LRNT 524: but Teacher i wrote . you cant prove .

“Robot sitting at a desk in a classroom” – Fotor Image Generator

For this task, we were asked to consider the impacts of a chosen learning innovation. I was unable to participate in the last activity in the course, where we researched a recent instance of innovation in instructional design. However, I read what my peers wrote, and  I am intrigued by the topic of AI content writers. I decided to examine that further in this task because I have found an upturn in written work submitted by my students where it is evident that they did not write what they submitted, but the work is not plagiarised. Students used an AI content writer.

AI content writers use artificial intelligence to produce written text based on the user’s input and generate “human-like” content (Choudry, 2022).  In an everyday scenario, we use AI content writers in the form of predictive text or often programs such as Grammarly (McKnight, 2021). They are not new, but the programs are becoming more advanced, and people’s awareness of the prevalence of their use is increasing (Thorley, 2021).  AI content writers are commonly used for marketing, blogs and online content generation (Choudry, 2022; Thorley, 2021). They are quick and easy to use. Using a selected app, the user can enter a topic/ detail they want to write about, and within seconds written text on that topic is produced. The key factor on the user’s side is that they must ensure that the details they enter are fit for purpose or that what is generated is not what is intended.

In the case of my students, for a paper they were tasked to write, I received well-written text meeting the word count that spoke very generally about the topic but did not include any citations. The work missed nuances of specific instructions, such as “use key terms from the text” or “why do you believe this to be true.” The submitted work was generic, lacked original thought and did not produce what was asked in the task. AI-content writers lack personality, writing style (Fyfe, 2022; McKnight, 2021), and originality (Mathew, 2022). You might wonder how I knew an AI content writer wrote it. Well, my students are ESL students whose written English is developing. When writing to a student to state I did not believe they wrote their perfectly written 800-word assignment, I received this response: “but Teacher i wrote . you cant prove .” After a conversation, they and others admitted what they had done.

In the case of my students, their reliance on easy-to-use technology impacts their learning. They feel they do not need to do the work or learn the content because they use an app to generate products to convey learning. There are limits to developing their written skills further if using an AI content writer rather than completing the task independently; however, there is an indication that grammar-checking apps can improve student mastery of grammar when instructors use it as a marking tool for feedback (Toncic, 2020).  AI content generation being submitted for text-based tasks has implications for the authenticity of work and academic integrity (Thorley, 2021). Suppose the other instructors in my program do not know what to look for in AI-generated content. Students may receive passing grades and graduate without learning or legitimately meeting the program’s demands. Teachers often unknowingly give better grades to students who use AI grammar-checking apps (Toncic, 2020), which brings forth discussions of access based on socioeconomic factors and digital literacy (Thorley, 2021). It also raises concerns about the academic rigour and program reputation of those who graduate and cannot compose a simple sentence without errors.  There are also implications on assessment choice, deliverables, design and ensuring that students learn what is intended.

In the traditional context of academia, the increase in AI content generators will force educators to consider multiple aspects of their practice. There will need to be considerations as to what writing is, what we place value on (Thorley, 2021) in teaching and assessment, and ensuring that students have equal access to the tools.  In terms of impact, there is much to consider. Still, it is important to note that there are cases where deliberately using AI content generators as co-creators of content integrated as part of a task can increase student writing competencies. For example, in a small study,  Kangasharju et al. (2022) had a group of students use the AI-based Poetry Machine as part of the draft process, using varying features and versions of their poems to create a final copy. The study found that the process of a co-writer increased student creativity and the quality of the product (Kangasharju et al., 2022).  Nazari et al. (2021) found that AI writing tools can improve self-efficacy and engagement while supporting self-directed learning in post-graduate non-native English speakers. While the results of these studies have limitations in scope and size, they demonstrate an opportunity to use AI content writers for good in a controlled environment. They present a positive outlook on how AI-writing tools can support student learning compared to a tool for cheating. Maybe as we re-think teaching and assessment of writing overall, we can find ways to integrate AI writing tools, and students will be less inclined to use them dishonestly. 

References

Choudry, M. W. (2022, October 18). What are AI content writing tools? (and should you use one?). HubSpot. https://blog.hubspot.com/website/ai-writing-generator#:~:text=AI%20writing%20tools%2C%20also%20known,businesses%20that%20conduct%20content%20marketing.

Fyfe, P. (2022). How to cheat on your final paper: Assigning AI for student writing. AI & SOCIETY. https://doi.org/10.1007/s00146-022-01397-z

Kangasharju, A., Ilomäki, L., Lakkala, M., & Toom, A. (2022). Lower secondary students’ poetry writing with the ai-based poetry machine. Computers and Education: Artificial Intelligence, 3, 100048. https://doi.org/10.1016/j.caeai.2022.100048

Mathew, V. (2022, August 3). Ai content writing risks – 8 unspoken consequences you need to know. ShoutMeLoud. https://www.shoutmeloud.com/ai-content-writing-risks.html

McKnight, L. (2021, March 9). To succeed in an AI world, students must learn the human traits of writing. The Conversation. Retrieved December 31, 2022, from https://theconversation.com/to-succeed-in-an-ai-world-students-must-learn-the-human-traits-of-writing-152321

Nazari, N., Shabbir, M., & Setiawan, R. (2021). Application of artificial intelligence powered digital writing assistant in higher education: Randomized controlled trial. Heliyon, 7(5). https://doi.org/10.1016/j.heliyon.2021.e07014

Thorley, J. (Host). (2021, June 1). Anticipating the impact of AI-based writing on education and assessment (No. 28) [Video podcast episode]. In Integrity matters. turnitin. https://youtu.be/oEvvi-0Wggc Toncic, J. (2020). Teachers, AI grammar checkers, and the newest literacies: emending writing pedagogy and assessment. Digital Culture and Education, 12(1), 26–51.