Ariffin, Adlina and Md Rahim, Nur Athirah and Ramli, Amizan Azim and Sulaiman, Suriani (2026) Decoding idiolect in the age of AI. In: 5th The 5th International Conference on Languages, Linguistics and Society, UMS Sabah.
![[img]](https://irep.iium.edu.my/style/images/fileicons/application_pdf.png) |
PDF
- Published Version
Restricted to Registered users only Download (1MB) | Request a copy |
Abstract
The rise of artificial intelligence globally has flipped the way education is being conducted. In recent years, newly improved AI chatbots can now cater to a wide variety of features, capable of generating original content with swift data retrievals that are individually curated to the user’s demand. In spite of these abilities, AI chatbots present complications in upholding academic integrity, by which they are capable of producing content that can deceptively go undetected by the conventional plagiarism detectors because the author’s linguistic fingerprints could not be distinguished. This study aims to identify the idiolect of human and AI writing patterns in academic writing. Using a corpus of 120 essays produced by 20 undergraduate students, the data were analysed using Python and AntConc to examine and categorize Hyland’s (2005) metadiscourse markers in identifying human idiolect and AI writing patterns between the two groups. The analysis reveals that human writings display zero overlapping idiolect pairs, reinforcing the idea that humans have unique writing patterns. The syntactical analysis demonstrates that human writing remains creative. Conversely, AI exhibits a contrasting pattern where the idiolect tends to overlap across the AI corpus. The syntactical analysis shows that syntactic structures in AI essays are structurally repetitive. These results demonstrate that human authors possess unique linguistic fingerprints, supporting the idea that an idiolect could not be replicated by other writers while AI portrays a distinct pattern across the corpus in which AI’s output remains rigid and systematic unlike the individualistic output of human writers.
| Item Type: | Proceeding Paper (Plenary Papers) |
|---|---|
| Uncontrolled Keywords: | Academic Text, Artificial Intelligence, Authorship Attribution, Plagiarism |
| Subjects: | T Technology > T Technology (General) > T173.2 Technological change |
| Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): | Kulliyyah of Islamic Revealed Knowledge and Human Sciences > Department of English Language & Literature |
| Depositing User: | AP Dr Adlina Ariffin |
| Date Deposited: | 02 Nov 2026 15:48 |
| Last Update: | 02 Nov 2026 15:48 |
| Queue Number: | 2026-10-Q5944 |
| URI: | http://irep.iium.edu.my/id/eprint/132383 |
| Indexed In: | UNSPECIFIED |
Actions (login required)
 |
View Item |