Open research questions in Academic integrity and plagiarism

What the literature leaves open

• Future work should address the integration of mobile application support, automated email and SMS notifications for submission status changes, optimisation of large-file processing, and the potential development of an in-house similarity algorithm to reduce external API dependency.

  A Secure Web-Based Research Management Framework Integrating Automated Plagiarism Detection for Higher Education · 2026 · DOI

• Future research could explore differences in perceptions across academic disciplines and career stages, as attitudes toward AI- 1 Future Recommendations The findings of this study open several openings for further investigation into the evolving relationship between generative AI and academic writing.

  Generative Artificial Intelligence and the Problem of Authorship and Personal Voice in Academic Writing · 2026 · DOI

• The findings of this study suggest that the problem exposed by GenAI is not simply that students can use new tools in unsupervised assessment. Rather, it is that many conventional take-home tasks can no longer be relied upon to verify authentic engagement or individual attainment when submitted as standalone artefacts under open conditions. At the same time, the findings do not support abandoning AI-integrated learning, nor do they justify a blanket return to examination-dominated assessment. Such a retreat would risk allowing assurance concerns to override educational design, narrowing the range of capabilities that higher education is able to foster and recognise. What is needed instead is a more deliberate repositioning of take-home assessment within a learning architecture in which assurance is embedded, designed, and demonstrated through fit-for-purpose forms of defensible judgement. On this basis, four recommendations are offered. 6.1. Recommendation 1: Reposition take-home assessment rather than treating it as standalone proof of learning The present findings indicate that take-home assessment in its conventional form should no longer be treated as sufficient evidence of individual attainment in AI-integrated higher education. Even where structured templates, AI logs, evaluation tables, and reflective components are mandated, the audit showed that traceability remained weak, internal consistency frequently broke down, and submitted artefacts often failed to verify the process they purported to document. These limitations do not mean that take-home assessment has lost its pedagogical value. They do mean, however, that its assurance role must be reconsidered. Accordingly, take-home tasks should be repositioned as important developmental components within the broader learning arc rather than as self-sufficient proof of learning. Their value lies in enabling exploration, drafting, iterative refinement, evidence use, and authentic engagement with the kinds of AI-supported workflows students are likely to encounter beyond university. Used in this way, take-home assessment remains central to contemporary pedagogy. What must change is not its existence, but the expectation that an unsupervised final artefact can, by itself, certify that the underlying thinking is genuinely the student’s own. 6.2. Recommendation 2: Treat structured AI pedagogy as the foundation, not the endpoint, of assurance The findings of the present study should not be interpreted as evidence against frameworks such as SAGE. On the contrary, SAGE remains pedagogically valuable because it teaches students to work with AI in a critical and evidence-based manner through generation, evaluation, refinement, critique, and reflection. The difficulty identified here is not that this pedagogical structure lacks value, but that pedagogical artefacts generated under unsupervised conditions do not automatically become credible assurance evidence merely because they are documented.

  Embedding assurance within learning: Empirical evidence from the SAGE framework for repositioning take-home assessment in AI-integrated higher education · 2026 · DOI

• reproducible searches, the review combined database- and journal-level searching (Page et al., 2020; Rethlefsen et al., 2021). Search strings combined controlled terms and keywords for integrity, ethics, fairness, leadership, governance, moral harassment, corruption, and AI ethics, with Boolean operators and truncation, and were adapted across databases search as reporting (Page et al., 2020; Rethlefsen et al., 2021). Searches were limited to 2024–2025 and to or studies public-integrity settings in Africa and Asia, with date and setting limits explicitly aligned to the eligibility criteria as advocated in PRISMA 2020 (Page et al., 2020; Rethlefsen et al., 2021).

  Academic Integrity and Fairness in Educational and Public Institutions: A Systematic Review · 2026 · DOI

• Kier and Ives support this systems perspective by gathering from students, staff, and tutors at an online Canadian university. Their content analysis of hundreds of open-ended 26 comments recommendations divided three broad categories: policy and procedures, compliance and commitment, and resources. Respondents desired clear, comprehensive policies developed through consistent enforcement, and sufficient support (e.g., writingskills lessons and academic resources).

  A Systematic Review of Academic Integrity and Respectful Interactions in Higher Education Institutions: Leadership, Systems, and AI Governance · 2026 · DOI

• The findings of this study have important implications for educational practice and policy. Addressing cheating requires solutions that target not only actual behaviors but also the temptation to engage in cheating practices, such as modifying exam settings and strengthening monitoring systems (Henderson, Chung, Awdry, Ashford et al., 2023). For instance, instructors can reduce the potential for cheating in online exams by designing exams that emphasize critical thinking rather than rote memory, such as open-book and application-based formats. Misconduct can be further discouraged by using rotated questions, test versions, and time restrictions (Spiegel & Nivette, 2023). Moreover, proctoring strategies should be continuously adapted to address integrity and student experience, emphasizing transparency and ethical use of technology (Maphalaa & Nkosi, 2025). Additionally, institutions should invest in implementing a supportive evaluation strategy that lessens the stress of a single, critical exam by utilizing a variety of assessment methods, including participation, projects, and performance tasks. However, educators and assessment specialists ought to collaborate to develop learning environments that provide students with academic and psychological skills, particularly self-efficacy and self-organization, which give them the confidence and skills they need to succeed with integrity.

  Self-Organization and Self-Efficacy as Predictors of Cheating Attitudes in Online Exams: A Self-Regulated Learning Perspective · 2026 · DOI

• Accordingly, the inconsistent findings are best understood as reflecting sensitivity to both contextual and methodological features, including differences in cheating indi- cators and their validity.

  The Effect of Performance Goals and Evaluation Standard on Cheating in an Academic Aptitude Test – An experimental series · 2026 · DOI

• The risks of AI misuse still need to be addressed directly, rather than being dismissed as paranoia among educators and researchers. AI detection tools are one way to mitigate these risks, but they must be part of a broader strategy that includes education, policy development, and assessment reform. However, detection tools alone are insufficient, as they can mistakenly flag legitimate content. Therefore, we recommend that institutions not only invest in detection technologies but also train faculty to recognize the nuances of AI-generated content. Moreover, AI-resistant assessments should focus on real-world applications and critical thinking, areas where AI is less effective at substituting human input. Looking forward, we believe the future of higher education lies in how well we integrate AI into our teaching and research practices while maintaining our commitment to academic integrity. We should not be afraid of AI; instead, we should become proficient in its use and comfortable with allowing it to represent our voice when appropriate. The key is to remain vigilant about where AI assists and where it overreaches, ensuring that our own intellectual contributions stay at the forefront. By doing so, the entire academic community, comprising administrators, instructors, and students, can harness AI’s potential to enhance the educational experience while upholding the values of integrity, trust, and originality that are central to higher education. 9 DECLARATION OF GENERATIVE AI AND AI-ASSISTED TECHNOLOGIES IN THE WRITING PROCESS Statement: During the preparation of this work, the author(s), Norman S. St. Clair and Pamela D. McCrau, used ChatGPT 4.0 in order to improve the readability of the document. 60 International Journal of Advanced Corporate Learning (iJAC) iJAC | Vol. 19 No. 1 (2026) Balancing Innovation and Integrity: Navigating the Challenges of Generative AI in Higher Education After using this tool/service, the author(s) reviewed and edited the content as needed, and we take full responsibility for the content of the publication. 10 REFERENCES A. Bandura, “Social cognitive theory of self-regulation,” Organizational Behavior and Human Decision Processes, vol. 50, no. 2, pp. 248–287, 1991. https://doi.org/10.1016/ 0749-5978(91)90022-L D. R. E. Cotton, P. A. Cotton, and J. Shipway, “Chatting and cheating: Ensuring academic integrity in the era of ChatGPT,” Innovations in Education and Teaching International, vol. 61, no. 2, pp. 228–239, 2024. https://doi.org/10.1080/14703297.2023.2190148 S. Joshi, “Comprehensive review of AI hallucinations: Impacts and mitigation strategies for financial and business applications,” International Journal of Computer Applications Technology and Research, vol. 14, no. 6, pp. 38–50, 2025.

  Balancing Innovation and Integrity: Navigating the Challenges of Generative AI in Higher Education · 2026 · DOI

• for Researchers Extend the model across countries and disciplines; examine longitudinal effects; test additional mediators (e.g., AI anxiety, institutional policy clarity) and mod- erators (e.g., year of study, assessment type); compare alternative SEM and causal designs.

  Academic Integrity and Students’ Ethical Use of ChatGPT in Higher Education · 2026 · DOI

• for Practitioners Embed digital literacy and ethical-AI training (transparency, anti-plagiarism, bias awareness, responsible use) into curricula; implement inclusive AI policies; and guide trust calibration and verification workflows to support responsible use.

  Academic Integrity and Students’ Ethical Use of ChatGPT in Higher Education · 2026 · DOI

• Keywords Conduct multi-institutional replications, experimental interventions on ethics/digital literacy training, and studies of assessment design that balance AI use with integrity (e.g., oral/ authentic assessments). artificial intelligence, academic integrity, ChatGPT, Gulf universities, PLS-SEM, student ethical responsibility, transparency, plagiarism avoidance, bias awareness, AI trust, digital literacy, AI usefulness, responsible use INTRODUCTION BACKGROUND The rapid integration of artificial intelligence tools such as ChatGPT in the education sector has attracted significant scholarly attention (Al-Jahwari & Yousif, 2025; X. Chen et al., 2020; Rejeb et al., 2024; Shishakly, 2025; Vieriu & Petrea, 2025; Zawacki-Richter et al., 2019). From the student perspective, prior research highlights the benefits of ChatGPT for writing, language learning, research, and administrative tasks (Dwivedi et al., 2023; Fitria, 2023; Lund & Wang, 2023; Shishakly et al., 2025). These systems provide real-time feedback on grammar, programming, and problem-solving by leveraging deep learning techniques to generate contextually relevant responses (Atlas, 2023; Baidoo- Anu & Owusu Ansah, 2023; Else, 2023; Herft, 2023; Kasneci et al., 2023; Qadir, 2022; Sallam, 2023; Sok & Heng, 2023; Susnjak, 2022; Vieriu & Petrea, 2025). Despite these advantages, scholars have raised substantial ethical concerns, including academic dishonesty, over reliance on AI, misinformation, and unfair assessment practices (Rudolph et al., 2023; Sok & Heng, 2023). Although ChatGPT can reduce instructional workload and foster pedagogical innovation (Cox, 2021), it also poses risks to academic integrity, responsible use, and algorithmic fairness (Farhi et al., 2023; Qadir, 2022; Welding, 2023). Ethical AI use is therefore expected to uphold fairness, transparency, privacy, and non-discrimination (Mhlanga, 2023). Plagiarism and contract cheating remain among the most pressing concerns (Cotton et al., 2024; Roe & Perkins, 2022), while transparency in disclosing AI assistance is increasingly emphasized as a foundation of academic credibility (Lamb, 2023; C. Lee & Cha, 2025; Tlili et al., 2023). RESEARCH GAP Although existing studies examine AI adoption and traditional academic misconduct, they offer limited empirical insight into how students conceptualize ethical responsibility when using generative AI tools. Most research focuses narrowly on plagiarism or cheating, with minimal attention to broader ethical dimensions such as transparency, responsible use, and algorithmic bias. Consequently, current academic integrity frameworks do not sufficiently account for AI-specific risks or student-level ethical decision-making. Furthermore, theoretical discussions often lack practical, evidence-based strategies to guide the ethical use of AI in educational contexts (Guerrero-Dib et al., 2020; Ramdani, 2018; Zawacki-Richter et al., 2019). Kumar et al.

  Academic Integrity and Students’ Ethical Use of ChatGPT in Higher Education · 2026 · DOI

• The research suggests that the balance between maintaining academic integrity and enforcing it among students, educators, and researchers when using AI tools remains insufficiently clarified.

  An Academic Text: The Balance Between Academic Integrity and Artificial Intelligence · 2026 · DOI

• Our future work is to improve it for more detection efficiency and less time complexity. We will consider the following work: (i) using word-k-grams instead of sentences, (ii) using a Lemmatiser instead of the stemmer to get better results from WordNet, and (iii) modifying the post-processing stage to gain more ideal granularity.

  A Linguistic Analysis of Undetected Plagiarism in Postgraduate Academic Papers · 2026 · DOI

• 1 Contextual and Scope Limitations This study was conducted within the HR administrative processes of a single institu- tional context, the United Arab Emirates University (UAEU). The empirical scope of the study was also limited to administrative and support functions rather than core mission activities. However, the sample size is insufficient for robust inferential testing.

  An AI-enabled framework for reducing administrative bureaucracy in higher education while maintaining governance and compliance · 2026 · DOI

• Future research should focus on tracking ethical chal- lenges over time as AI tools become more deeply integrated into research workflows. Persistent problems related to bias and lack of transparency suggest that ex- isting practices are insufficient to support responsible AI use across diverse research settings.

  Towards Ethical AI Adoption in Academic Research: Insights from a Systematic Literature Review · 2026 · DOI

• This study has several limitations, and the findings should therefore be viewed as indica- tive rather than exhaustive. Van Vlasselaer et al. International Journal for Educational Integrity (2026) 22:16 Page 17 of 20 First, the scope of our research was limited to 160 papers, four AI detection tools, and one GenAI model (GPT-4o Deep Research) for the fully AI-generated category. Although, these models were the most advanced at the time, the pace at which both GenAI tools and AI detection software are developed and updated means that our results are valid only for a specific snapshot in time. Further research should continue to evaluate available detection software, as well as some of the more recent genera- tive models such as GPT-5.2, Claude Opus 4.6, and Gemini Pro 3. Second, while our experimental design aimed to mimic realistic student writing behaviour, it could not fully replicate all the software-aided and manual refinements and adjustments that stu- dents may apply in practice. We tested only one humanisation strategy, a single prompt in GPT-4o, whereas students are likely to combine multiple techniques to evade detec- tion. This means that the ability of AI detection tools to identify this type of hybrid text may be lower than demonstrated in this study. Finally, no ground truth was available for the 1,163 master’s theses analysed. This analysis was exploratory in nature: it describes the distribution of Pangram’s flagging scores under real academic conditions rather than validating detection accuracy. As such, the flagging percentages cannot be interpreted as confirmed prevalence rates of AI use. Nevertheless, the analysis provided a valuable first indication of the extent to which generative AI may be present in real master’s the- ses, offering institutions a reference point for understanding current patterns of AI use among students. Future research should aim to develop designs that establish ground truth in authentic academic settings.

  Who wrote this? Evaluating the reliability of AI detection tools in higher education · 2026 · DOI

• Our methodology was limited but robust. While much of the work was manual, the findings were supported by a range of careful secondary work, including using Zaw et al. Research Integrity and Peer Review (2026) 11:26 Page 11 of 14 Fig. 2 Yearly trend of published cochrane reviews that included or cited potential fraudulent studies different search patterns and careful rechecking of the data by a second researcher to ensure validity. Within the confines of the research question, we are confident that we have identified every Cochrane review contaminated by the more notorious researchers who inhabit the upper echelons of the Retraction Watch database. This study provides a preliminary exploration into the prevalence and implications of unreliable studies in Cochrane systematic reviews. However, a notable limita- tion is the absence of statistical analysis to directly meas- ure the impact of these studies on the conclusions of the systematic reviews. Moreover, our focus exclusively on Cochrane reviews and studies authored by individu- als with a history of 24 or more retractions may limit the generalizability of our findings. Given the context, it is almost certain that Cochrane reviews contain many more problematic papers from authors who are less notorious. Other, less robust reviews will also likely contain addi- tional instances of fraud, given the extreme care that the Cochrane collaboration takes with their investigations. Additionally, limiting our study to authors with 24 or more entries into the database might not capture the broader range of issues presented by other systematic reviews from authors with fewer retraction histories, which could still affect the reliability of the evidence. Future research should employ rigorous quantitative methods to more definitively assess the impact of stud- ies by authors with multiple retraction histories. Expand- ing the scope to include a broader range of systematic reviews and evaluating the influence of studies from authors with varying levels of retraction records would further enhance our understanding of this critical issue. This may be solvable with a programmatic solution if there is the ability to access the Cochrane database directly.

  Research integrity within systematic reviews: investigating the prevalence of studies by authors with multiple retraction histories in Cochrane reviews · 2026 · DOI

• Future work could build on the present findings by (1) excluding formally retracted studies from meta-analyses and comparing results to assess how such retractions influence pooled effect estimates and overall conclu- sions and (2) conducting sensitivity analyses that exclude studies authored by individuals with multiple retraction records to evaluate the robustness of meta-analytic find- ings and the potential impact of research by authors with a history of retractions. Alternative thresholds for defin- ing highly retracted authors could also be explored to determine how different operational definitions influence conclusions. In addition, integrating retraction meta- data directly into systematic review software would allow reviewers to more efficiently identify and flag retracted or potentially fraudulent studies, enhancing transparency and methodological rigor. Collectively, these approaches could improve the reliability and interpretability of aggre- gated evidence while providing guidance for responsible meta-analytic practice.

  Research integrity within systematic reviews: investigating the prevalence of studies by authors with multiple retraction histories in Cochrane reviews · 2026 · DOI

• While students have always found ways to circumvent learning through plagiarism, generative AI systems’ ability to complete entire assignments means that past recommendations to discourage plagiarism are insufficient.

  Offloading Learning to a Stochastic Parrot: Generative AI’s Impacts on Academic Integrity and the Current State of Best Practices · 2026 · DOI

• While gender disparities in misconduct are widely documented, the underlying cognitive mechanisms explaining why males are more susceptible, and whether a supportive institutional environment can buffer the risks associated with academic pressure and negative attitudes, remain underexplored in the context of Chinese medical education.

  Institutional environment, academic attitude, and misconduct among medical students: a structural equation modeling and cluster analysis · 2026 · DOI

• Although this study offers substantive insights into the governance of GenAI in HE, certain limitations must be acknowledged. This study defines the core competencies for GenAI in HE. Intentionally, it does not list all the specific policies and guidelines in detail, as those details would rapidly become irrelevant and/or obsolete (Le, 2024). It is important that these be overarching and customizable to the individual institution. The rapid pace of GenAI innovation means that any policy or guideline recommendations risk becoming outdated as technologies, capabilities, and ethical implications continue to evolve. The findings, therefore, represent a detailed understanding of expert per- spectives at a specific point in time rather than a static or definitive account. While the Delphi method and collective writing approach ensured methodological rigor and inclu- sivity of expert perspectives, the participant pool, although geographically diverse, was limited to HE experts. The perspectives of other crucial stakeholders, such as students, policymakers, employers, and technology developers, were not directly represented. Their inclusion may have yielded additional insights into the lived experiences and insti- tutional consequences of GenAI adoption.

  Governing generative AI in higher education: a global Delphi study on policy and practice · 2026 · DOI

• The current research has investigated geography students’ perceptions of plagiarism at the University of KwaZulu-Natal. As both a lecturer and honours coordinator, I recommend that students learn to manage their time more effectively, as they receive an average of 40 assessments per semester. Thus, poor time management would increase pressure on each assessment and likely increase the appeal of academic dishonesty to save time. Second, digital resources and artificial intelligence should not be avoided; rather, they can be ethically embraced as educational aids that save time while teaching students critical thinking. Third, and most importantly, use all the resources available to you to assist you in succeeding in your studies; this includes consulting with your lecturer, visiting the library, reading and going beyond compulsory module readings. For lecturers, I recommend setting assessments that require originality- whereby students can practically see the implications of said theory- as this would likely encourage students not simply copy from the internet. Next, even if the university offers these workshops, establish relationships with the discipline librarian to arrange further, mandatory workshops for registered students. And though the massification of students has meant more work and less time to revise module content, it is highly encouraged for lecturers to annually revise their module content and formative and summative assessments, as this may increase class attendance and general interest in the class. Finally, higher education institutions are strongly encouraged to offer a compulsory, non-credit module on ethical academic behaviour that all registered undergraduate and postgraduate students must complete within their first year of studies. Second, personal experience has shown that both students and university staff are unfamiliar with the teaching and learning policies regarding plagiarism, the use of artificial intelligence, and their consequences. Thus, training workshops should be mandatory for academic staff to attend, to prevent disciplines and individuals from developing their own rules and to ensure that all actions are guided by the university’s policies. Being student-centred does not mean impunity; it is being concerned with an education that prioritises students’ inclusion, support and learning. With that being said, universities putting students first means helping students succeed academically while improving their scholarship and other skills. Therefore, universities have an obligation to protect their reputation, academic staff, and students by implementing fair and transparent processes that are unbiased towards any single stakeholder. 1 3N. P.

  Understanding Students' Perceptions of Plagiarism Within the Discipline of Geography in the University of KwaZulu-Natal · 2026 · DOI

• The anagram task used to measure cheating in this study has a benefit of being more aca- demic in nature in contrast to other objective measures of cheating such as a coin flipping task (Dickinson & McEvoy, 2021). However, concerns regarding ‘real’ and ‘non-words’ presented a limitation in the present study. Future studies could improve this measure by having participants define or provide an example of the words they generated. Although this would take extra time, it would provide a means to assess whether participants’ non- words were phonetic or spelling errors for real words. Additionally, this was a low stakes measure of cheating where students did not face the same pressures/motivations to succeed compared to real-life situations where students decide to engage in academic misconduct. It may be important to complement the use of this low-stakes measure of integrity with actual measures of academic misconduct collected through the university to better understand how this cheating measure relates to those engaging in academic misconduct. As mentioned above, the study includes self-report measures. Although self-report mea- sures are open to bias, in some cases self-reports are the best indicators of personal subjec- tive experiences. For example, knowing that students have greater perceived confidence regarding academic integrity is important. However, future studies may benefit from inclu- sion of objective measures to compare subjective experience of variables such as knowledge of academic integrity versus objective demonstration of that knowledge. It is important to note that this study represents an evaluation of a short-term academic integrity intervention with outcomes measured immediately after the intervention. Although effective for assessing the immediate questions posed in the present study this approach cannot prove the efficacy of these intervention types over the long-term nor does it address behaviours. Longer term interventions and measures that capture self-report, knowledge and behaviours are needed to more fully understand the impact of these interventions.

  Modules, Discussions, or Simulations? Comparing the Effect of Academic Integrity Interventions on University Students’ Perceptions and Cheating Behaviour · 2026 · DOI

• Based on the findings, the following recommendations are proposed for students, faculty, and higher education institutions: Guidelines for Responsible AI Use: Institutions should develop clear policies outlining how generative AI can be ethically used in academic work. Guidelines should include citation requirements, limits on direct content use, and encouragement of verification against authentic IKS sources. AI Awareness and Training: Conduct workshops and orientation sessions for both students and faculty on the responsible use of AI tools. Include practical exercises showing how AI can supplement learning without replacing primary sources of IKS knowledge. Integration with IKS Curriculum: Encourage faculty to design assignments that require critical engagement with IKS content, ensuring that students verify and interpret AI-generated information. Include evaluation criteria that reward original thinking and authentic understanding, not mere reproduction of AI outputs. Monitoring and Feedback: Use periodic surveys to monitor students‘ and faculty‘s perceptions and practices regarding AI use. Provide feedback and support mechanisms for students struggling to balance AI use with academic integrity. Promoting a Culture of Integrity: Foster a learning environment that emphasizes honesty, verification, and critical thinking when using AI. Recognize and reward responsible AI practices to reinforce positive behavior.

  Preserving the Authenticity of Indian Knowledge Systems: Student and Faculty Perspectives on Responsible Use of Generative AI · 2026 · DOI

• Recommender systems in educational contexts create filter bubbles by repeatedly suggesting similar content aligned with student initial interests, but the paper does not specify what algorithmic mechanisms (e.g., diversity constraints, serendipity injection, exploration-exploitation balancing) should be integrated into collaborative filtering and content-based recommendation algorithms to expose students to diverse viewpoints while maintaining personalization.

  Ethical challenges of artificial intelligence in education: A systematic literature review on bias, privacy, and academic integrity · 2026 · DOI

Most-cited papers in Academic integrity and plagiarism

• Chatting and cheating: Ensuring academic integrity in the era of ChatGPT · Innovations in Education and Teaching International · 2023 · 1,677 citations
• Academic Integrity considerations of AI Large Language Models in the post-pandemic era: ChatGPT and beyond · Journal of University Teaching and Learning Practice · 2023 · 513 citations
• Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI) · Journal of University Teaching and Learning Practice · 2023 · 370 citations
• Use of ChatGPT in academia: Academic integrity hangs in the balance · Technology in Society · 2023 · 311 citations
• Student perspectives on the use of generative artificial intelligence technologies in higher education · International Journal for Educational Integrity · 2024 · 245 citations
• ChatGPT versus engineering education assessment: a multidisciplinary and multi-institutional benchmarking and analysis of this generative artificial intelligence tool to investigate assessment integrity · European Journal of Engineering Education · 2023 · 238 citations
• Ethics of Artificial Intelligence in Education: Student Privacy and Data Protection · Science Insights Education Frontiers · 2023 · 219 citations
• How Common Is Commercial Contract Cheating in Higher Education and Is It Increasing? A Systematic Review · Frontiers in Education · 2018 · 206 citations
• Academic Integrity in Online Assessment: A Research Review · Frontiers in Education · 2021 · 203 citations
• A systematic review of research on cheating in online exams from 2010 to 2021 · Education and Information Technologies · 2022 · 168 citations

Most recent work

• The AI Guilt Complex: Moral Emotions and Ethical Dilemmas in Academic Technology Adoption · Journal of Academic Ethics · 2026
• AI Slop in Academic Publishing: History, Characteristics, Manifestations, Causes, and Mitigation Strategies · Internet Reference Services Quarterly · 2026
• Toward more Sustainable Plagiarism Policies in an AI Higher Education Environment: a Student-Informed Case Study · Journal of Academic Ethics · 2026
• Results from two decades of five-yearly plagiarism surveys: new insights into prevalence, understanding, attitudes, knowing vs naïve plagiarism, and generative artificial intelligence (genAI) · International Journal for Educational Integrity · 2026
• ‘I shouldn't be saying this’: Library-based student confessions about AI, cheating, and academic integrity · The Journal of Academic Librarianship · 2026
• How university staff evaluate generative AI: Cognitive and ethical perspectives on teaching, trust, and academic integrity · Human Technology · 2026
• AI Literacy Framework for Academic Writing in the Age of Artificial Intelligence · Internet Reference Services Quarterly · 2026
• Authorship Statement for Generative Artificial Intelligence: Assuring Trust and Accountability · Journal of University Teaching and Learning Practice · 2026
• Academic Integrity and Students’ Ethical Use of ChatGPT in Higher Education · Journal of Information Technology Education Research · 2026
• Challenges and recommendations for research security: Learning from research ethics and integrity · Research Ethics · 2026

Related topics in Social Sciences