11 Unhealthy GenAI narratives in higher education: Shortcuts, cheating and policing

Prologue

In this article I am capturing some thoughts relating to my online postgraduate students studying towards an MA in Digital Education in the School of Education at the University of Leeds and our use of GenAI, individually and collectively. I have been wondering if it is helpful to talk about shortcuts, cheating and policing when it comes to GenAI use by our students when what we really want is to help them on their path towards transformative learning.

Students are using GenAI, probably more than educators in higher education and more than educators know or want to know. Students may be told to avoid using it for learning and assessment explicitly; others are fearful of being labelled cheaters, penalised, and receiving lower marks for their work if they confess to having used GenAI. Some educators are perplexed when students state openly that they have used GenAI to support their learning (previous chapter). We are all still very new to GenAI in education.

A recent survey study at Harvard University showed that almost 90% of undergraduate students (n=326) who participated are using GenAI in their learning (Hirabayashi et al., 2024). The most frequent use, according to this study was for asking questions followed by getting writing support. Students expressed their concerns about equity, as not every student can afford to pay subscriptions to use the full functionality of a specific GenAI platform. The Digital Education Council (2024) Global AI student survey revealed similar results for undergraduate, postgraduate, and doctoral students (n=3839) from 16 countries. It was noted that for 80% of students, universities did not meet their expectations regarding AI; 71% expressed an interest to being involved in related decision-making; over 90% of study participants expect to have access to related development opportunities to become AI literate and AI to be used in their courses. Universities have an important role to play in this area. What are they going to do?

Shortcuts? I don’t think so, or who are my students

Postgraduate students engaging in online learning are intrinsically motivated. They are interest- and goal-driven. We know that this aids learning (Anggraeni et al., 2024). My students know why they are on a course and what they want to get out of it, and they are keen to apply what they are learning!

Taking shortcuts could be seen as strategic manoeuvres, perhaps. It has however negative connotations. Looking at learning from the angle of surface, strategic, and deep learning (Säljö, 1979; Entwistle et al., 2001) may be more helpful. I am referring to learning. Not learners. We all move along the surface, strategic, deep learning continuum depending on the situation we are in, our response to it and what we think matters the most in a specific moment in time. We may, of course, not take the right approach, or the approach that would help us learn the most. My students are no different. Could the same be suggested about learning with GenAI? Related research could reveal valuable insights, and we need empirical evidence how it all works and what it means going forward.

My students are from different parts of the world. They often work and have other responsibilities. They are professionals. They recognise the value of learning and development. They are often time-poor and have conflicting priorities. But they are good at task and time management, and committed to learning. They want to apply what they are learning. They are focused on achieving their goals. When challenges come their way, they don’t run away. Students persist and succeed (Shi et al., 2024). They are often confident in asking questions. They admit when they don’t understand something and seek help. They may not do this publicly. They may have been outside education and need to catch up with technologies used to support learning. And they are willing to do so! They are more than willing; they are committed.

Some of my students may not feel comfortable with being seen as feeling lost or not knowing something. They don’t want to waste their time. And they can get frustrated when they think this is happening. My students won’t wait for long for a response from their tutor. They will search for answers to their questions elsewhere. They are resourceful in coming up with solutions that will help them move forward in their learning and the assessment they mustdo. It is just a matter of time until they reach out to GenAI. I would love to trial Plato with my students, a chatbot developed by a graduate of the University of Leeds. Plato is transparent about sources which distinguishes it from many other chatbots (Dumitriuc, 2024). The paper by Alsafari et al. (2024) relating to intent-based chatbots to support students learning provides an inspiration and made me think about this further and what such a move could enable. Wondering what Weller (2024) would make of it, especially in relation to his thoughts about open licencing. I would like to find out how it links to open data, open education and open research more widely. How would my students engage with it? I am wondering and would love to find out.

For many students, GenAI can become a valuable study buddy to help them learn, to help them with their assessment, and to help them achieve their goals. I have seen my students using GenAI in diverse ways. Even in their dissertations. They demonstrate that they are inquisitive, inventive and critical, and that they will not take GenAI outputs at face value. They will question, they will critique, they will inquire! They will also reject.

Cheating? I don’t think so or how learning with GenAI looks like for my students

My students are willing to give GenAI a go. For them it is an extra tool in their learning toolkit. They know what matters is that they are learning and exploring ways to do this as best as they can. They don’t really look at GenAI as a tool to help them cheat (Gorichanaz, 2023). They know learning doesn’t happen through cheating and plagiarism. If I think they’re using genAI to cheat, that shows I haven’t understood my students and don’t trust them. Trust is everything. Dawson et al. (2024) encourage us to focus on validity instead of cheating when it comes to assessment.

My students are experimenting with GenAI to see if it can help them in any way. Not for shortcuts, but to help them learn. One of my students used GenAI in diverse ways for his dissertation, including data analysis. I experiment too with GenAI, and they know it. We also experiment together in the open. Many of them appreciate the values of open education, such as sharing, diversity and collaboration and contributed also to two crowdsourced open collections (Abegglen et al., 2024; Nerantzi et al., 2023). These students know that learning can be hard. Learning is hard. If they feel they are losing too much time with GenAI, they will quickly move on. If they get something out of the use of AI for their learning, they will continue using it, become more sophisticated and critical, and expand areas they will consider using it.

Remember, one idea generates another idea and another one. Ideas are like chains. A chain or network of ideas. Exactly as it happens in creative thinking.

The more we play and experiment with ideas, the more novel ideas we will come up with as we have opened our minds to not just the obvious connections but also the most unusual ones. This is also what happens when we use GenAI.

Not everything will work, but that is fine. My students recognise this and even if something doesn’t work, reflecting in and on the experience can provide useful insights. It may lead to a modified approach and change in direction based on these insights.

Policing? I don’t think so or how I support my students

Supporting committed students is a pure joy. Trusting them is important, as is being flexible and open. Enabling experimentation and modelling effective GenAI practices can make a real difference. I work with my students. We experiment with GenAI. For example, I used it to construct Problem-Based Learning scenarios and introduce feedback poetry. We share openly. We are all new to GenAI. We learn together. I also learn from my students. Human to human connections is and will remain fundamental to learning. Valuing these is important! Together we engage in open scholarship (Brew et al., 2023; Taylor et al., 2023). As an open educator and scholar, my practice is open, open also to GenAI.

I avoid policing. I avoid saying, “don’t use it.” We use a traffic light system for the use of GenAI in assessment at the University of Leeds. Red means don’t go near GenAI; orange is use it if you wish to support your learning; Green you must use it! Policing the use or prohibiting it is not the answer, at least for me it isn’t. What I think could work is being  flexible and elastic in our thinking. Just expecting it from our students doesn’t feel right.

Openness, transparency, collaboration, and questioning are key. I feel that what is important is helping our students develop critical AI literacy for responsible GenAI use while also developing our own as educators. Biases in the data, ethics, and legality of data harvesting, ownership, exclusion, accessibility, equity, trust, and sustainability remain big concerns. Williamson (2024) makes us think critically about GenAI and makes a case against its use in education, while Beetham (2024) invites us to consider helping our graduates develop AI resilience and take a critical, values-based stance that guides their engagement with GenAI. Could AI resilience be part of AI literacy? Utochkin (2024) reminds us that we can imaginatively shape alternative realities and futures we will live in and calls his institution “to cut the ‘AI’ bullshit”.

What are your thoughts on all this?

Let’s keep questioning!

Epilogue

In this article I explored the unhelpful and damaging narratives around shortcuts, cheating, and policing in relation to GenAI use by students and my role as an educator and how I support my students in their learning. Trust, openness, collaboration, and sharing are important to encourage, foster, and nurture responsible, critical, and creative use of GenAI based on our moral compass and values towards transformative learning. Together we explore, experiment, and learn, and I am excited with what we may discover and uncover together.

References

Abegglen, S., Nerantzi, C., Martínez-Arboleda, Karatsiori, M., Atenas, J. and Rowell, C. (Eds.) 2024. Towards AI literacy: 101+ creative and critical practices, perspectives and purposes.  #creativeHE. https://doi.org/10.5281/zenodo.11613520

Alsafari, B., Atwell, E., Walker, A. and Callaghan, M. 2024. Towards effective teaching assistants: From intent-based chatbots to LLM-powered teaching assistants. Natural Language Processing Journal. 8. https://doi.org/10.1016/j.nlp.2024.100101

Anggraeni, D., Wardani, D. K. and Noviani, L. 2024. Self-Regulated Learning, Grit, and Learning Motivation in Developing Learning Achievement: A Review. Formosa Journal of Multidisciplinary Research, 3(1), 135–148. https://doi.org/10.55927/fjmr.v3i1.7908

Beetham, H. 2024. What price your “AI ready graduates? Perhaps we should be promoting AI resilience instead. Imperfect offerings. 7 August 2024. https://helenbeetham.substack.com/p/what-price-your-ai-ready-graduates

Brew, M., Taylor, S., Lam, R., Havemann, L. and Nerantzi, C. 2023. Towards developing AI literacy: Three student provocations on AI in higher education, Special issue: Generative AI and implications for open online and distance education. Asian Journal of Distance Education, 18(2), 1-11. http://www.asianjde.com/ojs/index.php/AsianJDE/article/view/726

Dawson, P., Bearman, M., Dollinger, M. and Boud, D. 2024. Validity matters more than cheating. Assessment & Evaluation in Higher Education, 1–12. https://doi.org/10.1080/02602938.2024.2386662

Digital Education Council 2024. Global AI student survey. Digital Education Council. Digital Education Council Global AI Student Survey 2024. https://www.digitaleducationcouncil.com/post/digital-education-council-global-ai-student-survey-2024

Dumitriuc, N. 2024. Why Universities Need to Embrace Generative AI to Ensure Future Success. Plato. https://www.plato.ac/blog/why-universities-need-to-embrace-generative-ai-to-ensure-future-success

Entwistle, N., McCune, V. and Walker, P. 2001. Conceptions, styles and approaches within higher education: analytic abstractions and everyday experience. In:  Sternberg, R.J. and Zhang, L-F. (Eds.) Perspectives on Thinking, Learning and Cognitive Styles, London, Lawrence Erlbaum.

Gorichanaz, T. 2023. Accused: How students respond to allegations of using ChatGPT on assessments. Learning: Research and Practice. 9(2), 183-196.  https://doi.org/10.1080/23735082.2023.2254787

Hirabayashi, S., Jain, R., Jurkovic, N. and Wu, G. 2024. Harvard undergraduate survey on generative AI. An inaugural report commissioned by the Harvard Undergraduate Association. 8 August 2024. https://arxiv.org/pdf/2406.00833

Nerantzi, C. forthcoming. Educator appears perplexed with student’s statement in assignment that GenAI was used to support their learning. LoveLD, Magazine issue 6, October 2024, Association of Learning Developers in HE

Nerantzi, C., Abegglen, S., Karatsiori, M. and Martinez-Arboleda, A. (Eds.) 2023. 101 creative ideas to use AI in Education. #creativeHE. https://doi.org/10.5281/zenodo.8072949

Säljö, R. 1979. Learning about learning. Higher Education. 8(4), 443–51.

Shi, H., Zhou, Y., Dennen, V.P. and Hur, J. 2024. From unsuccessful to successful learning: profiling behavior patterns and student clusters in Massive Open Online Courses. Education and Information Technology. 29, 5509–5540. https://doi.org/10.1007/s10639-023-12010-1

Taylor, S., Brew, M., Lam, R., Havemann, L. and Nerantzi, C. 2023. AI in Higher Ed. A student-led panel discussion organised by the School of Education, University of Leeds and University College London, 8 March 23, online

Tumodóttir, A. 2024. Questions for consideration on AI & the Commons. 4 July 2024, Creative Commons. https://creativecommons.org/2024/07/24/preferencesignals/

Utochkin, D. 2024. Cut the ‘AI’ bullshit, UCPH. University Post, University of Copenhagen. 12 August 2024. https://uniavisen.dk/en/cut-the-ai-bullshit-ucph/

Weller, M. 2024. The darkish side of open licences. The Ed Techie. 23 August 2024. https://blog.edtechie.net/open-access/the-darkish-side-of-open-licences/

Williamson, B. 2024. AI in education is a public problem. code acts in education. 22 February 2024. https://codeactsineducation.wordpress.com/2024/02/22/ai-in-education-is-a-public-problem/