What do college students really think about ChatGPT?

We urgently need to talk with college students about how they are thinking about and using ChatGPT and generative AI more broadly

Image: Midjorney

I talk to my students a lot about emerging technologies and their potential impacts on society. And so it’s not surprising that we got sidetracked talking about ChatGPT in class last night.

What did surprise me though was the attitudes that were expressed in the admittedly informal discussion we had. These underlined a growing urgency to include students in conversations around how we navigate the use of generative AI in learning and education.

I should say up front that this was an extremely casual conversation with undergraduate students in my class on emerging tech and the the future. It wasn’t planned, there was no structure, and these are students who are used to talking about the potential downsides of irresponsible tech in society.

At the same, they’re students who represent a wide range of disciplines and backgrounds, from actuarial studies and architecture to psychology, political science, biology, economics, engineering, and a bunch of other areas. In other words, they are a pretty diverse group.

As we were chatting, three things emerged:

ChatGPT? Not a big deal

Despite all the reports and fears of students embracing ChatGPT and using it to find shortcuts to getting their grades, I saw little evidence of this in the class — although admittedly I was working with an n of very few. Rather, there was a distinctly “meh” attitude toward ChatGPT in the conversation.

Around a third — possibly less — of the class indicated that they use the platform regularly, but very few admitted to using it extensively. Many seemed to use it as an occasional resource, but not as a game-changing technology.

I’m probably over-interpreting here, but I got a clear sense of students normalizing ChatGPT and other text-based generative AI tools as small components in a growing landscape of apps that can be useful at times, but which are not necessarily transformative.

I came to college to learn … not to take shortcuts

A perspective that surprised me — but that shouldn’t have — was the number of students who saw using ChatGPT as antithetical to why they were attending college in the first place.

I had one student say (and I’m paraphrasing) “I came to college to learn — why would I ask a machine to do this for me?”

I got a clear sense here that the value propositions around ChatGPT from a student perspective do not align well with what we sometimes assume they are.

Of course, used in the right way, generative AI platforms can enhance learning. But I suspect we fall all too quickly into assuming that most students are just attending college for the grades and the degree. And in doing so, we forget that many are here because they want to learn and expand their understanding and perspectives.

My writing reflects who I am … I don’t want to be robbed of that

There were a few students in the conversation who clearly saw their writing (or their art) as expressions of themselves, and recoiled from the idea of an app robbing them of this.

I sympathize here. I typically don’t use ChatGPT myself when I write for exactly the same reason. The way I write is personal. It reflects who I am, and to relinquish that to a machine would be to diminish myself.

That this came up in conversation — unprompted — demonstrates a sophistication and an awareness around how at least some students are thinking about ChatGPT and generative AI that I suspect is at odds with the assumptions of many instructors.

We need to engage with students around ChatGPT — and fast

The upshot of this informal discussion is that it impressed on me how urgently we need to be including students in conversations around ChatGPT in learning and education.

While educators and educational establishments are still scrambling to respond to widespread access to powerful generative AI platforms, it seems that students are already integrating them into an increasingly busy landscape of apps, resources, and other demands on their attention — and in many cases, ChatGPT isn’t a top priority.

There are also indications that the ways students are thinking about and approaching generative AI are more sophisticated than we might initially assume.

I’d go so far as to say that the perspectives and attitudes I hear from my students sometimes come across as better informed, more sophisticated, and more mature, than those from some of my colleagues!

It’s also becoming increasingly clear that student attitudes toward ChatGPT and other AI platforms are very different now from what they were just 2 - 3 months ago. This isn’t surprising as the technology and it’s socialization are moving so fast. Yet most of the data on student behaviors around and attitudes toward ChatGPT date back to the initial surge of interest earlier this year.

In other words, we’re in danger of making decisions on ChatGPT and generative AI in learning and education that are based on out of date understanding of how students are thinking about and using the technology.

I also have a growing suspicion that we’ve been asking the wrong questions when it comes to exploring how students are thinking about ChatGPT. And as a result, the data that we do have don’t necessarily jive with the reality we’re seeing unfold.

I may be wrong of course. But the only way we can tell is by actually engaging and working with students now, rather than waiting until we feel we’ve got our own house in order as educators (by which time, it’ll be too late).

… and as a quick follow-up to this, there’s a reason why I invited two of my undergraduate students to be part of next week’s live streamed conversation around AI and the future of learning 😀