ERC Starting Grant for relAI Fellow Vincent Fortuin

Fellow

🎉Congratulations!

We are very excited to announce that relAI Fellow Vincent Fortuin has been awarded a European Research Council (ERC) Starting Grant for his project "AutoBayes",  with the title "Unlocking Reliable Small-data AI through Bayesian Deep Learning".

Most current deep learning models require large amounts of training data and tend to be overconfident in their predictions, undermining their reliability. Over the next five years, Vincent and his team will develop Bayesian deep learning methods to create AI models that are confident using limited data while also improving the quantification of the uncertainty in their predictions.

The ERC Starting Grant is one of the most competitive sources of funding available to early-career researchers; it enables them to pursue ambitious projects, to form their own teams and to gain independence at a crucial point in their careers. This year, the funding—amounting to €705 million in total—has been awarded to 421 early-career researchers throughout Europe.

For more information, see the most recent ERC press release.

relAI is proud to support the Philosophy of Machine Learning Conference (PhilML'26), which will take place in Munich from October 6 to 9. Tom Sterkenburg, a relAI Fellow, is one of the main organizers of this event.

The conference will address fundamental epistemological, ethical, and social questions related to machine learning through the lens of analytic philosophy. We welcome the following types of work:

1. Research that applies philosophical concepts and methods to gain insights into machine learning.

2. Research that critically reflects on the philosophical and ethical implications of machine learning.

To promote close and productive dialogue, PhilML brings together philosophers and machine learning researchers with a philosophical inclination, encouraging direct engagement with scientific and mathematical details.

Additionally, the main conference will be preceded by a graduate workshop on October 6. We invite submissions of extended abstracts for both the main conference and the graduate workshop. 

👉Registration link for the main conference and workshop: https://sites.google.com/view/philmlconference/registration

🎤 relAI fellow Valentin Hofmann was recently interviewed by the San Francisco Chronicle, the largest and most widely recognized newspaper in San Francisco and Northern California.

As an expert on social biases in AI models, Valentin was asked to comment on the controversial AI chatbot of state Senator Scott Wiener, which targeted Supervisor Connie Chan, Wiener's opponent in the 2026 California congressional election. The chatbot was accused of producing racist and sexist responses of Hong Kong-born Chan, who is a non-native English speaker.

In the interview, Hofmann explained that chatbots like Wiener’s are trained on large-scale internet data, which often contains racist and offensive stereotypes related to accents and citizenship. After reviewing the chatbot’s responses, Hofmann suggested that the model likely inferred from biographical and other available information that Connie Chan speaks with an accent. He noted that, in a failed attempt at satire, the chatbot exploited this information.

Hofmann's research demonstrates that linguistic stereotypes, particularly bias related to dialect and accent, are difficult to eliminate from AI models because they are encoded in subtle ways, in contrast to more explicit forms of racial prejudice. He also commented on the difficulties of eliminating problematic behaviour in AI chatbots and on current approaches to addressing it.

👉Link to Interview: https://www.sfchronicle.com/politics/article/pelosi-wiener-ai-chatbot-connie-chan-22378352.php

From July 28 to July 31, 2026, relAI hosted the second relAI International Summer School at LMU. This event focused on promoting educational exchange aimed at developing reliable AI and exploring current trends in the field.

The summer school welcomed a group of Chinese students and provided them with comprehensive insights into four key research areas of relAI: mathematical and algorithmic foundations, medicine & healthcare, robotics & interacting systems, and algorithmic decision-making. We would like to express our gratitude to relAI Fellows Prof. Dr. Eyke Hüllermeier, Prof. Dr. Michael Ingrisch, and Prof. Dr. Volker Tresp, as well as relAI PhD student Ian Huang, for their valuable contributions to the school.

We are happy to announce that Barbara Plank has joined relAI as a Fellow!

Barbara is Full Professor for AI and Computational Linguistics at LMU Munich, where she holds the Chair in AI & Computational Linguistics and co-directs the Center for Information and Language Processing (CIS). She also serves as Head of the Munich AI & NLP lab (MaiNLP) and visiting Professorship at the IT University of Copenhagen

Her research on robustness, domain shift, and human label variation aligns well with relAI’s Algorithmic Decision Making research area. This work explores how AI systems learn and make decisions in the face of uncertainty and disagreement. Additionally, her emphasis on interpretability, reasoning, and trustworthy evaluation provides essential foundations for developing reliable, fair, and transparent algorithmic decision systems.

At relAI, Barbara will contribute by participating in seminars, workshops, and panels, as well as offering career advice.

A warm welcome! 🤝

🎉 relAI is proud to announce that the International Association for Dental, Oral, and Craniofacial Research (IADR) has named relAI Fellow Falk Schwendicke as the recipient of the 2026 IADR Distinguished Scientist William H. Bowen Research in Dental Caries Award 🦷 .

IADR is a nonprofit organization dedicated to advancing dental, oral, and craniofacial research for global health and well-being. This esteemed IADR award recognizes exceptional and innovative contributions to our understanding of caries etiology and the prevention of dental caries. It is one of the 17 IADR Distinguished Scientist Awards and is considered one of the highest honors bestowed by the organization.

Falk Schwendicke’s Research Achievements

His early work focused on minimally invasive and evidence-based caries management, particularly regarding selective carious tissue removal and its economic evaluation. This research has laid the groundwork for contemporary treatment guidelines. His recent studies have increasingly emphasized the integration of emerging technologies to overcome challenges in caries detection and management. Notably, he has been a pioneer in employing advanced artificial intelligence (AI) applications for radiographic analysis, diagnostic support, and predictive modeling.

A significant achievement in his career was leading a randomized controlled trial that evaluated AI-assisted caries detection. This study set new standards for clinical research in the field and informed subsequent cost-effectiveness analyses. Schwendicke also participates in numerous editorial and review roles and has presented at the IADR General Session and various scientific meetings. He has authored over 500 peer-reviewed publications and 30 book chapters and is ranked among the top 1% of most-cited dental researchers worldwide, according to the Stanford global ranking.

👉 Information sources

https://www.iadr.org/about/news-reports/press-releases/falk-schwendicke-named-recipient-2026-iadr-distinguished

https://www.linkedin.com/posts/prof-dr-falk-schwendicke-9bb6271a1_iadr2026-activity-7444709023079350272-kiFG/?utm_source=share&utm_medium=member_ios&rcm=ACoAAAMg4egBnT-dMw4VyJR7tdTe0Z-9xhGUZZI

Welcome on board 🛳️ !

relAI Fellow Carsten Marr is Professor for AI in Cell Therapy and Hematology at the Medical Faculty and Clinics of the Ludwig-Maximilians-Universität München, as well as Director of the Institute of AI for Health at Helmholtz Munich.

In recent years, he has made significant contributions to AI-based hematological cytology. His focus on the interpretability of models trained on patient data to make predictions in a biomedical context 🩺 closely aligns with relAI's central themes of safety and responsibility. His innovative multiple instance learning models facilitate the investigation of relevant cells for disease prediction, while sparse autoencoders help correlate image features with diagnostic concepts. Additionally, his work on linking images and language enables direct comparisons between understandable human terms and cellular patterns within gigabyte-sized digital scans. At relAI, he will support students through lectures, mentoring, and participation in events.

🎉 Congratulations!

relAI is thrilled to announce that Frauke Kreuter, relAI Fellow and member of the relAI Steering Committee, has been elected a Fellow of the American Association for the Advancement of Science (AAAS). AAAS is the world's largest general scientific society and publisher of the journal Science. Founded in 1848, this non-profit international organization promotes scientific freedom, responsibility, education, and collaboration to improve humanity, serving over 120,000 members.

Being elected as a Fellow is a prestigious honor that recognizes individuals whose contributions to advancing science and its applications in service to society have distinguished them among their peers and colleagues.

👉 More Information: https://www.lmu.de/ai-hub/en/news-events/all-news/news/prof.-dr.-frauke-kreuter-wins-2026-waksberg-award.html


🎉 Congratulations to the relAI PhD student Johanna Topalis and relAI Fellow Prof. Michael Ingrisch!

🏆 The article they co-authored, “ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports”, has been awarded the Most Cited Article in European Radiology (Impact Factor 2024) by the European Society of Radiology! The work was presented at the European Congress of Radiology (ECR) 2026 in Vienna and honoured by the Editor-in-Chief of European Radiology, Prof. Bernd Hamm.

📖 The article presents the first exploratory case study evaluating the quality of simplified radiology reports generated by the large language model (LLM) ChatGPT. Radiologists rated the reports as generally high quality but also identified errors that could lead to harmful patient interpretations. The findings highlight both the potential and the limitations of early large language models in clinical communication: while simplified reports can enhance accessibility, medical expert supervision and domain-specific adaptation are vital to ensure patient safety.

💡 The study, first published as a preprint in December 2022, was among the earliest scientific assessments of ChatGPT's ability to simplify radiology reports for patients. Since then, a rapidly growing body of research has explored the role of large language models in medical text simplification.

👉 Publication: https://link.springer.com/article/10.1007/s00330-023-10213-1

      Preprint: https://arxiv.org/abs/2212.14882


We are thrilled to welcome Majid Khadiv as a Fellow at relAI!✨ He is an Assistant Professor at the School of Computation, Information and Technology (CIT) of the Technical University of Munich (TUM), where he holds the Chair of AI Planning in Dynamic Environments, and is Principal Investigator at the Munich Institute of Robotics and Machine Intelligence (MIRMI).

His lab focuses on the fundamental question of how to develop a scalable approach to building intelligent humanoid robots while also providing formal safety guarantees for reliable deployment in our daily lives. This research direction aligns with relAI's goal of creating safe and secure AI made in Germany. Moreover, his work on ethics in robotics 🤖 complements relAI's mission by emphasizing the importance of ethical considerations in the development of reliable AI.

As a fellow, he will contribute to the relAI curriculum by delivering lectures to students and helping them gain practical experience through internships.

A warm welcome! 🤝