Ethics of AI
Welcome to the Research Group Ethics and Critical Theories of AI!
Rainer is the head of the Ethics and critical theories of AI research group at the Institute of Cognitive Science at the University of Osnabrück. As a working group we develop new approaches for a contemporary critical philosophy of the digital, using our colloquium for regular discussions.
Our research group offers excellent students and doctoral candidates an inspiring environment for critical philosophy, ethics and social theory of the digital society – for example, in the context of bachelor’s and master’s theses as well as doctoral projects.
Together, the team contributes to the teaching program on the Ethics of AI offered at the University of Osnabrück. Our courses are generally held in English and mostly in hybrid formats (combining online/offline), so that they are also accessible to external guests.
NEWS
New open-access article : What does “delegation” mean in LLM use—and how can we distinguish its ethical degrees?
Rainer Mühlhoff has published a new article “From delegation to moral abdication: classifying large language model uses by judgment and epistemic control” in AI & Society and it is available open access.
The article develops a use-based taxonomy for large language model applications. Instead of classifying applications solely by sector, task, or abstract risk category, the framework focuses on concrete prompting practices and workflows along two dimensions: How much judgment is delegated to an LLM? And how much epistemic control do users retain over the knowledge base the LLM draws on when processing the prompt?
Annemarie Witschas : New essay
How fair are technologies that rely on precarious labor and the exploitation of natural resources to secure Western profits? Such power imbalances are the legacy of colonial dependencies which are now reinforced by patents and monopolies. But the path to a fair digital future does not lie in competing for a better place within this system, but in overcoming it. In her essay for the SUPERR Lab, Annemarie Witschas examines how countries can reclaim their resources and capacities in favor of needs-based development.
Paul Schütze : New publications
"The Transhumanist Anthropocene: From the climate crisis to upgrading humanity"
https://doi.org/10.1177/20530196261453840 - read the abstract
"AI and the climate crisis: power, ideology, and the politics of collapse" https://doi.org/10.4337/9781035342204.00044 - read the abstract
AI and democracy: the rule of law, civic participation, sustainability
Symposium followed by a public panel discussion
October 1, 2026 | 9 a.m. - 4:45 p.m. presentations, 5 p.m. - 6:30 p.m. public panel discussion
Venue: Einstein Center Digital Future, Robert-Koch-Forum, Wilhelmstraße 67, 10117 Berlin
Registration: by email
Contact: symposium@ethikderki.de
Latest information and updates
AI technologies are increasingly finding their way into key areas of political decision-making—from public administration and environmental policy to border management and economic governance. In doing so, they are not only changing procedures; they are also shifting power dynamics, accountability, and the scope for political action.
The symposium will explore AI as a systemic challenge to democracy and the rule of law. Building on a brief introduction to various focal points of our research, we offer a collaborative platform for analyzing and discussing the systemic costs of algorithmic governance for democratic structures. In doing so, we create an open space for dialogue with policymakers, NGO representatives, researchers, and anyone interested in the future of our digital society.
Event organized as part of the project “AI and Democracy: Rule of Law, Participation, Sustainability” at the University of Osnabrück, led by Prof. Rainer Mühlhoff, funded by the Federal Ministry of Research, Technology, and Space. In cooperation with the Einstein Center Digital Future, Berlin.
Spring 2027 - Transcript publishing : New Release: Anthology “AI and Crisis: Technological Power as a Threat to Liberal Democracies” (in German), edited by Paul Schütze and Rainer Mühlhoff
This volume examines—broadly along three thematic areas—the climate crisis, state sovereignty, and authoritarian politics—how current AI technologies reinforce authoritarian power structures, obscure political and economic responsibilities, and put pressure on democratic decision-making. The contributions provide a theoretically grounded assessment of the interplay between AI, authoritarian politics, and global crisis dynamics. The book is aimed at a broad audience and anyone who wants to understand why liberal democracies are facing their greatest crisis in decades—and what AI technologies have to do with it.
With contributions by: Anja Breljak, Daniela Hombach, Nora Lindemann, Roland Meyer, Peter Remmers, Hannah Ruschemeier, Eugenia Stamboliev, and Annemarie Witschas
Who are we?
An inclusive working atmosphere, gender equality, and consideration of the needs of young scholars with children are particularly important to us.
The research group "Ethics and Critical Theories of AI" consists of Prof. Rainer Mühlhoff and his team. Details of Rainer's ongoing work can also be found on Rainer's website.
Current topics in our working group are
- ethics and social philosophy of artificial intelligence,
- ethics and critique with regard to structural effects,
- privacy and data protection in the context of Big Data,
- intersectionality and anti-discrimination in digital technology,
- critical media philosophy of Human-Machine interaction
Events 2026
AI and democracy: the rule of law, civic participation, sustainability
Symposium followed by a public panel discussion
October 1, 2026 | 9 a.m. - 4:45 p.m. presentations, 5 p.m. - 6:30 p.m. public panel discussion
Venue: Einstein Center Digital Future, Robert-Koch-Forum, Wilhelmstraße 67, 10117 Berlin
Registration: by email
Contact: symposium@ethikderki.de
Latest information and updates
AI technologies are increasingly finding their way into key areas of political decision-making—from public administration and environmental policy to border management and economic governance. In doing so, they are not only changing procedures; they are also shifting power dynamics, accountability, and the scope for political action.
The symposium will explore AI as a systemic challenge to democracy and the rule of law. Building on a brief introduction to various focal points of our research, we offer a collaborative platform for analyzing and discussing the systemic costs of algorithmic governance for democratic structures. In doing so, we create an open space for dialogue with policymakers, NGO representatives, researchers, and anyone interested in the future of our digital society.
Event organized as part of the project “AI and Democracy: Rule of Law, Participation, Sustainability” at the University of Osnabrück, led by Prof. Rainer Mühlhoff, funded by the Federal Ministry of Research, Technology, and Space. In cooperation with the Einstein Center Digital Future, Berlin.
This program will be updated and expanded regularly.
| Preliminary program | |
|---|---|
| 9:15 a.m. - 10:00 a.m. | Opening Prof. Dr. Rainer Mühlhoff (Osnabrück University) |
| 10:15 a.m. - 11:30 a.m. | Bureaucracy as an ethos: on the limits of automating public administration Prof. Dr. Michael Bauer (European University Institute, Florenz), Nora Freya Lindemann (Eberhard-Karls University of Tübingen) & Prof. Dr. Rainer Mühlhoff (Osnabrück University) |
| 11:45 a.m. - 1:00 p.m. | AI at the EU's borders Anja Breljak (University of Potsdam) & Matthias Monroy (freelance journalist) |
| 1:15 p.m. - 2:15 p.m. | Lunch break |
| 2:15 p.m. - 3:30 p.m. | Scale, expand, accelerate: how AI reinforces growth paradigms Dr. Steffen Lange (University of Siegen) & Annemarie Witschas (Osnabrück University) |
| 3:30 p.m. - 4:45 p.m. | Sustainable technopolitics and the new authoritarianism Paul Schütze (Osnabrück University), Sandra Sieron (WZB Berlin/ HU Berlin) & Dr. Marie Wuth (KU Eichstätt-Ingolstadt) |
| 5:00 p.m. - 6:30 p.m. | Panel discussion Dr. Andreas Antić, Dr. Arian Henning, Prof. Dr. Rainer Mühlhoff, Dr. Dorothea Winter |
Nationwide event week : Wissenschaft gegen Faschismus
There were events and actions at many universities in Germany to highlight academia's responsibility against the rise of authoritarian politics and fascism.
We in the „Ethics and Critical Theories of AI“ group joined this week's actions with two events:
Rainer Mühlhoff : Lecture "Faschismus und KI"
Wednesday June 3, 2026, 7p.m., room 22/B01
Screening of the documentary “Ghost in the Machine”
As part of the “Science Against Fascism” campaign week, we cordially invite you to a screening of the documentary “Ghost in the Machine” by Valerie Veatch. The film explores how right-wing ideologies—ranging from social selection to eugenics—are deeply intertwined with the history of AI technology. The film will be shown in English (the availability of German subtitles is currently being reviewed).
Thursday, June 4, 7 p.m. - 9 p.m., in 15/E10 (EW building)
On the occasion of the exhibition “Computer Says No” by the !Mediengruppe Bitnik in the nave of the Kunsthalle Osnabrück, Rainer Mühlhoff presented his book *AI and the New Fascism*, published in 2025.
The new book examines the dangerous connection between tech companies and right-wing politics. The book poses questions: What role does AI play in this development? How can we recognize early on when these developments are directed against democracy—or even become fascist? In the U.S. as well as here in Europe.
After a brief introduction, Rainer Mühlhoff spoke with the artists of the !Mediengruppe Bitnik about artificial intelligence, power, and resistance in the digital age.
-
Invited presentation “Denke ich noch selbst? KI, ChatGPT & Co.” — phil.cologne 2025, session for students grade 9 and up. Wednesday, 25 June 2025, 10:00, COMEDIA Theater, Roter Saal, Vondelstraße 4–8, Cologne-Südstadt.
-
Invited panelist, podium discussion “Künstliche Intelligenz als Lehrkraft? Feedback und Bewertung durch die Maschine” — Europa-Universität Flensburg, 19 June 2025.
-
Invited talk “Updating Purpose Limitation for AI – a normative approach to AI regulation” (with Hannah Ruschemeier) at the Oxford Internet Institute, University of Oxford. Monday, 16 June 2025, 12:00–13:00 UK / 13:00–14:00 CET.
-
Invited keynote “Vom Mythos zur Machtfrage: Die gesellschaftliche Bedeutung von KI” — HSBI KI-Kongress 2025 “Chancen und Herausforderungen durch KI”, Hochschule Bielefeld, 3 June 2025.
-
Invited talk “The AI State: Why We Need To Save The Rule Of Law From ‘Effectiveness’” (with Hannah Ruschemeier) — IViR Lecture Series, University of Amsterdam, 16 May 2025.
-
Invited presentation “Hausaufgaben mit KI korrigieren!?” (with Marte Henningsen) — University Future Festival 2025, 14 May 2025.
-
re:publica 25, Berlin — Talk “Digitaler Faschismus: Wie KI-Ideologie die Demokratie untergräbt und wie wir sie verteidigen können” (with Aline Blankertz), 28 May 2025, 17:30–18:00, Stage 5.
-
Invited keynote “Die Macht der künstlichen Intelligenz” — Humboldt-Lecture 2025, Humboldt-Gymnasium, Berlin, 24 March 2025.
-
Invited panelist, Session 3.3 “Chancen und Risiken von Künstlicher Intelligenz – Sichere Nutzung intelligenter Technologien” — Nationale Konferenz IT-Sicherheitsforschung 2025, BMBF (Bundesministerium für Bildung und Forschung), 19 March 2025, 10:45–12:15.
-
Conference presentation “Prediction Power and Cybernetic AI” (with Hannah Ruschemeier) — STS-Hub 25, Berlin, 11 March 2025.
-
Invited talk “Reading as Assessing. Automated Homework Grading With LLMs” — Workshop “New Reading Scenes: On Machine Reading and Reading Machine Learning Research”, Institute for Advanced Study in the Humanities (KWI) Essen, in cooperation with SFB Virtuelle Lebenswelten, Ruhr-Universität Bochum, organized by Katia Schwerzmann. Bochum, 28 February 2025.
Publications
In her essay for the SUPERR Lab, Annemarie Witschas examines how countries can reclaim their resources and capacities in favor of needs-based development.
How fair are technologies that rely on precarious labor and the exploitation of natural resources to secure Western profits? Such power imbalances are the legacy of colonial dependencies which are now reinforced by patents and monopolies. But the path to a fair digital future does not lie in competing for a better place within this system, but in overcoming it.
Rainer Mühlhoff's new article “From delegation to moral abdication: classifying large language model uses by judgment and epistemic control” has been published in AI & Society and is available open access.
The article develops a use-based taxonomy for large language model applications. Instead of classifying applications solely by sector, task, or abstract risk category, the framework focuses on concrete prompting practices and workflows along two dimensions: How much judgment is delegated to an LLM? And how much epistemic control do users retain over the knowledge base the LLM draws on when processing the prompt? Taken together, these dimensions help distinguish forms of cognitive offloading that may look similar at the interface level but differ significantly in their ethical implications. It analyzes prompts as “delegation artifacts” that make visible how cognitive and normative labor is distributed between users and systems.
Through two case studies—automated grading in education and contract termination in public administration—the article shows how interfaces and institutional framing can obscure extensive judgment delegation and the resulting responsibility-related vulnerabilities.
Its central normative claim is that ethically defensible LLM use must remain structured as tool use. Human users need to retain epistemic control over the underlying knowledge base, independent judgment, and the ability to justify and take responsibility for decisions. Where these conditions are absent, delegation threatens to become moral abdication.
Our team member Paul Schütze published this article:
"The Transhumanist Anthropocene: From the climate crisis to upgrading humanity"
https://doi.org/10.1177/20530196261453840
Abstract: This paper identifies a significant shift in the current dynamics of the Anthropocene driven by transhumanism. While the Anthropocene has traditionally emphasized humanity’s impact on Earth’s systems, transhumanist visions now direct this transformative force toward the biological and cognitive constitution of the human. This has two main implications for the Anthropocene dynamics: first, it marks an expansion of human action toward the technical control of human biological and cognitive capacities; second, it generates unprecedented living conditions that redefine what it means to be human. To capture this development, we introduce the notion of the “Transhumanist Anthropocene.” With it, we point to a novel trajectory within the overall Anthropocene dynamics. This is characterized by a reinforced anthropocentrism, the rise of a techno-religion shaped by transhumanist thought, a reduction of the perceived urgency of the climate crisis, and an increasing reliance on techno-solutionist models of environmental governance. We argue that this significantly complicates efforts toward a sustainable future in the face of the climate crisis.
Citation: Schütze, P., & Latzer, M. (2026). The Transhumanist Anthropocene: From the climate crisis to upgrading humanity. The Anthropocene Review, 0(0).
Our team member Paul Schütze published this article:
"AI and the climate crisis: power, ideology, and the politics of collapse"
https://doi.org/10.4337/9781035342204.00044
Abstract: In this chapter I argue that AI's climate impact extends beyond its material footprint. Its ideological influence is equally critical. I show this by looking at two cases: First, I show how AI technologies enable a continuation of the status quo of climate breakdown by promoting a false sense of security and control. This is the result of what I call AI futurism, an ideology focused on technological solutions to climate issues while distracting from necessary societal changes. Second, I explore how AI reproduces the power structures and the logics of racial capitalism. AI advances a system of division and exploitation that undermines genuine societal, political, or economic relations. Both AI's false promise of control and its racial capitalist character hinder effective politics addressing planetary collapse.
Citation: Schütze, P. (2026). "30: AI and the climate crisis: power, ideology, and the politics of collapse". In Handbook on Anthropology and Artificial Intelligence. Cheltenham, UK: Edward Elgar Publishing. Retrieved Aug 17, 2026
We test two widely used tools
AI tools promise to relieve teachers by supporting the correction, feedback, and grading of student texts. In our new open-access article in SEMINAR – Lehrerbildung und Schule, Sean Quägwer and I test two widely used systems for automated feedback and AI-supported grading: FelloFish and Edaira.
The study follows up on our earlier study of Fobizz’s AI grading assistant and asks whether the problems observed there are limited to a single tool or point to structural limits of grading systems based on large language models. The result is clear: FelloFish and Edaira also show substantial variation for identical inputs, a lack of reproducibility in their assessments, and problematic effects when students revise texts on the basis of automated feedback. One particularly troubling finding is that verbatim adoption of automatically suggested formulations is sometimes rated more highly than independent revisions that are equivalent in content.
Our conclusion: such tools currently cannot be recommended as autonomous systems for grading student work. If they are used at all, they should be used only as supporting tools, transparently and under critical human supervision.
Rainer Mühlhoff and Sean Quägwer, 2026. „Automatisierte Korrektur mit KI? Wir testen zwei verbreitete Tools“. SEMINAR – Lehrerbildung und Schule 2/2026: 62–76. doi:10.3278/SEM2602W007. Material appendix (in German)(PDF, 6,72 MB) and full text (in German) are available.
Our team member Peter Remmers published this chapter in the German book ‘ Handbuch der Künstlichen Intelligenz’, together with two co-authors.
Award ceremony
The book has been awarded with the this year's “The Political Book” prize of the Friedrich-Ebert-Stiftung. The ceremony took place on May 21, 2026 in Berlin.
Available now (in German language). Order from the publisher for 8,00 EUR (print) Ebook available on all known plarforms.
“Very impressive […]. Incredibly exciting to read, but also very concerning […].” — Benjamin Knödler ― Deutschlandfunk Kultur
“In his concise, opinionated treatise, Mühlhoff not only dismantles the rhetoric of the AI prophets but also exposes their questionable ideology.” — Matthias Dusini ― Falter
“Rainer Mühlhoff is one of the most interesting thinkers in German philosophy of technology.” — Jana Ballweber ― KNA
Publisher’s announcement: Elon Musk and Donald Trump are firing masses of public sector employees in order to build an AI state. Tech CEOs sell artificial intelligence as a savior for humanity’s biggest problems, even though the industry behind it is based on exploitation and contempt for human dignity. Why is the public distracted by speculation about salvation or annihilation through AI, instead of paying attention to the real and significant harms AI is causing in the present? How can we identify the increasingly fascist tendencies emerging from the alliance between the tech industry and the new right? Rainer Mühlhoff develops answers and discusses possible solutions.
There’s no AI ethics without addressing power.
In this book, Rainer Mühlhoff lay the foundation for a Power-Aware Ethics of AI – bridging critical philosophy, Science and Technology Studies, and media theory to rethink ethics in the age of networked media and data-driven AI.
- he introduces the concept of Human-Aided AI
- Explores how power, UX design, and subjectification intertwine
- Revisits the history of AI and cybernetics
- Critiques the idea that all biases can be “fixed” by technical means
- Makes the case for collective ethics and responsibility
- Calls for stronger regulation of AI
- And concludes with a “Manifesto for a Power-Aware Ethics of AI”
Open Access, published by Bristol University Press. PDF full text available as well.
This project grew out of his lecture series at the University of Osnabrück – heartfelt thanks to all the students for their invaluable engagement and enthusiastic feedback.
Rainer Mühlhoff and Hannah Ruschemeier, International Journal of Law and Information Technology
In recent years, big tech and AI companies have put much effort into promoting their vision of rapid AI advancement as the only path forward. But is this truly the only option? What interests lie behind these grand narratives, and which alternative futures are being sidelined?
This new article investigates how the AI industry shapes collective visions of the future to bolster its economic power, drawing parallels to the tactics used by fossil fuel companies in the 20th century. The AI-driven futures they promote often lead to exclusionary and unsustainable "elitist futures", characterized by power consolidation and persistent inequalities. Furthermore, these visions echo a colonial legacy of extractive expansion, portraying the future as a frontier to be conquered.
It is urgent and necessary to demystify how the AI industry constructs these collective future visions for its own benefit. This allows to better inform transformative actions and strive for truly sustainable and inclusive futures.
Witschas, Annemarie. Prefabricated futures: AI imaginaries between elitist visions and social justice claims. Nomos Verlagsgesellschaft mbH & KG, 2025.
Read the Nomos article
Rainer Mühlhoff, Verfassungsblog
On the Internet, we generate data every day that we often think is harmless – our likes, our location or our browser history. But companies that own this data can use AI algorithms to derive intimate information from it. Do we suffer from mental illness? Do we abuse substances? Are we rich or poor? Such algorithms are used to treat people differently in an automated way – when it comes to the price of insurance, the selection of applicants in hiring processes, decisions about credit or immigration. The combination of Big Data and AI is currently producing a new technology of population management. This development leads to a new form of power in our societies, to which we all contribute through our data and which we therefore need to discuss critically and ethically.