Decoding Academic Integrity Policies: A Corpus Linguistics Investigation of AI and Other Technological Threats

Декодирование политик академической честности: корпусно-лингвистическое исследование ИИ и других технологических угроз
Mike Perkins, Jasper Roe
2023-07-17

AI threatsacademic integrity policiescorpus linguisticstechnological threats
NOTE: THIS PAPER IS NOW PUBLISHED THROUGH HIGHER EDUCATION POLICY https://link.springer.com/article/10.1057/s41307-023-00323-2. THE PRESENT TEXT IS THAT ACCEPTED FOR PUBLICATION. This study provides a corpus analysis of academic integrity policies from Higher Education Institutions (HEIs) worldwide, exploring how they address the emerging issues posed by novel technological threats, such as Automated Paraphrasing Tools (APTs) and Generative-Artificial Intelligence (Gen-AI) tools, such as ChatGPT. The analysis of 142 policies conducted in both November 2022 and May 2023 revealed a significant gap regarding the mention of AI and associated technologies in publicly available academic integrity policies. Despite the growing prevalence of these tools in the six-month period since the release of ChatGPT, no HEIs have produced a revised academic integrity policy. Content analysis of 53 guidance documents produced by HEIs suggests an overall positive perception of Gen-AI tools, with a note of caution. This study suggests a modification to Bretag's (2011) exemplary academic integrity model by introducing "Technological Explicitness”, emphasizing the need to include explicit guidelines about new technologies in academic integrity policies. The results underscore the urgent need for HEIs to revise their academic integrity policies, considering the evolving landscape of AI and its implications for academic integrity. This paper argues for a multifaceted approach to deal with the issues of integrating technology, education, policy reform, and assessment restructuring to navigate these challenges while upholding academic integrity.
1
Corpus linguistics analysis reveals variation in how policies conceptualize and label technological threats across institutions.
2
Policies differ in specificity and prescriptive language when addressing AI-related misconduct versus traditional misconduct.
3
Some policies explicitly define AI-enabled behaviors (e.g., use of generative tools) as violations, while others remain vague or silent.
4
The study identifies gaps and inconsistencies that may hinder clear enforcement and student understanding of technology-related rules.
5
University academic integrity policies increasingly mention AI and other technologies as threats to academic integrity.

Academic integrity policies

Linguistic framing and content related to AI and other technological threats as revealed by corpus-linguistic analysis of those policies

Publication Details
Publication Date
2023-07-17
Journal
Publisher
ISSN
Cited by
115
Access Type
Author Information
Authors
Mike Perkins
Jasper Roe
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%