The perversion of democracy has advanced somuch that it has become a caricature of itself.Many of the features associated with authoritarian regimes—nepotism and corruption, major inequalities and abuses of human rights, as well as the provision of a havenfor war profiteersand business criminals—are also associated with officially-labeled democratic environments. Even though democracy is superior to other regimes in terms of “self-correction” as prompted by the principles of “evaluation, political competition, and freedom of expression,” democracies have regularly trapped themselves in policies that expose bigotryanddouble standards. For example, while the West (the US, in particular) supported non-democratic regimes in its fight against Communism during the Cold War, the EU’s involvement in the Arab Spring has opened questions about whether it has eventually assisted authoritarian instead of democratic rule. Therefore, it seemsperfectly fine that the controversial FIFA awardsthe hosting of the World Cup to both democratic and authoritarian regimes, or that theNorwegian Nobel Committee repeatedly awards the Peace Prize to individuals or supranational entities with dubious performance vis-à-vis democratic postulates.
Category: 4/ Fellow”s Projects
Aula Fellow Project
-
Data Journalism Appropriation in African Newsrooms: A Comparative Study of Botswana and Namibia
Data journalism has received relatively limited academic attention in Southern Africa, with even less focus on smaller countries such as Botswana and Namibia. This article seeks to address this gap by exploring how selected newsrooms in these countries have engaged with data journalism, the ways it has enhanced their daily news reporting, and its impact on newsgathering and production routines. The study reveals varied patterns in the adoption of technology for data journalism across the two contexts. While certain skills remain underdeveloped, efforts to train journalists in data journalism have been evident. These findings support the argument that in emerging economies, the uneven adoption of data journalism technologies is influenced by exposure to these tools and practices.
-
IndicMMLU-Pro: Benchmarking Indic Large Language Models on Multi-Task Language Understanding
Known by more than 1.5 billion people in the Indian subcontinent, Indic languages present unique challenges and opportunities for natural language processing (NLP) research due to their rich cultural heritage, linguistic diversity, and complex structures. IndicMMLU-Pro is a comprehensive benchmark designed to evaluate Large Language Models (LLMs) across Indic languages, building upon the MMLU Pro (Massive Multitask Language Understanding) framework. Covering major languages such as Hindi, Bengali, Gujarati, Marathi, Kannada, Punjabi, Tamil, Telugu, and Urdu, our benchmark addresses the unique challenges and opportunities presented by the linguistic diversity of the Indian subcontinent. This benchmark encompasses a wide range of tasks in language comprehension, reasoning, and generation, meticulously crafted to capture the intricacies of Indian languages. IndicMMLU-Pro provides a standardized evaluation framework to push the research boundaries in Indic language AI, facilitating the development of more accurate, efficient, and culturally sensitive models. This paper outlines the benchmarks’ design principles, task taxonomy, and data collection methodology, and presents baseline results from state-of-the-art multilingual models.
-
Mediatized discourses on Europeanization in Spain
Political and media polarization has had a detrimental impact on democratic principles and democratic processes on a
global scale. In Europe, such polarization has eroded the trust in national and European institutions and has challenged the
basic values that stand at the heart of the European integration project. The aim of this study is to analyze Spanish media discourses on Europeanization, with an attempt to identify key areas in which polarizing narratives related to Europeanization
are more prevalent. To conduct our study, six national media outlets were selected based on four criteria: media format,
ownership, ideology, and consumption. A final sample of 540 news items collected between July 2021 to March 2022 was
selected for analysis. Using a qualitative methodological approach, the study was carried out in two stages. In the first
phase, we conducted a content analysis to identify the main topics discussed in relation to the European Union and the
actors represented in them. This led to the identification of polarizing narratives and discourses emerging in the context
of the discussed topics. In the second phase, we used critical discourse analysis to analyze polarizing discourses. -
Potential and perils of large language models as judges of unstructured textual data
Rapid advancements in large language models have unlocked remarkable capabilities when it comes to processing and summarizing unstructured text data. This has implications for the analysis of rich, open-ended datasets, such as survey responses, where LLMs hold the promise of efficiently distilling key themes and sentiments. However, as organizations increasingly turn to these powerful AI systems to make sense of textual feedback, a critical question arises, can we trust LLMs to accurately represent the perspectives contained within these text based datasets? While LLMs excel at generating human-like summaries, there is a risk that their outputs may inadvertently diverge from the true substance of the original responses. Discrepancies between the LLM-generated outputs and the actual themes present in the data could lead to flawed decision-making, with far-reaching consequences for organizations. This research investigates the effectiveness of LLM-as-judge models to evaluate the thematic alignment of summaries generated by other LLMs. We utilized an Anthropic Claude model to generate thematic summaries from open-ended survey responses, with Amazon’s Titan Express, Nova Pro, and Meta’s Llama serving as judges. This LLM-as-judge approach was compared to human evaluations using Cohen’s kappa, Spearman’s rho, and Krippendorff’s alpha, validating a scalable alternative to traditional human centric evaluation methods. Our findings reveal that while LLM-as-judge offer a scalable solution comparable to human raters, humans may still excel at detecting subtle, context-specific nuances. Our research contributes to the growing body of knowledge on AI assisted text analysis. Further, we provide recommendations for future research, emphasizing the need for careful consideration when generalizing LLM-as-judge models across various contexts and use cases.
-
Power Preservation, No Matter the Means: Populism and Conspiracy Theory as Instruments of Political Consolidation in Serbia
This article examines the Serbian political leadership—the president and government alike—by addressing the dominant political figures’ narratives. We communicate with the theoretical aspects in the study of populism and conspiracy theories as this nexus enables us to examine the specific nature of the domestic politics in Serbia. In our view, the ruling elite complements its populist discourse with conspiracy theory to ensure its survival in power, by regularly generating fear about the threat posed to Serbian statehood and lack of apprehension for Belgrade’s geopolitical preferences and exploration of foreign policy alternatives. Our analysis fills a major gap in the literature, since there has been only sporadic research on this topic and none of it has focused on the merger of populism and conspiracy theory. The findings we have reached—largely those of the elite’s self-victimization narratives and their dissemination of anti-Western sentiments—provide for a fresh contribution to the debate concerning the power struggle and the state of democracy in Serbia, especially given the fact that the key political stakeholders draw heavily on pro-regime media outlets to readily disseminate their self-serving accounts.
-
AI as a New Public Intellectual?
In a dialogue with ChatGPT, I asked if it could be considered a public intellectual.
-
Data Journalism, Accountability and Transparency in Zimbabwe’s ‘New Dispensation’: Some Empirical Reflections
In this chapter, we explore the intersection of data journalism practices with issues of (governance) transparency and accountability. We advance the argument that data journalism can be instrumental in helping journalists seek accountability in opaque regimes that have an uneasy relationship with watchdog journalism. We use the Zimbabwe’s post-coup regime to demonstrate that at the centre of political authoritarianism, is a refusal to account, and a culture of non-transparency. Faced with such, the media can utilise publicly available sources of data journalism to exercise their responsibility. Data journalism is, hence, critical as a media practice that provides avenues for journalists in semi-authoritarian regimes to continuously pursue their mandates as accountability seekers. Our chapter contributes to emerging literature on data journalism in Africa, especially in semi-authoritarian contests like that of Zimbabwe.
-
Louvre Abu Dhabi as a State Project: A Museum, a New Ministry, and the Politics of Culture
As the first universal museum in the Arab world, Louvre Abu Dhabi (LAD) aims to cover all aspects of human history, promoting inclusivity and openness. Despite being hailed as a symbol of tolerance and intercultural dialogue, some shortcomings and valid criticisms are evident. This study explores LAD’s artistic and political roles, as well as its contribution to social cohesion and the United Arab Emirates’ (UAE’s) readiness to address broader issues in line with the UAE Centennial Plan 2071. The analysis also delves into the responsibilities of cultural initiatives, the challenges of operating within an authoritarian context, and the necessity for policy reconsideration to achieve the UAE’s ambitious visions. The fact that LAD has undoubtedly served state interests highlights its limitation in engaging with the broader mission of museums—to address often-unpleasant political and socioeconomic matters. Moreover, the article asserts that constructive exchanges about critical issues should be embraced as opportunities rather than threats; suppressing these discussions through censorship and fear undermines the potential for growth and understanding. Thus, the study calls for a balanced approach where cultural institutions like LAD can foster dialogue and contribute meaningfully to the social and political landscape of the UAE.
-
Towards regulating AI : A natural, labour and capital resources perspective
Policymakers who are looking at artificial intelligence (AI) applications are thinking about what we as a society want to achieve and what we need to protect, yet it is not commonly known that AI apps require intensive natural resources, labour and capital.
-
Trainees’ perspectives and recommendations for catalyzing the next generation of NeuroAI researchers
At this critical juncture in the development of NeuroAI, we outline challenges and training needs of junior researchers working across AI and neuroscience. We also provide advice and resources to help trainees plan their NeuroAI careers.
-
Evaluating Online AI Detection Tools: An Empirical Study Using Microsoft Copilot-Generated Content
Our findings reveal significant inconsistencies and limitations in AI detection tools, with many failing to accurately identify Copilotauthored text. Examining eight freely available online AI detection tools using text samples produced by Microsoft Copilot, we assess their accuracy and consistency. We feed a short sentence and a small paragraph and note the estimate of these tools. Our results suggest that educators should not rely on these tools to check for AI use.