December 13, 2024
Thanks to Alissa Simon, HMU Tutor, for today’s post.
A group of researchers recently held a webinar for the Modern Language Association (MLA) on the topic of AI integration into higher education. Part of a joint task force, the speakers came from a diverse background of large and small schools from around the country. They presented data from their progress on “Building a Culture for Generative AI Literacy in College Language, Literature, and Writing.” This critical conversation is vital not just to higher ed, but to all educators. Though higher education is already feeling AI’s impacts, it is important to continually educate ourselves about the pros and cons of an AI integrated world.
What follows are a handful of key takeaways for me. Clearly I’m not a tech expert and my gut reaction is against AI rather than towards it. Having said that however, I do see real reasons to incorporate it into work and research. How we do that, though, requires great skill and experience. The panelists offered real advice garnered from their own experiences. And I want to underscore something that they repeated multiple times: whether you incorporate it or not, do so openly and for good reason. Do not blindly accept or reject generative AI.
First, it is important to look at AI in two ways. AI is both a tool and a product. Referring to it in this way might help students better understand that the results they receive are not bias free, but part of a larger product. In addition, platforms are expanding with intentions of connecting to specific markets. In this way, it is a product and therefore, users should identify this lens. More commonly, AI is referred to as a tool. It certainly can be a tool for students as they research. But the tool is not thoughtful, discerning, or critical. It regurgitates and compiles. So, while it may be a great research tool, it might also be worth checking the AI’s work before you submit any essays, articles, or papers. Critical thinking should come from the user.
This leads into the second point, that AI literacy implies criticality. If we intend to incorporate AI into educational spaces, daily lives, and everything else, we must critique its output. Know when to use it and when not to use it. Learn AI’s flaws. If you choose not to use it, do so with reasons, not just gut refusal. Some questions we might ask ourselves are: What do you know and what do you not know about generative AI? What are your learning outcomes for class; what are your goals with generative AI? (Using your responses, you can scaffold AI into various areas of class, or remove it altogether). How does generative AI use a student’s data and intellectual properties? This final question is important for those implementing AI in classrooms, but it’s also important for the users. Many of us are both students and teachers. We should be knowledgeable on the ways in which AI collects data. Sometimes privacy settings can be changed, but knowing how it collects data may alter your use of a specific company’s tool.
Finally, understand that AI is inherently different from human communication. AI doesn’t have context so it does not (yet) speak to an audience. It is also endlessly patient. It can return the same response a thousand times. It can read the same question every day for a year. It does not understand human impatience and cannot feign impatience. Having a conversation with AI might be enlightening, exhausting, exhilarating, liberating, boring, etc. But only the human experiences these emotions.
As the panelists concluded, the speakers offered these key ideas for integrating AI into our world:
1] Be transparent. If you use an AI, say so. If you don’t, also own that. Transparency goes both ways: teacher to student, and student to teacher. In order to make this clear to students, it might be worth having a conversation about values. For example: what do you value in education and are these the same things that students value?
2] Recognize that generative AI erases linguistic bias. Linguistic diversity should be encouraged. Dialects and unique speech should be encouraged. AI has not mastered tones or dialects, which runs the risk of prioritizing one style of speech over all others.
3] Emphasize shared governance, if possible.
4] Advocate for students to opt out of policies that ask for student’s private information. Protect student information and intellectual data at all times.
5] Focus on AI literacy versus plagiarism narratives. Looking for plagiarism reinforces attitudes of punishment and surveillance rather than offering educational supports. Basically, teach students (and ourselves) to think critically, to use AI when it benefits us and makes sense to do so.
Thanks to the webinar presenters: Holly Hassel of Michigan Technological University; Antonio Byrd of University of Missouri, Kansas City; Leonardo Flores of Appalachian State University; and Jen William of Purdue University, West Lafayette.
Photo credit: Shutterstock/ 1st footage
