Synopsis: In this abstract, I reflect on my experience as a Year 12 student who was falsely flagged by an AI detection tool in coursework assessment. I explore how this affected my confidence, writing practices, and trust in the fairness of assessment processes. Through this reflection, I argue that student voices must be central to conversations about academic integrity in an AI-influenced education system.
Academic integrity has always been something I associated with honesty and effort in my own learning. As a student, it shapes how I approach written work, assessments, and responsibility for what I submit. These ideas became especially important to me as I moved into senior secondary school, where coursework plays a major role in assessment.
As a 17-year-old Year 12 student, I am submitting this abstract to share my experience of being affected by the use of AI detection tools in school assessments. Using auto-ethnography, I reflect on how being flagged by an AI detection tool despite not using generative AI shaped my confidence, writing practices, and trust in assessment processes, and why I believe student voices need to be more central in discussions about academic integrity as artificial intelligence becomes more present in education.
My BTEC Coursework Experience
I experienced the impact of AI detection tools firsthand through my BTEC Business coursework. Unlike exam-based subjects, BTEC assessment is largely assignment-based, and students are evaluated on content, structure, application of knowledge, and written communication. Teachers are also supposed to use the similarity and AI detection reports through Turnitin, which are intended to be used as supporting information rather than definitive evidence of misconduct.
In one of my assignments, a common AI detection tool indicated that the majority of my work was AI-generated. I had not used any form of generative AI in completing the assignment. When I informed my teacher of this, I was instructed to rewrite and resubmit the work. This decision was made despite the tool’s own guidance stating that AI detection scores should not be relied upon in isolation due to their limitations and potential inaccuracies.
This experience left me feeling confused and discouraged. I had written the assignment in the way my school had taught me to write: using clear structure, appropriate academic language, correct grammar, and well-developed sentences. I found it difficult to understand how demonstrating these skills could result in my work being seen as artificial. Rather than engaging in a discussion about my writing process or evidence of learning, the assessment process focused primarily on reducing the AI detection percentage.
In order to resubmit the assignment successfully, I felt pressured to alter my original writing. I used so-called “humanising” tools to modify my own words, not to improve clarity or quality, but to satisfy the expectations of the AI detection system. As a student who strongly values academic integrity, I found it strange that I had to manipulate my own work in order for it to be perceived as honest.
To better understand whether my experience was isolated, I drew on informal conversations with Year 13 students, who described strategies they had used to reduce AI detection scores, such as simplifying vocabulary, removing conjunctions, reducing punctuation, and using less formal sentence structures. While these methods often resulted in lower or even zero AI percentages, they involved intentionally weakening the quality of writing. This raised a significant concern for me: we as students were being indirectly encouraged to write poorly in order to avoid suspicion and get good grades.
Impact on me as a student
This experience left me with a loss of trust in the very education system that had taught me how to write well. Over the past twelve years of schooling, I had been encouraged to develop clear structure, formal academic language, and strong written communication skills. Yet when I applied those skills in my BTEC assignment, they became the reason my work was questioned. Being told that my writing appeared “too advanced” or “college level” felt unfair, as I was being punished for meeting and exceeding the expectations set by my teachers and curriculum.
What made the situation more difficult was the lack of an opportunity to defend my work. Despite repeatedly explaining that the assignment was my own, the decision remained unchanged, with greater emphasis placed on the AI detection score than on my explanation or writing process. I ultimately felt compelled to alter my work, not to improve its quality, but to reduce suspicion. Learning to deliberately “dumb down” my own writing, simplifying sentences, removing structure, and lowering the standard of expression, felt illogical, and demotivating, and conflicted with the skills I had been encouraged to develop. Instead of feeling proud of my abilities, I began to question whether writing well was something I should avoid. This experience affected not only how I approached that assignment, but how I viewed assessment, trust, and fairness within my education. Ultimately, I feel this experience raises broader questions about fairness, transparency, and the role of AI detection tools in assessment practices
To conclude, I believe it is crucial that students` voices are actively heard, ensuring fairness, transparency, and trust, rather than allowing AI detection scores alone to dictate the perceived integrity of our work.
Keywords: Artficial intelligence, AI Detection, Student Voice, AI Detection Tools