Potentials and Challenges of AI in Crisis Management Across Austria and Europe

Artificial intelligence (AI) holds great promise for improving how societies prepare for and respond to crises. Yet real-world implementation is far from straightforward. It is shaped by structural, regulatory, and ethical considerations that can constrain both AI’s functionality and its adoption in high-stakes contexts. Drawing on recent Austrian research projects, this paper examines recurring challenges in AI-supported communication and decision-making in public health and crisis management. These include data fragmentation, lack of model interpretability, regulatory uncertainty, and insufficient stakeholder alignment. While rooted in the Austrian context, these challenges reflect broader patterns seen across Europe, particularly in light of evolving regulatory
frameworks such as the EU AI Act. Addressing them requires more than technical innovation. It necessitates inclusive design, long-term institutional strategies, and strong ethical governance to align technological development with societal needs. The paper further positions AI literacy as a foundational capability for sustainable deployment, enabling informed engagement, accountability,
and trust across all levels of adoption. When grounded in human-centred principles, AI systems can become a meaningful support in building resilience and responsiveness to future crises.