Opinion: AI- It’s Not Fair!

By Martha Schenk ( she/her)

Is Turnitin really using our submissions to train its commercially generative AI? Hot on the trail of this burning question, my incendiary investigation was quickly snubbed out by the Director of Academic Development: Professor Stephen Marshall. While Marshall assured me that the universities contract with Turnitin explicitly prevents the plagiarism detection provider from utilising student work for the training of artificial intelligence systems since mid-last year, he raised another more poignant, less considered AI issue: “The biggest problem in all of this presupposes that everybody has equitable and equal access to these tools and technologies”.

It turns out that, in the wake of AI’s swift inauguration to the education sector, some students are being left behind. Similarly to the introduction of personal devices like laptops in learning environments, accessibility to AI programs due to financial constraints is predicted to become a substantial issue in the coming years. “People with the resources to spend $50-$100 a month can get access to reasonably better AI tools, meaning that they use their time more productively, and the superficial qualities of what they produce in terms of writing will be better”, Marshall explained. “We know those things will have an effect on peoples’ grades, their capacity to be successful in their courses”. 

Unlike the introduction of personal devices—the gradual development of which has allowed for decades of adaptation—AI is advancing at breakneck speed. “Everything’s compressed, we’re having to respond to these things much more rapidly”, was the message. This means that in today’s unprecedented technological era, swift action is required to ensure that vulnerable students are not disadvantaged in current and future educational settings. Marshall stresses that “every student should have an equal opportunity to demonstrate success in our courses” and that the university has “a responsibility as a formal education provider to provide a learning environment where that is achieved”. But what exactly is the solution?  

One approach being considered is that of a revisited cyber-commons model. The university could host its own AI locally, making educational software both free and readily available for both students and staff. Marshall believes that there will be “some collective effort by the universities in New Zealand to find a way where we can have locally operated AI environments that will make it possible to put tools into courses equitably” but admits “I don’t know what that will look like yet, it’s complicated and will involve spending a lot of money”.  

While this could make basic AI tools more accessible, the approach negates the imminent presence of better and more advanced outside programs that will always be purchasable by those with the resources to do so. Think, you’re sitting in lecture haggling with VicGPT, which may or may not be about to explode, while your neighbour flaunts their subscription to a program which can transcribe your lecturers’ thoughts, probably. While your classmate works away with infinite knowledge at their fingertips, your laptop is still booting up. 

Worries about the environmental, employment, and ethical elements of AI use aside, the forecasted lack of equity in educational AI might seem enough to tip you over the edge. But all hope is not lost. 

While many believe that artificial intelligence in education is not a matter of ‘when’, but rather ‘how’, and indeed programs like Grammarly, ChatGPT, OpenAI, Google Gemini, Duolingo, and even Kahoot! can already be considered mainstream in educational spheres, English Programme Director Dr Adam Grener says it’s important to remember that we have agency. “Part of the thing that worries me is the assumption from society, or certain leaderships, that AI introduction to learning spaces is somehow inevitable”, he explained. “Teaching is about interacting with people and ideas”, and “the more we automate things and outsource thinking, we aren’t considering the purpose of the actual university”. 

Grener isn’t alone in his stance. A press release from August 1^st published by a network of New Zealand academics working in Communication and Media Studies across Aotearoa universities claimed that “there is a tendency in our universities to be resigned to AI as an unstoppable and unquestionable technological force” and that “AI is not an inevitable technological development which must be incorporated into higher education”. Their reasoning included arguments of intellectual property theft, research integrity, student anxiety, techno-capitalism, data-sovereignty, clashes with Te Tiriti, and even data colonialism.

Equity issues could just be the beginning. As AI gains traction in the educational sector, it’s vital to remember that we have the power to shape its path.

Sources:

(https://www.scoop.co.nz/stories/ED2508/S00002/statement-on-ai-in-universities-from-aotearoa-communication-media-scholars-network.htm#google_vignette)