top of page
KIWIT Research Class

Research class led by Prof. Dr. Stefan Kühl, with staff members Bernd Eckstein and Dennis Düllmann, as well as Prof. Dr. Marcel Schütz.

Offer for researchers and early-career researchers

Digitalization processes in organizations
classify and understand

  • Four-semester course at Bielefeld University

  • Weekly focused text discussions online

  • Regular information on texts and authors is provided on this page

In this research class, designed to run over four semesters, approximately 100 key texts on the topic of organization and digitalization are discussed on a weekly basis. The research class is part of the Bielefeld Graduate School in History and Sociology and is embedded in the KIWIT research group. Due to participation from multiple locations, the seminar is conducted online via Zoom.

For each session, all participants read a selected text, which is then discussed jointly. Prior reading is required for meaningful participation. Attendance is flexible, and participants may join any sessions or discussions that are of interest to them. The research seminar is primarily aimed at doctoral researchers at Bielefeld University and its project partners who are actively engaged in relevant research, but it is also open to interested master’s students and advanced bachelor’s students who have already completed the introductory module in organizational sociology. External researchers from other universities as well as other interested participants with a thematic focus are also welcome. Registration with brief information about one’s background and motivation is required (see information below)

04.05.2026: Elena Esposito (2023): Does Explainability Require Transparency? Sociologica 16(3) und (2021): Transparency versus explanation: The role of ambiguity in legal AI. Journal of Cross-Disciplinary Research in Computational Law 1(2).

 

Gallery | click/swipe​​​

In der Research Class am 04.05.2026 diskutieren wir einen Beitrag der Soziologin Elena Esposito, der eine der zentralen Debatten der KI-Forschung neu bewertet: das Verhältnis von Transparenz und Erklärbarkeit. Ausgangspunkt ist das Dilemma, dass moderne Algorithmen zwar immer leistungsfähiger werden, gleichzeitig aber als „Black Boxes“  undurchsichtig bleiben. Esposito hinterfragt die gängige Forderung, dass Erklärbarkeit zwingend die Offenlegung technischer Prozesse, also Transparenz, erfordern müsse.


Unter Rückgriff auf die Systemtheorie Niklas Luhmanns schlägt Esposito vor, Erklärungen nicht als die Enthüllung interner Abläufe, sondern als sozialen Kommunikationsprozess zu verstehen. So wie Menschen in Gesprächen nicht ihre neuronalen Prozesse offenlegen, sondern Gründe liefern, die für das Gegenüber anschlussfähig sind, können auch KI-Erklärungen als kommunikative Rekonstruktionen begriffen werden. Dieser Ansatz ermöglicht es, Erklärbarkeit (Explainability) von der technischen Durchsichtigkeit (Transparency) zu entkoppeln.


Damit eröffnet der Text eine eigene Perspektive auf das Design von KI: Wenn Erklärungen eigenständige kommunikative Leistungen sind, verschiebt sich der Fokus von der technischen Analyse des Quellcodes hin zur Frage, wie Maschinen so „sozialisiert“ werden können, dass ihr Handeln für Menschen verstehbar bleibt. Esposito zeigt, dass die Leistungsfähigkeit von KI gerade in ihrer Opazität liegen kann und dass wir lernen müssen, mit „untransparenten“ Partnern sinnvoll zu kommunizieren.


Elena Esposito studierte Politikwissenschaften, Philosophie und Soziologie bei Umberto Eco und Niklas Luhmann an den Universitäten Bologna und Bielefeld, wo sie seit 2016 Professuren innehat. Sie erforscht u.a. die Berechenbarkeit der  Zukunft am Beispiel Künstlicher Intelligenz und algorithmischer  Vorhersagen. An der Universität Bielefeld läuft derzeit ihr  umfangreiches EU-gefördertes „Projekt The Social Consequences of Algorithmic Forecast in Insurance, Medicine and Policing. Derzeit nimmt sie ein Fellowship an der Stanford University wahr.  


Duration
Winter Semester 2025/2026 – Winter Semester 2027/2028

Schedule
Mondays, 4:15 p.m. (via Zoom)

Seminar Guidelines

  • Sessions begin punctually at 4:15 p.m.

  • The KIWIT Research Class, planned as a two-year program, meets weekly during the semester on Mondays from 4:15 to 6:00 p.m. and is conducted online via a consistent Zoom link (provided upon registration). Each session is based on a text that all participants are expected to have read in advance. Participation without prior engagement with the assigned—at times demanding—text is strongly discouraged.

  • Participants attending for the first time are asked to briefly introduce themselves in the chat, addressing the following points: institutional affiliation, specific interests in the field of digitalization (and, where applicable, artificial intelligence), and current research projects.

  • The seminar does not follow a formal moderation structure; instead, contributions are organized organically via Zoom’s raise-hand function.

  • When more than five participants are present (which is typically the case), microphones should remain muted. In sessions with a particularly large number of participants, a small-group discussion phase of approximately 20 minutes may be scheduled to facilitate closer engagement with the text. The guiding questions remain constant: Which arguments are convincing? Where do doubts or unresolved issues remain? During small-group discussions, participants are encouraged to keep their microphones switched on.

  • If authors of the assigned texts are present, they are kindly asked to refrain from participating in the discussion during the first 60 minutes. This will be followed by an opportunity for extended commentary and reflection.

Registration

Those interested in participating in the Research Class are requested to send an email with brief information about their academic background and motivation to Prof. Dr. Stefan Kühl (stefan.kuehl[at]uni-bielefeld.de). Registered participants will be added to the mailing list

Aktueller Text hier (Bild anklicken)

Reinhören: Unser neuer Science Podcast ist online.

Aktualisiert am: 29.04.26/jb

Bildschirmfoto 2025-08-22 um 17.27.14.png
To the website of the Federal Ministry of Research
To the website of the Federal Ministry of Research
To the website of the Federal Ministry of Research

Privacy Policy

Accessibility statement

imprint

KIWIT is an interdisciplinary and inter-university research network.

Funded by the Federal Ministry for Research, Technology and Space (BMFTR).

© 2025 NBS Northern Business School for KIWIT. Email: forschungsgruppe.kiwit[at]uol.de

bottom of page