Vortrag im Rahmen des University of Vienna Physics Colloquiums
Artificial intelligence is changing not only the tools, but the underlying conditions of academic teaching: how students practice, generate texts and solutions, receive feedback - and how we can assess performance validly. In this talk, I outline established findings from educational research on effective learning (clarity of learning goals, active practice, quality of feedback, misconceptions, motivation, collaboration, interaction) which, in my view, are now more relevant than ever. From these, I derive how course and curriculum design can be realigned for the AI era. The focus is on three levers: assessment (validity, authenticity, assessment formats between product and process), feedback (targeted, timely, learning-effective; AI as support), and competencies (which abilities graduates should demonstrably master in the future). The second part is designed as a discussion: what adjustments to curricula, assessment architecture, policies, and faculty development are needed to improve teaching sustainably, rather than merely reacting with AI-driven ad hoc solutions? No one, however, will have final answers.
A light buffet will be offered before the lecture at around 16:00.
