Design principles for WP3
The analysis of both sessions not only provides insights into professional autonomy within AI-integrated professional contexts, but also concrete design principles for curriculum development within WP3. Both domains demonstrate that AI integration requires not only new technical skills, but above all a redefinition of how study programmes develop and demonstrate professional understanding, autonomy, responsibility and active steering.
Design Principle 1: In-depth understanding remains the foundation of professional autonomy
Students must not only be able to use AI output, but also be able to understand, assess, reconstruct and justify its content. Operational AI literacy or technically proficient use of AI is not, in itself, sufficient as the ultimate goal of professional degree programmes.
Both sessions make it clear that active AI-steering without sufficient in-depth understanding of the domain and processes can lead to false competence: situations in which students produce professionally appearing output without truly understanding the underlying logic, quality or implications.
Programmes therefore continue to invest in conceptual foundations, system understanding and in-depth domain knowledge, even as AI takes over a large proportion of the operational work.
Design Principle 2: Making underlying understanding explicitly visible
AI systems are making it increasingly difficult to assess whether professional competences are genuinely present based on end products alone. Students can successfully complete assignments and produce convincing output without the underlying expertise necessarily being equally well developed.
Programmes therefore provide learning environments in which students must explicitly analyse, justify choices, reconstruct lines of reasoning and investigate errors, even when AI is actively used within the learning process.
The aim here is not to exclude AI, but to reveal whether in-depth understanding, professional judgement and substantive autonomy are genuinely present.
At the same time, such learning activities offer opportunities to reveal the underlying professional dispositions, such as the willingness to reflect critically, to take responsibility and to explicitly justify professional judgement.
Design Principle 3: Assessment shifts from production to professional decision-making
As AI increasingly supports or automates routine tasks, professional added value shifts away from mere production and towards interpretation, argumentation, quality assessment, active steering and accountability.
Assessment therefore focuses less on the mere production of output and places greater emphasis on making professional reasoning, substantive choices, argumentation, the reconstruction of thought processes and the ability to justify decisions within complex AI-integrated contexts visible.
Design Principle 4: Students learn to deal with AI output that is not always consistent, predictable or factually correct
In-depth understanding and professional autonomy become particularly evident when AI output falls short, contexts become ambiguous or systems produce unexpected results.
Programmes therefore deliberately include learning activities in which students critically examine AI output, actively adjust it, reconstruct errors and analyse deviations. It is precisely these situations that reveal whether students are consciously steering AI or remain primarily dependent on generated output.
Design Principle 5: AI literacy is explicitly developed as a professional management competence
Active AI-steering does not develop spontaneously. Students must explicitly learn how AI systems function, what limitations and risks are associated with them, and how AI output can be purposefully guided through context, prompting, workflow design and iterative refinement.
Both sessions make it clear that low steering often arises when students use AI primarily via the standard operation of systems without sufficient explicit guidance or critical monitoring of the output.
AI literacy is therefore developed as a professional form of process management, not merely as the use of technical tools.
Design Principle 6: Study programmes actively resist knowledge erosion and dependency
Both sessions demonstrate that dependent use of AI does not necessarily arise suddenly, but often grows gradually as AI output becomes increasingly convincing and underlying expertise is utilised less actively.
Time pressure, the drive for efficiency and the increasing reliability of AI systems can lead to a gradual decline in active reasoning, critical analysis and conscious steering. Training programmes therefore not only develop the necessary knowledge and skills, but also create learning environments in which professional dispositions are consciously developed and practised. Professional autonomy within AI-integrated professional contexts therefore requires not only the ongoing development of knowledge and skills, but also of professional dispositions that ensure professionals continue to think critically, take responsibility and steer AI in a purposeful manner, even as AI becomes increasingly autonomous and convincing.