10–13 November 10-13, 2026
Online
VI MeLCi Lab Autumn School 2026
Organised by CICANT: MeLCi Lab, AISIC, and InTouch Labs | Lusófona University, Portugal
Website: https://melcilab.cicant.ulusofona.pt/training/vi-melci-lab-autumn-school-2026-advanced-school-on-ai-research-practice-in-media-and-communication/
---
Researchers in communication and media studies now face a structural tension. Artificial intelligence - particularly large language models - has entered the research pipeline as a tool for applications such as literature search, data annotation, audience segmentation, and discourse analysis. At the same time, AI has become an object of inquiry: a force reshaping civic cultures, media ecologies, and the conditions under which publics form. These two roles demand different competencies. Using AI as a method requires technical skill, prompt design, and validation protocols. Studying AI as a societal force requires critical frameworks drawn from political theory, media literacy, and the ethics of datafication. Most training programmes address one side or the other. This school addresses both and the friction between them.
The VI MeLCi Lab Autumn School invites applications from PhD students, postdoctoral researchers, and early-career scholars for a four-day intensive online programme. The school combines keynote lectures with hands-on workshops, structured around two complementary themes. Participants will work with media-specific datasets, confront the interpretative challenges particular to communication research, like bias in content classification, the instability of AI-generated annotations, and the opacity of recommendation systems, and develop both the technical and critical capacities the current research landscape requires.
No prior experience with AI or data science is assumed. Introductory modules provide the necessary foundations.
---
Theme 1: AI in Research Practice: Foundations, Methods, and Ethics
AI tools have entered research workflows faster than the methodological standards needed to govern their use. Zero- and few-shot prompting now enables researchers with no computational training to perform tasks that previously required supervised classifiers or teams of human coders (Gilardi et al., 2023; Grossmann et al., 2023; Ziems et al., 2024). The accessibility is genuine. So are the risks: prompt instability, opaque model behaviour, and the absence of agreed reproducibility standards mean that convenience can outpace accountability (Barrie et al., 2025). This theme equips participants with the methodological foundations, practical skills, and ethical orientation to use AI tools rigorously.
1.1 Foundations of Current AI Tools
Large language models have transformed what is computationally tractable in text-based research. Prompting techniques that require no training data have achieved annotation accuracy comparable to - and in some cases exceeding - expert human coders. But the same flexibility that makes LLMs accessible also makes them fragile: minor prompt adjustments can shift outputs in ways that compromise replicability. This sub-track addresses the theoretical architecture of contemporary AI tools, the methodological principles governing their responsible use, and the best practices emerging for transparent, accountable deployment in communication research.
1.2 Accountable Literature Search Using AI Tools
AI-powered platforms such as SciSpace and Litmaps have accelerated literature discovery, enabling researchers to map citation networks, identify thematic clusters, and surface relevant work at a pace that manual search cannot match. The efficiency gain, however, introduces a new accountability burden. AI-assisted searches can silently exclude relevant literature, privilege certain databases, or present coverage as comprehensive when it is partial. This sub-track develops strategies for validating AI-generated search results, assessing coverage boundaries, and maintaining the transparent documentation practices that methodological rigour demands.
1.3 AI-Assisted Data Annotation in Research Pipelines
Data annotation anchors most empirical research pipelines. Where this task once relied exclusively on human coders, AI-based annotation now offers a viable and often highly effective alternative - particularly at scale. The central challenge is consistency. Barrie et al. (2025) demonstrate that prompt stability, i.e., the degree to which semantically equivalent prompts produce equivalent annotations, remains a significant source of variability. This sub-track introduces participants to AI-driven annotation workflows, focusing on practical approaches to assessing and improving annotation reliability through frameworks such as Prompt Stability Scoring (PSS) and integrating responsible validation practices into research design.
Theme 2: Communication, Audiences, and Civic Cultures in the Age of AI
AI does not only reshape how researchers work. It reshapes the media environments researchers study. Algorithmic recommendation determines what the public sees, platform architectures mediate how citizens engage, and the datafication of everyday life raises questions about equity, inclusion, and democratic participation that existing frameworks struggle to answer. This theme addresses AI not as a methodological resource but as a structural force within media ecologies - one that demands critical engagement from researchers who study communication, audiences, and civic cultures.
2.1 Civic Cultures and Artificial Intelligence
AI-driven platforms and recommendation algorithms now mediate core dimensions of civic life: how citizens encounter information, how activist networks form, and how media literacy is exercised or undermined (Sarafis et al., 2025). This sub-track examines the opportunities and challenges AI introduces for civic engagement, exploring how algorithmic mediation reconfigures the conditions under which publics participate in democratic processes.
2.2 Digital Citizenship and Media Literacy in an AI-Mediated World
The competencies required for informed participation in AI-mediated environments remain poorly defined. Critical media literacy now extends to skills that existing frameworks have not yet systematised: recognising AI-generated content, understanding how recommendation systems shape information exposure, and assessing the epistemic status of machine-produced outputs (Chiu et al., 2024). This sub-track examines what digital citizenship demands in an environment shaped by misinformation, deepfakes, and opaque algorithmic curation.
2.3 Data Ethics, Equity, and Inclusivity in AI Research
AI technologies carry biases embedded in their training data, design choices, and deployment contexts. The ethical implications of using these tools for knowledge production: who is represented, whose categories are imposed, and whose communities bear the risks of misclassification, remain insufficiently examined (Ferrara, 2024; Ntoutsi et al., 2020). This theme moves beyond the binary framing of AI as either a technological panacea or an existential threat. It addresses responsible research practice, equitable research design, and the specific obligations researchers hold when working with data from or about underrepresented communities.
Application Details
Deadline for submission: 15 September 2026
Notification of acceptance: 12 October 2026
Registration deadline: 28 October 2026
Interested participants should submit their application (in English) by 15 September 2026, including:
1. An updated curriculum vitae (max. 3 pages)
2. A research statement describing their doctoral dissertation or current research project, including research questions and methods (max. 2 pages)
3. A motivation letter describing their current engagement with AI, specific concerns or interests regarding AI's role in media research and practice, and their preferred theme (max. 2 pages)
Applications should be submitted as a single ZIP file to melci.lab@ulusofona.pt with the subject line: "Application for the VI MeLCi Lab Autumn School".
The school will be conducted online and in English.
For enquiries, please contact: melci.lab@ulusofona.pt
---
References
Barrie, C., Palaiologou, E., & Törnberg, P. (2024). Prompt stability scoring for text annotation with large language models. arXiv preprint arXiv:2407.02039. https://doi.org/10.48550/arXiv.2407.02039
Chiu, T. K., Ahmad, Z., Ismailov, M., & Sanusi, I. T. (2024). What are artificial intelligence literacy and competency? A comprehensive framework to support them. Computers and Education Open, 6, 100171. https://doi.org/10.1016/j.caeo.2024.100171
Ferrara, E. (2024). Fairness and bias in artificial intelligence: A brief survey of sources, impacts, and mitigation strategies. Sci, 6(1), 3. https://doi.org/10.3390/sci6010003
Gilardi, F., Alizadeh, M., & Kubli, M. (2023). ChatGPT outperforms crowd workers for text-annotation tasks. Proceedings of the National Academy of Sciences, 120(30), e2305016120. https://doi.org/10.1073/pnas.2305016120
Grossmann, I., Feinberg, M., Parker, D. C., Christakis, N. A., Tetlock, P. E., & Cunningham, W. A. (2023). AI and the transformation of social science research. Science, 380(6650), 1108–1109. https://doi.org/10.1126/science.adi1778
Ntoutsi, E., Fafalios, P., Gadiraju, U., Iosifidis, V., Nejdl, W., Vidal, M., ... & Staab, S. (2020). Bias in data-driven artificial intelligence systems — An introductory survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 10(3). https://doi.org/10.1002/widm.1356
Sarafis, D., Karamitsios, K., & Kravari, K. (2025). AI and civic engagement: A brief exploration of applications and opportunities. 2025 International Conference on Advancement in Data Science, E-learning and Information System (ICADEIS), 1–6. https://doi.org/10.1109/icadeis65852.2025.10933183
Ziems, C., Held, W., Shaikh, O., Chen, J., Zhang, Z., & Yang, D. (2024). Can large language models transform computational social science? Computational Linguistics, 50(1), 237–291.
