Mirjam Glessmer
Centre for Engineering Education at Lund University
Responses
Sometimes it feels as if AI arriving has sucked all the joy out of being a teacher.
Hi, that sounds terrible! Without knowing anything about you or your context, first let me tell you that you are not alone. The presence of GenAI is a challenge many teachers struggle with, and it has caused many to question their own role as a teacher now that it seems that knowledge is even more readily available than before and students might feel that they can get personalised instruction from GenAI 24/7. I recently read a study by Paxton et al. (2026) on writing instruction — probably the subject that is affected the most by the presence of GenAI — which I think it is relevant far beyond, where they ask what the presence of GenAI does to teacher emotions. They motivate their study by writing “[t]he emotional well-being of frontline educators has far-reaching consequences, not only for faculty job satisfaction and retention but also for student success (Madigan and Kim, 2021)" (a study in which they find “preliminary evidence that teacher burnout can affect the students they teach"!). Teachers do emotional labour to regulate and care for emotions — their own as well as their students’. Paxton et al. (2026) interviewed eight writing teachers and find teachers emotional responses to GenAI are curiosity, but also anxiety over how writing skills will be valued in the future, and a sense of moral injury (because their jobs as teachers were made so much more difficult, but also because students would now never learn the same skills as before, and also not experience the joy that comes from being a skillful writer, especially in a second language). teachers respond to students’ potential use of GenAI by increased emotional and cognitive labour, and more time working with assessment, especially when GenAI use is suspected. teachers perceive a lack of institutional guidance on how to deal with GenAI and suspected misuse. teachers feel less professional accomplishment compared to “before AI”. From their findings, Paxton et al. (2026) develop recommendations that — of course — include that universities need to provide clearer guidance for how to deal with GenAI and with suspected cheating, but also to “expand the range of supports offered to faculty in the wake of GenAI, specifically addressing the emotional and interpersonal pedagogical effects of the technology. Whereas workshop sessions on integrating GenAI tools into teaching and assignment design proliferate, guidance on addressing issues of trust-building in the [teacher student relationship] would constitute valuable additions” and relatedly, to offer more opportunities to talk about their experiences: “formal support for communities of practice in which instructors can share their experiences with each other for mutual support and acquire guidance from experts". And at least in my context at LU, we are working on all of this! There is maybe not so much you can do about how much support you receive from your university (but sometimes even contacting people whose job it is to support you, for example in educational development units, can change the support you get -- both because you are then in direct contact with them, and also because they can use your request to document the demand and lobby for it higher up, which might ultimately give them more resources to support you and others), but the last recommendation about the communities of practice is something that you can somewhat drive yourself. We recently wrote an article (Glessmer et al., 2026) where we present four grassroot initiatives that teachers had created to talk about teaching in different contexts, and there we share how we interpret design principles for such communties and the success factors and difficulties we identified. Maybe you can initiate a community to talk with colleagues and support each other? Last not least, maybe feeling like teaching is not fun any more is also an opportunity to seek out conversations with students. It is probably not the best approach to tell them that you feel that there is no joy left, but rather to ask them how they feel about the presence of GenAI. Maybe once you create some connections and realise that both you and your students do actually want to talk with, and learn from, actual humans, things will start to feel better? It won't make GenAI go away completely, nor all the challenges that its presence brings to teaching, but it might bring back at least some of the joy in learning and teaching, both for you and your students! -- Glessmer, M. S., Daae, K., Thoni, T., Lundmark, A. M., Dunnett, K., and Herde, H. (2026). Creating grassroots communities of practice to discuss teaching. Oceanography 39(3), https://doi.org/10.5670/oceanog.2026.e300. Madigan, D. J., & Kim, L. E. (2021). Does teacher burnout affect students? A systematic review of its association with academic achievement and student-reported outcomes. International journal of educational research, 105, 101714. Paxton, A., Kang, P., & Yim, A. (2026). The Affective Impact of Generative Artificial Intelligence on Instructors of First-Year Writing Courses. Teaching and Learning Inquiry, 14, 1-17.
What happens to data that I (involuntarily) provide to a GenAI tool?
Here is an interesting case of GenAI being used to solve the Navier-Stokes problem, one of the "Millenium Prize Problems", where there is at least the suspicion that storing ongoing work in a model might have contributed to its training -- without the knowledge of the actual humans who stored their ongoing work there: https://www.theguardian.com/science/2026/sep/08/openai-claims-to-have-solved-maths-problem-that-stumped-humans-for-decades I don't know what kind of contracts they had with the GenAI providers, but it's at the least interesting to be aware that this kind of thing might happen, and concern about data privacy seems to not be such an unreasonable reaction.
How good is generative AI at maths?
Here is an interesting case of GenAI being used to solve the Navier-Stokes problem, one of the "Millenium Prize Problems": https://www.theguardian.com/science/2026/sep/08/openai-claims-to-have-solved-maths-problem-that-stumped-humans-for-decades
What do I need to tell my students about what they are (not) allowed to do with GenAI in my course?
Hi! First: It is definitely good to share some kind of guidance with your students. Most students are very concerned about accidentally breaking a rule that they weren’t aware of, so they are very keen on knowing exactly what is and is not allowed in your course, especially since it can vary wildly between different teachers as you describe. At Lund University, Rachel Forsyth & team have developed a “GenAI guidance wizard” (see canai.education) where you pick and choose between different ways to approach GenAI. Based on many different ways students report on using GenAI, you for example respond to “May students use GenAI to help them with accessing materials, for instance transcribing text to speech or vice versa, creating alt text for images, or creating summaries or images?” by clicking "yes" or "no". After you have gone through a whole list of use cases and decided on whether or not students are supposed to use GenAI in those cases, you receive a .docx document. In that document, the language has flipped from the teacher perspective (“May students … ?”) to address student directly (“You may … !”). You can then go in and edit that .docx to make sure it fits your course and ideas in all details. You can add to it or remove sections that aren’t applicable. Since the GenAI guidance wizard was developed for Lund University specifically and based on the GenAI framework in place here, it contains a section about what students at Lund University are never allowed to do (e.g., share work that is not the students' own, e.g. other students' work or materials that your teacher has created, unless the students have permission to do so). This part might need to be adapted depending on the context in which you want to use the guidance. And as I wrote above, you need to carefully edit the document to make sure that it really expresses exactly what you want it to do. But this is a great tool, and a very good first place to start when you are writing a guidance from scratch. And maybe your colleagues will want to generate a guidance document of their own for their own courses, and you can discuss on that basis? So long story short: I cannot tell you what to tell your students, but there is a great tool that can help you articulate it yourself. And then, of course, it is important to discuss the document with students to make sure you have the same understanding of what it means (see this discussion on what it means to use GenAI to “polish” a text)
Should a student use GenAI to polish their text?
Reading the quote in Rachel's response, I am wondering what exactly was meant by "AI". Word has in-built spell and grammar checks, using which I would agree are accepted academic practice. Also tools like grammarly seem to be widely accepted. But while those use AI, they are not LLMs. For me, personally, acceptable "polishing" would mean to look at a text on a sentence-by-sentence basis and make individual decisions about whether a suggested word or phrase actually improves readability and the point I want to make -- or not, similarly to how one would work with the tools I mention above. Copying and pasting a whole suggested rewritten paragraph (or more!) from a LLM would, in my perception, not count as polishing any more but rather would be having a LLM rewrite. And there I worry that nuance (or even the meaning in general if I am trying to say something that isn't mainstream in the LLMs training data) is getting lost, and I also would hate to loose the feeling of ownership of my text. But I agree that this is a conversation that needs to be had between supervisor and supervisee in their specific context, and that probably also needs to be revisited as tools develop and with them our perception of what is acceptable and what isn't.
How good is generative AI at maths?
My expertise is certainly not in mathematics, but I found this recent substack post "Are we just collecting lots of AI-proof-shaped stamps?" by Adam Kucharski very interesting: Even if AI can be used to generate proofs of problems that have puzzled mathematicians for centuries (and there are apparently examples of this happening), what is the actual value of having said proof if nobody can explain how to get there? Can the field still build on it and integrate it in the accepted body of knowledge without being able to explain it in the traditional ways of however people are usually working with proofs? Maybe mathematics is about to face a paradigm shift, who knows. Or maybe AI-proofs are going to be rejected until someone can explain the actual logic behind them, step-by-step. I think these might be interesting questions to discuss with students, and then also in extension what it means for them themselves to use AI to generate proofs...
Pros and cons of using AI-generated podcasts as teaching materials?
I have two points right away I think are worth considering: First, you write that “students aren’t reading anymore”. Is that something you are willing to accept? If not, maybe it is worth looking into what can make reading more attractive, for example providing the texts instead of having them buy expensive textbooks, or by working with social annotation (there are plugins that work with Learning Management Systems) so you can scaffold opportunities for students to practice looking for the relevant information; marking what they find most interesting, noteworthy, questionable; asking questions and providing answers to peers; linking to more information; and much more, so that they exercise the skills that might make it more likely that they start reading again. If you do want to use podcasts, carefully quality check them before recommending them to your students. Rettberg (2026) finds (for one specific software, and at one point in time) that podcasts misunderstand niche-topic texts (which, probably, a lot of our teaching content is) and, by trying to match a typical podcast format, introduce biases and make everything sound mainstream. They also warn that by being whispered to directly into our ears (since we often listen to podcasts on headphones, in contrast to sitting in a lecture theatre or having a radio playing in the background in the kitchen), podcasts might appear more intimate and students might then be less critical of potential bias or misinformation. So it is important that both you and your students are aware that this might potentially be happening. I do understand the appeal of the idea, though. We experimented with this three years ago and where very positively surprised back then. But I have since become a lot more critical of AI based on all the literature that shows how biased AI models are, and would now be very careful with using them to generate teaching materials.
How can I approach a student if I suspect that they have used AI in ways that aren’t allowed (but I don’t have proof)?
You are completely correct about AI detection software not being reliable (read more about that here), and I would also be very careful with the tell-tale signs that you think you see and that you interpret as AI – your gut is as unreliable as the detection softwares!* I would start by considering what it is exactly that you hope to gain from confronting the student, and what you stand to lose by doing it. It is possible that the student will admit to using AI in ways that you explicitly did not allow (but check your course policy first – is it really so clear that you can say that you are not allowing a specific use, like using AI to polish language or support editing?). When they admit to it, what happens then? Will you ask them to redo the work, or fail them on the task, or notify an oversight committee? And can you be 100% sure that they did not just admit to it to get out of an uncomfortable confrontation, and now they are carrying consequences for something they did not, in fact, do? On the flip side of the coin, what happens if they do not admit to it? If they did not do it, you have now created a situation where they feel wrongly accused, probably wonder if you will always be looking out to catch them cheating even though they never did, so your relationship with that student has suffered. If other students hear about you wrongly accusing someone, that will also influence your relationship with them. And will you actually believe them if they say that they did not do it, or will you also be looking differently at them after this? It is of course also possible that your hunch is right, they did use AI, but they still don’t admit to it. Since you don’t have proof either way, where does it leave both of you now? So in a nutshell, I think that you really don’t have anything to gain from confronting the student. What you could, however, do, is address the topic more generally with the whole class. Not that you suspect someone is going against your policy, but going over your policy again, ask students what questions they have regarding that policy, if they know what that means in practice. Or, depending on what specifically looked like AI to you, addressing that you have seen a lot of this stye of writing, but that another style would be more appropriate; that you would appreciate if they edit their drafts in certain ways to avoid repetition; ...; meaning addressing specifically what it is that you want to be different in the texts they produce next time, rather than making it about whether AI wrote it or your students did. *Three special cases that I can think of: 1) You detect plagiarism. That is academic misconduct and needs to be addressed as such, no matter whether it was the student themselves or AI and the student did not notice. 2) You find made-up references or clear misrepresentation of facts. You still don’t know completely for sure that they come from AI, but that does not matter. Those are obvious problems with the content of the essay, and you can address them as such, for example by asking the student to send you the (non-existent) article, or explain how they found the wrong “facts”. 3) The student forgot to edit the AI output to the extent that you find things like “Certainly! Here is the response to the essay question using scientific language”. In that case, you might just ask the student to explain why they added that sentence in the context.
How to avoid getting lazy using these tools and losing skills to AI
AI use can indeed be a slippery slope! One study that shows that is the one by Poulidis et al. (2025), who use a 12-week home online chess training provided to 216 members of chess clubs (with at least a year of training, so already demonstrably motivated and engaged, and about half of them 18 or younger). Participants were invited to participate and reminded to go practice on that platform by their coaches, and there were financial incentives in place (10$ as base incentive and bonuses up to 150$). On the platform, students were assigned two conditions: Either they received automated tips in “critical moments” but couldn’t ask for help, or they received the same kind of tips but also could request additional help by clicking a button. And clicking the button they did! And that had consequences on learning: Performance gain in the self-regulated group were a lot lower than in the other group. In the group where students could not ask for hints but received targeted feedback at points in the process determined by the algorithm, students had to go through productive struggle which contributed to their learning, whereas in the other group, based on survey results, “students knowingly over-relied on AI assistance, thereby diminishing their sense of accomplishment. Despite recognizing these drawbacks, they not only continued to rely on it but did so increasingly over time”. The size of that effect is moderated by student motivation (but not skill!): “motivation moderates the learning losses induced by self-regulated AI use, with more motivated students experiencing substantially smaller learning losses”. Students in the self-regulated group also played 24% fewer training games than students in the other group. When asked for a preferred training type for hypothetical future training, the largest group of self-regulated students (40%) picked “no tips”! But Poulidis et al. (2025) also compared learning gains against students who had not been part of the study and thus did not have access to the platform at all, and overall, students learned more on the platform, no matter the experimental condition they were under. So you can learn from AI! However, “[g]iving students control over when to receive assistance can substantially hinder learning; effective AI tutors should therefore target help to moments when it best supports learning rather than providing assistance on demand“. Otherwise it is easy that students fall in the “agency trap: when highly accurate solutions are easily accessible, even students who genuinely want to learn over-rely on AI assistance.” Poulidis et al. (2025) conclude that “students recognized that overuse would harm their learning yet still relied heavily on AI assistance—awareness alone cannot prevent misuse”. Emphasis in that quote is mine because I find it to be so important: even though participants were motivated to learn and knew that by clicking the button, they were harming their own learning, they still could not resist the temptation, and 40% therefore (or at least that’s my interpretation) would wish to not even be led into temptation in hypothetical future training by completely removing the option to get feedback. Poulidis et al. (2025) close by writing “[a]s self-regulated AI use becomes increasingly ubiquitous in education and the workplace, preventing harm to long-term learning and the atrophy of human capabilities is a central design challenge”. This is a really interesting article with super relevant results. It seems likely that what they find — it is really difficult to resist an apparent AI shortcut, even if people know it is going to backfire on them — will work similarly in situations where AI can so clearly give a correct answer that results in clearly marked “wins” (a win does not get much clearer than winning a game of chess). The reason I am posting this whole novel in response to your question, though, is that I think being aware of such studies might help you and your students to make the decision to put measures in place to remove temptation for themselves, so as not to even step on that slippery slope at all, which is also my own solution to the problem. But I am curious to read other approaches! Reference: Poulidis, S., Bastani, H., and Bastani, O. (October 01, 2025). Self-Regulated AI Use Hinders Long-Term Learning. The Wharton School Research Paper, Available at SSRN: https://ssrn.com/abstract=5604932 or http://dx.doi.org/10.2139/ssrn.5604932
Can I use AI to analyse students’ free-text responses?
There are several points I think are worth considering here. First, those free-text answers are the students’ intellectual property and might contain sensitive, personal data. I would be very careful with uploading that type of text to a LLM (although LU’s CoPilot license allegedly protects our data). If you run an AI model locally on your computer, that might not be a concern. Second, I would be very wary of the quality of the outputs. In my own research, we find that LLMs are not suitable for qualitative research for many reasons (Glessmer & Forsyth, 2025) and Nguyen & Welch (2025) come to the same conclusion using a very different approach. Jowsey et al. (2025), in the piece “We reject the use of generative artificial intelligence for reflexive qualitative research” signed by 419 qualitative researchers from 32 countries, write very directly that they reject GenAI in reflexive qualitative research because 1. GenAI is incapable of meaning-making; 2. Qualitative research needs to be done by humans; and 3. GenAI is too harmful to the environment and to the human workers who need to filter out toxic content. Of course, all those three articles are focussed on scholarly use of qualitative data, but still relevant. There is more anecdotal evidence that nicely shows problems with AI analysing qualitative data that is directly relevant also to analysis that is done “just for a quick look” and does not need to have research quality: a very nice Bluesky post by Sasha Gusev with the probably shortest write-up of a study in the history of AI: “I assigned random gender/ethnicity labels to scientific abstracts from the literature and then asked Claude to do a thematic analysis. Claude identified a clinical versus computational split for female/male authors and a DEI focus for Black/URM authors. All in completely random data.” Full prompt and output available on Github. “Real signals or artificial stereotypes? Adventures with a cultural Copilot” by Adam Kucharski on Substack:**** An artificial dataset (200 responses assigned the label “UK”, plus the same 200 responses assigned the label “US” in the same file) and AI “finds” lots of cultural differences in expressiveness, language style, emotional framing, cultural tone that is clearly not in the dataset. Another dataset about career aspirations in 5 countries (again, with identical data for each) and the output is about lots of cultural differences. So if people haven’t gotten the message yet: Do not use LLMs for analysis of qualitative data! Lastly, I would think about what it says about the relationship between you and your students if they put in the effort to give you written feedback and you chose to not read it yourself, and instead read an artificially created summary. Maybe it is worth the effort to actually read it yourself? References: Glessmer, M. S., & Forsyth, R. (2025). “Superficially Plausible Outputs from a Black Box: Problematising GenAI Tools for Analysing Qualitative SoTL Data”. Teaching and Learning Inquiry 13 (January):1–9. https://doi.org/10.20343/teachlearninqu.13.4 Nguyen, D. C., & Welch, C. (2025). Generative Artificial Intelligence in Qualitative Data Analysis: Analyzing—Or Just Chatting?. Organizational Research Methods, 10944281251377154. Jowsey, T., Braun, V., Clarke, V., Lupton, D., & Fine, M. (2025). We reject the use of generative artificial intelligence for reflexive qualitative research. Qualitative Inquiry, 10778004251401851.