By David Lund Herum, Junior Researcher
Being a student in this particular time and space probably means you have encountered ChatGPT in relation to your studies. I at least have. Therefore, I also chose to write a project about its possible affects in education. More specifically at ITU. But this blogpost will not focus on the findings of this. Instead, I will try and indulge you in the confusing process I had of narrowing down my scope.
How it began
I first remember ChatGPT being discussed by a lecturer at the beginning of one of my courses. We were allowed to use ChatGPT in our projects, as long as we referenced it. The lecturers were also interested in learning more of its capabilities. Having no idea what this chat thing was I decided to visit it myself after the lecture. I was quite intrigued by what I found. I remember asking it to explain Edward Saidâs concept of âOrientalismâ in no more than five sentences. I had written my Bachelorâs project using this concept and sitting in front of the computer it was the first thing which popped into my mind:
âCertainly! Edward Said’s concept of Orientalism refers to the Western academic and cultural tradition that constructs and represents the East (Middle East, Asia, etc.) as a homogeneous, exotic, and inferior “other.” This ideology perpetuates stereotypes, generalizations, and power imbalances between the West and the East, shaping Western understanding and dominance over Eastern cultures. Said argued that Orientalism isn’t merely a scholarly field but a system of knowledge intertwined with colonialism, imperialism, and the exercise of power, where the East is portrayed as irrational, backward, and in need of Western intervention. Said aimed to expose and challenge this biased portrayal, highlighting its role in sustaining unequal power dynamics between the West and the East.â (ChatGPT, 2023).
I was intrigued that this technology could explain something which I had spent a semester reading up on, in such a quick and precise manner. Was this going to make my life easier as a student, by changing how I studied?
As time went by, I started consulting ChatGPT when I needed help with my studies. Especially courses which required some kind of programming skills. But one day, my partner introduced me to the German sociologist Hartmut Rosa (2013) and his book âSocial Acceleration: A New Book of Modernityâ. In this book Rosa explains how we (his arguments mainly base themselves in Western society) experience social acceleration. Technology is here to benefit people. It helps make our everyday tasks easier and faster. But to what purpose, asks Rosa. Rosa believes that in its current state, technology simply provides us with the time to complete more work. Technology is therefore not providing us with time to be individuals independent from our work. It obscures our relation to our work by creating processes in which we have more of it.
Seeing that ChatGPT was proclaimed to be a game changer within education (other areas as well) I wanted to investigate if it actually just led to more work? Seeing how well and fast it explains theoretical concepts, like Orientalism, is it actually just resulting in courses becoming harder and more demanding of students?
A project takes shape
This landed me on a project regarding the impacts of ChatGPT at ITU. Through the guidance of my supervisor, I quickly found out that students were not the only ones who was impacted by this technology. ChatGPT had also made its presence clear to educators, by shaping both the students’ learning outcomes and exam processes.
As I started preparing for the interviews and conducting my general literature research, I found that my initial idea of analysing ChatGPT in relation to Rosas social acceleration might not be the most pressing issue. Instead, I found it more pressing to introduce the environmental impacts of using ChatGPT. I also believed that its data collection practices, and the possible re-production of inequality were necessary to mention. When students and educators use these technologies, they (unconsciously?) become part of these concerning practices. I therefore believe these concerns are important to highlight and reflect in the use of ChatGPT at ITU. I therefore decided to include these issues in my interview guide.
I conducted six interviews for the project. Here my limited experience with interviewing showed itself. I found it hard to conduct the actual interviews. I believe I might have deviated from the interview guide a bit too much. Whenever I felt a question was irrelevant, I skipped it. Afterwards I realised that not having the answers I deemed irrelevant, obviously made it much harder to compare possible opposing opinions as I might have lacked those opposed opinions.
What I did find was different experiences of how ChatGPT influences the current practice of teaching at ITU. Some of these included concerns with its presence and others included opportunistic views of how this technology might force us to re-evaluate how we assess students.
I was therefore left with some interesting insights into the effects of this technology as well as different research papers arguing for its potential negative impacts.
Upon analysing these findings, I suddenly found Rosa relevant again. Not only because of his notion of social acceleration but also because he discusses resonance. He uses this concept to explain how we as subjects interact with our surrounding world and evolve from this. A process which is uncontrollable (Rosa, 2022). Could this concept maybe highlight a potential lack in ChatGPTâs capabilities of shaping subjects at universities, as resonance might not be achievable in this meeting? But I also found Catherine DâIgnazio and Lauren F. Kleinâs concept of âData Feminismâ relevant. Here I saw a framework which could help me discuss possible solutions in ensuring transparency from companies such as OpenAI and also further highlight issues of inequality in models like ChatGPT (DâIgnazio & Klein, 2020). But what about capitalism? Can ChatGPT exist outside of this concept and its focus on profit? Does this not affect how this technology is currently being utilised, which in turn affects the students and educators using it?
Where it left me
I was simply left with too many paths to take. I usually have issues with narrowing down my scope, when working with any topic. But not to this extend.
Through my studies I have read critiques of technological determinism. That technology is political. But I have never felt it to such a great extend as in this project. Writing about the influence of ChatGPT at ITU is just the beginning. There are so many more paths that needs to be researched, and I believe this showcases the complex nature of technologies such as ChatGPT. Sure, it has potentials. But it seems that each potential comes with a handful of risks, concerns and unintended consequences. I also realised that this is not unique to ChatGPT. Many of the online services or technological devices we all use, have some grey areas which can be hard to fully comprehend.
These areas need continuous research. We must understand how these technologies impact our societies, even though many paths can be taken. We cannot trust its creators to unfold them with intentions of benefiting all of humanity.
Bibliography
Brevini, B. (2022). Is AI good for the planet? Polity Press.
DâIgnazio, C., & Klein, L. F. (2020). Data feminism. The MIT Press.
OpenAI. (2023). ChatGPT (December 15. GPT 3.5) [Large language model]. https://chat.openai.com
Rosa, H. (2022). Det ukontrollerbare (H. Sander, Trans.). Eksistensen.
Rosa, H., Trejo-Mathys, J., & Rosa, H. (2015). Social acceleration: A new theory of modernity (Paperback ed). Columbia University Press.