Magdalena Styś | A Bachelor’s student has no need for GenAI
For students who use AI during their studies, columnist Magdalena Styś is highly critical. “If you’re not planning on doing any learning on your own, isn’t it wasteful to spend three years, thousands of euros and all that housing-related anxiety just to have a computer do all the studying?”
Keeping your mouth shut in conversations with new acquaintances can lead to surprisingly vulnerable confessions. In my two years at the UvA, I’ve met students who happily admit to using large language models to do their quizzes and write their essays, or – in an extreme case – pretend they’re getting advice from historical figures.
I’d be happy to concede that I don’t have experience with high-level research in most academic disciplines. I’m willing to imagine that the world of academia hides really productive uses for GenAI that I’m unaware of. (The use of AI in cancer detection comes to mind here.) I do, however, have a pretty good idea of what level of effort and skill a Bachelor’s degree entails, and I don’t believe there’s a good explanation for relying on large language models or other forms of GenAI at that level.
Calculator
Academically, it’s akin to using a calculator in elementary-school mathematics: sure, you’re outsourcing a skill that a machine can do for you for the rest of your life, but it misses the entire point of why you’re asked to perform the task in the first place. The most meaningful difference between the two is that a calculator is really, really unlikely to hallucinate that 2+2=5.
Much like in the case of long division or multiplication of fractions, you’re not asked to write reading responses because you’re expected to spend the rest of your life writing reading responses; every time you pawn off an assignment to an LLM, you’re robbing yourself of not only knowledge, but also a chance to develop critical thinking skills. This is especially baffling when you consider the fact that you’re paying tuition to study here; if you’re not planning on doing any learning on your own, isn’t it wasteful to spend three years, thousands of euros and all that housing-related anxiety just to have a computer do all the studying? Surely, forging a diploma is more economically efficient at that point?
Ethical reasons
Morally, using most forms of GenAI means a voluntary provision of data (or money in the case of paid subscriptions) to models managed by companies that are, to name a few, funding far-right extremism and cutting hundreds of thousands of trees to make space for data centers. Even if you’re skeptical about the extent of damage to the environment that GenAI produces, the industry’s enmeshment with questionable political projects makes the ethical considerations go way beyond the environmental concerns. A common counter-criticism to this line of argument is that most of our daily life is similarly morally dubious: however, the main difference between that and using GenAI is that the latter is a new thing you actively introduce to your life, which is significantly easier to avoid than getting rid of things already deeply woven into the fabric of our lives.
As long as people will study, they will try to find ways to minimize their workload. Next time, perhaps consider doing so with a tool that isn’t involved in election interference. Or at least one that doesn’t make up sources.
