested in public discourse, and how language and discourses can have so many different meanings and interpretations. Traditional content analysis methods often try to capture this complexity – whether in large-scale media narratives, open-ended survey responses, or political discourse on social media. Computational approaches offer new possibilities, but also raise critical concerns about validity, interpretability, and methodological rigor (Baden et al., 2022; Boumans & Trilling, 2016).
Large Language Models (LLMs) have become increasingly central to communication research, with their capacity to process vast amounts of text, identify underlying patterns, and assist in qualitative coding (Chew et al., 2023; Alizadeh et al., 2025). Researchers have explored their use in a variety of tasks, from detecting polarization in open-ended survey responses to analyzing media frames and political discourse (DiGiuseppe & Flynn, 2025; Marino & Giglietto, 2024). However, the integration of LLMs into content analysis presents challenges: How well do they align with human interpretations? Can they enhance research beyond automation? And what role should they play in investigating contested or ambiguous meanings (Pilny et al., 2024; Gunes & Florczak, 2025)?
This panel moves beyond discussions of mere optimization and accuracy, instead critically and deeply discussing the use of LLMs to engage with conflict, disagreement, and interpretive diversity. Instead of seeing them as tools to impose consensus, we ask how researchers can use and interact with them to reveal tensions, challenge assumptions, and contribute to new methodologies (Dai et al., 2023; De Paoli, 2024). Across four studies, we assess LLMs mediating, amplifying, or reframing scholarly debates on the methods to analyse contentious political and social issues. Our discussion examines both the benefits and risks of these approaches, raising questions about the role of AI in media and communication methodologies.
Freitag, 17. Oktober 2025
11:00-12:30 Uhr, Room 11a – Groundfloor
Blake Hallinan, CJ Reynolds, Rebecca Scharlach, Dana Theiler, Isabell Knief, Omer Rothenstein, Yehonatan Kuperberg, Noa Niv: „From community guidelines to industry standards: Mapping the policy priorities of mainstream, alternative, and adult live content platforms“
Despite growing concern over the standardization of content moderation, there has been little empirical investigation beyond mainstream social media. We developed a novel approach to compare rules and policy priorities within Community Guidelines based on categories from the Trust and Safety Professionals Association. We focused on livestreaming, a particularly challenging format to moderate, and asked: what policies govern content? And how do mainstream, alternative, and adult content platforms differ? We analyzed 12 platforms and identified four orientations towards industry standards: the mainstream ideal, the regulatory competitor, the alternative ethos, and the overlooked concerns. These orientations partially map onto divisions between mainstream, alternative, and adult livestreaming platforms, allowing us to pinpoint different factors driving the adoption of industry standards. Finally, we discuss the tradeoff between free expression and sexual expression, highlight epistemological considerations regarding the use of policy documents, and conclude with an agenda for future comparative research.
16:00-17:30 Uhr, Room 11d – 2nd Floor
Rebecca Scharlach, Taylor Annabell, Blake Hallinan, Emillie de Keulenaar, Thales Lelo: „Where Do We Go From Here? Part II – Lessons from the AoIR Flashpoint Symposium on AI & Platform Governance“
The AoIR Flashpoint Symposium in June 2025, hosted by the Platform Governance, Media & Technology Lab (ZeMKI, University of Bremen), brought together over 100 researchers and policy experts from more than 30 countries and 60 institutions to discuss emerging challenges and opportunities of tech governance. The two-day event featured over 50 presentations, including three plenary talks by leading scholars in platform and AI governance, a workshop on large language models as tools and objects of study, and showcased a wide range of global and interdisciplinary perspectives.
The governance of platforms and (generative) AI has become increasingly critical as these technologies profoundly shape public discourse, societal norms, and policymaking. The advent of tools such as ChatGPT continues and intensifies longstanding challenges from social media, such as misinformation, bias, and hate speech. With ongoing shifts in the tech landscape, platform governance research is at a turning point. Growing frustration with corporate platforms has spurred a wave of alternatives, including alt-tech, reactionary platforms, and federated systems. Meanwhile, emerging technologies such as generative AI present platforms with a host of new challenges, as well as significant regulatory scrutiny. We see public attention and daily routines move from social media to generative AI. We also observe a powerful and concerning alignment of technology companies and the US administration. At the same time, the European Union and governments around the world aim to reign in both the excesses as well as the freedoms of technological advancements.
Samstag, 18. Oktober 2025
9:00-10:30 Uhr, Room 10b – Groundfloor
Andreas Hepp, Wiebke Loosen: „Why the future of journalism is not a technological rupture: On the imagination of the societal needs of public communication and innovation in pioneer journalim“
The dominant discourse in journalism often frames its future as shaped by “technological rupture” and innovation imperatives. This paper challenges such narratives, arguing instead for an understanding of journalism’s transformation as a broader structural change. Through the concept of “pioneer journalism,” we analyze how journalists who experiment with new practices and imagine possible futures shape the field’s transformation. Based on a media- ethnographic study in Germany, we examine how pioneer journalists imagine the societal needs for public communication and what implications they derive from this for innovating journalism for a better future. Our findings reveal that their imaginations are strongly rooted in democratic values, emphasizing the need for independent information, foster a basic consensus and civic engagement, among other things. Paradoxically, however, their discourse on innovation often mirrors Silicon Valley’s categories, raising critical questions about whether such frameworks can adequately address the imagined needs for public communication. Finally, we discuss the question of whether the idea of such societal needs should not necessitate an alternative understanding of innovation.
11:00-12:30 Uhr, Room 8g – 2nd Floor
Axel Bruns, Jennifer Stromer-Galley, Jill Karia, Felix Victor Münch, Philipp Kessling, Jakob Ohme, Lion Wedel, Nico Pfiffner, Thomas N. Friemel, Samantha Vilkins, Katherine M. FitzGerald, Tariq Choucair, Daniel Angus, Caroline Gardam, Kunal Chand, Laura Vodden, Klaus Gröbner, Katharina Esau, Carly Lubicz-Zaorski, Ehsan Dehghan, Kate Susan O’Connor- Farfan: „Social Media in Elections: Evidence from the United States, Germany, and Australia“
The year 2024 was marked by an unprecedented confluence of elections around the world, with more than 50% of the world’s population called upon to vote on the future of their governments; perhaps most important of these was the presidential election in the United States in November 2024, which saw Donald Trump returned to the Presidency – an outcome whose immense consequences are already being felt strongly around the world, mere months into the new administration’s term.
In particular, Trump’s victory and his immediate upending of the rule of law at home and world order abroad impacts directly on major subsequent elections elsewhere in the world, including in closely allied nations like Germany (whose federal election took place in February 2025) and Australia (where a federal election is scheduled for April or May 2025). The resurgence of Trumpism emboldened extreme right parties like Germany’s Alternative für Deutschland (AfD), which attracted over 20% of the popular vote in the 2025 election; and Trump’s threats of import tariffs and wavering support for international alliances are emerging as a key topic in the 2025 Australian election campaign.
These developments are further exacerbated by substantial changes in online campaigning environments and strategies. Social media platform operators like Mark Zuckerberg and Elon Musk have closely aligned themselves with the Trump administration (or in Musk’s case, joined it outright), in part to seek protection against European Union and other regulations that require action against disinformation, abuse, and hate speech, and enforce transparency and researcher data access; they have dismantled their content moderation and fact-checking teams; and (in Musk’s case) are actively disseminating disinformation, hate speech, and extremist content. This has also opened the door for other political agitators and influence operators to push problematic materials, including conspiracy theories and AI-generated disinformation.
Finally, the changing platform landscape – marked by the gradual decline of Facebook, a steady exodus from X under Musk’s leadership, the rapid rise of TikTok, and the emergence of federated Twitter alternatives like Mastodon and Bluesky – also necessitates substantial changes in electoral campaigning on the one hand, and in campaign research methods on the other. This panel brings together five papers from major research teams that trace these developments through the US, German, and Australian elections of 2024 and 2025. They provide new insights into the changing electoral campaigning environments of the present moment, and offer new approaches for how we can conduct such research under these changed circumstances.