Research Profile
Rainer Mühlhoff’s research area is ethics and social philosophy of artificial intelligence. His work centers on how AI technologies transform social power relations — for example, through new forms of knowledge production, social selection, techno-political ideologies, and the delegation of decisions.
His approach combines ethics, critical social philosophy, political philosophy, philosophy of technology, and media studies. Rather than treating AI as an isolated technical tool, he analyzes it as part of sociotechnical systems in which companies, public institutions, infrastructures, data practices, and social ideologies interact.
Power, Democracy, and Techno-Authoritarianism
One current focus is the relationship between AI, democracy, and political power. Rainer Mühlhoff examines how AI companies and digital infrastructures are transforming public decision-making processes, public discourse, and democratic institutions.
This includes analyzing what he calls “tech ideologies”, digital authoritarianism, and new fascist tendencies, as well as questions concerning the rule of law, social participation, and democratic control of technological infrastructures.
Bestselling book “AI and the New Fascism” Project “AI and Democracy”
Power-Aware Ethics of AI
Rainer Mühlhoff develops a power-aware ethics of AI. This approach extends conventional technology ethics through a systematic analysis of social power relations. Ethical problems do not arise only from faulty decisions made by individual systems; they are already embedded in the institutional, economic, political, and broader structural conditions through which AI is intertwined with society.
A particular interest lies in the historical transformation of judgment, knowledge, and responsibility in the context of AI understood as a sociotechnical human–machine system. The use of generative AI and large language models raises the question of when the delegation of individual tasks turns into the delegation of judgment and, ultimately, into the moral abdication of responsibility. As an affiliated researcher at the Institute of Medical Informatics at Charité – Universitätsmedizin Berlin, Rainer Mühlhoff also examines these questions in the field of medical AI applications, where the delegation of judgment is particularly closely connected to professional responsibility, epistemic control, and consequences for patients.
Manifesto for a Power-Aware Ethics of AI The book “The Ethics of AI”
Predictive Knowledge, Data Protection, and Regulation
Another focus is the ethical and political significance of predictive knowledge. AI systems can predict sensitive information about people even when the individuals concerned have not disclosed that information themselves. Such predictions frequently rely on data from many other people and create new forms of collective vulnerability.
Medical AI systems are an important application area. Medical data are not only particularly sensitive in themselves; they can also generate predictive insights and consequential privacy harms for third parties. This makes purpose limitation and an updated approach to data protection for the AI era particularly relevant.
Through the concepts of predictive privacy and purpose limitation for AI models, Rainer Mühlhoff and legal scholar Hannah Ruschemeier develop new approaches to data protection and AI regulation. Their work is particularly concerned with responsibly constraining the social power generated by trained models, large datasets, and the possibility of their subsequent repurposing.
Data Protection and AI Regulation Project “Predictive Knowledge is Power”
AI, Education, and Societal Futures
Within the Ethics and Critical Theories of AI research group, Rainer Mühlhoff also investigates the role of AI in education and schools. This work includes studies of automated assessment and feedback systems, AI literacy projects, and educational formats that enable young people to engage critically with AI.
One current topic is AI resignation as a diagnosis of our time: the feeling among young people that they are powerless in the face of technological developments and the visions of the future promoted by large tech corporations. This research asks how education can foster not only individual adaptability, but also democratic agency and alternative visions of society’s future.
Theoretical Foundations and Interdisciplinary Collaboration
The current work builds on longstanding research into power, subjectivation, affect, and digital media. Concepts such as Human-Aided AI, automated inequality, digital disempowerment, and affective subjectivation demonstrate that digital technologies are always embedded in social relations and collective practices.
Rainer Mühlhoff collaborates across disciplines with scholars in law, computer science, medicine, and education, as well as with partners from industry and civil society. He combines his training in mathematics and computer science with approaches from philosophy, critical theory, and media studies.