Kim M. Hajek and Paul Trauttmansdorff are Postdoctoral Fellows with the EDI and the Chair for Philosophy and History of Science and Technology at TUM. They co-taught the ‘data clinic’ course with Sabina Leonelli.

Nathalia Uribe is a master’s student in Data and Society at TUM, and student assistant in the Friedrich Schiedel Fellowship Program for Technology in Society, where she works with Paul Trauttmansdorff in the project “Coding Faces in Medicine”. Her academic interests centre on the ethical, legal, and societal dimensions of data and AI, with particular focus on ethical dimensions of data and AI applications.

Lena Sindel is a master’s student in Responsibility in Science, Engineering and Technology (RESET) and a student assistant at the Chair for Philosophy and History of Science and Technology at the TUM School of Social Sciences and Technology.

“how academic learning about data ethics could be applied to real-world problems” (Lena)

“taking weight out of the narrative that we should rely on data for everything” (Nathalia)

Making data ethics material practical and accessible to learners is key to the EDI’s goal of promoting active ethical reflection at every stage of data work. At the Technical University of Munich (TUM), we put this principle into practice in summer semester 2025 through the ‘data clinic’ format, an interactive problem-solving and teaching format designed to bridge theory and practice in data governance and ethics. As part of the master’s course ‘Data Ethics and Governance’, students worked in small groups on real-world ethical issues linked to the use of artificial intelligence (AI) in financial technology (fintech) in three African countries, Rwanda, Kenya and Nigeria. This was a ‘challenge’ issue proposed by our partner for the data clinic, the Center for Law and Innovation, part of the CERTA Foundation in Rwanda. The clinic thus created a rare opportunity for both students and partners to enter a space of co-learning and problem-solving, guided by EDI team-members in the Chair of Philosophy and History of Science and Technology.

The challenge posed in this data clinic was to examine particularly the implications for transparency, fairness, and other core ethical principles of the current boom in fintech innovation, which sees AI tools being used in many financial applications, such as credit scores, loans, and fraud detection. Informed by a practice-oriented approach to data ethics, we encouraged students to look beyond ‘data’ and ‘AI’ as either quick technical fixes to social problems, or even as unitary objects of study. This meant exploring how theoretical concepts discussed in lectures play out in concrete social, political, and technological contexts relevant to the use of fintech apps in three African countries, while recognising the limits of our situated perspectives as outsiders—to the fintech industry, as well as to African societies.

In this way, one student group decided to focus on the specifics of data collection and use in fintech apps used for loans decisions, trawling through the fine print of loan providers’ websites to determine exactly what kinds of data are currently collected for use in AI lending algorithms. Other groups characterised and analyzed key risks entailed by AI use in fintech, and proposed ways in which current regulatory and governance challenges could be transformed into opportunities that improve more equitable outcomes for governments and users. EDI members are synthesising the students’ work into a report on the challenge together with the Center for Law and Innovation, with the aim of contributing insights for their current work. Students, for their part, were enabled to handle complex data governance and ethics issues in a real-world setting.

Here’s what they thought:

Lena (Masters’ student, Responsibility in Science, Engineering and Technology): I was curious to see how academic learning about data ethics could be applied to real-world problems, and the data clinic offered the perfect opportunity to do so. It was refreshingly practical: unlike many theory-heavy courses at university, it put us straight into producing a deliverable that could have real value for an external partner, which made the learning experience more tangible. For me, this was a great motivator and the element that drew me most to the course. The challenge that the partner organisation presented to us was very open, which initially made it difficult to define a clear focus and revealed some differences between what the partner expected and what our lecturers emphasized. That openness forced us out of a “plan-and-execute” mindset and into an agile, iterative way of working. We had to constantly revise and refine our topic, goals, and approach based on new feedback, information, and discussions. Working closely with other students and lecturers throughout those feedback loops, I realized how valuable it is to stay flexible and adjust direction instead of trying to stick rigidly to an initial plan when tackling messy, real-world problems. This was the first time I experienced this to such an extent in the process of writing an academic report, and it turned out to be a really valuable learning experience.

Through these iterations, my group gradually developed our focus and decided to explore the ethical risks of using AI in fintech across Kenya, Nigeria, and Rwanda. The topic was challenging because there is limited literature and research on AI fintech in these countries, which made it difficult to move beyond a simple list of potential ethical risks. It made me more aware of how uneven the global academic landscape is, and how much context matters when discussing technology and ethics. To make our work more concrete and relevant, we created short, fictionalized “risk case stories” that illustrated how these risks might play out in everyday situations, an approach inspired by the ‘data stories’ from lectures. Writing these stories was new for me and surprisingly challenging, as I had to balance insights from real reports and articles with a sense of creative empathy. It reinforced for me how essential it is to reflect on one’s own perspective, background, and assumptions, and how they influence one’s work, especially when conducting research in contexts different from one’s own.

Nathalia (Masters’ student, Data & Society): I enjoyed the class so much because it allowed me to think about data not as a ground truth or fact, taking weight out of the narrative that we should rely on data for everything. First, we need to think about how data comes into being and what context and decisions shape it. Through the critical analysis of African LendTech companies, we were able to identify strategic points of action for policymakers—to advocate for transparency, fairness, and a better contextual understanding of credit assessment with alternative data sources. We could see how discourses promoting the inclusivity of financial products aimed at often-excluded groups can be undermined by actual data practices that reinforce existing exclusions and don’t support communities. Finally, we came to understand that for development we have to go beyond the normative discourse around data and algorithms, looking at the social decisions and implications that shape these technologies, which in turn determine our experiences.

We believe data clinics offer a valuable space to learners exploring challenges in data ethics and governance and we are constantly trying to improve this space. If you think your organization could collaborate with us by providing a meaningful data challenge for our data clinic, we invite you to have a look at our information sheet and reach out to us using the following links:

Accompanying poster: https://doi.org/10.5281/zenodo.17251905

Data clinic infosheets: https://doi.org/10.5281/zenodo.15182263