14. 10. 2026
16.45 - 18.15
Seminarraum VI (Theologie) | Karl-Rahner-Platz 3, 1. Stock
As AI technologies spread and develop rapidly, experts and developers claim that systems based on artificial neural networks are expanding the scope of machine understanding, even though no consensus exists on what machine understanding should look like. This talk offers a critical review of accounts of machine understanding. Indeed, experts use evidence of advancements and limitations to defend claims about the explainability, safety, and intelligence of AI systems. Nonetheless, defending such claims requires an account of machine understanding. To do so, we must identify what constitutes understanding and how to evaluate it. In this talk, I will present model- and ability-based accounts of understanding, evaluate etiological and phenomenological requirements, and reflect on proposals for conceptual frameworks for machine understanding.
Giovanni Galli is currently a PhD student at the University of Innsbruck. He completed his B.A. and M.A. at the University of Bologna. After earning his first PhD in philosophy of science from the University of Urbino, where he worked as a research collaborator, he has been a Postdoctoral Fellow at the University of Teramo. His research interests include epistemology, philosophy of science, and philosophy of artificial intelligence, with a particular focus on understanding and scientific understanding. As a local administrator, he is also interested in science’s public and social role.