Nikoletta TOLNER & Monika POGÁTSNIK
The rapid advancement of technology, particularly the availability of artificial intelligence (AI) tools, has fundamentally transformed higher education assessment, creating new challenges in academic integrity (Kell et al., 2025; Stanoyevitch, 2024). The International Center for Academic Integrity (ICAI, 2025) identifies six fundamental values for maintaining academic integrity: honesty, courage, trust, respect, responsibility, and fairness. These principles serve as guiding standards across all areas of learning, teaching, and research, forming the foundation of many institutional codes of ethics. The spread of online education and assessment has not only altered the learning environment but also opened new opportunities for cheating (Zdravkova, 2023), while students’ motivations (Maleki, 2025) and attitudes have undergone significant change (Amzalag, Shapira & Dolev, 2022; Lancaster & Cotarlan, 2021). Detecting and preventing cheating in online examinations (Corrigan-Gibbs et al., 2015) has become increasingly difficult, especially with the rise of AI-based tools, chatbots, and automated text generators, which introduce a new dimension to the issue of academic honesty. According to the ICAI, it is not AI itself that threatens integrity, but rather its unethical and non-transparent use. Institutions therefore have the responsibility to teach students how to apply AI responsibly, ensure transparency in its use, and address inequalities arising from differences in access (ICAI, 2025). For universities, it is an urgent task to develop strategies that are technologically, pedagogically, and ethically sound to reduce the propensity to cheat and preserve the integrity of examinations, particularly in the online environment (Kell et al., 2025; Rüth, Jansen, & Kaspar, 2024).
The aim of this study is to explore students' attitudes at Obuda University’s Alba Regia Faculty toward misconduct in online examinations, with a special focus on the impact of the availability of AI-based tools. The research seeks to determine the extent to which access to AI technologies influences students’ propensity to cheat and their moral and practical perceptions of such behavior. To this end, we conducted a questionnaire-based survey and analyzed the responses using statistical methods.
AI tools, such as ChatGPT, are capable of generating complex texts that can be used in exam situations and essay writing (Oravec, 2023), thereby raising new questions regarding academic integrity (Asogwa, Isiwu & Nwakpadolu, 2022; Cotton, Cotton & Shipway, 2023; Lund et al., 2025). An increasing number of studies are examining methods for detecting AI-based cheating, such as text style analysis and the development of plagiarism-detection algorithms (Fraser, Dawkins & Kiritchenko, 2025). Therefore, universities must be prepared to introduce appropriate regulations and tools to prevent such types of misconduct (Luo, 2024).
The security of online examinations also requires special attention. While online testing offers a convenient and flexible solution, the risk of cheating may increase if appropriate proctoring software or monitoring methods are not applied (Cantiello & Geschke, 2024). Some studies suggest that AI-driven cameras and behavior-analysis algorithms used in online proctoring can significantly reduce the likelihood of cheating, yet they also raise ethical concerns (Coghlan, Miller & Paterson, 2021). Therefore, educators and institutions must develop strategies that minimize opportunities for cheating while respecting students’ privacy and comfort (Mutimukwe et al., 2025).
The use of artificial intelligence in higher education cannot be excluded, as both students and faculty increasingly rely on these tools. The challenge lies not in banning AI altogether, but in integrating it in an ethical, transparent, and responsible manner (Oravec, 2023). Some research calls for adapting assessment practices by introducing strategies that support authentic learning, safeguard integrity, and provide direction for future educational policy and pedagogical reforms (Garcia et al., 2025). Education must aim to teach students how to properly document and reference the possibilities offered by AI, as mastering these skills contributes to preserving academic integrity and preparing for a professional environment increasingly interwoven with AI.
Understanding students’ propensities and motivations for cheating is crucial to ensuring the security of online examinations. Research indicates that the main factors behind cheating include performance pressure, time constraints, and a lack of trust in online proctoring systems (Waltzer & Dahl, 2022). In addition, social norms can influence cheating behavior; when students observe their peers cheating regularly, they are more likely to engage in this behavior (Malesky et al., 2021). Another important factor is the perception of AI tools: while some students regard them as learning aids, others use them to cheat (Lund et al., 2025).
A useful starting point for interpreting students’ attitudes toward cheating is the theory of behavioral economist Dan Ariely (2012), which suggests that most people are not entirely honest or dishonest, but rather seek a compromise that allows them to benefit from cheating while maintaining a positive moral self-image (Ariely, 2012). According to Ariely (2012), people generally cheat only a little, as long as they can justify it to themselves. This so-called "fudge factor" means that decision-making is not merely a matter of weighing risks (getting caught vs. gaining), but involves an interplay between identity and self-justification. Another important finding in Ariely’s work is that the example set by the social environment (for instance, if multiple people in a group cheat) significantly increases the likelihood of cheating.
Further nuance for understanding students’ attitudes toward cheating is provided by the research of Gino and Ariely (2012), which suggests that creative thinking can serve not only innovation but also self-justification. Their findings indicate that individuals with higher creativity are more likely to find acceptable explanations for rule-breaking, thereby maintaining a positive moral self-image. This means that cheating does not necessarily stem from a lack of moral standards, but from the cognitive ability to “reframe” one’s own actions as morally acceptable.
Thus, Ariely’s work provides an important theoretical framework for a deeper understanding of student behavior observed during online exams. The blurring of boundaries between permissible and prohibited conduct, especially with the advent of digital tools and AI, presents new challenges for maintaining academic integrity.
What significant differences and distinct resilience profiles can be identified between Generation Z and Generation Alpha, considering the influence of gender and the moderating effects of relevant demographic variables? Among quantitative research methods, this study employed a questionnaire survey to examine factors influencing the propensity to cheat, perceptions of the use of AI tools in online exams, and the relationships between demographic characteristics and attitudes toward cheating. The questionnaire was completed by students from the fields of engineering, economics, and geodesy at our university. The survey was conducted online (via Google Forms) during the second semester of the 2024/25 academic year, with the assistance of teaching staff who also provided a brief introduction to the students. A total of 189 valid responses were collected. The topic was analyzed from multiple perspectives (Figure 1).
Source: Author's own compilation
Demographic data: e.g., age, gender, type of residence, level of study, and major. This helps to understand how students’ backgrounds may influence their propensity to cheat.
Attitudes toward cheating: assessment of statements that map students’ beliefs about cheating (e.g., “It is easier to cheat in online exams than in traditional exams”).
Likelihood of cheating methods: examined the probability of various forms of cheating (e.g., use of secondary devices, internet searches, assistance from friends).
University rules and instructor role: questions also addressed how well students know and consider university rules to be fair, and whether clearer expectations from instructors or emphasizing the relevance of the course could reduce cheating tendencies.
Use of AI tools: the questionnaire specifically investigated how acceptable students consider the use of AI tools and how frequently they believe other students use them.
Below, the demographic characteristics of the respondents are presented (Figure 2).
Source: Author's own compilation
The respondents had an average age of 24.1 years (ranging from 18 to 50), with 78.3% falling between 18 and 24 years old and 15.9% being over 30. Among the demographic variables, young (18–24 years old) full-time male students are significantly overrepresented in the sample, accurately reflecting the gender distribution of students in our faculty. Although our findings provide useful insights into students’ perceptions of online exam cheating and the use of AI tools, they should be interpreted with caution. We conducted the survey among students from a single faculty, and participation was voluntary, so our sample is not representative of the wider university population at either the national or international level. We interpret our results primarily as exploratory findings reflecting the characteristics and attitudes of this specific student group. Future research should involve larger and more diverse samples across multiple institutions to examine whether the patterns identified in this study are observable in broader higher education contexts.
Exam mode preference: The majority of respondents (111 students, 58.7%) prefer online exams, while 78 students (41.3%) prefer in-person exams.
Participation in online exams: 104 students (55.0%) participated in online exams 1–5 times, 54 students (28.6%) never participated, 19 students (10.1%) participated 6–10 times, and 12 students (6.3%) participated more than 10 times.
Perceived security of protection: In response to the question, “How secure do you feel the protection against cheating is during online exams compared to traditional exams?” (1 = “Much less secure,” 5 = “Much more secure”), 41.8% of respondents gave a score of 3, 35.4% scored 1–2, and 22.7% scored 4–5. Thus, most respondents rated the security of online protection as moderate or weak.
Discomfort caused by security measures: The majority of respondents (75.1%) reported experiencing discomfort due to proctoring measures (“Yes, sometimes” 29.6%, “Yes, often” 24.3%, “Rarely” 21.2%), while 24.9% never experienced such discomfort.
Likelihood of attempting to cheat (undetected): In response to the question of how likely they would be to attempt cheating if it remained completely unnoticed (1 = “very unlikely,” 5 = “very likely”), 34.9% of respondents gave a score of 3, 40.8% scored 4–5 (19.6% scored 4; 21.2% scored 5), and 24.3% scored 1–2.
Providing and receiving help: 56.6% of respondents (107 students) reported that they had helped someone achieve a better result in an online exam, while 43.4% (82 students) had not. Similarly, 54% (102 students) had received help from others, while 46% (87 students) had not.
Failing an online exam: 137 students (72.5%) had never failed an online exam, 50 students (26.5%) had failed a few times, and 2 students (1.0%) had failed multiple times.
Online vs. traditional exams: 68% of respondents believed that it is easier to cheat in an online exam compared to a traditional one.
For the Likert-scale statements, respondents mostly selected values close to neutral or above. Notably, 64% (scoring 4–5) agreed with the statement “In the hope of receiving a higher scholarship, many students do not shy away from cheating.” Similarly, 72% considered the statement “Out of fear of failing, several students have already resorted to cheating during exams” to be likely (4–5 points). In contrast, agreement was less pronounced with statements such as “Cheating is justifiable if the exam format or content is too difficult” (32% gave 4–5 points) and “If others cheat, it puts me at a disadvantage” (33% gave 4–5 points). A majority of respondents (58%) believe that most instructors strictly punish cheating. Furthermore, 57% of students indicated that lack of time is one of the reasons for cheating, while 58% stated that the instructor’s personality influences their willingness to cheat. With respect to the statement “Cheating is motivated by the need to meet parental or external expectations,” the respondents were nearly evenly divided, with 66 agreeing and 61 disagreeing.
Forms of cheating: For the statements referring to specific forms of cheating (use of unauthorized notes, asking for help, internet searches, use of messaging applications, secondary devices), the most frequently indicated as likely were asking friends or classmates for help (43.4% scored 4–5) and searching for answers on the internet (44.0%). The use of unauthorized notes was considered likely by 41.3% of respondents. Other methods (messaging, secondary devices) were also reported with considerable likelihood (34–40%). Overall, respondents generally considered multiple methods of cheating to be possible.
AI tools in online exams: A majority of respondents (60%) agreed (4–5 points) with the statement “Using AI tools constitutes cheating,” while 16% disagreed. Regarding the likelihood that “other students use AI tools,” 62% of respondents agreed (4–5 points). Concerning the statement “I consider the use of AI tools acceptable,” 34% disagreed (1–2 points), while only 40% regarded it as acceptable (4–5 points). This ambivalence illustrates well the mechanism of “self-justified cheating” described by Ariely and Gino (2012), which suggests that cheating is often not the result of a fully conscious and explicit decision but rather a rationalized behavior shaped by situational context and internal narratives. It was also observed that among students who considered the use of AI tools acceptable (40%), 78.6% (56 individuals) did not strictly regard it as cheating, but rather as a form of learning support. This supports the notion that the further a practice is perceived to be from traditional cheating practices (e.g., crib notes), the easier it becomes to justify it morally. Based on the findings, the online environment, the relatively looser monitoring, and the ambiguous status of AI tools collectively contribute to students perceiving cheating more as a rationalized behavior than as a consciously unethical act. When asked whether universities should more strictly monitor the use of AI in online exams, only 28% agreed, while 39% disagreed. Furthermore, we examined the relationship between AI use and the propensity to cheat. The results showed a strong association: the use of AI tools was closely related to cheating propensity (χ² = 63.30, df = 16, p = 1.44×10⁻⁷), indicating that students who are more inclined to cheat are also more likely to use AI tools.
At the end of the questionnaire, two open-ended questions were included. One asked students about the reasons for cheating, and the other about possible ways to reduce it. We first analyzed the responses related to the causes of cheating using thematic coding. The most frequently occurring words in the texts referring to the reasons for cheating are illustrated in the word cloud in Figure 3.
Source: Author's own compilation
The responses were first labeled with short, descriptive tags (“codes”), which were then grouped into broader themes (Figure 4). The aim of the analysis was to explore the factors underlying online exam cheating and to understand how students rationalize their own behavior.
Source: Author's own compilation
Based on the content analysis of student responses, several interrelated factors can be identified behind online exam cheating. The qualitative data indicate that cheating is not solely the result of individual decisions but also the outcome of organizational, pedagogical, and environmental influences.
Motivational factors show that students decide to cheat under the influence of internal pressures and constraints. The most common reason is lack of time and overload: many prepare for several exams at once while also dealing with other obligations (e.g., work, family). In addition, fear of failure and the need to meet expectations are strongly present, further intensified by the desire to maintain a scholarship or meet family expectations. These factors suggest that for students, cheating is often not an end in itself but rather a forced means of survival and meeting performance requirements.
Environmental factors mainly stem from the specific characteristics of online exams. Students feel that the risk of getting caught is lower in the online environment, as direct supervision and control are missing. The technical features of the online format, such as the simultaneous availability of multiple digital devices, also facilitate opportunities for cheating. In addition, the sense of a safer environment and the normalization of cheating within the community (others doing the same) further increase the willingness to cheat. These factors indicate that cheating is not solely an individual decision but also a consequence of the structural weaknesses of the given exam environment.
Pedagogical factors refer to the educational circumstances that contribute to students’ willingness to cheat. One of the most frequently mentioned reasons is the perceived irrelevance or uselessness of the curriculum. Students often do not see the practical value of certain subjects, which reduces their motivation to prepare honestly. Closely related to this is the perception of unrealistic expectations, when requirements seem disproportionately difficult or excessively high compared to the time available for preparation. Another significant factor is the attitude of instructors. Several students reported that they do not receive enough explanations, support, or practical examples. This creates uncertainty, making cheating a kind of substitute tool. Unclear requirements and insufficient communication further exacerbate the problem, as students often do not know exactly what is expected of them in exams. Pedagogical factors highlight that cheating is often not merely an individual moral decision but a symptom of deficiencies in the educational environment. Transparent requirements, relevant curriculum, and a supportive teaching attitude could reduce the willingness to cheat.
The analysis clearly shows that students’ willingness to cheat is influenced not only by individual decisions but also by the learning environment, teaching methods, and system-level factors. Therefore, measures aimed at reducing cheating require a comprehensive approach. In addition to technical control of the exam environment, increasing the relevance of the curriculum, improving instructors’ pedagogical approaches, and strengthening ethical education are all important tasks.
The following recommendations may help to create a more comprehensive, secure, and student-friendly examination environment:
Thoughtful scheduling and flexibility of exam periods: providing multiple trial and retake opportunities so that students do not perceive failure as an irreversible risk.
Psychological and learning-management support: integrating training, workshops, and stress-management techniques into the educational program.
Security solutions: secure exam browsers, randomization of tasks, screen sharing, and camera use, which reinforce the seriousness of the examination situation.
Honor code: emphasizing students’ ethical responsibility and strengthening community norms.
Relevant and application-oriented curriculum: prioritizing learning content and exam tasks directly related to future professional activities.
Clear exam requirements: communicating transparent and straightforward criteria to reduce misunderstandings and unrealistic expectations.
Varied forms of assessment: introducing project assignments, oral exams, or small-group solutions where simple copying is less applicable.
Instructor attitude and role modeling: delivering material in an understandable and engaging way and emphasizing ethical behavior to increase student commitment.
Students justify cheating not only with rational arguments but also with internal narratives that help them maintain a positive moral self-image. These narratives often serve as self-justifications, allowing students to perceive their actions as morally acceptable while minimizing feelings of guilt. According to Ariely’s “fudge factor” theory, people tend to rationalize rule-breaking in ways that preserve their moral values and self-concept, which aligns well with the narratives surrounding online exam cheating. It is important to emphasize that these narratives are not necessarily conscious but emerge within the context of the situation. Based on the responses, the following student narratives were identified (Figure 5).
Source: Author's own compilation
Students perceive that they have no other choice, or that cheating is necessary given the circumstances. This narrative often builds on lack of time, overload, and fear of failure. “Fear of failing, expectations, stress, lack of time.”
Students compare their behavior to that of others, justifying it by noting that others do the same or that the environment makes cheating easier. This narrative serves to rationalize actions based on social norms and situational factors. “Others are doing it, so why not?”
Students rationalize cheating through a cost–benefit logic, where the advantages of the action (better grades, easier passing) outweigh the risks. “Low risk, high reward.”
In this narrative, students question the relevance of the course content and justify cheating by arguing that the subject or instruction does not warrant the effort invested. “Most courses have no professional value, so cheating carries less weight.”
Students use the opportunities provided by modern tools and the online environment as self-justifications. This includes, for example, the use of AI tools as learning support and the easier access to online exam materials. “Easier solution, doesn’t require studying.”
Student narratives indicate that cheating is not merely about circumventing rules, but is shaped by internal stories and self-justifications. For educational institutions, this means that combating cheating does not end with the implementation of strict rules and supervisory measures. More effective strategies may include ethical dialogue, understanding students’ motivations and internal narratives, instructors’ role modeling, and enhancing the relevance and transparency of the curriculum. In this way, institutions can not only reduce the opportunities for cheating but also influence students’ internal acceptance, fostering genuine learning and responsible behavior.
The other open-ended question asked students for their ideas on how to reduce cheating in online exams. The most frequently occurring words related to this topic in the responses are shown in the word cloud in Figure 6.
Source: Author's own compilation
Students approach the reduction of cheating from several perspectives (Figure 7). These can be grouped into three main categories:
Technological solutions
Exam organization and format
Instructor attitude and pedagogy
Source: Author's own compilation
Based on students’ opinions, reducing cheating in online exams requires the combined application of several factors (Figure 8). Technical tools, such as screen sharing, webcams, or secure browsers, are useful but not sufficient on their own. Proper exam organization and task design are also important, including optimizing time limits, using varied task types, and incorporating project-based and oral exams. Additionally, the pedagogical approach is crucial: fostering honesty and designing assessments that measure genuine knowledge are at least as important as technical controls. The responses also indicate that completely eliminating cheating is nearly impossible; therefore, rethinking the purpose and value of exams, improving the quality of the curriculum, and authentically assessing competencies are more effective long-term strategies for reducing cheating and maintaining the integrity of education.
Source: Author's own compilation
We examined whether students’ background has any influence on cheating. Figure 9 shows which demographic characteristics are significantly related to specific aspects of cheating.
Source: Author's own compilation
The results show that among male students, a significantly higher proportion consider cheating acceptable under certain circumstances (e.g., when the exam is perceived as too difficult). In contrast, female students tend to reject the justifiability of cheating to a greater extent, with a larger share firmly opposing it. This suggests that male students are more likely to view cheating as a rational, situational decision, while female students demonstrate stronger adherence to normative attitudes. Thus, gender differences are significant in the moral evaluation of cheating.
The type of residence is also significantly associated with perceptions of the feasibility of online cheating. Students living in smaller settlements are more likely to believe that cheating online is easier, whereas students residing in the capital tend to be more skeptical in this regard. This difference may be explained by variations in educational experiences and access to online examinations. Students from smaller towns or villages are less exposed to the technological and organizational control mechanisms that are more common at larger universities, which may reinforce their perception that online exams are “easier to circumvent.”
Students’ propensity to cheat—particularly in the hypothetical scenario where the risk of being caught is completely eliminated—shows a significant association with parental educational attainment. Among students whose parents have lower levels of education, a higher proportion indicate that they would likely cheat under such circumstances, whereas children of parents with higher education more often display a rejecting attitude. This finding suggests that the family’s cultural capital and value system regarding learning may play a crucial role in shaping students’ ethical decisions. Children of highly educated parents are presumably more likely to bring stronger internalized norms into the educational environment, which in turn reduces their inclination to engage in cheating.
Based on the analysis, attitudes toward cheating are shaped by several demographic factors. Male students tend to adopt a more flexible moral stance on cheating, while female students are more consistent in rejecting it. Students from smaller settlements are more likely to believe that cheating is easier in an online environment compared to those living in the capital. Moreover, the children of parents with lower educational attainment are more inclined to cheat in a risk-free situation, whereas those whose parents hold higher degrees are more likely to reject such behavior. These differences highlight that attitudes toward cheating are not merely the result of individual decisions, but are closely linked to social background and processes of socialization.
We conducted a hierarchical cluster analysis based on the Likert-scale variables related to cheating attitudes, willingness to cheat, and the use of AI tools. This allowed us to distinguish three groups of students (Figure 10).
Source: Author's own compilation
Members of this group relatively strictly reject cheating and the acceptance of aids (including AI). On average, they tend to disagree with statements suggesting that cheating is acceptable or easier in online exams (the corresponding means are around 2–3). They also regard AI tools as cheating and do not consider their supervision important (low means: AI cheating = 2.84; monitoring = 2.02). Overall, this group shows the lowest willingness and acceptance regarding both cheating and the use of AI.
Students in this group are highly aware that cheating is easier in online exams (mean = 4.48) and perceive external factors as influential (e.g., if others cheat, mean = 3.83). At the same time, they strongly value rules and strict monitoring, believing that instructors strictly punish cheating (mean = 3.81), and they firmly support monitoring the use of AI tools (mean = 3.88). Interestingly, this group clearly considers AI tools to be cheating (mean = 4.55), but they acknowledge that others’ use of AI does not increase their own willingness to use it. Overall, this cluster shows moderate acceptance of cheating but is characterized by a strict rule-following orientation.
This largest group shows the highest averages on statements indicating a willingness to cheat. They strongly agree that others’ cheating increases their own likelihood of cheating (mean = 3.49), that fear intensifies the tendency to cheat, and that cheating is more likely in higher-credit courses (means around 3.5–4). Regarding AI tools, they are relatively lenient. Although they acknowledge that “using AI is cheating” (mean = 4.08) and see others’ use as an incentive, they do not view monitoring as urgent (mean for AI monitoring = 2.70). Overall, this cluster is the most permissive toward both cheating and the use of technological aids.
These distinct profile patterns indicate that students can be grouped into three clearly differentiated clusters based on their attitudes toward cheating and the use of technological aids. This insight can support the development of more tailored anti-cheating strategies and educational oversight. The findings suggest that institutions should apply differentiated approaches. The first group can be reinforced in their current stance, the second group benefits from stronger regulation and transparency, while the third group requires motivational and ethical education tools, since strictness alone is likely to be less effective.
The spread of online examinations is redefining the concept of cheating in higher education. With the emergence of artificial intelligence, students now have access to tools whose use is often difficult to separate from the application of genuine knowledge. The results of the questionnaire survey show that attitudes toward cheating are not uniform. 40% of respondents (74 students) interpret AI tools as a means of supporting learning, while 60% (115 students) tend to or fully reject their use, regarding them as cheating. This division illustrates how, in the new technological environment, the concept of cheating is becoming increasingly nuanced, and ethical boundaries more uncertain, particularly in the case of online examinations.
The demographic analysis revealed that attitudes toward cheating are also shaped by factors such as gender, type of residence, and parents’ educational attainment. Male students tend to evaluate the morality of cheating more flexibly, while female students are more consistent in their rejection of it. Students from smaller towns and villages are more likely to believe that cheating is easier online than those living in the capital. Similarly, students whose parents have lower educational attainment are more likely to admit they would cheat in risk-free situations, whereas the children of parents with higher education tend to reject such behavior. This highlights that attitudes toward cheating are closely linked to social background and patterns of socialization.
Cluster analysis identified three clearly distinguishable student groups:
“Strict Rejecters” (58 students), who show the lowest willingness to cheat and the least acceptance of AI.
“Rule-Oriented Realists” (42 students), who recognize the opportunities for cheating but believe in rules and strongly support strict monitoring.
“Permissive Pragmatists” (89 students), who are the most tolerant toward cheating and the use of AI tools, and are less supportive of monitoring.
These profiles suggest that differentiated strategies are required to address student attitudes toward cheating and technological aids. The first group should be supported in maintaining their current stance, the second group requires stronger regulation and transparency, while the third group may benefit more from motivational and ethical education measures, as stricter control alone is unlikely to be effective.
This diversity and uncertainty point to the need for higher education institutions to rethink academic integrity norms in light of the challenges posed by the digital era. Clarifying the perception of AI use, redefining ethical frameworks, and shaping student attitudes are essential for safeguarding the credibility and value of knowledge in higher education. Universities must proactively respond to the challenges of technological development by developing comprehensive guidelines that balance the encouragement of innovation with the preservation of academic integrity. This includes the responsible integration of AI-based tools into education, the updating of examination procedures and regulations, and the regular training of both students and instructors in the ethical and conscious use of technology.
Based on the findings, three key directions can be identified for higher education institutions:
Developing clear institutional policies regarding the acceptable use of AI tools in assessments.
Redesigning assessment methods toward more authentic and competence-based evaluation.
Strengthening ethical education and dialogue about responsible technology use.
The findings contribute to a better understanding of how university students perceive online exam cheating and the role of artificial intelligence in academic dishonesty in digital assessment environments.
Amzalag, M., Shapira, N., & Dolev, N. (2022). Two sides of the coin: lack of academic integrity in exams during the corona pandemic, students' and lecturers' perceptions. Journal of Academic Ethics, 20, 243–263. https://doi.org/10.1007/s10805-021-09413-5 (Last download: 09/15/2025)
Ariely, D. (2012). The (Honest) Truth About Dishonesty – How We Lie to Everyone – Especially Ourselves? HarperCollins Publishers, 2012, 304 pp.
Asogwa, V. C., Isiwu, E. C., & Nwakpadolu, G. M. (2022). Artificial Intelligence in Education: Benefits and Risks. Propellers Journal of Educational Research and Theory, 1(1), 1–18.. https://ijvocter.com/pjert/article/view/8 (Last download: 09/15/2025)
Cantiello, J., & Geschke, R. H. (2024). Preventing Academic Dishonesty in Online Courses: Best Practices to Discourage Cheating. Journal of Health Administration Education, 40(2), 205–230.
Coghlan, S., Miller, T., & Paterson, J. (2021). Good Proctor or “Big Brother”? Ethics of Online Exam Supervision Technologies. Philosophy & Technology, 34, 1581–1606. https://doi.org/10.1007/s13347-021-00476-1 (Last download: 09/10/2025)
Corrigan-Gibbs, H., Gupta, N., Northcutt, C., Cutrell, E., & Thies, W. (2015). Deterring Cheating in Online Environments. ACM Trans. Comput.-Hum. Interact., 22(6), Article 28, 23 pages. https://doi.org/10.1145/2810239 (Last download: 09/15/2025)
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2023). Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International, 61(2), 228–239.
Fraser, K. C., Dawkins, H., & Kiritchenko, S. (2025). Detecting AI-Generated Text: Factors Influencing Detectability with Current Methods. Journal of Artificial Intelligence Research. 82. 2233-2278. https://doi.org/10.1613/jair.1.16665 (Last download: 09/15/2025)
Garcia, M. B., Rosak-Szyrocka, J., Yilmaz, R., Metwally, A. H. S., Acut, D. P., Ofosu-Ampong, K., Erdoğdu, F., Fung, C. Y., & Bozkurt, A. (2025). Rethinking Educational Assessment in the Age of Generative AI: Actionable Strategies to Mitigate Academic Dishonesty. Pitfalls of AI Integration in Education: Skill Obsolescence, Misuse, and Bias, 1-24. https://doi.org/10.4018/979-8-3373-0122-8.ch001 (Last download: 09/15/2025)
Gino, F., & Ariely, D. (2012). The dark side of creativity: Original thinkers can be more dishonest. Journal of Personality and Social Psychology, 102(3), 445–459. https://doi.org/10.1037/a0026406 (Last download: 09/15/2025)
Kell, C. M., Thandar, Y., Bhundoo, A. K., Haffejee, F., Mbhele, B., & Ducray, J. (2025). Academic integrity in the information age: insights from health sciences students at a South African University. Journal of Applied Research in Higher Education, 17(7), 16–28. https://doi.org/10.1108/JARHE-12-2023-0565 (Last download: 09/10/2025)
International Center for Academic Integrity (ICAI). (n.d.). Home Page. URL: https://www.academicintegrity.org/aws/ICAI/pt/sp/home_page (Last download: 09/10/2025)
Lancaster, T., & Cotarlan, C. (2021). Contract cheating by STEM students through a file sharing website: a Covid-19 pandemic perspective. International Journal for Educational Integrity, 17(3).
Lund, B. D., Lee, T. H., Mannuru, N. R., & Arutla, N. (2025). AI and academic integrity: Exploring student perceptions and implications for higher education. Journal of Academic Ethics. 23, 1545–1565. https://doi.org/10.1007/s10805-025-09613-3 (Last download: 09/15/2025)
Luo, J. (2024). A critical review of GenAI policies in higher education assessment: A call to reconsider the “originality” of students’ work. Assessment & Evaluation in Higher Education, 49(5), 651–664. https://doi.org/10.1080/02602938.2024.2309963 (Last download: 09/15/2025)
Maleki, A. (2025). Mindset Matters More than You Think: Investigating Psychological Reasons Behind Online Exam Cheating Behaviors among EFL Learners in Higher Education. Journal of Academic Ethics, 23, 405–422. https://doi.org/10.1007/s10805-024-09591-y (Last download: 09/15/2025)
Malesky, A., Grist, C., Poovey, K., & Dennis, N. (2021). The Effects of Peer Influence, Honor Codes, and Personality Traits on Cheating Behavior in a University Setting. Ethics & Behavior, 32(1), 12–21. https://doi.org/10.1080/10508422.2020.1869006 (Last download: 09/12/2025)
Mutimukwe, C., Viberg, O., McGrath, C., & Cerratto-Pargman, T. (2025). Privacy in online proctoring systems in higher education: Stakeholders’ perceptions, awareness and responsibility. Journal of Computing in Higher Education. https://doi.org/10.1007/s12528-025-09461-5 (Last download: 09/12/2025)
Oravec, J. A. (2023). Artificial Intelligence Implications for Academic Cheating: Expanding the Dimensions of Responsible Human-AI Collaboration with ChatGPT and Bard. Journal of Interactive Learning Research, 34(2), 213-237. https://doi.org/10.70725/304731gmmvhw (Last download: 08/15/2025)
Rüth, M., Jansen, M., & Kaspar, K. (2024). Cheating behaviour in online exams: On the role of needs, conceptions and reasons of university students. Journal of Computer Assisted Learning (JCAL), 40(5), 1987-2008.
Stanoyevitch, A. (2024). Online assessment in the age of artificial intelligence. Discov Educ, 3, 126. https://doi.org/10.1007/s44217-024-00212-9 (Last download: 09/12/2025)
Waltzer, T., & Dahl, A. (2022). Why do students cheat? Perceptions, evaluations, and motivations. Ethics & Behavior, 33(2), 130–150.
Zdravkova, K. (2023). Evolution of academic dishonesty in computer science courses. 9th International Conference on Higher Education Advances. https://doi.org/10.4995/HEAd23.2023.16081 (Last download: 09/15/2025).