Artificial intelligence is already reshaping education at TU/e, and it makes no sense to either ban it or ignore it. The question is no longer whether students will use AI, but how we as a university can guide that use responsibly. That is why I have been asked to take the lead in embedding AI into all aspects of education at TU/e. My task is to steer this transformation so that, by 2030, AI becomes a natural, thoughtfully integrated part of how we learn and teach. Establishing a clear vision is essential to ensure that both lecturers and students understand how AI can support learning while maintaining academic integrity.
Achieving this change requires more than minor adjustments: it demands rethinking entire degree programs, learning objectives, and assessment methods. I often compare this process to turning an oil tanker: it takes time and distance to shift direction. In the short term, our focus is on providing tools and guidelines for both students and instructors. In some cases, AI can be a legitimate aid, while in others, such as assessing fundamental understanding, it might be necessary to return to pen-and-paper or oral exams. What matters is moving away from testing facts that AI can easily provide, and instead encouraging deeper conceptual understanding and complex problem-solving. Challenge-based, multidisciplinary learning will be key to achieving that.
In the coming years, TU/e must become a place where AI enhances rather than replaces learning.
For more information, take a look at the interview in Cursor (thank you, Leoni Andriessen).
This post presents my perspective on the transformative potential of artificial intelligence (AI) in reshaping educational practices, particularly at the university level, through constructive alignment. It was developed using a multi-stage, AI-assisted process that combined human insights with interactions across various AI systems. This collaborative approach aimed to ensure clarity, coherence, and relevance while significantly reducing the effort required to structure and articulate ideas effectively. The creation process involved leveraging Gemini and ChatGPT to identify pertinent references from current academic discourse. The AIgenerated outputs were converted and underwent manual revisions and refinements.
