The promotion of AI competence is increasingly recognized as a key prerequisite for managing digital transformation processes and is situated within the broader debate on future skills. Based on the AIComp model, the scenario-based self-assessment instrument AICompAss was developed to capture AI competence across twelve competence domains and to represent it on three proficiency levels. In total, the instrument comprises 72 action-oriented scenarios derived from learning outcomes aligned with the revised Bloom’s taxonomy. Its purpose is to make differentiated competence profiles visible and to systematically identify developmental needs. The development followed a design-based research approach consisting of three phases: conceptual design, validation by an interdisciplinary expert panel, and pilot implementation. The pilot was conducted in spring 2025 with eight employees from an Austrian university administration (response rate: 89%). The results reveal an overall low baseline level with pronounced heterogeneity among participants. Distinct polarizations emerged across specific competence domains, while other areas remained consistently low. The additionally collected AI Activity Index (KIX) indicated three distinct initial positions: two participants with high scores, four with medium, and two with low. These findings underline that usage intensity alone is not a sufficient indicator of competence levels but requires differentiation through scenario-based assessment methods. Based on the aggregated competence profiles, need clusters were identified, and an exemplary training concept was developed that combines practical, reflective, and ethical dimensions. AICompAss bridges diagnostic and didactic perspectives, offering an evidence-based foundation for future skills development in organizational learning contexts.
Karl Brandstetter (Thu,) studied this question.