TLDR: Curtin University has announced it will disable Turnitin’s artificial intelligence writing detection feature starting January 1, 2026. This decision stems from a commitment to fostering trust and clarity in academic culture, and to ensure assessments are fair and relevant, acknowledging ongoing debates about the reliability of such AI detection tools. Traditional text-matching for plagiarism will remain active.
Curtin University has declared its intention to deactivate the artificial intelligence (AI) writing detection feature within Turnitin, a widely used text-matching tool, effective January 1, 2026. This change will be implemented across all campuses and study periods, marking a significant shift in the university’s approach to academic integrity in the age of generative AI .
The university stated that this move is aimed at “fostering trust and clarity within a modern academic culture and continuously improving our assessments to ensure they are secure, fair, relevant and future-ready” . This decision comes amidst a broader, ongoing debate within the academic community regarding the accuracy and reliability of AI detection software. Concerns have been raised about the potential for false positives, where legitimate student work might be incorrectly flagged as AI-generated, leading to undue stress and accusations of misconduct .
While the AI writing detection will be disabled, Curtin University emphasized that the regular text-matching originality checks, which identify unoriginal material and potential plagiarism, will remain fully active . Students will continue to submit their assignments through Turnitin as usual, and their work will still be reviewed for adherence to academic integrity instructions, ensuring fair and consistent assessment .
The unreliability of AI detection tools has been a point of contention among students and educators. Discussions on platforms like Reddit highlight experiences where students were falsely accused of AI use, even when they did not utilize such tools . One user on Reddit commented on the “garbage” nature of the service, stating, “Tested it (don’t ask how) and it never picked up shit” . Another shared a personal account of being flagged multiple times, describing the experience as “incredibly frustrating” and questioning if their academic writing style could be misinterpreted by the AI .
Also Read:
- University Students Express Anxiety and Confusion Over AI Integration in Classrooms
- AI’s Impact on Academic Dishonesty: A Shift in Methods, Not Overall Prevalence
Curtin University currently uses Turnitin’s GenAI writing indicator to check for undisclosed use of generative AI software, and for documents with 20% or more GenAI detection scores, staff review a detailed report. However, to prevent false positives, no score or highlights are shown for GenAI detection scores in the 1% to 19% range . The university supports the philosophy of teaching students to appropriately use GenAI technology ethically and responsibly, preparing them for future professional environments . Students are advised to be transparent about any AI use and follow assessment instructions, with Curtin Grammarly being the only accepted version of Grammarly for use in assessments that will not be flagged by Turnitin .
