SRHE Blog

The Society for Research into Higher Education

Tag Archives: AI


by Leave a comment

Enhancing student learning through innovative scholarship

by Jenny Burnham, Matt Mears, Kennedy Offor, Jane Woodin and Tim Herrick

This blog is a report on the Enhancing Student Learning Through Innovative Scholarship (ESLTIS) conference held at the University of Sheffield in July 2026. The conference involved participants from 24 different universities and nearly 50 sessions.

When higher education faces real pressure, coming together to reimagine how we teach and learn is more than just professional development – it is a vital act of community. In July 2026, 97 educators, researchers, and student partners gathered at the University of Sheffield for two energising days at the Enhancing Student Learning Through Innovative Scholarship (ESLTIS) conference.

Across two inspiring keynotes, 39 talks and oral bites, and seven hands-on workshops, the atmosphere was filled with practical optimism, bold experimentation, and genuine connection. At a time when financial constraints weigh heavily across the sector, ESLTIS 2026 proved that an exceptional, high-impact academic gathering does not have to cost the earth. By keeping delegate fees as low as possible without compromising on quality, the conference demonstrated that transformative scholarship can remain accessible, inclusive, and deeply rewarding for everyone involved.

To capture the spirit of those two days, this blog brings together the key ideas, personal reflections, and practical insights that defined the event. We explore the overarching themes shaping the sector, followed by firsthand perspectives from our keynote speaker, insights from a session chair navigating AI in the curriculum, and the invaluable contributions of our student partners.

Key Conference Themes

With the tone of ‘Dealing with Disruption’ set by Professor Sam Nolan, the conference explored how learning and teaching in higher education is responding to and being shaped by the outside influences bringing both adversity and opportunity into the sector. A recurrent theme of innovations drawing on AI technology encompassed support for diverse student needs, its use in assessment, and as a teaching aid. The conference mirrored the acknowledgement of the evolving higher education demographic, with discussions covering inclusivity, decolonisation, engagement, identity, and self-worth, with structured gap-filling support, and breaking down of participation barriers, busting out of disciplinary silos to support students’ belonging and wellbeing. Wide-ranging talks covered: curriculum both hidden and carefully designed, strategies to augment students’ development of their skills and their readiness for their post-graduation world, and experiences of collaboration with and between students, staff, and professional services teams

Underpinning it all was the continuing ethos of the ESLTIS community that recognises and celebrates the education-focus of academia, exploring this with talks and workshops on academic identity – how it grows and develops, the things that shape it, and how it can be described and evidenced. The conference was not one thing for all delegates, it had something for everyone; areas with which we are familiar and aspects provoking thought and reflection, from which we could learn.  Yet again, I have attended a conference and come away with more than I expected. 

Reflections from a keynote speaker 

Dr Jenny Burnham is a Senior University Teacher in Chemistry at the University of Sheffield. Here she writes about her career and how she shared her experience at the conference.

I am interested in everything and my career highlights have been similarly eclectic. Rather than focus on a single speciality, I used myself as the common thread to weave a story that would be useful, relevant, or inspirational for a diverse audience of enthusiastic and accomplished educators. Taking inspiration from the conference name and conference themes (academic identity, community in learning and teaching, SoTL, supporting student success, and learning and teaching possibilities), my keynote made a virtue of the meanders of my academic career so far.  

My academic journey started in chemistry, shaping my relationship with information (evidence is key) and placing the primacy of experimental investigation at the centre of my disciplinary awareness. My early scholarship interests grew from this, exploring the development of inquiry skills through laboratory education. Spending a lot of time teaching students in the lab and enjoying their individuality shaped a desire to see them all succeed. This developed into an education philosophy that students are clever and capable with enormous potential, which permeates my work. An MEd drawing on what can be learned from sharing practice and insight (Rowland, 1999) transformed my approach to education. The benefits I felt in sharing my work with people who were interested motivated me to extend my networks. These have been a source of inspiration, opportunity and support, and allowed me to define my leadership identity; supporting, facilitating, promoting, and enabling collegial interaction and growth within my community.

My keynote aimed to provide a little of something for everyone, and evidence that I succeeded in this is captured in three different post-keynote interactions. I was thanked for owning my nerves: “it helps” she said. I had lunch with a PhD chemist considering an academic career; and I received an enquiry about a Learning Landscapes initiatives I worked on. There was something for me too: a final, unexpected result of my talk was that pulling it together reminded me of the cool stuff I have done.  Celebrating successes is definitely something I intend to do more of.  

Reflections from a session chair 

Dr Kennedy Offor was a member of the local organising committee and reflects on his experiences at ESLTIS in the role of session chair.

Chairing the ‘AI in the Curriculum’ session gave me with a unique perspective on how colleagues from different disciplines are responding to the rapid emergence of generative AI in higher education. Although the eight presentations ranged from institutional policy to engineering, mathematics, business, humanities, and language learning, a common thread emerged throughout the session – how learning, assessment, and curriculum design should evolve to ensure that students continue to develop critical judgement, disciplinary understanding, and professional responsibility. The discussion reflected a noticeable shift from reacting to AI as a disruptive technology towards designing learning environments that make purposeful and educationally meaningful use of it.

Laurie Wilson opened the session by introducing the University of Sheffield’s developing ‘Common Approach to AI in the Curriculum’, emphasising “harmonisation rather than homogenisation” and arguing that AI should support good disciplinary teaching rather than drive it. Joanne Irving-Walton then offered a thought-provoking perspective by examining the “gaps” where judgement develops, suggesting that AI increasingly participates in the uncertainty, rehearsal and reflection through which students and educators learn. 

Panos Kloukinas, Nick McCullen and Will Roberts built on this by arguing that mathematics and computation education should increasingly prioritise application, interpretation and problem formulation over procedural execution, while Dawn Whitton demonstrated how students could critically evaluate AI itself through a staged business assessment that required them not only to use AI, but also to critique its assumptions, biases and wider societal implications.

The engineering perspective was represented through our own presentation, which used longitudinal Graduate Teaching Assistant reflections to show how observations of student uncertainty, teamwork and AI use informed iterative redesign of an MSc engineering design module. Mireilla Bikanga Ada’s graduate study provided an interesting counterpoint by showing that, despite widespread AI adoption, teamwork remained the strongest predictor of graduates’ perceptions of degree relevance. 

Rachel Johnson illustrated how a simple, low-stakes classroom activity could encourage students critically to examine AI-generated information and its biases, while Franziska Collier demonstrated how AI avatars could provide authentic, low-risk opportunities for students to practise high-stakes language assessments.

The discussions reinforced my own view that the key educational question is no longer whether students should use AI, but how curricula can help them develop the professional judgement needed to use it responsibly. Rather than diminishing the role of educators, the session demonstrated that AI makes thoughtful curriculum design, assessment, and disciplinary expertise more important than ever, and highlighted an encouraging direction of travel for the Scholarship of Teaching and Learning community.

Students as Partners 

Dr. Jane Woodin was a member of the local organising committee and offers her thoughts on working with students as volunteer partners 

The conference ethos of inclusive dialogue was also evident in the organisation of student volunteers who were invited to participate in whatever they wished, from visiting and advising on the venue beforehand, supporting registration and general enquiries, and co-chairing sessions together with a staff volunteer. Responses from the eight volunteers ranged from tentative interest in helping out to requests to ‘have a go at everything’, and others who did the minimum in terms of active volunteering because they found the topics of the talk so interesting and important.

It was heartening to hear one volunteer offer to sit at the front of the large auditorium to oversee a presentation session, handling the Q &A masterfully, including interventions from the co-chair. Another volunteer started out shyly offering to help at the registration desk, but seeing fellow students taking the chairing role, threw herself into it enthusiastically. 

These experiences remind us how the processes surrounding education are as important as the education itself – or perhaps are or should be inseparable, as the ‘way that you do it’ melds seamlessly into ‘what you do’; offering an example of SoTL in action at its best.  

Final Reflections

Dr Matt Mears is a Senior University Teacher in Physics at the University of Sheffield, and was chair of the Local Organising Committee for ESLTIS2026

As Conference Chair, reflecting on the energy across those two days fills me with immense pride and a distinct sense of gratitude. While hosting an event of this scale undoubtedly takes work, the dividend it pays back to the host team and the wider community is extraordinary. Watching colleagues reconnect, draw genuine inspiration from such a rich variety of presenters, and spark new collaborations in real time reminded me why these gatherings are so vital for our sector.

If you have ever toyed with the idea of organising a conference, or perhaps bringing ESLTIS to your own institution, I cannot encourage you enough to take the leap. The effort involved is far outweighed by the connection, camaraderie, and fresh momentum it generates. Whether as an attender, a presenter, or a future host, I look forward to seeing where our community goes next.


Dr Jenny Burnham is a chemistry university teaching specialist with 20 years’ experience teaching and learning about education at the University of Sheffield. 

Dr Matt Mears is a Senior University Teacher in Physics in the School of Mathematical and Physical Sciences at the University of Sheffield. His research focuses on inclusive education practices and laboratory-based teaching.

Dr Kennedy Offor is an engineering educator and researcher with experience teaching undergraduate and postgraduate engineering at the University of Sheffield’s School of Electrical and Electronic Engineering and, more recently, apprenticeship and technical engineering education at the AMRC Training Centre.  

Dr. Jane Woodin is a Senior University Teacher in the School of Languages, Arts and Societies at the University of Sheffield. She has taught and researched in intercultural communication, focusing on peer interaction and dialogue approaches.

Dr Tim Herrick is a Senior University Teacher in the School of Education, and a National Teaching Fellow. He was part of the ESLTIS conference committee and helped co-ordinate this blogpost.

Reference Rowland, S (1999) ‘The role of theory in a pedagogical model for lecturers in higher education’ Studies in Higher Education, 24(3), 303–314. https://doi.org/10.1080/03075079912331379915


by Leave a comment

When assessment goes local, who keeps what students bring back?

by Carlene Kyeremeh

A student sent me a photograph of a handwritten waybill. Dated November 2022, it records a unit price for printed sachet film. It also carries the names of a supplier and a customer and the given name of the person who issued it. The waybill appeared in a term paper on fiscal consolidation as evidence of a cost increase that no statistical series I have found records.

It is in my files. Not the department’s. Mine.

I describe the waybill here but do not reproduce it, identify those named, or give the figure it records. The students who wrote the paper agreed to this account. I have no record that the supplier, the customer or the person named agreed to the document’s retention or wider use, so I have minimised the identifying detail, including the price, which is the most commercially sensitive thing on the page.

I have argued elsewhere that the curriculum can help to thicken a thin documentary record. Ask each student to obtain one item of locally sourced evidence: an interview, a quoted price, a document from a firm or public office, and, over successive cohorts, a course can assemble material that no database previously held. This is no longer only my ambition. Programmes are already redesigning assessment in response to generative AI, and one attractive move is to require evidence a model cannot retrieve: a price obtained in a market, an interview conducted locally, a document held by a firm or public office. That redesign brings third-party material into course files at scale, often without a parallel decision about what should happen to it after assessment.

That is the promise. The waybill in my files is the problem that follows it.

What universities already know how to govern

Research ethics and commercial confidentiality are not routinely ignored in student work. Professionally oriented programmes have long had to decide what an organisation may disclose, who may see it and when it must be returned. For example, McMaster University’s course-based research guidance explicitly covers undergraduate, MBA and graduate courses. The University of Toronto provides both a course-based ethics route and a classroom confidentiality agreement for third-party proprietary information. In the UK, the University of Manchester’s ethics guidance asks researchers to establish first whether an activity is research training conducted as coursework, described as collecting and analysing a small amount of data to gain experience in method, or research proper. These arrangements show that collection, consent, confidentiality and use within a defined project can be governed. My argument is therefore not that universities lack such protections. It begins where a normal project would end.

The problem begins when the course ends

If material is returned, destroyed or retained only with the assessed work, its lifecycle has a clear end. But the design I am proposing depends on the evidence outliving the assignment. A document collected by one student in 2026 should be available to another cohort in 2028 or 2029, alongside prices, interviews and documents gathered in the intervening years. The course is not only assessing students. It is building a local source collection.

That changes the questions. What exactly does the university hold after the assessment period? Who is its custodian? How long may it be retained? May another cohort consult it? What fresh permission is needed if it is later used for research or publication? These procedures can help answer those questions, but not automatically. An agreement that properly confines information to one course may rule out the very reuse on which the proposed archive depends.

The evidence is not the assignment

The student’s analysis is coursework. The material used to produce it is not necessarily the same thing. A waybill may contain personal or commercially sensitive information. Interview notes are harder still: they are neither the source’s own record nor the student’s analysis, but one person’s written account of what another person said, made for an assignment and now held by a lecturer.

Where the activity is classified as teaching rather than research, this third-party material can still raise a lifecycle question: do institutional rules cover its custody, retention and reuse after the assessment for which it was collected? The Manchester distinction is instructive here. A course collection built to outlive its cohorts is not a small amount of data gathered to practise a method, but neither does it fit neatly within research as that definition intends. It is a lifecycle problem created by a particular curricular ambition.

The reuse is the point

Ghana’s Data Protection Act, 2012 (Act 843) requires personal data to be collected for a specific, explicitly defined and lawful purpose, limits retention and requires further processing to be compatible with the original purpose. Material obtained for one student’s term paper and later made available to another cohort may involve further processing. Nothing in the waybill I received establishes permission for it to become part of a standing course collection.

The Act contains provisions for historical, statistical and research uses where the relevant conditions are met. Those provisions offer a governed route, not a retrospective blank cheque. They do not turn a lecturer’s folder into an institutional archive. The underlying principle is not unique to Ghana: UK guidance likewise treats purpose, storage and research-related reuse as connected but distinct questions.

Ordinary assignment governance is adequate for ordinary assignments. The problem appears when coursework is asked to do something additional: to enlarge a locally available documentary record over time.

The archival value and the governance risk arise from the same feature. An archive that cannot be consulted next year is not an archive. A collection that can be consulted next year without the source having been told is not a defensible one.

A small institutional answer

The answer need not be a new bureaucracy. It could begin with a short institutional instrument that separates use in a student’s assessment from retention in a course source collection. Before collection begins, it should state the teaching purpose, what may be collected, who may later consult it, how long it will be kept, which unit is responsible and what choices the source has. Research use or publication should require a separate decision rather than being hidden inside the phrase ‘educational purposes’.

The remaining requirement is custody. What one lecturer can require, only an institution can keep. A named custodian, controlled access, a retention and disposal schedule, and a record of what the course has produced would convert a personal folder into an accountable institutional practice. Existing MBA, ethics and confidentiality processes provide models; the task is to extend their discipline to the intended afterlife of the evidence.

Proportionality matters. If the process becomes heavier than the assessment it supports, colleagues may stop asking students to gather local evidence. But a department unable to provide a clear collection notice, a retention rule and somebody accountable for what returns should not require the fieldwork. Otherwise, institutional risk is transferred to students and to the businesses that agree to speak with them.

Readers can test this in ten minutes. Take your own institution’s course-based ethics route or classroom confidentiality agreement and look for two things: whether either says anything about third-party material after the assessment period, and who is named as responsible for it. If neither appears, ask where else the institution governs the evidence after the assignment ends. If there is no answer, the arrangements may govern the assignment without governing its evidentiary afterlife.

Before we ask for more

The curriculum can enlarge the archive. The qualification is that the archival ambition creates an additional governance obligation. The question is not only whether the evidence was collected ethically and kept confidentially. It is whether its intended custody, retention and reuse were made explicit before collection began.

I have a waybill in my files and no good institutional answer to those questions. That is not an argument against asking students to go and find out. It is an argument for deciding what we will do with what comes back, before we ask them to go.

Dr Carlene Kyeremeh is Associate Professor and Vice President, University Advancement, Recruitment and Research at All Nations University, Koforidua, Ghana. Her research addresses curriculum decolonisation, higher education policy in Africa and the Caribbean, and the political economy of educational technology.This post develops a governance question raised in Issues 14 and 15 of The Decolonized Curriculum, her LinkedIn newsletter.


by Leave a comment

Learning spaces, educational technology and the future of the physical learning environment

by James Rutherford

To look into the near future, it would be sensible to start by avoiding this typical question: ‘What technology do we need to install in these rooms?’ We would be wise to ask instead: ‘What do we want our education to look like?’ Future learning spaces and associated educational technologies require an approach that supports teaching and student collaboration, with accessibility features and environmental adaptations embedded within the design of learning spaces rather than delivered through standalone audio-visual equipment.

Research on learning spaces highlights the importance of aligning technology decisions with pedagogy and intended learning activities, rather than allowing the space and in-class technology to dictate educational practice. (Radcliffe et al, 2008) This principle resonates strongly with my experience working in HE learning environments, where the most successful spaces are those that enable effective communication and participation for all learners. The technical and physical space itself must therefore support diverse forms of engagement, ensuring that staff can teach effectively and that students can participate fully through listening, discussion, collaboration, practice and observation. (Jamieson, 2003)

Audiovisual technology

At City St George’s, University of London, the educational technology allows staff to enter any learning space, connect a laptop, teach, collaborate, record or stream sessions with minimal effort. This has largely been achieved through the learning spaces project (Designing Active Learning Initiative (DALI), 2026) and subsequent upgrade work, with room controls and user experiences standardised across the institution to reduce complexity and support requirements.

Design for learning and teaching

However, this can mean that technical decisions are made before the educational intent has been planned and shared with all stakeholders. It is advisable that educational technology supports the intended teaching activities rather than determines them, with the educational strategy and signature pedagogy of the university driving the learning environment design (Rudd, 2006; Hasper, 2024). Spaces ought to be designed around the learning and teaching needs of an institution, making it straightforward for staff to teach effectively and for students to engage successfully.

Environmental prerequisites

In many learning environments, audio quality often receives less attention than display technology, but it is obvious that students need to hear clearly, be able to participate in discussions and access lecture capture recordings. It is increasingly recognised as a key contributor to students’ learning outcomes, and there is evidence that speech intelligibility and the control of background noise are fundamental requirements for effective learning. This is because the design of each space directly influences effective communication. (Acoustic Design for Schools, DFE, 2015) This includes unwanted noise from corridors, neighbouring rooms, ventilation and external sources, as extraneous noise can be a real hindrance to learning and teaching capabilities.

Therefore, acoustic treatment needs to be considered as fundamental, now and in the future, not least with accessibility at the forefront of our minds. Universities can often improve existing spaces through acoustic treatment and interventions designed to reduce reverberation, although the feasibility and cost will vary by building.

Controllable lighting, appropriately designed and ergonomic furniture, thermal comfort and excellent sightlines all contribute to an effective learning experience. Classroom furniture is now expected to be comfortable, robust and appropriate for the type of teaching taking place. Spaces need to accommodate different body sizes, the requirements of student movement, with diversity and accessibility alongside different timetabled patterns of use and perhaps involving longer teaching sessions.

Accessibility is fundamental

Future learning spaces will need to be designed with diverse learners in mind, consistent with the principles of Universal Design for Learning (Universal Design for Learning｜CAST, 2026). Spaces are required to include captioning, recordings, remote participation and flexible methods of seating and student engagement. Across HE, accessibility is increasingly being incorporated into space planning from the outset through features such as captions, accessible controls, suitable furniture for students with access needs, clear audio and barrier-free access. Real inclusion is developing into a measure of quality rather than an afterthought.

Design for hybrid delivery

Hybrid delivery could be seamlessly integrated into learning space design while ensuring that the on-campus experience remains the priority. Spaces need to support clear sound and appropriately intelligent camera angles. Video conferencing tools permit content sharing, recording and online participation as standard, but hybrid teaching must not feel like an add-on; rather, it should be a flexible and inclusive addition to the learning approach (Beatty, 2007; Melcher et al, 2025). In hybrid spaces, students need to hear the lecturer, hear questions from the room, follow discussions and revisit recordings afterwards.Consequently, excellent microphones, clear loudspeakers, robust voice support, as well as captions and assistive listening are expected to be core requirements.

AI tools

AI is likely to support areas such as transcription, captioning, camera tracking controls, fault detection and remote monitoring for the service teams. Its role would be to improve accessibility, as well as provide data analysis of how the space is performing environmentally and reduce workload rather than add complexity. Such developments must, however, consider how AI tools relate to privacy, ethics, and data governance, which will remain important factors in their implementation.

Support and maintenance

Learning spaces will become easier to monitor, maintain and update through centralised management tools that do exist, but with emerging AI capabilities that could enhance these processes through predictive maintenance, intelligent system monitoring, automated troubleshooting, and data-driven insights. Support teams would then be able to identify and resolve many issues remotely, improving reliability and assisting with planning of new environments.

The transformation of accessibility and inclusion

Accessibility is no longer regarded purely as a matter of compliance, but as an essential component of the learning environment. Future learning spaces ought to be planned and designed on the basis that:

  • some students will need captions
  • some students will rely heavily on lecture capture recordings
  • some students will be neurodivergent
  • some students will join remotely
  • some students will be working in noisy or shared environments
  • some students will need language support

This is key because it changes accessibility from a ‘special provision’ into part of the normal learning experience, in particular the essential role that lecture recordings play for neurodivergent and disabled students (Horlin, Hronska and Nordmann, 2024).

The AI-supported room

AI technology is likely to be adopted within learning spaces with support for:

  • automatic camera framing
  • lecturer tracking
  • capture of student questions
  • reduction of background noise in lecture capture
  • live transcription
  • searchable lecture capture recordings
  • automatic chapters in those recordings
  • summaries of sessions
  • early warning of equipment faults
  • better support information for technical teams
  • information about how rooms are being used

There are obviously numerous risks with adopting AI: too much monitoring; weak governance; data protection and privacy concerns; intellectual property questions; and technology that looks clever but does not help teaching. However, as two recent literature reviews illustrate, the opportunity that AI tools can bring is undoubtedly significant, with the potential for less pressure on academics to operate the technology in a classroom and by offering better access for students (Crompton and Burke, 2023; Adamakis and Rachiotis, 2025).

Conclusion

If the last twenty years were defined by the digitisation of teaching, the next decade could be defined by the redesign of learning itself. Will higher education across the UK embrace this opportunity? Can it afford not to in the current climate of financial uncertainty and geopolitical vulnerability? Accessibility, hybrid participation, intelligent systems and engaging learning environments can no longer be regarded as enhancements, as they will surely become the standard to meet the expectations of students and staff. The institutions that thrive will not be those with the most technology, but those that use it most purposefully. The future of learning spaces is not about what technology we install; it is about what we enable.

James Rutherford is a Senior Educational Technologist at City St George’s, University of London, specialising in learning spaces, educational technology, and hybrid delivery. With over 35 years’ experience in higher education, he has worked as a learning space designer, AV project manager, and video producer, leading initiatives that support innovative and inclusive learning and teaching. James holds a Master’s in Advanced Educational Practice from the UCL Institute of Education and is a Fellow of Advance HE.


by Leave a comment

Vibe-coding and co-creation: human-AI collaboration in higher education

by Andrew Williams

The debate around artificial intelligence (AI) in higher education has largely focused on students. Are they using ChatGPT in assessments? How do we maintain academic integrity? How should teachers and curricula adapt?

Much less attention has been given to academics working with generative AI (GenAI) as a creative collaborator or co-pilot, rather than merely attempting to mitigate its impact. The potential for productive and effective AI collaboration deserves greater consideration because it challenges longstanding assumptions about expertise, creativity and educational design.

Ever since the release of ChatGPT in November 2022, I have been fascinated by how intuitive the interaction with GenAI tools can be and the ease with which ideas can be explored through ordinary conversation. GenAI is a sophisticated technology that allows users to test possibilities, refine outputs, and rapidly turn preliminary concepts into usable material.

My background is in biological sciences, with no grounding or expertise in computer science, software engineering or coding. For me, this technology has helped me overcome technical barriers and allowed me to explore new ways to teach and to develop bespoke educational resources. I envisioned a learning experience for my undergraduate medical science students (studying human genetics and evolution) that moved beyond static textbook diagrams and dry explanations to a more interactive and visually stimulating format. Realising that vision required a simulation that let students tweak parameters, observe real‑time population dynamics, and grasp complex evolutionary ideas.

Traditionally, such a project would rely on grants, external software developers, or months of self‑taught coding. For most of us, the gap between an innovative pedagogical idea and a digital reality was confined by technical expertise. Today, through iterative collaboration with GenAI, natural language can function as a design interface, allowing educators to transform ideas into working educational tools.

I was also intrigued by the ever improving, and often publicised, capabilities of GenAI tools as coding agents. With no coding experience myself, I decided to try my hand at ‘vibe-coding’.

Natural language as a design interface

‘Vibe-coding’ is an emerging term within technology communities that describes a process where functional intent is communicated via natural language prompts, allowing AI to generate and refine code. This allows the user to focus on what the tool should do rather than how it must be coded. Essentially, ‘vibe-coding’ involves directing the AI to write and co-develop code aligned with your vision and pedagogical objectives. In other words – no code, no problem!

My recent research describes an iterative prompting strategy and the outcomes of a human-AI collaboration to create an interactive simulation that models certain aspects of evolution (Williams, 2026a). Instead of learning JavaScript or working with software developers, I co-created an application with ChatGPT using only natural language prompting. The result was a browser-based simulation capable of supporting inquiry-based learning and formative assessment in undergraduate education.

Rather than blindly generating content based on loosely defined instructions, the human-AI collaboration involved a structured iterative prompting strategy, careful testing of AI-generated output and re-prompting until the desired aims of the project were met. This enabled the creation of an interactive simulation in which students could investigate biological systems, manipulate variables, export datasets and construct evidence-based explanations (figure 1). This allowed students to move beyond consuming information, to generating and interpreting data themselves in real-time.

Figure 1. Snapshot of ‘Survival Island’ – an interactive simulation.

The Illusion of automated expertise

There is a prevailing narrative that AI democratises expertise, in that it allows anyone to do anything more quickly and with less effort. My experience of ‘vibe-coding’ suggested something different. The  credibility of the final simulation depended on actively embedding disciplinary expertise directly into the co-design process.

While the AI had removed the technical barrier (the coding), it had increased the importance of my academic expertise and oversight. The AI could generate a thousand lines of code in seconds, but it couldn’t tell me if the code represented a plausible scientific simulation. My epistemic judgement was required to verify the AI-generated output. The final student-ready simulation required multiple rounds of iteration, testing and re-prompting, until the collaboration achieved a credible and biologically grounded application suitable for teaching.

Rather than replacing my expertise, ChatGPT provided capabilities I do not possess: writing HTML; JavaScript; debugging code. I contributed disciplinary knowledge, pedagogical intent and continuous evaluation. Every feature reflected educational decisions rather than technical ones. The simulation succeeded because the educational need was clearly defined before AI generated the code.

Higher education has often treated GenAI as an automated assistant or as a problem to be mitigated. Increasingly, it may be more productive to think of AI as a collaborator whose contributions depend upon human judgement. Expertise remains indispensable, as humans ultimately remain responsible for judging whether outputs are educationally, scientifically and ethically sound.

Beyond the prompt: the AI literacy gap

Universities have understandably invested considerable effort in developing student AI literacy. This project led me to a broader appreciation of the need to promote teacher AI literacy across higher education, a critical area often overlooked.

AI literacy for teachers is not just about knowing how to write a good prompt or choosing the best online GenAI tool. Rather, it is a multi-dimensional competency that considers foundational knowledge (prompting, tool use), ethical, emotional and technical proficiencies. For educators to co-create teaching resources in collaboration with AI, for example through vibe-coding, more than technical curiosity is required.

The iterative ‘vibe-coding’ process I followed mirrors the ‘critical evaluation’, ‘epistemic’ and ‘innovation’ dimensions of an integrated AI literacy framework for higher education teachers (Williams, 2026b). This recently published conceptual framework comprises eight core AI literacy dimensions, which you can access in full using the link.

In the current context, the important  AI literacy dimensions include:

  • Critical evaluation: the ability to treat the AI model as a “black box” and rigorously verify outputs against authoritative sources.
  • Epistemic authority: a clear understanding of the probabilistic nature of GenAI and the domain-specific expertise to judge AI output, thereby retaining human agency during interactions with AI.
  • Innovation mindset: the willingness to move from being a consumer of tools to a co-creator of bespoke learning environments, including an understanding of how to collaborate effectively with GenAI.

These are becoming essential academic capabilities rather than optional technical skills. This transition presents considerable challenges, as academic practice has historically been predicated on either individual resource creation or the outsourcing of technical development to third parties. GenAI now enables academics to become directors of increasingly sophisticated creative processes. They define the educational challenge, AI performs the technical implementation, and academics evaluate and verify the final outcome.

As technical production becomes easier with GenAI, educational judgement becomes increasingly important. Our value is no longer found in the technical labour of production, but in our ability to define the educational challenge, critique the iterations, and curate the final outcome. The “vibe” we are coding is our pedagogical intent.

Key takeaways

  • AI is a tool, not a replacement.
  • Human-AI collaboration enables pedagogical creativity.
  • ‘Vibe-coding’ uses AI’s coding capabilities but demands epistemic oversight.
  • Teacher AI literacy is multi‑dimensional.
  • Institutional support is required to build that AI literacy.

GenAI is redefining the landscape of higher education, not through incremental improvements to traditional teaching materials, but by enabling entirely new educational content to be created. However, despite the promises of this new technological revolution, we still need robust evidence about student learning gains, engagement, accessibility and the sustainability of AI-generated educational resources.

It is easy to provide AI tools for HE educators to use. The challenge is to give educators the technical, critical and ethical literacy required to use them effectively.

Andrew Williams is a Professor (Teaching) in the Faculty of Medical Sciences at UCL. He teaches immunology and cell and molecular biology to undergraduate and postgraduate students. His research aims to explore the impact of artificial intelligence (AI) on assessment and learning in higher education and the ways in which we can promote student and staff AI literacy.


by 1 Comment

A narrator Is not a witness: what AI got right — and wrong — across three Ghanaian MBA classrooms

by Carlene Kyeremeh

It was late, and I was building slides on fiscal policy for a class of twenty, most of whom had never taken economics before. I asked an AI tool for a worked example of expansionary policy — something concrete enough to anchor the mechanism. It answered in seconds: a government cuts taxes and raises infrastructure spending; roads get built; contractors hire; demand rises. Clean. Usable. American. The mechanism travelled, but the roads were interstate highways, and the government spending the money was not one my students would ever petition. The example had arrived so quickly, and so confidently, that I nearly pasted it in. For a moment, convenience almost became curriculum.

Then, in the same session, the same tool did something I could not have done as well alone. I asked it to scaffold the vocabulary — a glossary that assumed nothing and met students who did not yet have the words. It was patient in a way I am not always patient at that hour. It found analogies; it sequenced the terms so that each rested on the one before. That slide was better for the machine’s help.

Within a single sitting, the tool had lowered the barrier for my students and quietly imported the exact default my work exists to help people notice. Across three MBA courses this term, I began to see the same pattern: the tool improved many aspects of how I prepared to teach, and became least dependable where the teaching required local knowledge, current evidence or an institution named correctly. The help landed on form. The failures landed on specificity.

In my international trade finance course, I extend a running case — Adom Naturals Ltd, a Kumasi agribusiness I built so that trade finance stops being abstract and acquires a location, a product and a currency exposure. I asked the tool to situate the firm against current conditions: what the African Continental Free Trade Area (AfCFTA) changes for a business like this, and where COCOBOD now sits in the picture. It answered in assured paragraphs and told the familiar story: Ghana grows some of the world’s finest cocoa, ships much of it out with limited processing, and watches the greater share of value accrue downstream. Fluent, orderly, and a season out of date.

The confident narrator

It missed what I happened to be holding in a government source that week: in February 2026, Cabinet directed that, from the 2026/27 crop season, a minimum of 50 per cent of Ghana’s cocoa beans should be processed locally. For Adom Naturals, that reform changes the opportunity set. Cocoa liquor, butter, cake and other processed products can retain more value within Ghana and may qualify for preferential treatment in African markets where the relevant AfCFTA rules of origin and tariff requirements are satisfied. The tool had narrated the extractive arrangement in the present tense and missed the policy intended to change it.

Nothing marked the claim as stale. The tool warns that it can make mistakes, in a line printed beneath every answer, but a caveat attached equally to everything is not calibration; it is the absence of it, dressed as candour. A colleague says: I am sure of this; check me on that. The tool says it might be wrong about anything, then says everything in the same even voice. The danger was never that it made mistakes. Every source makes mistakes. The danger was that it sounded exactly as certain when it was wrong as when it was right.

The wrong institution

Financial regulation showed me the problem from another angle. Ask a general-purpose tool about capital adequacy, disclosure or market conduct and it is fluent, because the published record is thick with Basel standards, US and UK regimes, and decades of commentary. Ask it to route the same questions through Ghana’s regulatory architecture and the fluency thins.

When I asked which body supervises an insurer in Ghana, and then which oversees a securities offering, it reached both times, confidently, for the Bank of Ghana. The answer was plausible because the central bank is prominent in Ghana’s financial system. It was nevertheless wrong. Insurance supervision belongs to the National Insurance Commission under the Insurance Act, 2021; securities-market regulation belongs to the Securities and Exchange Commission under the Securities Industry Act, 2016, as amended. When I named the specific commissions, the tool corrected itself at once. The information was retrievable; it was not the default.

Three defaults, one voice

Set the three moments side by side and a more complicated pattern emerges. The fiscal-policy example exposed a geographical default; the cocoa case, a temporal one; and the regulation case, an institutional one. These were different failures, but they arrived in the same confident voice.

I cannot inspect the tool’s training archive, so I cannot attribute every error to missing data alone. A stale policy claim may reflect a knowledge cut-off or the absence of live search. A regulatory error may reflect weak retrieval, poor weighting or the greater prominence of a general institution over a specialised one. What I can observe is an asymmetry of retrieval: general and North Atlantic formulations arrived unprompted, while Ghanaian specificity had to be named, sourced and verified into view.

That asymmetry belongs in the larger conversation about AI and epistemic justice — about whose knowledge is dense enough, accessible enough and prominent enough to be retrieved fluently, and whose is thin enough to be flattened, displaced or missed. The tool did not invent the hierarchy of whose knowledge counts. It inherited a record shaped by that hierarchy and can reproduce it at scale, in fluent prose. Better models may reduce some errors, but model improvement alone cannot repair knowledge that remains absent, inaccessible or systematically underrepresented.

I develop that argument more formally elsewhere, in work currently under review. Here, I want only to report what it looks like from inside three classrooms, at the point where defaults become examples and examples become curriculum.

Verification is the work

I use these tools daily and they earn their place, so let me be honest about the difficulty. The answer is not refusal; refusing the help is not a decolonial act, only less help. The answer is the discipline I have argued for all along, now turned on the machine: no sentence enters the curriculum until it points to a source I can hold. Verification is not the friction that slows the tool down. With a tool like this, verification is the work.

I have also begun turning that work into a learning activity. I place selected AI outputs beside the relevant primary or institutional sources and ask students to identify what the model has generalised, dated or assigned to the wrong body. Verification becomes not only my quality-control procedure but part of the curriculum itself.

Perhaps that is the graduate skill this moment now asks for. Not simply how to find information, the tool is generous with information, but how to test information whose presentation gives no sign whether it has earned our trust. The scarce skill is no longer retrieval. It is discernment. Teaching has always required two kinds of expertise: explaining ideas well, and knowing where they belong. The tool is becoming remarkably good at the first; the second is still ours. It narrates beautifully — but a narrator is not a witness, and decolonising the curriculum now includes learning to interrogate the archive that speaks back.

If your tool has ever been confidently wrong about your own institution, your own regulator or your own country’s data, I would like to know what it got wrong — and whether a student would have caught it.

Dr Carlene Kyeremeh is an Associate Professor and Vice President, University Advancement, Recruitment & Research, at All Nations University, Ghana, where she teaches managerial economics and international trade and finance on the MBA programme. Her research examines decolonial curriculum reform, gender equity and academic mobility in African higher education, with the African Continental Free Trade Area as a recurring empirical anchor. She is currently researching the reintegration of diaspora-return faculty in Ghanaian universities. She writes The Decolonized Curriculum, a newsletter on curriculum decolonisation in African higher education. 

LinkedIn [https://www.linkedin.com/build-relation/newsletter-follow?entityUrn=7412983304175534080]

Author’s note: This article is adapted and substantially expanded from Issue 13 of The Decolonized Curriculum.


by 3 Comments

When papers become currency

by Carolina Guzmán Valenzuela

Over the last few months, I have found myself discussing academic publishing with young researchers in Germany and Chile. What struck me in both places was not excitement about ideas, journals or scholarly debates. It was anxiety.

At workshops on publishing strategies and scholarly journals (one at the HoFoNa Conference in Germany and another at the Institute of Education at the University of Chile) the conversations quickly moved away from writing itself and towards something far more pragmatic: survival.

Young researchers spoke about publishing in highly strategic terms. Which journals were ‘safe’? Which ones counted for postdoctoral applications and funding schemes? Which journals’ turnarounds were quick enough? Which journals offered the highest probability of acceptance?

Several participants openly discussed how they calculated publication decisions in relation to career survival. The question was often not where their work fitted best intellectually, but which journals were fast enough, prestigious enough and predictable enough to maximise their chances of securing a postdoctoral position, grant or future contract.

What I found quite revealing was how naturally many early-career academics now speak the language of optimisation. Quartiles, Article Processing Charges (APCs), turnaround times, indexing systems, impact factors and publication strategies are discussed with remarkable fluency. Many young researchers are being socialised into academia through the logic of strategic productivity before they have had the opportunity to develop a slower intellectual voice of their own.

And who can blame them?

Across many universities today, academic life has become increasingly precarious and accelerated. Temporary contracts, short-term postdoctoral positions, uncertain funding, metric-driven evaluations and intense competition have transformed publishing into something far more strategic than it was. In systems (such as Chile’s), where academic careers and funding schemes heavily depend on publications indexed in Web of Science (WoS) and Scopus, papers increasingly function as academic currency.

Under these conditions, it is not surprising that publishers promising continuous publication, high-volume output and relatively predictable editorial processes have expanded rapidly. This is one reason why publishers such as MDPI and Frontiers have become so deeply embedded within contemporary academic life. Some of their journals are indexed in WoS and Scopus, which count towards grants and promotions and contribute directly to institutional rankings and evaluation systems. In other words, they are not operating outside the university system. They are increasingly part of how the system itself functions.

It is fair to say, though, that some established journals are reporting turnaround times that are not radically different from publishers such as Frontiers or MDPI. Elements of acceleration and compressed editorial timelines are also becoming increasingly visible across the wider publishing ecosystem, suggesting that these dynamics are no longer confined to specific publishers.

In any case, average turnaround statistics do not fully capture broader differences in selectivity, publication scale, editorial oversight and peer review intensity. During my years as Coordinating Editor of Higher Education, manuscripts frequently went through several rounds of major revision over many months. Reviewer disagreement, editorial discussion and substantial intellectual reshaping were often central parts of the process. The deeper question, then, may not simply be speed itself, but whether meaningful scholarly judgement, rigorous peer review and sustained intellectual critique can realistically be maintained under conditions of industrial-scale publication.

This becomes particularly important in a publishing ecosystem increasingly organised around scale. Large editorial boards, continuous publication models and relatively low desk-rejection rates create a parallel publishing universe in which the formal conventions of academic publishing are maintained, but where critique, filtering and editorial curation risk becoming increasingly compressed and uneven.

APCs have also reshaped inequalities within academic publishing. Publishing open access in prestigious journals often depends on fees that remain inaccessible for universities and researchers in Latin America, Africa and other underfunded academic systems unless institutions are able to absorb the costs.

At the same time, substantial APCs are no longer confined to traditionally prestigious journals. Many high-volume publishing models also involve significant publication fees, suggesting that what many early-career researchers are increasingly paying for is not simply open access, but speed, predictability and reduced temporal uncertainty within an academic system governed by short-term contracts, funding deadlines and metric-based evaluations. In short, they are purchasing life chances.

The irony is difficult to ignore. The same universities and funding systems that demand constant productivity often make meaningful participation in prestigious publishing circuits governed by international rankings, indexing systems and performance metrics.

Meanwhile, reviewers, the invisible infrastructure sustaining the entire system, are visibly exhausted. Every week, I receive multiple emails asking me to review manuscripts within ten or fourteen days. Some are oddly urgent in tone, politely aggressive, reminding reviewers about editorial targets and turnaround times. Many read as though they were generated from the same automated template. At this stage, I rarely even reply. I simply delete them. What concerns me is not only the pressure itself, but how normalised this publication culture has become.

Sometimes the system reaches extremes. In Chile, recent controversies surrounding publication incentives revealed academics producing absurd numbers of WoS indexed papers while receiving substantial productivity-linked salaries and bonuses. In some cases, this meant publishing well over one hundred papers in a single year. The issue here is not simply individual behaviour. The deeper question is structural: what kind of national research funding system rewards institutions according to publication output and, in turn, transfers these pressures directly onto academics?

Artificial intelligence is likely to intensify these dynamics even further. AI did not create the culture of academic hyperproduction, but it is accelerating it dramatically. Faster writing. Faster reviewing. Faster summarising. Faster publishing. More output everywhere.  And less time for, or even interest in, thought as a result.

None of this means that traditional academic publishing represented an egalitarian system. It has long been shaped by oligopolistic publishers, exclusionary gatekeeping and profound global inequalities. Open access has unquestionably expanded the circulation of knowledge and enabled greater visibility for scholars outside elite institutions. Some forms of academic closure deserved to be challenged. But something important may also be getting lost.

During those workshops in Germany and Chile, I observed that many young researchers seemed caught between two incompatible temporalities: the time required for meaningful scholarship and the accelerated pace demanded by contemporary academic careers. That tension, more than any individual journal or publisher, surely represents one of the defining conditions of academic life today.

Carolina Guzmán-Valenzuela serves on the SRHE Governing Council. She is a Serra Húnter Fellow at the Universitat Autònoma de Barcelona (Spain) and Senior Research Fellow at the Universidad de Tarapacá (Chile). Her work focuses on higher education, epistemic justice, decolonial perspectives, and inequalities in global knowledge production, particularly in Latin America.


by Leave a comment

From ad hoc to constructive: the ABC levels of GenAI integration in business education

by Qianqian Chai and Xue Zhou

Introduction – the challenge of GenAI integration in business education

Since the release of ChatGPT, Generative Artificial Intelligence (GenAI) has rapidly entered higher education. Business schools, with their strong ties to industry and emphasis on applied skills, provide a particularly important setting for examining GenAI’s role in curriculum design. Yet, while adoption has expanded quickly, the educational outcomes of GenAI integration have not been consistent (Kurtz et al, 2024). Across cases, educators identified both benefits and risks, including engagement, skills development, and overreliance. This unevenness suggests that rather than reflecting a single trajectory of adoption, early practice appears to involve different approaches to integration. The central issue is not whether GenAI is used, but how different approaches shape outcomes.

This blog draws on our recent study of GenAI integration in business modules at a UK Russell Group university (Zhou et al, 2026). Through a qualitative analysis of 17 educator cases across 24 modules, we examined how GenAI was incorporated into curriculum design, and how different approaches were associated with distinct benefits and challenges. Using the lens of constructive alignment (Biggs, 1996), we identified three patterns of integration: ad hoc, blended, and constructive, which together form the ABC levels for understanding how GenAI is integrated into curriculum practice. We use these levels to explore why some approaches appear more educationally effective than others. In particular, this blog will offer research-informed insights into how GenAI can be integrated more effectively and sustainably in business higher education. While the cases are drawn from business education, the patterns identified and principles of constructive integration have wider relevance across disciplines where GenAI is increasingly embedded in curriculum design.

ABC levels of GenAI integration in the business curriculum

Our analysis identified three levels of GenAI integration: ad hoc, blended, and constructive. Table 1 outlines these distinctions across key dimensions.

Table 1 ABC levels of GenAI curriculum integration

Constructive integration represents a qualitatively different approach, grounded in constructive alignment, where intended learning outcomes, teaching activities, and assessment are deliberately designed to develop and evaluate students’ ability to use GenAI critically and effectively. At this level, GenAI is not an optional or supporting tool, but an integral component of disciplinary learning, with a clear pedagogical purpose and coherent role across the curriculum.

By contrast, ad hoc integration is characterised by occasional and isolated use, where GenAI is introduced as an optional or experimental tool without being planned into the broader curriculum design. Blended integration moves beyond this by incorporating GenAI into selected learning activities or tasks, giving it a more purposeful pedagogical role, but its use remains only partially embedded. Both approaches therefore fall short of the coherence and strategic alignment that define constructive integration.

The distinction between these patterns is therefore not simply a matter of more or less GenAI use, but of how GenAI is positioned within the curriculum: as an experiment, as a support, or as a capability to be deliberately developed. Although developed from business education contexts, this typology offers a lens that can be applied more broadly to understand how GenAI is positioned within different disciplinary curricula.

Why constructive integration matters

Across the cases, GenAI integration generated benefits and challenges across students, educators, and institutions. At the student level, reported benefits included stronger engagement, confidence, and employability-related skills, while the main risks centred on overreliance, inequality, ethical concerns, and ineffective outputs. For educators, benefits included efficiency gains, professional learning, and improved teaching performance, but these were accompanied by increased workload and the need to redesign activities and assessments. At the programme level, GenAI enhanced curriculum relevance but raised concerns about academic standards.

Figure 1 shows that these benefits and challenges were not distributed evenly across the three patterns of integration. Constructive integration displayed the strongest and broadest benefits, while ad hoc and blended approaches showed narrower gains alongside more exposed challenges. In other words, the issue is not whether GenAI brings value or risk, but how curriculum design shapes the balance between them.

Figure 1 Trade-offs of GenAI integration: challenges (red) vs benefits (green)

What makes constructive integration different is not the removal of challenge, but the stronger presence of educational value. In the study, constructively integrated cases were linked more clearly to student engagement, capability development, employability, and curriculum relevance because GenAI was embedded through aligned outcomes, activities, and assessment, rather than added on as a tool or support. Importantly, these cases also showed stronger educator development, including pedagogical reflection and confidence, despite workload pressures. This suggests constructive integration enhances both student outcomes and educator learning by embedding AI within coherent curriculum design.

How constructive integration is achieved

Table 2 presents examples from the modules in this study, showing how GenAI was constructively integrated into existing pedagogical strategies without requiring curriculum redesign.

Table 2 Constructive GenAI Integration into Existing Pedagogical Strategies

Taken together, the cases suggest several practical principles for integrating GenAI more coherently within the curriculum. These principles are not specific to business education, but reflect broader curriculum design considerations that can be adapted across disciplines with different pedagogical traditions.

  • Integration builds on existing pedagogical strategies: GenAI should be embedded within approaches already familiar to the discipline, such as project-based or simulation-based learning, without requiring curriculum redesign (Chugh et al, 2023).
  • Sharpen the role of GenAI by disciplinary purpose: In different contexts, GenAI supported strategic analysis, research and synthesis, reflective thinking, or data interpretation. Its value depends on alignment with module aims (Zhou & Milecka-Forrest, 2021).
  • Make AI use purposeful through assessment and evaluative tasks: In stronger cases, GenAI was connected to tasks that required students to interpret, justify, compare, or critique AI-supported outputs, rather than simply using AI to complete tasks (Biggs & Tang, 2010).
  • Support deeper student engagement through scaffolding: Structured guidance, such as prompting strategies, comparison activities, and reflective tasks, enabled more critical and purposeful use (Cukurova & Miao, 2024).

Overall, constructive integration is less about introducing new tools than about redesigning existing curriculum elements so that GenAI is meaningfully aligned with disciplinary learning.

Conclusion

The ABC levels developed in our study show that GenAI integration in business education does not follow a single trajectory but ranges from ad hoc and blended use to constructive integration. The key difference lies in approach: constructive integration embeds GenAI through aligned outcomes, activities, assessment, and scaffolding. The challenges observed across GenAI integration practices suggest an urgent shift from ad hoc GenAI integration toward strategic and constructive integration in business education. In this way, higher education can support students’ employability and capability development, strengthen educators’ professional and pedagogical confidence, and enable institutions to sustain coherent, future-facing curricula.

Dr Qianqian Chai is a Lecturer in Business and Management at Queen Mary University of London and Chair of the AI in Education Innovation Sub-committee in the School of the Arts. Her research focuses on AI in higher education, including curriculum design, academic integrity, and policy. q.chai@qmul.ac.uk

Professor Xue Zhou is a Professor in AI in Business Education and Dean of AI at the University of Leicester. Her research interests include digital literacy, digital technology adoption, cross-cultural adjustment, and online professionalism. xue.zhou@le.ac.uk


by 1 Comment

Judgement under pressure: generative AI and the emotional labour of learning

by Joanne Irving-Walton

What AI absorbs and why that matters

Most debates about generative AI in higher education fixate on what it produces: essays, summaries, answers, paraphrases. I find myself increasingly interested in something else – what it absorbs. Over the past year, as conversations about AI have threaded through seminars and tutorials, a pattern has gradually become visible. In those discussions, students rarely begin with content production; instead, they talk about how it helps them get started and steadies them enough to keep going. They use it when the blank page paralyses, when feedback stings and when uncertainty feels exposing. One student described asking AI to “make it feel possible”. Another spoke of feeding tutor comments into the system so they could be “explained more kindly”. A third reflected, almost apologetically, “I don’t want it to do my work… I just need something to push against before I say it out loud and risk looking stupid”.

In each case, AI is not replacing thinking. It is absorbing part of the emotional labour involved in it, and as that labour is redistributed, the texture of judgement shifts. Academic judgement does not tend to emerge from comfort. It develops in the stretch between knowing and not knowing, when confidence dips, stakes feel heightened, and your sense of competence is quietly tested (Barnett, 2007). Staying in that stretch long enough for thinking to clarify demands more than intellectual effort; it requires emotional steadiness, time, space and the capacity to tolerate uncertainty without rushing to resolution (Biesta, 2013). Traditionally, that steadying work has been shared across learning relationships: tutors reframing feedback, peers normalising confusion, supervisors encouraging persistence through doubt. Generative AI now occupies part of that terrain.

I do not think this is inherently a problem. For some students, it is transformative. It marks a shift in where the labour of learning takes place and that change deserves examination rather than alarm.

Four modes of engagement and emotional labour

When students talk about how they use AI, their practices tend to cluster into four overlapping orientations. These are not moral categories so much as shifts in where emotional and cognitive labour is undertaken.

Instrumental engagement appears when students use AI to summarise readings, refine phrasing or impose structure. Here the friction lies in form-making and shaping thought into something communicable. The judgement at stake is procedural: what is proportionate or efficient in this context?

Dialogic engagement emerges when students test interpretations or rehearse arguments. AI becomes a low-stakes sounding board, absorbing some of the vulnerability of articulating something half-formed. The question beneath it is interpretive: what does this mean, and how far do I trust my reading and myself?

Metacognitive engagement is evident when students ask AI to critique their reasoning or compare approaches. What is absorbed here is evaluative tension and the discomfort of examining one’s own argument. The judgement in play is comparative and strategic: which option is stronger, and why? And then there is affective-regulatory engagement. Here, AI absorbs the anxiety that precedes judgement itself. It breaks tasks into steps, softens feedback, lowers the threshold for beginning, offers reassurance before submission and quietens the internal ruminations and rehearsals of everything that might go wrong. This is not peripheral to learning. It is increasingly central.

Figure: Where the labour of learning now lives

Accessibility, safety and the risk of smoothing too much

For many students, particularly those navigating anxiety, executive dysfunction, neurodivergence or heavy external commitments, this emotional buffering is not indulgence but access (Rose & Meyer, 2002). Breaking tasks into steps or privately rehearsing ideas before speaking can widen participation rather than diminish it.

We should not romanticise struggle. Nor should we imagine that institutional structures have ever been able to hold every student perfectly. For some learners, AI offers another place to rehearse thinking, one that sits alongside, rather than replaces, human dialogue.

But there is a tension here. If AI consistently absorbs the strain of uncertainty before ideas encounter resistance, if feedback is softened before it unsettles, if structure replaces the slow work of wrestling thought into form, then something quieter begins to shift. Much of this work happens privately, in browser tabs and late-night prompts, in spaces students do not always feel comfortable admitting to. That makes it harder for us to see what is being strengthened and what may be thinning. The danger is not comfort, but the quiet disappearance of formative strain.

By formative strain, I do not mean suffering for its own sake, nor simply the “desirable difficulties” described in cognitive load theory (Bjork & Bjork, 2011) or the stretching associated with a Vygotskian zone of proximal development (Vygotsky, 1978). I am referring to the lived experience of remaining with ambiguity, critique and partial understanding long enough for judgement to consolidate; the emotional as well as cognitive work of staying with a problem. If that work is always pre-processed, it may narrow the rehearsal space where judgement forms.

Scaffold or substitute

Much depends on whether AI remains a scaffold or begins to function as a substitute. Used as scaffold, it lowers the emotional threshold just enough for deeper engagement, absorbing anxiety without displacing judgement. Used as substitute, it reduces not only strain but evaluation itself; the work of deciding and committing shifts elsewhere. The distinction lies less in the tool than in how it is woven into the learning environment.

Individual awareness and institutional responsibility

It would be easy, and unfair, to frame this as a matter of individual discernment. Students already carry a great deal. But nor is this simply a matter of institutional correction. We are all navigating new terrain in real time, without a settled script.

If we are serious about judgement formation, then responsibility is shared — and it is evolving. This is less about detection or prohibition than about openness. AI engagement is happening whether we discuss it or not. The question is whether we bring it into the light. That might mean inviting students to reflect on how they used AI in a task, not as confession, but as analysis. It might mean modelling, in our own teaching, what it looks like to question or refine an AI response rather than accept it wholesale. It certainly means acknowledging the emotional labour of learning openly (Newton, 2014), recognising that starting can be harder than finishing and that this, too, is part of learning.

At a structural level, we also need some candour. Systems built on speed, metrics and visible output inevitably amplify the appeal of friction-reducing tools. If polish is rewarded more consistently than process, we should not be surprised when students bypass the stretch between uncertainty and articulation. Cultivating discernment, then, is not a matter of allocating blame. It is a collective project of making the shifting terrain of AI use visible, discussable and educative.

Where the emotional work now lives

Generative AI has not diminished the importance of human judgement. If anything, it has made visible how emotionally mediated that judgement has always been (Immordino-Yang & Damasio, 2007). The interior work of learning – the hesitation, the rehearsal, the private negotiation of uncertainty – has never been fully observable. It has always unfolded, at least in part, elsewhere.

What AI changes is not the existence of that interior space, but its texture. Some of that labour now takes place in dialogue with a system that can stabilise, extend or subtly redirect thinking. That creates an opportunity: we are at a juncture where the emotional dimensions of learning can be surfaced and examined more deliberately than before.

It also carries risk. Students can disappear down an AI rabbit hole just as easily as they once disappeared into rumination. The question is not whether the interior work exists, but how it is shaped and whether it ultimately strengthens judgement or thins it.

References

Barnett, R (2007) A will to learn: Being a student in an age of uncertainty Open University Press

Biesta, GJJ (2013) The beautiful risk of education Paradigm Publishers

Bjork, EL & Bjork, RA (2011) ‘Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning’ in MA Gernsbacher, RW Pew, LM Hough & JR Pomerantz (eds), Psychology and the real world: Essays illustrating fundamental contributions to society (pp. 56–64) Worth Publishers

Newton, DP (2014) Thinking with feeling: Fostering productive thought in the classroom Routledge

Vygotsky, LS (1978) Mind in society: the development of higher psychological processes Harvard University Press Rose, DH & Meyer, A (2002) Teaching every student in the digital age: universal design for learning ASCD

Joanne Irving-Walton is a Principal Lecturer at Teesside University, working across learning and teaching and international partnerships. She is particularly interested in how academic judgement and professional identity develop through the emotional realities of higher education.


by Leave a comment

Teaching students to use AI: from digital competence to a learning outcome

by Concepción González García and Nina Pallarés Cerdà

Debates about generative AI in higher education often start from the same assumption: students need a certain level of digital competence before they can use AI productively. Those who already know how to search, filter and evaluate online information are seen as the ones most likely to benefit from tools such as ChatGPT, while others risk being left further behind.

Recent studies reinforce this view. Students with stronger digital skills in areas like problem‑solving and digital ethics tend to use generative AI more frequently (Caner‑Yıldırım, 2025). In parallel, work using frameworks such as DigComp has mostly focused on measuring gaps in students’ digital skills – often showing that perceived “digital natives” are less uniformly proficient than we might think (Lucas et al, 2022). What we know much less about is the reverse relationship: can carefully designed uses of AI actually develop students’ digital competences – and for whom?

In a recent article, we addressed this question empirically by analysing the impact of a generative AI intervention on university students’ digital competences (García & Pallarés, 2026). Students’ skills were assessed using the European DigComp 2.2 framework (Vuorikari et al, 2022).

Moving beyond static measures of digital competence

Research on students’ digital competences in higher education has expanded rapidly over the past decade. Yet much of this work still treats digital competence as a stable attribute that students bring with them into university, rather than as a dynamic and educable capability that can be shaped through instructional design. The consequence is a field dominated by one-off assessments, surveys and diagnostic tools that map students’ existing skills but tell us little about how those skills develop.

This predominant focus on measurement rather than development has produced a conceptual blind spot: we know far more about how digital competences predict students’ use of emerging technologies than about how educational uses of these technologies might enhance those competences in the first place.

Recent studies reinforce this asymmetry. Students with higher levels of digital competence are more likely to engage with generative AI tools and to display positive attitudes towards their use (Moravec et al, 2024; Saklaki & Gardikiotis, 2024). In this ‘competence-first’ model, digital competence appears as a precondition for productive engagement with AI. Yet this framing obscures a crucial pedagogical question: might AI, when intentionally embedded in learning activities, actually support the growth of the very competences it is presumed to require?

A second limitation compounds this problem: the absence of a standardised framework for analysing and comparing the effects of AI-based interventions on digital competence development. Although DigComp is widely used for diagnostic purposes, few studies employ it systematically to evaluate learning gains or to map changes across specific competence areas. As a result, evidence from different interventions remains fragmented, making it difficult to identify which aspects of digital competence are most responsive to AI-mediated learning.

There is, nevertheless, emerging evidence that AI can do more than simply ‘consume’ digital competence. Studies by Dalgıç et al (2024) and Naamati-Schneider & Alt (2024) suggest that integrating tools such as ChatGPT into structured learning tasks can stimulate information search, analytical reasoning and critical evaluation—provided that students are guided to question and verify AI outputs rather than accept them uncritically. Yet these contributions remain exploratory. We still lack experimental or quasi-experimental evidence that links AI-based instructional designs to measurable improvements in specific DigComp areas, and we know little about whether such benefits accrue equally to all students or disproportionately to those who already possess stronger digital skills.

This gap matters. If digital competences are conceived as malleable rather than fixed, then AI is not merely a technology that demands certain skills but a pedagogical tool through which those skills can be cultivated. This reframing shifts the centre of the debate: away from asking whether students are ready for AI, and towards asking whether our teaching practices are ready to use AI in ways that promote competence development and reduce inequalities in learning.

Our study: teaching students to work with AI, not around it

We designed a randomised controlled trial with 169 undergraduate students enrolled in a Microeconomics course. Students were allocated by class group to either a treatment or a control condition. All students followed the same curriculum and completed the same online quizzes through the institutional virtual campus.

The crucial difference lay in how generative AI was integrated:

  • In the treatment condition, students received an initial workshop on using large language models strategically. They practised:
  • contextualising questions
  • breaking problems into steps
  • iteratively refining prompts
  • and checking their own solutions before turning to the AI.
  • Throughout the course, their online self-assessments included adaptive feedback: instead of simply marking answers as right or wrong, the system offered hints, step-by-step prompts and suggestions on how to use AI tools as a thinking partner.
  • In the control condition, students completed the same quizzes with standard right/wrong feedback, and no training or guidance on AI.

Importantly, the intervention did not encourage students to outsource solutions to AI. Rather, it framed AI as an interactive study partner to support self-explanation, comparison of strategies and self-regulation in problem solving.

We administered pre- and post-course questionnaires aligned with DigComp 2.2, focusing on five competences: information and data literacy, communication and collaboration, safety, and two aspects of problem solving (functional use of digital tools and metacognitive self-regulation). Using a difference-in-differences model with individual fixed effects, we estimated how the probability of reporting the highest level of each competence changed over time for the treatment group relative to the control group.

What changed when AI was taught and used in this way?

At the overall sample level, we found statistically significant improvements in three areas:

  • Information and data literacy – students in the AI-training condition were around 15 percentage points more likely to report the highest level of competence in identifying information needs and carrying out effective digital searches.
  • Problem solving – functional dimension – the probability of reporting the top level in using digital tools (including AI) to solve tasks increased by about 24 percentage points.
  • Problem solving – metacognitive dimension – a similar 24-point gain emerged for recognising what aspects of one’s digital competences need to be updated or improved.

In other words, the AI-integrated teaching design was associated not only with better use of digital tools, but also with stronger awareness of digital strengths and weaknesses – a key ingredient of autonomous learning. Communication and safety competences also showed positive but smaller and more uncertain effects. Here, the pattern becomes clearer when we look at who benefited most.

A compensatory effect: AI as a potential leveller, not just an amplifier

When we distinguished students by their initial level of digital competence, a pattern emerged. For those starting below the median, the intervention produced large and significant gains in all five competences, with improvements between 18 and 38 percentage points depending on the area. For students starting above the median, effects were smaller and, in some cases, non-significant.

This suggests a compensatory effect: students who began the course with weaker digital competences benefited the most from the AI-based teaching design. Rather than widening the digital gap, guided use of AI acted as a levelling mechanism, bringing lower-competence students closer to their more digitally confident peers.

Conceptually, this challenges an implicit assumption in much of the literature – namely, that generative AI will primarily enhance the learning of already advantaged students, because they are the ones with the skills and confidence to exploit it. Our findings show that, when AI is embedded within intentional pedagogy, explicit training and structured feedback, the opposite can happen: those who started with fewer resources can gain the most.

From ‘allow or ban’ to ‘how do we teach with AI?’

For higher education policy and practice, the implications are twofold.

First, we need to stop thinking of digital competence purely as a prerequisite for using AI. Under the right design conditions, AI can be a pedagogical resource to build those competences, especially in information literacy, problem solving and metacognitive self-regulation. That means integrating AI into curricula not as an add-on, but as part of how we teach students to plan, monitor and evaluate their learning.

Second, our results suggest that universities concerned with equity and digital inclusion should focus less on whether students have access to AI tools (many already do) and more on who receives support to learn how to use them well. Providing structured opportunities to practise prompting, to critique AI outputs and to reflect on one’s own digital skills may be particularly valuable for students who enter university with lower levels of digital confidence.

This does not resolve all the ethical and practical concerns around generative AI – far from it. But it shifts the conversation. Instead of treating AI as an external threat to academic integrity that must be tightly controlled, we can start to ask:

  • How can we design tasks where the added value lies in asking good questions, justifying decisions and evaluating evidence, rather than in producing a single ‘correct’ answer?
  • How can we support students to see AI not as a shortcut to avoid thinking, but as a tool to think better and know themselves better as learners?
  • Under what conditions does AI genuinely help to close digital competence gaps, and when might it risk opening new ones?

Answering these questions will require further longitudinal and multi-institutional research, including replication studies and objective performance measures alongside self-reports. Yet the evidence we present offers a cautiously optimistic message: teaching students how to use AI can be part of a strategy to strengthen digital competences and reduce inequalities in higher education, rather than merely another driver of stratification.

Concepción González García is Assistant Professor of Economics at the Faculty of Economics and Business, Catholic University of Murcia (UCAM), Spain, and holds a PhD in Economics from the University of Alicante. Her research interests include macroeconomics, particularly fiscal policy, and education.

Nina Pallarés is Assistant Professor of Economics and Academic Coordinator of the Master’s in Management of Sports Entities at the Faculty of Economics and Business, Catholic University of Murcia (UCAM), Spain. Her research focuses on applied econometrics, with particular emphasis on health, labour, education, and family economics.


by 1 Comment

Widely used but barely trusted: understanding student perceptions on the use of generative AI in higher education

by Carmen Cabrera and Ruth Neville

Generative artificial intelligence (GAI) tools are rapidly transforming how university students learn, create and engage with knowledge. Powered by techniques such as neural network algorithms, these tools generate new content, including text, tables, computer code, images, audio and video, by learning patterns from existing data. The outputs are usually characterised by their close resemblance to human-generated content. While GAI shows great promise to improve the learning experience in various disciplines, its growing uptake also raises concerns about misuse, over-reliance and more generally, its impact on the learning process. In response, multiple UK HE institutions have issued guidance outlining acceptable use and warning against breaches of academic integrity. However, discussions about the role of GAI in the HE learning process have been led mostly by educators and institutions, and less attention has been given to how students perceive and use GAI.

Our recent study, published in Perspectives: Policy and Practice in Higher Education, helps to address this gap by bringing student perspectives into the discussion. Drawing on a survey conducted in early 2024 with 132 undergraduate students from six UK universities, the study reveals an impactful paradox. Students are using GAI tools widely, and expect their use to increase, yet fewer than 25% regard its outputs as reliable. High levels of use therefore coexist with low levels of trust.

Using GAI without trusting it

At first glance, the widespread use of GAI among students might be taken as a sign of growing confidence in these tools. Yet, when students are asked about their perceptions on the reliability of GAI outputs, many express disagreement when asked if GAI could be considered a reliable source of knowledge. This apparent contradiction raises the question of why are students still using tools they do not fully trust? The answer lies in the convenience of GAI. Students are not necessarily using GAI because they believe it is accurate. They are using it because it is fast, accessible and can help them get started or work more efficiently. Our study suggests that perceived usefulness may be outweighing the students’ scepticism towards the reliability of outputs, as this scepticism does not seem to be slowing adoption. Nearly all student groups surveyed reported that they expect to continue using generative AI in the future, indicating that low levels of trust are unlikely to deter ongoing or increased use.

Not all perceptions are equal

While the “high use – low trust” paradox is evident across student groups, the study also reveals systematic differences in the adoption and perceptions of GAI by gender and by domicile status (UK v international students). Male and international students tend to report higher levels of both past and anticipated future use of GAI tools, and more permissive attitudes towards AI-assisted learning compared to female and UK-domiciled students. These differences should not necessarily be interpreted as evidence that some students are more ethical, critical or technologically literate than others. What we are likely seeing are responses to different pressures and contexts shaping how students engage with these tools. Particularly for international students, GAI can help navigate language barriers or unfamiliar academic conventions. In those circumstances, GAI may work as a form of academic support rather than a shortcut. Meanwhile, differences in attitudes by gender reflect wider patterns often observed on academic integrity and risk-taking, where female students often report greater concern about following rules and avoiding sanctions. These findings suggest that students’ engagement with GAI is influenced by their positionality within Higher Education, and not just by their individual attitudes.

Different interpretations of institutional guidance

Discrepancies by gender and domicile status go beyond patterns of use and trust, extending to how students interpret institutional guidance on generative AI. Most UK universities now publish policies outlining acceptable and unacceptable uses of GAI in relation to assessment and academic integrity, and typically present these rules as applying uniformly to all students. In practice, as evidenced by our study, students interpret these guidelines differently. UK-domiciled students, especially women, tend to adopt more cautious readings, sometimes treating permitted uses, such as using GAI for initial research or topic overviews, as potential misconduct. International students, by contrast, are more likely to express permissive or uncertain views, even in relation to practices that are more clearly prohibited. Shared rules do not guarantee shared understanding, especially if guidance is ambiguous or unevenly communicated. GAI is evolving faster than University policy, so addressing this unevenness in understanding is an urgent challenge for higher education.

Where does the ‘problem’ lie?

Students are navigating rapidly evolving technologies within assessment frameworks that were not designed with GAI in mind. At the same time, they are responding to institutional guidance that is frequently high-level, unevenly communicated and difficult to translate into everyday academic practice. Yet there is a tendency to treat GAI misuse as a problem stemming from individual student behaviour. Our findings point instead to structural and systemic issues shaping how students engage with these tools. From this perspective, variation in student behaviour could reflect the uneven inclusivity of current institutional guidelines. Even when policies are identical for all, the evidence indicates that they are not experienced in the same way across student groups, calling for a need to promote fairness and reduce differential risk at the institutional level.

These findings also have clear implications for assessment and teaching. Since students are already using GAI widely, assessment design needs to avoid reactive attempts to exclude GAI. A more effective and equitable approach may involve acknowledging GAI use where appropriate, supporting students to engage with it critically and designing learning activities that continue to cultivate critical thinking, judgement and communication skills. In some cases, this may also mean emphasising in-person, discussion-based or applied forms of assessment where GAI offers limited advantage. Equally, digital literacy initiatives need to go beyond technical competence. Students require clearer and more concrete examples of what constitutes acceptable and unacceptable use of GAI in specific assessment contexts, as well as opportunities to discuss why these boundaries exist. Without this, institutions risk creating environments in which some students become too cautious in using GAI, while others cross lines they do not fully understand.

More broadly, policymakers and institutional leaders should avoid assuming a single student response to GAI. As this study shows, engagement with these tools is shaped by gender, educational background, language and structural pressures. Treating the student body as homogeneous risks reinforcing existing inequalities rather than addressing them. Public debate about GAI in HE frequently swings between optimism and alarm. This research points to a more grounded reality where students are not blindly trusting AI, but their use of it is increasing, sometimes pragmatically, sometimes under pressure. As GAI systems continue evolving, understanding how students navigate these tools in practice is essential to developing policies, assessments and teaching approaches that are both effective and fair.

You can find more information in our full research paper: https://www.tandfonline.com/doi/full/10.1080/13603108.2025.2595453

Dr Carmen Cabrera is a Lecturer in Geographic Data Science at the Geographic Data Science Lab, within the University of Liverpool’s Department of Geography and Planning. Her areas of expertise are geographic data science, human mobility, network analysis and mathematical modelling. Carmen’s research focuses on developing quantitative frameworks to model and predict human mobility patterns across spatiotemporal scales and population groups, ranging from intraurban commutes to migratory movements. She is particularly interested in establishing methodologies to facilitate the efficient and reliable use of new forms of digital trace data in the study of human movement. Prior to her position as a Lecturer, Carmen completed a BSc and MSc in Physics and Applied Mathematics, specialising in Network Analysis. She then did a PhD at University College London (UCL), focussing on the development of mathematical models of social behaviours in urban areas, against the theoretical backdrop of agglomeration economies. After graduating from her PhD in 2021, she was a Research Fellow in Urban Mobility at the Centre for Advanced Spatial Analysis (CASA), at UCL, where she currently holds a honorary position.

Dr Ruth Neville is a Research Fellow at the Centre for Advanced Spatial Analysis (CASA), UCL, working at the intersection of Spatial Data Science, Population Geography and Demography. Her PhD research considers the driving forces behind international student mobility into the UK, the susceptibility of student applications to external shocks, and forecasting future trends in applications using machine learning. Ruth has also worked on projects related to human mobility in Latin America during the COVID-19 pandemic, the relationship between internal displacement and climate change in the East and Horn of Africa, and displacement of Ukrainian refugees. She has a background in Political Science, Economics and Philosophy, with a particular interest in electoral behaviour.