Overview

Emerging general-purpose artificial intelligence (GPAI) systems that can tackle multiple tasks at near-human levels are showing great promise. Our mission is to explore how current and future GPAI systems can radically accelerate scientific research, transform biomedical discovery and healthcare delivery, and ultimately contribute to human health and well-being.

Our work focuses on two critical areas of research:

Global AI Strategy. We aim to understand how GPAI can have the most profound impact on scientific progress, uncovering opportunities, barriers, and risks associated with global GPAI-driven research ecosystems. Building on our deep understanding of both cutting-edge AI and biomedical research, we are developing strategies for the effective and responsible deployment of GPAI in biomedicine.

Scalable Oversight & Trustworthy AI. We devise techniques for the responsible development, deployment, and oversight of advanced AI systems (e.g., large language models) in time- and safety-critical settings, such as medical decision making. Our research investigates robust mechanisms for post-deployment monitoring and explores the potential of trusted “safeguard AI” systems. These systems assist human users in detecting errors and misalignment of external frontier AI systems, ensuring their safe and reliable operation.

Our group is part of the Institute of Artificial Intelligence at the Medical University of Vienna, Austria.

People

Assoc.-Prof. Matthias Samwald, PhD (principal investigator) →  Short bio
Konstantin Hebenstreit, MSc (PhD student)
Mihai Truta, BSc (visiting MSc student)
Thomas Prade (MD student, medicine)
Theresa Plumer, BSc (MSc student)
Stefan Huber, BSc (visiting MSc student)
Constantin Convalexius 
(MD student, project assistant)