Moving the conversation beyond assessment, three academic developers, Akile Ahmet, Marina Franchi, and Lee-Ann Sequeira, highlight the not-so-obvious and arguably insidious implications of generative AI in higher education – whitewashing writing, disrupting relationships, and the role of learning technologists and academic developers.
In recent months, ChatGPT and other generative artificial intelligence (AI) tools have taken up significant place within the university community, sparking concerns among educators and students. Besides resulting in a moral panic about cheating, the debate oscillates between AI being a panacea and throwing HE, as we know it, into chaos.
In this blogpost, three academic developers from the Eden Centre for Education Enhancement at the London School of Economics deliberately step away from the noise around AI and assessment and consider the not-immediately-obvious implications of generative AI in higher education.
It's crucial to ask whether AI tools like ChatGPT will further force students to write and think through someone else’s imagination
Whitewashing diverse writing styles
Akile Ahmet
Instead of focusing solely on the potential for misconduct, we should be examining the broader implications of AI technology in education. As scholars such as Ruha Benjamin and Sennay Ghebreab have noted, AI tools are not neutral and can reinforce existing power dynamics and social hierarchies, particularly those related to race. In light of recent calls to decolonise universities, it’s crucial to ask whether AI tools such as ChatGPT will further force students to write and think through someone else’s imagination and through the dominant lens of whiteness that prevails in UK universities.
As AI becomes more integrated into education, it’s essential that we take a step back and consider its impact on learning, thinking, and writing. We need to move beyond simply trying to prevent misconduct and consider how AI technology will shape our universities and the knowledge they produce alongside the interactions between teachers and students. We must critically examine the ways in which AI reinforces existing power dynamics and take steps to ensure that the tools we use are not further perpetuating already harmful and oppressive systems.
It is … important to investigate … (AI’s) potential to dent the trust between educators and students and between students and their peers
Disrupting trust in the classroom
Marina Franchi
Besides the consequences of AI, it is worth considering the effects of the debate on AI on relationships that are so carefully built and nurtured in the classroom. Students are depicted as engaging with generative AI from a variety of perspectives – from it being another tool for them to enhance productivity and support their communication skills to using it to outsmart one another and potentially outsmart teachers, who in turn are asked to react and prevent misconduct. At the same time, it is possible to foresee students questioning academic authorship of teaching and learning materials. It is therefore important to investigate the effects AI might have on the classroom and, in particular, the potential to dent the trust between educators and students and between students and their peers.
The work that educators have done to ensure that the classroom is a principled, non-judgemental environment can clash with an assessment design heavily predicated on the assumption of students’ misconduct. Further, ongoing debates about the threat of academic misconduct are bound to inform students’ perceptions of their peers and impact on their experiences of assessed group work. As academics and educators, we ought to constantly consider the ways in which what is discussed outside the classroom, in the public sphere, and within institutional walls, has a profound effect on the relationships that are lived and embedded in our classroom.
As universities rely more on algorithms and their purveyors, to what extent are complex data systems … determining what counts as learning?
What is our role in this sci-fi film?
Lee-Ann Sequeira
Higher education today feels like a retro sci-fi film. Instead of teaching and learning being done by humans, it is now being taken over by sophisticated software and AI. Large language models and generative AI are used to write essays and text- and code-matching software is used to detect plagiarism and AI assistance. Deus ex machina or The Matrix?
Leaving aside the theatrical, dystopian references, it’s important to understand the impact of the medium on the message aka learning and not dismiss such questions with mealy-mouthed platitudes. Are our HE policies and practice designed to get students to swerve AI (banning its use), to rely on an instrumental use of AI (using Turnitin to detect ‘plagiarism’), or to engage more critically with it?
A second question that goes beyond the transactional nature of AI in assessment, is what the emergence of ChatGPT reveals about the edutech-industrial complex. As universities rely more on algorithms and their purveyors, to what extent are complex data systems – virtual learning environments, grading software, learning analytics – determining what counts as learning? And what is the role of learning technologists and academic developers in this situation – cheerleaders, handmaidens, digital wardens, or intellectual pessimists?
_______________________________________________________________________________________________________________________
This post is opinion-based and does not reflect the views of the London School of Economics and Political Science or any of its constituent departments and divisions.
_________________________________________________________________________________________________________________________
Image: Movie poster of the film, The Matrix Reloaded (2003)
