How AI is rewriting the skillset of researchers
Artificial intelligence takes over routine tasks like data analysis and copywriting, but never replaces critical judgment and creativity, writes Judith Künneke. ‘AI is unlikely to replace researchers; it will push us to develop faster.’
In the past few years, AI has moved from being a tool in computer science to a presence across academia that we have not seen before. Today, it quietly “sits” next to us as we write, analyze, and brainstorm. What used to be a mostly solitary process of juggling data, words, and ideas is increasingly becoming a collaboration (although the solitary environment might still hold). This trend is not just about efficiency, but it is changing what it means to be a skilled researcher.
AI tools now assist with literature searches, coding, data cleaning, and even writing paper drafts or reviewer responses. At first glance, this seems pretty positive: more time for creativity, less for routine work. However, this transformation raises some questions that we should probably reflect on: If machines can perform parts of our intellectual work, what is our added value? And what does this imply for raising the next generation of researchers?
Traditionally, academic training was and still is heavy on methods: learning how to collect, structure, and analyze data. But as AI automates these foundational skills, researchers must increasingly train judgment and design thinking. That is, the ability to ask meaningful questions, to assess whether AI-generated results make sense, and to see the broader implications of findings. The art of thinking about thinking, also known as metacognition, becomes more central to our work.
This is a rotating column from the Tilburg Young Academy (TYA). Each month, a different TYA member highlights developments in the academic world.
At the same time, the ease AI provides bears a nontrivial risk of intellectual complacency. When text generation, analysis, and visualization become costless, the temptation grows to accept outputs without deeper, critical engagement with the output. The never-ending struggle with theory, data, and writing can start to feel unnecessary. Yet, it is precisely that struggle that builds depth, originality, rigor, and discourse.
For early-career researchers, the demands appear to increase. They have to learn traditional academic methods while also developing digital and ethical literacy, and at the same start investing in their higher-order skills such as ethical reasoning, synthesis, and creation of new solutions.
These are the skills that AI – at least until some point in the future – cannot replicate and that define our intellectual independence. Understanding AI does not seem optional because it is part of being a responsible scholar. But so is maintaining the curiosity and healthy skepticism that define great research.
As supervisors, we carry a special responsibility in this transition. PhDs are entering a world fundamentally different from the one we knew. They need structured opportunities to develop higher-order skills: to evaluate evidence critically, to reason ethically, and to translate complexity into knowledge that matters.
Universities play a crucial role by creating mentoring programs, interdisciplinary collaborations, and training modules that explicitly focus on reflection, reasoning, and judgment. We should normalize conversations about how we think, not only what we produce.
In the end, AI is unlikely to replace researchers; it will push us to develop faster. The real question is not whether AI changes academia but how we, as researchers and educators, will change ourselves in response.
Judith Künneke works as Associate Professor of Managerial Accounting at TiSEM. She is interested in how AI-supported decision environments change managerial judgment, and what the use of AI implies for skill formation, evaluations, and career trajectories.
Advertentie.
