Illustrates how students’ objective for what they need to learn and which tools they need to master are challenged by their concerns about future job market demands, this contradiction prompting some students to use GenAI. Credit: Journal of Computing in Higher Education (2026). DOI: 10.1007/s12528-025-09488-8

Studies from the Department of Communication and Learning in Science show that AI tools such as ChatGPT are not merely being used as support in students’ studies. In fact, they may be reshaping how students perceive knowledge and learning—a perspective that is not always shared by teachers and universities.

Generative AI has, some argue, become a natural part of students’ everyday lives. While public debate often becomes stuck on issues of cheating, assessment and control, research indicates that the transformation runs far deeper than that.

Today, students use AI to search for information, explain connections, summarize literature and test ideas—often without involving teachers. Alongside this, some students have begun to question which skills are truly relevant when so much academic work can be carried out with, and by, AI. At the same time, many express concern that higher education does not provide sufficient training in how to use AI in ways that meet the expectations of working life and society.

What is the value of higher education institutions?

These changing behaviors and expectations challenge the traditional role of universities, which has historically been built on specific assumptions about what it means to learn, to know and to perform. When both knowledge production and problem-solving increasingly take place in interaction with AI, universities need to redefine, articulate and justify their learning outcomes and pedagogy in ways that feel relevant to today’s students.

"The question then becomes: What is it that higher education actually offers that cannot be replaced by AI, and how can this value be communicated and realized in practice?" says Tiina Leino Lindell, postdoctoral researcher, who together with Professor Christian Stöhr has conducted several studies on AI in higher education.

The studies, based on interviews with both teachers and students, also show that students increasingly use AI to prioritize their time. Tasks perceived as boring, repetitive or irrelevant to future careers are often delegated to AI tools, while activities considered important or personally developing are prioritized. This, in turn, creates tensions, as teachers express concern that students may miss foundational elements or practice certain skills too little.

Academic integrity over technology

Another challenge is that while technological development and student behavior are changing very rapidly, university organizations are designed for stability and long-term planning. Discussions about digitalization therefore often become reactive: technology changes practice first, and guidelines are formulated afterwards. The researchers argue that universities need to work in a more principle-based rather than technology-based manner.

"Guidelines that are strongly tied to individual technologies risk becoming outdated quickly as both tools and patterns of use evolve. A recurring theme in the material is therefore the need for guidelines grounded in overarching pedagogical principles, such as academic integrity, transparency, clear learning outcomes and responsibility. In this way, the need to rewrite guidelines every time technology changes is reduced," says Christian Stöhr.

He emphasizes that questions of AI and pedagogy cannot rest on individual teachers alone, but must become a natural part of universities’ collective work on educational goals and quality.

"Our studies do not show how AI should be used in higher education. However, they do show that both students and teachers are already engaging with the technology in ways that challenge established norms. This makes questions of goals, responsibility and pedagogy difficult to postpone," says Christian Stöhr.

The aim of the study was to investigate how generative AI affects norms, roles, and practices in engineering education from the students’ perspectives. It is based on interviews with 25 engineering students who actively use generative AI in their studies. The paper is published in the Journal of Computing in Higher Education.

Four key themes in the study

Students’ self-directiveness and efficiency: Students use generative AI to solve practical problems, such as translating languages, understanding theory, and debugging code. They view the tool as a fast and constantly available mentor, in contrast to traditional digital tools that are often perceived as slower or insufficient.

The objectives of learning are challenged: Many students feel that mastering generative AI is necessary to be prepared for working life and therefore adapt their learning to what they perceive as future demands. However, this is also an area where they feel that education does not provide sufficient support.

The role of teachers is changing: This role changes when students’ actual AI use does not always align with teachers’ formal rules. Students often turn to generative AI for simpler questions, which reduces interaction with teachers. Differences between teachers’ and students’ views on how generative AI use should be restricted also create new challenges for the teaching role.

The ethics of cheating are challenged: These are challenged when a high workload is combined with the efficiency of AI tools, which can lead to boundary-drawing practices that conflict with academic integrity.

The second study presented in Learning, Media and Technology was conducted in two phases. The first phase consisted of interviews, both individual and group-based, with engineering students from 13 different programs. The questions focused on whether, how and why students use generative AI. One aim was to capture their views on the advantages and disadvantages, as well as their thoughts on what rules and guidelines might be needed. The students’ input was then grouped into five themes.

In the second phase, university teachers, postdoctoral researchers and educational developers used these five themes as a starting point to explore different possible futures, rather than to predict a single outcome. This scenario planning approach is an effective way to identify challenges and imagine future directions. A two-year timeframe was chosen to be both realistic and forward-looking.

More information

Tiina Leino Lindell et al, The AI disruption in engineering education: an analysis of changing student norms through cultural historical activity theory, Journal of Computing in Higher Education (2026). DOI: 10.1007/s12528-025-09488-8

Tiina Lindell et al, Navigating generative AI in higher education—six near future scenarios, Learning, Media and Technology (2025). DOI: 10.1080/17439884.2025.2562405

Citation: AI challenges established norms in higher education (2026, February 3) retrieved 3 February 2026 from https://phys.org/news/2026-02-ai-norms-higher.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.