Illustration by Allyson (Aven) Rivas
Google’s Gemini prompts us to “Ask anything,” so naturally this is a question we’re encouraged to pose. Maybe even forced to pose, even as we guiltlessly make Gemini polish our final papers in the same instance. At first glance, this reads like another high-minded, attention-grabbing AI criticism.
What we’re faced with instead is something of an uncomfortable truth: college is preparing us for a future no one really knows much about. College is no longer the safety net that people, including university administration, understood it to be. The skills from our degree are not something employers necessarily want or even have the know-how to employ amidst AI incorporation now.
The prospects of a new tool, an urge to keep up with work trends and to be the forerunner of new ones are driving universities to rapidly implement AI programs. Tailoring its range of capabilities to useful and reliable tasks is the goal. Establishing its pitfalls is also pivotal to create a university that is preparatory and also pedagogical.
From study aids to task delegation, AI usage has flown its way into Red Hawk Country for most students in one way or another.
One usage concern that is being addressed is academic dishonesty. Copying the work of a classmate or that of professional programmers, cheating by any other perpetrator would violate all the same. It also happens that this concern existed before Gemini and will still be here with every new technology to come.
Be as it may, cheating needs a long-term solution. According to Frontiers in Artificial Intelligence, “…AI-generated content tests traditional academic quality control mechanisms and poses new requirements for evaluating academic output reliability.” Rigorous metrics like “transparency requirements and clarifying responsibility attribution” are achieved if we have a concise framework.
Data privacy, a more contemporary issue, warrants rapid action. It’s something that’s always brought up in roundtable discussions and for good reason. ASU’s agreement with ChatGPT, the first college to partner with an AI company, states that student data will not be used to train the models.
How a legal framework can protect both student information and AI’s user input reliance, while accounting for changing AI models, is unclear. Timing, or how long these goals can be simultaneously achieved, is also something there’s little direction to start answering.
As eager as we are to optimize AI models for school usage, there is also a drastic trend in making and executing its monitoring. A UNESCO News survey, outlines universities’ approach to sustainability, mentioning things like mandatory AI literacy courses and “systematic consultation and engagement with students and faculty.”
Current laws may suffice for the AI we have today, but universities need to still move forward with planning for the future. We should have anticipatory protocols for what tomorrow’s AI might accomplish, and scaffolds readying us for the reasons it will be advanced.
It may be awkward and a bit unnerving to realize we don’t fully understand its current capabilities to make substantial laws. Yet, that’s even more of a reason to create vigorous institutional policies: those very capabilities are changing fast, and so are the uses and products coming from it.
While not perfect, bylaws for university-wide AI usage will serve us by minimizing current threats and foreseeing future ones to some extent.
The threat we as students are most familiar with is our tendency to accept without questioning or just to assume. But in an odd way, this fact can bring us comfort. As a population that’s historically “most susceptible” to this threat, we probably have the most wisdom to tackle it.
Students taking part in building AI policies strengthens the school in every respect. It’s also an incredibly sensible decision. With Montclair State’s newly minted “Working Group on Artificial Intelligence”, multidisciplinary opportunities for student input should be commonplace if we truly believe that AI will become total commonness.
Advancing projects, making current processes more efficient, having the knowledge to co-work with AI and above all, securing safety for university personnel is what every school hopes for. Tokenizing the planning process with students, a demographic who are largely affected by it, reiterates the ultimate philosophy of keeping it pedagogical, keeping it real and keeping it human.
That in all honesty (albeit academic) is a prompt worth asking.
