As part of the Research Training Group KEMAI – Knowledge Infusion and Extraction for Explainable Medical AI, funded by the Deutsche Forschungsgemeinschaft (DFG), a PhD position in the field of digital ethics is available from January 2027 onward. The project is situated at the intersection of philosophy, computer science, and medicine and investigates the role of explainable AI (XAI) in clinical decision-making contexts.
The PhD project will be jointly supervised by Prof. Dr. Rebekka Hufendiek and Dr. Hauke Behrendt.
The KEMAI team aims to combine the benefits of knowledge-based and learning-based systems in order not only to achieve state-of-the-art accuracy in medical diagnosis, but also to communicate predictions to physicians in an interpretable and clinically meaningful way, while taking ethical implications within medical decision-making processes into account.
KEMAI’s central goal is to provide interdisciplinary doctoral training in explainable medical AI for researchers from computer science, medicine, and philosophy. The RTG offers a structured doctoral program that creates an environment in which early-career researchers can conduct high-level interdisciplinary research in the field of medical AI.
We invite highly motivated candidates with a strong interest in research and a desire to contribute to an interdisciplinary academic environment to apply. The position is funded for 3+1 years and comes with a 65% TV-L E13 salary.
The deadline for applications is the 30th of June 2026
For further information, please visit the RTG website: https://kemai.uni-ulm.de/
In addition to the requested documents, please attach a writing sample (e.g., a chapter from your MA-thesis or a draft paper)
A possible PhD project could focus, for example, on medical scenarios in which AI systems generate diagnostic or therapeutic recommendations and thereby become involved in potentially high-stakes decisions. The central question is under what conditions such recommendations can be explained in a way that enables responsible clinical judgment.
Explainability is not primarily understood as mere technical transparency of models, but rather as a practice-oriented form of justification. The project develops an approach that conceives of explanations as context-, stakeholder-, and risk-sensitive practices. Explanations must therefore be aligned with specific decision-making situations, the epistemic roles of the actors involved, and the potential consequences and risks associated with the decision.
The project assumes that the requirements for adequate explanations increase with the stakes involved: the greater the potential impact on affected individuals, the greater the justificatory burden explanations must satisfy.
A particular focus lies on guideline-based explanations, which enable the assessment of AI outputs in light of established medical guidelines and diagnostic criteria. In this way, the project seeks to strengthen the active role of clinical decision-makers, reduce overreliance on AI systems, and improve the attribution of responsibility within socio-technical decision-making contexts.
While the project is primarily conceptual in orientation, it explicitly aims at interdisciplinary integration. The criteria developed within the project will be further specified in close collaboration with participating disciplines within the KEMAI cluster and, in the longer term, translated into technical and empirical research contexts.