Generative AI for Academic Purposes: Comparing Chinese and Malaysian University Students’ Knowledge,Willingness,and Concerns

Abstract

This study explores Chinese and Malaysian university students’ perceptions of generative AI (GenAI) tools for academics, with a focus on their knowledge, willingness to use, and concerns. A quantitative, cross-sectional survey design was employed, and data were collected from 400 students through a structured online questionnaire. Both groups demonstrated high awareness of GenAI, with cultural differences shaping their perceptions. Knowledge is positively correlated with usage, whereas willingness is primarily influenced by concerns. Chinese students emphasized ethical considerations, particularly academic integrity, whereas Malaysian students prioritized the reliability of AI-generated content. Common concerns included the impact of GenAI on critical thinking, human interaction, and responsible academic use. The selection of China and Malaysia provides insights into two distinct and evolving educational contexts. To promote responsible GenAI use, institutions should address both its advantages and ethical implications while adapting policies to diverse cultural contexts. The implications and directions for future research are discussed.

Plain Language Summary

How Chinese and Malaysian University Students View AI in Education: Knowledge, Willingness, and Concerns

This study explores how university students in China and Malaysia view the use of generative AI (GenAI) tools in their studies. It looks at what they know about GenAI, how willing they are to use it, and what concerns they have. A survey of 400 students (200 from each country) showed that while both groups are familiar with GenAI, their views differ because of cultural influences. Students with more knowledge about GenAI tend to use it more often. However, their willingness to rely on it depends on their concerns. Chinese students worry more about ethical issues such as cheating and fairness, while Malaysian students focus on whether AI-generated information is accurate and trustworthy. Both groups share common worries about GenAI affecting students’ critical thinking, reducing real human interaction, and being used irresponsibly in academics. By comparing these two groups, the study shows that universities should create clear and fair rules for using AI in education. Teachers and administrators need to balance the benefits of GenAI with ethical concerns and adapt these rules to different cultural settings. Overall, the findings help guide how AI can be used responsibly and effectively in higher education.

Keywords

GenAI perceptions knowledge willingness cross-country comparison

Introduction

The rapid advancement of generative artificial intelligence (GenAI) tools, such as ChatGPT is reshaping knowledge production and academic practices in higher education. These tools enable students to generate ideas, refine writing, and enhance productivity. At the same time, their increasing use has raised concerns regarding ethical implications (H. Wang et al., 2024), potential misuse (Kutty et al., 2024), and their impact on critical thinking skills (Premkumar et al., 2024). As GenAI becomes embedded in university learning environments, understanding how students perceive and engage with this technology has become an important issue for educational innovation and policy.

Existing research has primarily focused on pedagogical benefits (Farrelly & Baker, 2023), ethical challenges of GenAI adoption (Kumar et al., 2024), and equity-related challenges (Samala et al., 2025). While studies highlight issues such as plagiarism (Hughes et al., 2025), and overreliance on algorithmic feedback (Kohnke, 2024), they remain largely context-specific and do not adequately examine how national policy environments and educational cultures shape students’ perception and usage intentions. In particular, research investigating how students’ knowledge of GenAI, willingness to use it for academic purposes, and concerns about its implications vary across different higher education systems, is scarce. This represents a significant gap, as perceptions of emerging technologies are often influenced by policy regulation, institutional norms, and socio-cultural attitudes toward innovation.

China and Malaysia provide a meaningful basis for comparative analysis. China’s proactive national AI agenda and digital education reforms have accelerated the integration of AI-driven tools in universities (Cheng & Zeng, 2023; Khanal et al., 2025), supported by policy frameworks such as the “Education Informatization 2.0 Action Plan,” which emphasize both innovation and ethical governance (Ma, 2025). In contrast, Malaysia’s higher education system is gradually incorporating GenAI through initiatives aligned with the Malaysia Artificial Intelligence Roadmap 2021–2025, focusing on AI literacy and adaptive learning (Jaaffar et al., 2021). These differing policy trajectories and levels of institutional integration provide a valuable opportunity to examine whether and how national contexts influence students’ knowledge, willingness, and concerns regarding GenAI use.

This study addresses the identified gap by comparatively examining university students’ knowledge of GenAI, willingness to use it for academic purposes, and concerns. By analyzing similarities and differences across these two contexts, the study offers some empirical evidence on how policy environments and educational cultures may shape student engagement with emerging AI technologies. The findings extend current literature beyond single-country investigations and offer context-sensitive insights for policymakers and higher education institutions (HEIs) seeking to develop responsible and effective GenAI integration strategies.

Literature Review

Opportunities and Challenges of GenAI Tools

Integrating generative AI (GenAI) into higher education presents both pedagogical opportunities and significant challenges. Existing literature recognizes the potential of GenAI to support personalized learning (Laak et al., 2024), facilitate formative assessment (Mills et al., 2025), and foster creative problem-solving (Hoßbach & Isaksen, 2025). When implemented responsibly, these tools may improve efficiency and expand access to academic resources.

At the same time, concerns persist regrading academic integrity, ethical use, and overreliance on AI systems (Adamakis & Rachiotis, 2025). Scholars caution that improper implementation may compromise originality, reinforce algorithmic bias (Francis et al., 2024), and weaken independent thinking (Gonsalves, 2026). These tensions suggest that GenAI integration requires institutional frameworks that balance innovation with ethical governance and responsible AI literacy.

Ethical and Academic Integrity Concerns

Academic dishonesty remains a dominant concern in GenAI adoption. Yparrea and Hernández-Rodríguez (2024) noted that GenAI tools may facilitate plagiarism by generating content that students may misrepresent as original work. However, Duah and McGivern (2024) argued that many students use GenAI as a learning aid rather than for misconduct, suggesting that the issue may lie more in institutional regulation than in the technology itself.

The absence of clear policy frameworks further complicates ethical implementation. Shailendra and Rudolph et al. (2024) highlighted that unclear guidelines create uncertainty among students and faculty, leading to inconsistent practices. Moreover, Zlotnikova and Hlomani (2025) emphasized data security risks and algorithmic bias, underscoring the need for transparent governing mechanisms. Together, these studies indicate that ethical concerns surround GenAI are not uniform but mediated by institutional preparedness and regulatory clarity.

Impact on Learning and Student Engagement

Despite ethical concerns, GenAI tools have demonstrated tangible academic benefits. Lee and Moore (2024) reported improvements in research efficiency and resource access, while Liu et al. (2024) found that tools such as ChatGPT support academic writing, grammar, and reading development. However, Liu et al. (2024) also observed limited effectiveness in fostering higher-order thinking, suggesting that GenAI’s strengths lie primarily in procedural and linguistic enhancement rather than deep cognitive engagement.

Chambers and Owen (2024) further noted that AI-powered tools assist students in structuring academic work and understanding course materials. Nonetheless, Almassaad et al. (2024) cautioned that excessive reliance may reduce critical thinking and interpersonal interaction. Collectively, the literature suggests that, GenAI functions most effectively as a complementary tool rather than a substitute for human mentorship and reflective learning.

Students’ Perceptions of GenAI Tools

Students’ perceptions of GenAI range from enthusiasm to skepticism. Duah and McGivern (2024) found that students generally perceive GenAI as enhancing academic performance rather than facilitating misconduct. However, limited understanding of accuracy and bias remains a concern (Zhai et al., 2024).

Similarly, Liu et al. (2024) identified a dual perception: students value efficiency and support but question reliability. Daher and Hussein (2024) emphasized efficiency as a key driver of positive attitudes, particularly among students managing heavy workloads. Kim et al. (2025) reported that college students, particularly those in writing-intensive courses, appreciate GenAI tools for idea generation and content structuring, whereas students in research-oriented disciplines express reservations about it limitations in handling complex, abstract, or interdisciplinary topics. These findings suggest that perceptions are shaped not only by technological functionality but also by academic demands.

Disciplinary Differences

Disciplinary context significantly shapes GenAI usage. K. D. Wang et al. (2025) reported that STEM students frequently rely on AI tools for problem-solving, generating solutions, summarizing readings, and exploring related topics, viewing these tools as essential technical aids. In contrast, students in language-focused programs, including those in English for Academic Purposes (EAP) programs (Kohnke, 2024), use GenAI for linguistic clarification and communication support.

Similarly, Tantivejakul et al. (2024) and Suonpää et al. (2024) found that communication and business students leverage GenAI for translation, writing refinement, and concept clarification. In creative disciplines, however, skepticism prevails due to concerns about originality and privacy (Shen et al., 2025). These variations demonstrate that GenAI’s perceived value depends on disciplinary priorities, reinforcing the need for context-sensitive integration strategies.

Cultural and Contextual Effects

Beyond disciplinary differences, cultural and national contexts influence perceptions of GenAI. Aldossary et al. (2024) found that Saudi students viewed GenAI as a tool for enhancing learning outcomes and self-efficacy, reflecting positive institutional attitudes toward educational technology. Chan and Hu (2023) emphasized that adoption patterns vary across educational systems, shaped by cultural norms, teacher authority structures, and national technology agendas. However, despite growing interest in contextual influences, comparative research examining how students’ knowledge, willingness, and concerns vary across national systems remains limited.

Theoretical Foundation

This study draws on the Technology Acceptance Model (TAM; Davis et al., 2024) and the Unified Theory of Acceptance and Use of Technology (UTAUT; Venkatesh et al., 2003) to frame its examination of students’ knowledge, willingness, and concerns. These frameworks explain how perceptions of usefulness, ease of use, and social influence shape technology adoption.

In this study, knowledge reflects students’ awareness and understanding of GenAI’s capabilities, while willingness represents their readiness to integrate it into learning and academic work. Meanwhile, concerns function as potential barriers linked to ethical risks, academic integrity, and fairness considerations. Integrating TAM and UTAT enables a socio-technical interpretation of GenAI adoption, supporting comparative analysis across educational contexts.

Research Gap and Questions

Although prior studies have examined GenAI’s benefits, risks, disciplinary variations, and contextual influences, most investigations remain single-country descriptive. Given that policy environments and educational cultures shape technology adoption, a cross-national comparison could offer insights on how contextual factors influence students’ engagement with GenAI.

China and Malaysia provide contrasting yet comparable cases, characterized by differing AI policy trajectories and educational governance models. Examining students’ perspectives in these contexts addresses a significant gap in comparative GenAI research.

Based on these considerations, this study addressed the following research questions (RQs):

(1) What are the perceptions of (a) knowledge, (b) willingness, and (c) concerns regarding GenAI tools among Chinese and Malaysian students?

(2) What are the (a) benefits and (b) challenges perceived by students regarding the use of GenAI tools? How do these perceptions differ between Chinese and Malaysian students?

(3) What insights can be drawn from the data comparing the perspectives of Chinese and Malaysian students on the use of GenAI tools in their academic lives?

By addressing this gap, this study provides a nuanced understanding of students’ perceptions of GenAI tools, enriching the ongoing discourse on their integration into higher education. Furthermore, it offers insights into how these perceptions differ across cultural and educational contexts, providing implications for the design and implementation of AI-supported academic frameworks.

Methodology

This study employed a survey design to collect data from university students in China and Malaysia. The data collection was conducted through an online questionnaire, which was adapted from Chan and Hu’s (2023) study. To ensure linguistic and contextual appropriateness, the questionnaire was administered in two language versions: English for Malaysian students and Chinese for students in Mainland China. The Chinese version was translated from the original English instrument using a forward–backward translation procedure involving two bilingual experts, with minor wording adjustments made to ensure cultural clarity and conceptual equivalence.

The final instrument comprised 24 items, consisting of 18 structured items on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree,” four items on demographic information and GenAI usage, and two open-ended questions. The survey covered key topics, including students’ knowledge of GenAI tools such as ChatGPT, their willingness to use these tools, and their concerns regarding the same. Cronbach’s alpha was computed only for the 18 Likert-scale. The demographic and open-ended questions were excluded from the reliability analysis.

Prior to the main data collection, pilot testing was conducted separately for both versions (n = 30 for each group) to examine reliability and comprehensibility. The Cronbach’s α values for all constructs were calculates in SPSS (see Table 1).

Table 1.

Instrument Reliability (English and Chinese Versions).

Language	Cronbach’s alpha	No. of items
English	.876	18
Chinese	.865	18

As seen in Table 1, Cronbach’s Alpha values for both instruments are over .80, indicating satisfactory internal consistency for both the English and Chinese instruments. Content validity for both versions was established through expert review by three specialists in educational technology and higher education, who assessed the relevance, clarity, and representativeness of the items. These steps provided strong evidence of the instruments’ reliability and validity for use among university students in both contexts.

To facilitate data collection, the questionnaire was created via Microsoft Forms and Google Forms, as the latter is not accessible in China. Corresponding QR codes were generated to streamline the distribution process among students. The target population comprised undergraduate students from a private university in China and a private university in Malaysia, focusing on the Faculty of Education. These institutions were selected purposively based on their active engagement in digital and technology-enhanced learning initiatives, making them relevant contexts for examining students’ interaction with emerging technologies such as GenAI. Private universities in both countries have demonstrated relative institutional flexibility in piloting and adopting innovative educational technologies, thereby providing an appropriate setting to explore students’ knowledge, willingness, and concerns regarding GenAI use.

The two universities were selected to ensure structural comparability in terms of institutional type (private), disciplinary focus (Education), and student demographic profiles (e.g., age range, academic background, and program structure). By controlling for these institutional and disciplinary characteristics, the study aimed to reduce extraneous variability and strengthen the validity of cross-national comparison. Thus, the comparison focuses on how students within similar institutional environments respond to GenAI under different national policy and cultural conditions.

Furthermore, convenience sampling method was adopted. While this provided a practical means of obtaining data from two comparable contexts, it also limits the representativeness and generalizability of the findings. As such, the results should be interpreted with caution. Specifically, the data were collected solely from Education students at one private university in each country; therefore, the findings are indicative of these specific institutional and disciplinary contexts rather than all university students in China and Malaysia. Accordingly, the study should be viewed as a comparative case study offering contextual insights rather than generalizable conclusions. The authors also acknowledge that the reliance on private universities may not fully represent the broader student population, particularly in China where public institutions dominate the higher education landscape. However, the findings provide valuable preliminary insights into university students’ awareness, willingness, and concerns regarding GenAI serving as a foundation for future research involving larger and more diverse samples.

Recruitment was conducted through professional and extended networks of the researchers within their respective universities. An information sheet and a consent form were embedded in the survey, requiring participants to select either “agree” or “disagree” to proceed. Participation in the study was entirely voluntary. No personal identifiers such as names, student IDs, or email addresses were collected, and no face-to-face interaction occurred between the researchers and respondents.

Formal ethics approval from an institutional review board was not required, as the study did not involve any intervention, or collection of sensitive or identifiable personal data. Furthermore, the research was conducted in full accordance with established ethical principles for educational research, including respect for participants’ autonomy, integrity, confidentiality, and anonymity.

Finally, 400 undergraduate students participated in the study, with 200 respondents from each country. Descriptive analysis was employed to analyze the survey data, while thematic analysis was used to examine the open-ended responses. NVivo 14 was used to facilitate the thematic analysis. Codes were initially generated, and in instances of disagreement between the researchers, discussions were held to resolve discrepancies and reach a consensus. A codebook was then developed on the basis of this agreement and used to code the remaining responses. The findings are presented in the next section.

Findings

This section presents the findings of the study, highlighting key insights into students’ knowledge, willingness, and concerns regarding the use of GenAI tools in academic contexts. The results also shed light on the correlation between these factors and how they vary across different student groups.

Demographic Information

Table 2 presents the demographic information of the respondents. The characteristics include gender, academic year, and the frequency of using GenAI tools like ChatGPT.

Table 2.

Demographic Information of Respondents.

	China		Malaysia
Characteristics	n	%	n	%
Gender
Male	98	49.00	83	41.50
Female	102	51.00	117	58.50
Academic year
Year 1	11	5.50	9	4.50
Year 2	29	14.50	31	15.50
Year 3	58	29.00	64	32.00
Year 4	102	51.00	96	48.00
How often do you use GenAI tools like ChatGPT
Never	1	0.50	0	0.00
Rarely	28	14.00	8	4.00
Sometimes	43	21.50	32	16.00
Often	98	49.00	99	49.50
Always	30	15.00	61	30.50

Note. n represents the number of respondents.

As shown in Table 2, among the Chinese respondents, the majority were female (51%, n = 102), whereas males constituted 49% (n = 98). Similarly, the Malaysian sample showed a greater proportion of females (58.50%, n = 117) than males (41.5%, n = 83). This indicates a similarity in gender representation between the two groups, with a higher proportion of female respondents than male respondents among both Chinese and Malaysian university students.

The distribution of respondents across academic years was notably similar. Both China and Malaysia, Year 4 students made up the largest group, accounting for 51% (n = 102) and 48% (n = 96) of the respondents, respectively. They were followed by Year 3 students (29% and 32%), Year 2 students (14.5% and 15.5%), and Year 1 students (5.5% and 4.5%). This pattern suggests a comparable representation of academic levels among respondents in both countries, focusing predominantly on students in their final years of study.

The frequency of using GenAI tools such as ChatGPT showed notable differences between respondents from China and those from Malaysia. In both groups, most respondents reported using GenAI tools “often,” with 49% (n = 98) in China and 49.5% (n = 99) in Malaysia. However, the second most common response differed between the two groups. In China, 21.5% (n = 43) indicated that they use GenAI tools “sometimes,” whereas in Malaysia, 30.5% (n = 61) reported that they “always” use these tools, reflecting a higher frequency of consistent use among Malaysian respondents. Conversely, the proportion of respondents who “rarely” used GenAI tools was more common in China (14%, n = 28) than in Malaysia (4%, n = 8). Notably, only 0.5% (n = 1) of the respondents in China indicated that they “never” use GenAI tools, whereas no respondents in Malaysia selected this option. These trends suggest that while GenAI tools are widely adopted by both groups, Malaysian respondents show a slightly stronger inclination toward frequent and consistent use.

Perceptions of Knowledge, Willingness, and Concerns

This section addresses RQ 1: What are the perceptions of (a) knowledge, (b) willingness, and (c) concerns regarding GenAI tools among Chinese and Malaysian students?

Knowledge of GenAI Tools

To address RQ 1(a), mean scores were interpreted following Chan and Hu’s (2023) scale, where mean values of 1.00 to 1.80 indicate strongly disagree, 1.81 to 2.60 as disagree, 2.61 to 3.40 as neutral, 3.41 to 4.20 as agree, and 4.21 to 5.00 as strongly agree.

As shown in Table 3, both Chinese and Malaysian students generally reported high levels of agreement regarding their knowledge of GenAI tools. Malaysian students reported slightly higher mean scores across most statements than their Chinese counterparts.

Table 3.

Students’ Knowledge of GenAI.

Statement	China		Malaysia
Statement	Mean	SD	Mean	SD
I understand generative AI technologies like ChatGPT has limitations in its ability to handle complex tasks	4.42	0.60	4.51	0.63
I understand generative AI technologies like ChatGPT can generate an output that is factually inaccurate	4.20	0.70	4.17	0.70
I understand generative AI technologies like ChatGPT can generate outputs that are out of context or inappropriate	4.19	0.73	4.27	0.75
I understand generative AI technologies like ChatGPT can exhibit biases and unfairness in their output	4.08	0.75	4.21	0.71
I understand generative AI technologies like ChatGPT may rely too heavily on statistics, which can limit their usefulness in certain contexts	4.15	0.80	4.21	0.77
I understand generative AI technologies like ChatGPT have limited emotional intelligence and empathy, which can lead to output that is insensitive or inappropriate	4.12	0.81	4.24	0.80

Furthermore, both groups of respondents strongly understood GenAI tools, as reflected by mean (M) scores ranging from 4.08 to 4.42 for Chinese students and 4.17 to 4.51 for Malaysian students. These scores indicate a high level of agreement with the statements assessing knowledge about GenAI tools.

Specifically, students from both groups had the highest mean score for the statement, “I understand that generative AI technologies such as ChatGPT have limitations in their ability to handle complex tasks” (M = 4.42 for Chinese students, M = 4.51 for Malaysian students), highlighting a shared recognition of this limitation.

However, differences emerged in the statements receiving the lowest mean scores. Among Chinese students, the lowest score was associated with the statement, “I understand generative AI technologies such as ChatGPT can exhibit biases and unfairness in their output” (M = 4.08), suggesting comparatively less agreement with this aspect. In contrast, Malaysian students reported the lowest score for the following statements: “I understand generative AI technologies such as ChatGPT can generate output that is factually inaccurate” (M = 4.17), indicating a slightly lower perception of this limitation compared to other statements.

Furthermore, as seen in Table 4, the results revealed a positive but modest correlation between students’ knowledge of GenAI tools and their frequency of use in both Chinese and Malaysian contexts. For Chinese students, the Pearson correlation coefficient is .230, with statistical significance at the .01 level (p < .05), indicating that as their knowledge of GenAI tools increases, and their frequency of use also tends to rise modestly. Similarly, for Malaysian students, the Pearson correlation coefficient is .212, which is also statistically significant at the .01 level (p = .01), suggesting a comparable pattern where greater knowledge of GenAI tools is modestly associated with increased frequency of use.

Table 4.

Correlation Between Knowledge of GenAI Tools and Frequency of Use.

Students	r	p-Value	Interpretation
Chinese	.230	.001	Positive, modest correlation
Malaysian	.212	.010	Positive, modest correlation

Note. Significance at p < .05.

Lastly, an independent-samples t-test was conducted to compare students’ knowledge of GenAI tools between Chinese and Malaysian participants.

As seen in Table 5, an independent-samples t-test indicates no statistically significant difference (p = .123) in knowledge of GenAI between Chinese (M = 4.200, SD = 0.454) and Malaysian students (M = 4.269, SD = 0.430). The effect size is small (d = 0.160), suggesting that the difference between groups is negligible.

Table 5.

Comparison of Knowledge of GenAI Between Chinese and Malaysian Students.

Variable	Group	N	Mean	SD	t(df)	p	Cohen’s d
Knowledge	China	200	4.200	0.454	–1.545 (398)	.123	0.160
	Malaysia	200	4.269	0.430			0.160

Willingness to Use GenAI Tools

To address Research Question 1(b), the mean scores were analyzed to assess the willingness of Chinese and Malaysian students to use GenAI tools, as presented in Table 6. Both groups display a generally high level of willingness, with mean scores ranging from 4.05 to 4.41 for Chinese students, and 4.13 to 4.45 for Malaysian students, indicating a strong interest in integrating GenAI tools into their academic practices.

Table 6.

Students’ Willingness to Use GenAI.

Statement	China		Malaysia
Statement	Mean	SD	Mean	SD
I envision integrating generative AI technologies like ChatGPT into my teaching and learning practices in the future	4.35	0.62	4.38	0.62
Students must learn how to use generative AI technologies well for their careers	4.18	0.56	4.13	0.58
I believe generative AI technologies such as ChatGPT can improve my digital competence	4.05	0.92	4.30	0.76
I believe generative AI technologies such as ChatGPT can help me save time	4.30	0.63	4.25	0.65
I believe AI technologies such as ChatGPT can provide me with unique insights and perspectives that I may not have thought of myself	4.33	0.62	4.40	0.62
I think AI technologies such as ChatGPT can provide me with personalized and immediate feed- back and suggestions for my assignments	4.31	0.68	4.35	0.71
I think AI technologies such as ChatGPT is a great tool as it is available 24/7	4.41	0.60	4.45	0.68
I think AI technologies such as ChatGPT is a great tool for student support services due to anonymity	4.31	0.62	4.29	0.68

Notably, the highest mean score for both groups was found in the statement, “I think AI technologies such as ChatGPT is a great tool as it is available 24/7” (M = 4.41 for Chinese students, M = 4.45 for Malaysian students), suggesting that students highly value the accessibility and availability of GenAI tools.

For the lowest mean scores, both groups showed relatively similar response patterns. Chinese students reported the lowest mean score for the statement, “I believe generative AI technologies such as ChatGPT can improve my digital competence” (M = 4.05), while Malaysian students indicated a slightly lower score for the statement, “Students must learn how to use generative AI technologies well for their careers” (M = 4.13).

These findings suggest that while both groups see the value of GenAI tools in enhancing their academic experiences, Chinese students were less confident in the tools’ potential to improve digital competence, whereas Malaysian students had a slightly lower perception of the importance of learning how to use these technologies for future career success.

Furthermore, as seen in Table 7, for Chinese students, the data indicated a statistically significant correlation (r = .152, p = .03) between frequency of use and willingness to use GenAI tools, suggesting that frequent use slightly increases willingness, although the relationship is limited. Conversely, no significant correlation was found between knowledge of these tools and willingness, indicating that knowledge alone does not enhance the willingness to use them. For Malaysian students, neither the frequency of use nor had showed a significant influence on their willingness, suggesting that familiarity and usage habits do not strongly impact their engagement with GenAI tools in academic settings.

Table 7.

Correlation Between Frequency, Knowledge, and Willingness to Use GenAI Tools.

Students	Relationship	r	p-Value	Interpretation
Chinese	Frequency → Willingness	.152	.031	Weak, significant
Chinese	Knowledge → Willingness	-	.991	Not significant
Malaysian	Frequency → Willingness	-	.784	Not significant
Malaysian	Knowledge → Willingness	-	.883	Not significant

Note. Significance at p < .05.

Lastly, an independent-samples t-test was conducted to compare Chinese and Malaysian students’ willingness to use GenAI tools.

As seen in Table 8, an independent-samples t-test indicates no statistically significant difference (p = .122) in willingness to use GenAI between Chinese (M = 4.276, SD = 0.284) and Malaysian students (M = 4.312, SD = 0.229). The effect size is small (d = 0.155), suggesting that the difference between groups is negligible.

Table 8.

Comparison of Willingness Between Chinese and Malaysian Students.

Variable	Group	Mean	SD	t(df)	p	Cohen’s d
Willingness	China	4.276	0.284	–1.548 (398)	.122	0.155
	Malaysia	4.312	0.229			0.155

Concerns Regarding GenAI Tools

To address Research Question 1(c), the mean scores were analyzed. As shown in Table 9, both groups displayed relatively moderate levels of concern, with mean scores ranging from 2.51 to 3.26 for Chinese students, and 2.51 to 3.26 for Malaysian students, indicating a general apprehension about certain aspects of using these tools.

Table 9.

Students’ Concerns.

Statement	China		Malaysia
Statement	Mean	SD	Mean	SD
Using generative AI technologies such as ChatGPT to complete assignments undermines the value of university education	2.80	1.40	2.51	1.35
GenAI technologies such as ChatGPT will limit my opportunities to interact with others and socialize while completing coursework	2.63	1.07	2.71	1.10
GenAI technologies such as ChatGPT will hinder my development of generic or transferable skills such as teamwork, problem-solving, and leadership skills	2.65	1.19	2.71	1.19
I can become overreliant on generative AI technologies	3.20	1.40	3.26	1.39

The highest mean score in both groups was for the statement, “I can become overreliant on generative AI technologies” (M = 3.20 for Chinese students, M = 3.26 for Malaysian students), suggesting that students are cautious about the potential for becoming overly dependent on GenAI tools.

For the lowest mean scores, Chinese students expressed the least concern regarding the statement, “Generative AI technologies such as ChatGPT will limit my opportunities to interact with others and socialize while completing coursework” (M = 2.63), whereas Malaysian students had slightly lower concerns about “Using generative AI technologies such as ChatGPT to complete assignments undermine the value of university education” (M = 2.51). These results indicate that, while both groups recognize potential drawbacks, their concerns are somewhat divergent, with Chinese students less concerned about social isolation, and Malaysian students are less concerned about the impact of GenAI on the value of their education.

Furthermore, as seen in Table 10, for Chinese students, a statistically significant negative correlation (r = –0.209, p = .003) was observed between concerns and willingness to use GenAI, implying that higher apprehension about GenAI’s ethical or academic implications slightly reduces their readiness to adopt it. However, students’ knowledge levels did not significantly affect their concerns, as evidenced by non-significant correlation values (r = −.063, p = .374). The frequency with which students use GenAI tools did not significantly correlate with their concerns (r = −.055, p = .441), suggest that the amount of use does not have a strong effect on their concerns.

Table 10.

Correlation Between Concerns, Knowledge, Frequency, and Willingness to Use GenAI Tools.

Students	Relationship	r	p-Value	Result
Chinese	Concerns → Willingness	−.209	.003	Weak, significant
	Knowledge → Concerns	−.063	.374	Not significant
	Frequency → Concerns	−.055	.441	Not significant
Malaysian	Concerns → Willingness	−.231	.001	Weak, significant
	Knowledge → Concerns	−.006	.928	Not significant
	Frequency → Concerns	.076	.284	Not significant

Note. Significance at p < .05.

Similarly, for Malaysian students, concerns were negatively correlated with willingness (r = –0.231, p = .001). Although modest, this relationship suggests that as students’ concerns about these tools increase, their willingness to use them decreases. However, there was no significant relationship between knowledge of GenAI tools and their concerns (r = −.006, p = .928), indicating that knowledge does not reduce concerns. Also, the statistically insignificant relationship between tool usage and concerns (r = .076, p = .284) suggests that the frequency of GenAI use does not meaningfully affect students’ apprehensions regarding its academic or ethical implications.

Overall, for both Chinese and Malaysian students, concerns about GenAI tools appeared to be a significant factor in reducing their willingness to use them. However, neither the frequency of use nor students’ knowledge of the tools showed a strong relationship with their concerns. The primary factor influencing students’ willingness to use GenAI tools in both groups is their level of concern, with Malaysian students showing a slightly stronger negative correlation between concerns and willingness to use tools than Chinese students.

Lastly, an independent-samples t-test was conducted to compare Chinese and Malaysian students’ concerns about using GenAI tools.

As seen in Table 11, an independent-samples t-test indicates no statistically significant difference (p = .998) in concerns between Chinese (M = 2.815, SD = 1.020) and Malaysian students (M = 2.796, SD = 1.036). The effect size is small (d = 0.019), indicating virtually no practical difference between the two groups.

Table 11.

Comparison of Concerns Between Chinese and Malaysian Students.

Variable	Group	Mean	SD	t(df)	p	Cohen’s d
Concerns	China	2.815	1.020	0.187 (398)	.998	0.019
	Malaysia	2.796	1.036			0.019

Benefits of GenAI

This section addresses RQ2 (a): What are the benefits perceived by students regarding the use of GenAI tools? How do these perceptions differ between Chinese and Malaysian students?

The responses to the open-ended questions were analyzed, and corresponding themes were generated on the basis of students’ feedback. These themes are presented (see Tables 12 and 13) in this section.

Table 12.

Coding Framework for the Benefits of GenAI Use among Students.

Country	Initial codes (excerpts of participants’ responses)	Sub-themes	Overarching rhemes
China	“GenAI helps me study by myself when I don’t understand the teacher.”	Self-directed learning	Self-study Support
	“I want to use AI to improve my grades and meet expectations.”	Academic motivation and performance pressure	Navigating the Competitive Environment
	“Competition is very high; AI helps me perform better.”	Academic motivation and performance pressure	Navigating the Competitive Environment
	“AI gives me answers quickly, which saves time before exams.”	Reducing workload and managing time	Saving Time
	“AI translates difficult English readings.”	Language assistance	Language Support
	“I use AI to check my translation and understand English articles.”	Language assistance	Language Support
Malaysia	“AI saves me a lot of time when I do assignments.”	Quick task completion	Efficiency
	“I use GenAI to improve my essays and projects.”	Academic quality enhancement	Enhancement of Academic Work
	“It helps me get new ideas for group projects.”	Idea generation	Brainstorming Ideas
	“AI gives me new opinions on different topics.”	Exposure to diverse viewpoints	Broadening Intellectual Perspectives

Table 13.

Coding Framework for the Challenges of Using GenAI.

Chinese students	Malaysian students	Key themes
Expressed fear that relying too much on GenAI may lead to laziness and reduced learning motivation.	Concerned that overuse of GenAI could discourage independent effort.	Diminished Personal Efforts
“If I always use AI to finish my work, I will lose the habit of thinking by myself.”	“It makes things too easy, and sometimes I don’t feel like trying to solve problems on my own.”	Diminished Personal Efforts
Reported feelings of guilt or discomfort, viewing GenAI use as conflicting with personal academic values.	–	Guilt
“I feel guilty using it because it feels like I’m not doing the real work.”	–	Guilt
Worried that GenAI might compromise honesty and originality.	Concerned about maintaining authenticity in their work.	Academic Integrity
“I am afraid my assignments may not reflect my true ability if I use AI too much.”	“I worry about plagiarism because I can’t always tell if the content is original or AI-generated.”	Academic Integrity
Believed that frequent GenAI use could weaken critical and creative thinking.	Similar concerns about creativity loss.	Hindered Critical Thinking
“AI gives ready answers, so I might think less deeply about topics.”	“I might depend too much on AI’s ideas and forget to think in my own way.”	Hindered Critical Thinking
Feared reduced teacher-student communication due to GenAI dependence.	–	Reduced Human Interaction
“Using AI means I talk less to my teachers about assignments.”	–	Reduced Human Interaction
–	Expressed doubts about the reliability of AI-generated content.	Accuracy and Reliability Concerns
	“Sometimes AI gives wrong or outdated answers, and I can’t always tell which is correct.”
	“It generates random information, which I am not sure is accurate. I feel stressed about fact-checking every single thing.”

Firstly, students responded to the open-ended question: What are the reasons for your willingness to use GenAI tools in your academic work?

The open-ended responses were analyzed using thematic analysis, following Braun and Clarke’s (2006) six-phase approach. Initial codes were generated inductively based on participants’ statements, which were then refined into themes representing perceived benefits and challenges. To ensure the rigor of analysis, coding consistency was cross-checked by both authors, and a codebook was developed after resolving discrepancies through discussion. Illustrative quotes were also included to capture students’ authentic voices and to demonstrate how codes were derived from raw data.

Table 12 presents the coding framework developed from participants’ responses regarding the benefits of using GenAI tools in academic work. It illustrates how initial codes were condensed into sub-themes and overarching themes, supported by representative excerpts from students’ responses in both China and Malaysia.

As shown in Table 12, the analysis revealed both convergent and divergent patterns in how Chinese and Malaysian students perceive the benefits of GenAI in their learning.

For Chinese students, the most prominent theme, Self-study Support, reflects the cultural emphasis on independent learning and academic self-reliance. Many respondents expressed that GenAI tools enable them to study autonomously, particularly when classroom explanations were unclear. Closely linked to this is the theme of Navigating the Competitive Environment, which captures the students’ desire to excel within China’s highly competitive academic landscape. These learners viewed GenAI as a strategic aid to meet high performance expectations. The sub-theme of Language Support also emerged strongly, with students noting that AI tools helped them overcome linguistic barriers, particularly when engaging with English academic materials. Additionally, Saving Time was identified as a recurring benefit, as students appreciated AI’s ability to provide quick assistance, thereby reducing stress associated with exams and heavy workloads.

In contrast, Malaysian students placed greater emphasis on practical and creative dimensions of GenAI use. The theme of Efficiency was dominant, with students highlighting the ability of AI tools to streamline academic tasks and save time. Enhancement of Academic Work also emerged as a key theme, indicating that students use AI not merely for convenience but to improve the overall quality of their assignments and written outputs. Moreover, Brainstorming Ideas and Broadening Intellectual Perspectives illustrate Malaysian students’ orientation toward collaborative learning and intellectual exploration. They valued AI for its potential to generate new ideas and expose them to diverse viewpoints, reflecting a more exploratory and interactive approach to learning compared to their Chinese counterparts.

These thematic patterns indicate that while both groups recognize the functional and academic advantages of GenAI, Chinese students frame its use within a context of performance, discipline, and linguistic challenge, whereas Malaysian students focus on efficiency, creativity, and intellectual expansion.

Challenges of GenAI

This section addresses RQ2 (b): What are the challenges perceived by students regarding the use of GenAI tools? How do these perceptions differ between Chinese and Malaysian students?

Similarly, the students were asked another open-ended question: What are the primary concerns or challenges you face when using GenAI tools in your academic work?

Table 13 presents the coding framework that summarizes the challenges associated with using GenAI tools as reported by Chinese and Malaysian students.

As seen in Table 13, for Chinese students, the most salient theme was Diminished Personal Efforts. Many participants expressed apprehension that dependence on GenAI could lead to laziness and a decline in self-discipline, echoing the deep-rooted cultural values of perseverance and hard work within the Chinese education system. Relatedly, several respondents reported experiencing Guilt when using AI tools, perceiving it as a potential shortcut that might undermine their personal learning efforts. The theme of Academic Integrity also emerged strongly, with students voicing fears that GenAI might compromise honesty and originality in academic work. Furthermore, the concern over Hindered Critical Thinking was frequently mentioned, as students worried that excessive use of AI-generated responses could reduce opportunities for independent analysis and problem-solving. A few participants also mentioned Reduced Human Interaction, noting that overreliance on GenAI may limit meaningful engagement between students and teachers, which traditionally forms an integral part of learning in China.

In contrast, Malaysian students shared some overlapping concerns but also highlighted different priorities. Like their Chinese counterparts, they were apprehensive about Overreliance on GenAI diminishing personal effort and self-initiative. However, their responses placed greater emphasis on Academic Integrity, specifically regarding the authenticity and originality of AI-generated work. Additionally, Hindered Critical Thinking emerged as an important issue, with students questioning whether extensive use of AI could weaken their capacity to develop original ideas or construct arguments independently. A distinctive theme among Malaysian students was Accuracy and Reliability Concerns, many expressed skepticism toward the factual correctness of AI-generated information and cautioned against uncritical acceptance of such outputs.

Collectively, these findings reveal that while both groups recognize the potential drawbacks of GenAI use, Chinese students tend to view the challenges through a moral and self-disciplinary lens, focusing on guilt, diligence, and teacher-student relationships. In contrast, Malaysian students adopt a more pragmatic stance, concentrating on concerns of reliability, originality, and the cognitive consequences of AI use. These differences underscore the influence of cultural and educational contexts in shaping students’ perceptions of GenAI and highlight the need for pedagogical interventions that foster critical, ethical, and reflective use of AI tools in both settings.

Comparing GenAI Perspectives: China and Malaysia

This section addresses RQ3: What insights can be drawn from the data comparing the perspectives of Chinese and Malaysian students on the use of GenAI tools in their academic lives? The key insights regarding the perspectives of Chinese and Malaysian students on the use of GenAI tools in their academic lives, drawn from the findings which are presented in this section.

Difference in Reasons for Using GenAI

Chinese students primarily view GenAI tools as a support mechanism for self-study, aiding in navigating competitive academic environments and saving time. Their emphasis is on using these tools to meet high academic expectations and ease the pressure of assignments and exams, reflecting a cultural focus on individual responsibility for learning.

Malaysian students, on the other hand, are more focused on enhancing academic work and using GenAI tools to brainstorm ideas, improve the quality of assignments, and gain broader intellectual perspectives. This suggests that Malaysian students may see these tools as a way to enhance the depth of their academic output rather than just a tool for efficiency.

Concerns About Overreliance and Academic Integrity

Chinese students expressed concerns about becoming overly reliant on GenAI, leading to diminished personal learning efforts, and fear the potential negative impact on their ability to develop critical thinking and problem-solving skills. Students also expressed apprehension about the potential impact on teacher-student interaction if GenAI tools are overused.

Malaysian students shared similar concerns about the loss of personal touch in their work and the potential accuracy and reliability issues with GenAI-generated content. They also expressed strong ethical concerns regarding academic integrity, fearing that the use of GenAI could undermine originality and creativity in academic work.

Cultural Context and Educational System Influence

Chinese students, who are accustomed to a highly competitive education system, viewed GenAI tools as an aid to cope with academic pressure, especially in the face of intensive exams and assignments. This could explain their focus on timesaving and supporting self-study as key reasons for adopting GenAI tools.

Malaysian students, while also facing academic pressures, seemed more inclined to use GenAI tools to augment their academic capabilities by improving the quality of work and gaining different perspectives. The usage of GenAI could be seen as a way to enhance their intellectual curiosity and academic output.

Ethical Concerns and Academic Integrity

Both groups of students expressed cautiousness about using GenAI tools because of potential threats to academic integrity. While Chinese students were more concerned about becoming overly reliant on these tools, Malaysian students expressed concerns about ethical implications, such as plagiarism and the potential loss of originality. This difference could reflect varying educational philosophies and cultural emphasis on academic honesty in each country.

Cultural Attitudes Toward Technology

Chinese students seemed more instrumental in their approach to GenAI tools, utilizing them to manage academic pressure and optimize personal learning efforts. Meanwhile, Malaysian students embraced a more holistic view, using these tools not only for practical assistance but also for enhancing their creativity and intellectual growth.

The insights from comparing Chinese and Malaysian students’ perspectives on GenAI tools highlight both similarities and differences in their motivations, concerns, and usage patterns. While both groups acknowledged the benefits of efficiency and academic enhancement, Chinese students tend to focus more on timesaving and self-study support, while Malaysian students prioritized improving academic quality and broadening intellectual perspectives. Additionally, both groups expressed concerns about the ethical implications of using GenAI tools and the potential for overreliance. These findings suggest that cultural contexts, educational systems, and academic pressures play significant roles in shaping students’ attitudes toward the use of GenAI tools in academic settings.

Discussion

This study explored and compared Chinese and Malaysian university students’ perceptions of GenAI tools in academic contexts. The findings reveal that both shared understandings and context-specific differences in students’ knowledge, willingness, and concerns regarding GenAI. By examining these perceptions cross-nationally, the study contributes original insights into how cultural, ethical, and educational factors shape students’ engagement with AI in higher education.

Knowledge and Frequency of Use

Both groups demonstrated high awareness and frequent use of GenAI tools, with a positive correlation between knowledge and usage (see Table 4). This aligns with Chan and Hu’s (2023) finding, that increased familiarity leads to more frequent usage. The positive correlation between knowledge and frequency of use is not surprising, as familiarity often encourages greater engagement with new technologies. Notably, the absence of statistically significant differences (see Table 5) and the small effect size for knowledge (d = 0.160) further indicate that students in both countries possess a similar level of GenAI awareness. This convergence suggests that the adoption of GenAI technologies in higher education may be occurring at a relatively uniform pace across these contexts, potentially driven by global digital exposure rather than localized educational differences alone.

However, the results also extend this understanding by showing that awareness alone does not guarantee meaningful academic integration. Factors such as institutional support, access, and culturally rooted attitudes toward innovation appear to influence usage highlighting the need for a more nuanced understanding of AI literacy in higher education.

The Limited Role of Knowledge Versus The Powerful Role of Concerns

Chinese and Malaysian students were generally open to adopting GenAI, though the factors shaping willingness differed. This interpretation is supported by the negligible between-group differences in willingness (d = 0.155; see Table 8), indicating that the overall level of willingness toward GenAI remains largely consistent across both student populations. This adds new evidence that knowledge alone plays a limited role in shaping students’ willingness to use GenAI, whereas concerns rooted in cultural and ethical values exert a stronger influence. Contrary to previous studies (e.g., Chan & Hu, 2023), which posited a linear relationship between knowledge and willingness, the findings suggest that students’ decisions to engage with GenAI are more strongly shaped by socio-cultural expectations and ethical awareness than by mere exposure or technical familiarity.

The Cultural Duality of Concerns (Ethics vs. Reliability)

Concerns about GenAI varied between contexts, with Chinese students expressing ethical guilt and anxiety related to academic honesty, while Malaysian students emphasized the accuracy and reliability of AI-generated information. Both groups shared apprehensions regarding plagiarism, decline in critical thinking, and reduced human interaction echoing prior research by Liu et al. (2024) and Almassaad et al. (2024). However, it is important to contextualize these differences within the statistically non-significant findings and extremely small effect size for concerns (d = 0.019; see Table 11), which indicate that the overall intensity of concerns is nearly identical across groups. This suggests that the variation lies not in the magnitude of concern but in its underlying orientation .that is, ethical versus functional, highlighting qualitative rather than quantitative differences. From a cross-cultural perspective, this study extends existing literature by demonstrating that the roots of AI-related concerns are not universal but culturally mediated, shaped by local educational values and institutional expectations.

Instrumental Versus Holistic Usage Patterns

Chinese students viewed GenAI as a means for self-study support, navigating academic competition, and language translation, while Malaysian students associated it with efficiency, brainstorming, and broadening learning perspectives. These findings suggest that GenAI adoption aligns with each country’s educational priorities self-reliance and performance in China versus collaboration and creativity in Malaysia. Common challenges included fears of diminished personal effort, compromised integrity, and a decline in critical thinking, consistent with prior literature. Yet, the finding that Chinese students reported guilt while Malaysian students questioned reliability provides a novel cross-context insight into culturally specific apprehensions toward AI.

Summary of Contributions

This study contributes three original insights:

(1) It highlights that students’ willingness to adopt GenAI is shaped by cultural and ethical factors rather than exposure alone.

(2) It reveals how educational contexts influence perceived benefits, self-reliant learning in China versus collaborative enhancement in Malaysia.

(3) It provides comparative evidence that concerns about AI stem from different cultural values: ethical responsibility in China and informational reliability in Malaysia.

These findings broaden current understandings of GenAI adoption in education, emphasizing the importance of culturally responsive AI literacy frameworks and responsible integration strategies.

Implications

This study contributes to a deeper understanding of how cultural and contextual differences shape students’ engagement with GenAI tools. While previous research focused primarily on general acceptance, our findings highlight nuanced concerns, including ethical dilemmas in China and reliability issues in Malaysia. These insights provide a foundation for developing targeted and culturally responsive AI integration strategies in higher education. Importantly, the comparative design of this study demonstrates that students’ knowledge, willingness, and concerns are not uniform across national contexts, underscoring the need for policy and pedagogical approaches that are sensitive to local educational cultures and governance frameworks.

Given the varying perceptions and concerns expressed by Chinese and Malaysian students, it is evident that a one-size-fits-all approach to the adoption of GenAI tools may not be effective. Educational strategies should therefore be tailored to address the specific concerns and cultural contexts of students in different regions. For instance, while both groups share concerns about academic integrity, Chinese students appear more focused on the ethical implications of using GenAI tools, whereas Malaysian students are more concerned with the reliability of the content generated. This distinction illustrates how national policy environments and academic norms may shape students’ evaluative frameworks toward emerging technologies, reinforcing the practical value of cross-national comparison. Based on these insights, three actionable implications are proposed:

(1) Integrate structured AI literacy programs within university curricula that explicitly teach students how to use GenAI tools ethically, evaluate AI-generated content, and understand plagiarism and authorship boundaries.

(2) Develop institutional guidelines and honor codes that clearly define acceptable and unacceptable uses of GenAI tools in coursework and research, ensuring consistent practices across departments.

(3) Provide hands-on workshops and case-based discussions to help students critically evaluate GenAI outputs, identify biases, and apply AI responsibly in problem-solving and writing tasks.

For Chinese institutions, emphasis could be placed on ethical reasoning and integrity training, addressing moral dilemmas and fairness in AI-assisted learning. For Malaysian institutions, more focus could be given to evaluating accuracy, verifying AI-generated content, and building students’ confidence in cross-checking digital information.

Furthermore, the study emphasizes the importance of developing strategies to enhance students’ understanding of the potential benefits of GenAI tools while simultaneously addressing their concerns. Both Chinese and Malaysian students highlighted the tools’ efficiency, support for self-study, and potential for improving academic work. Institutions could therefore design AI-support systems and mentorship models where educators guide students in integrating AI into their learning processes without diminishing creativity or critical thinking. By demonstrating how similar benefits coexist with context-specific concerns, the findings contribute to a more balanced and evidence-based approach to AI governance in higher education.

As suggested by Zlotnikova and Hlomani (2025) and Cogo et al. (2024), integrating GenAI tools into education should involve comprehensive training frameworks and continuous monitoring to ensure responsible and equitable use. By combining ethical education, practical training, and institutional regulation, universities can build a culture of responsible AI engagement that strengthens both academic integrity and innovation. Overall, the comparative outcomes of this study extend existing research by moving beyond single-country perspectives and offering actionable insights for designing context-sensitive, culturally responsive GenAI policies in higher education systems.

Limitations

One limitation of this study is its focus on only two countries; China and Malaysia limiting the generalizability of the findings to other cultural or regional contexts. The study provides valuable insights into the perceptions of students in these two countries, the results may not reflect the perceptions of students in other parts of the world, where different cultural, educational, and technological factors may shape their experiences with GenAI tools.

Although the findings offer meaningful insights into student perceptions of GenAI tools in China and Malaysia, the findings are constrained by its reliance on convenience samples from two private universities, which may not fully reflect the diversity of higher education contexts in either country. Future research could include larger and more diverse samples to validate and extend these findings. Since this study relied on self-reported data, future studies could include mixed methods to reduce potential biases. Expanding the geographical scope to include students from different cultural and educational contexts would offer deeper insights into how these factors shape GenAI adoption. Longitudinal studies could also track how perceptions and concerns evolve as GenAI becomes more integrated into academic life. Finally, examining the effects of interventions such as workshops on ethical AI use could reveal strategies to enhance students’ confidence and promote responsible engagement with AI tools.

Conclusions

In conclusion, the findings of this study highlight the intricate relationship between students’ knowledge, concerns, and willingness to engage with GenAI tools in their academic activities. Both Chinese and Malaysian students demonstrated a strong general awareness of these tools, yet their willingness to adopt them was influenced by a variety of factors beyond mere knowledge. Cultural attitudes play a significant role, with each group emphasizing different aspects of GenAI’s potential benefits and challenges. These variations reflect how cultural contexts shape students’ perceptions of technological advancements, suggesting that their educational experiences and societal values impact their readiness to embrace new technologies in academic settings.

Additionally, the study underscores the importance of addressing students’ concerns about the ethical implications of using GenAI tools, particularly in relation to academic integrity and the reliability of AI-generated content. While both student groups shared similar concerns regarding the potential for overreliance on these tools and their effects on critical thinking, slight differences emerged in the nature of their apprehensions. Chinese students were more concerned about feelings of guilt associated with the use of AI tools, whereas Malaysian students focused more on the accuracy and reliability of the content generated. These findings suggest that promoting a responsible and effective integration of GenAI tools in academic practices, educational strategies must consider both universal concerns and culturally specific issues, ensuring that students are equipped to use these tools ethically and confidently.

Footnotes

Acknowledgements

The authors are thankful to all the students who participated in the study.

ORCID iDs

Aarti Pushp Rawal

Jia Liu

Ethical Considerations

Ethical approval was not required as participation was fully anonymous with no risks involved.

Consent to Participate

Informed consent was obtained from all participants.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data supporting this study are available from the corresponding author upon request.

References

Adamakis

Rachiotis

(2025). Artificial Intelligence in Higher Education: A state-of-the-art overview of pedagogical integrity, artificial intelligence literacy, and policy integration. Encyclopedia, 5(4), 180. https://doi.org/10.3390/encyclopedia5040180

Aldossary

A. S.

Aljindi

A. A.

Alamri

J. M.

(2024). The role of generative AI in education: Perceptions of Saudi students. Contemporary Educational Technology, 16(4), ep536. https://doi.org/10.30935/cedtech/15496

Almassaad

Alajlan

Alebaikan

(2024). Student perceptions of Generative Artificial Intelligence: Investigating utilization, benefits, and challenges in Higher Education. Systems, 12, 385. https://doi.org/10.3390/systems12100385

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Chambers

Owen

W. J.

(2024). The efficacy of GenAI Tools in Postsecondary Education. Brock Education Journal, 33(3), 57–74. https://doi.org/10.26522/brocked.v33i3.1178

Chan

C. K. Y.

(2023). Students’ voices on generative AI: perceptions, benefits, and challenges in higher education. International Journal of Educational Technology in Higher Education, 20(1), 1–18. https://doi.org/10.1186/s41239-023-00411-8

Cheng

Zeng

(2023). Shaping AI’s future? China in global AI governance. Journal of Contemporary China, 32(143), 794–810. https://doi.org/10.1080/10670564.2022.2107391

Cogo

Patsko

Szoke

(2024). Generative artificial intelligence and ELT. ELT Journal, 78(4), 373–377. https://doi.org/10.1093/elt/ccae051

Daher

Hussein

(2024). Higher Education Students’ perceptions of GenAI tools for learning. Information, 15, 416. https://doi.org/10.3390/info15070416

10.

Davis

F. D.

Granić

Marangunić

(2024). The technology acceptance model. In The technology acceptance model: 30 years of TAM, (pp. 1–57). Springer. https://doi.org/https://link.springernature.com/home/

11.

Duah

J. E.

McGivern

(2024). How generative artificial intelligence has blurred notions of authorial identity and academic norms in higher education, necessitating clear university usage policies. International Journal of Information and Learning Technology, 41, 180–193. https://doi.org/10.1108/ijilt-11-2023-0213

12.

Farrelly

Baker

(2023). Generative Artificial Intelligence: Implications and considerations for Higher Education Practice. Education Sciences, 13, 1109. https://doi.org/10.3390/educsci13111109

13.

Francis

N. J.

Jones

Smith

D. P.

(2024). Generative AI in Higher Education: Balancing Innovation and Integrity. British Journal of Biomedical Science, 81, 14048. https://doi.org/10.3389/bjbs.2024.14048

14.

Gonsalves

(2026). Generative AI’s impact on Critical Thinking: Revisiting Bloom’s taxonomy. Journal of Marketing Education, 48, 4–19. https://doi.org/10.1177/02734753241305980

15.

Hoßbach

Isaksen

(2025). AI-augmented approaches to creative problem-solving: A metacognitive perspective. Creativity and Innovation Management, 34, 854–869. https://doi.org/10.1111/caim.70003

16.

Hughes

Malik

Dettmer

Al-Busaidi

A. S.

Dwivedi

Y. K.

(2025). Reimagining higher education: Navigating the challenges of generative AI adoption. Information Systems Frontiers. Advance online publication. https://doi.org/10.1007/s10796-025-10582-6

17.

Jaaffar

A. R.

Abd Rani

S. H.

Zakaria

(2021). Education and innovation growth: Establishing Entrepreneurial University Framework in Malaysia. In Sergi

B. S.

Jaaffar

A. R.

(Eds.), Modeling Economic Growth in contemporary Malaysia (pp. 17–29). Emerald Publishing Limited. https://doi.org/10.1108/978-1-80043-806-420211003

18.

Khanal

Zhang

Taeihagh

(2025). Development of New Generation of Artificial Intelligence in China: When Beijing’s global ambitions meet local realities. Journal of Contemporary China, 34(151), 19–42. https://doi.org/10.1080/10670564.2024.2333492

19.

Kim

Detrick

(2025). Exploring students’ perspectives on generative AI-assisted academic writing. Education and Information Technologies, 30, 1265–1300. https://doi.org/10.1007/s10639-024-12878-7

20.

Kohnke

(2024). Exploring EAP students’ perceptions of GenAI and traditional grammar-checking tools for language learning. Computers and Education: Artificial Intelligence, 7, 100279. https://doi.org/10.1016/j.caeai.2024.100279

21.

Kumar

Rao

Singhania

Verma

Kheterpal

(2024). Will artificial intelligence drive the advancements in higher education? A tri-phased exploration. Technological Forecasting and Social Change, 201, 123258. https://doi.org/10.1016/j.techfore.2024.123258

22.

Kutty

Chugh

Perera

Neupane

Jha

Gunathilake

Perera

N. C.

(2024). Generative AI in higher education: Perspectives of students, educators and administrators. Journal of Applied Learning and Teaching, 7(2), 47–60. https://doi.org/10.37074/jalt.2024.7.2.27

23.

Laak

K. J.

Abdelghani

Aru

(2024). Personalisation is not guaranteed: The challenges of using generative AI for personalised learning [Conference session]. International Conference on Innovative Technologies and Learning. https://doi.org/10.1007/978-3-031-65881-5_5

24.

Lee

S. S.

Moore

R. L.

(2024). Harnessing generative AI (GenAI) for automated feedback in higher education: A systematic review. Online Learning, 28(3), 82–106. https://doi.org/10.24059/olj.v28i3.4593

25.

Liu

Park

McMinn

(2024). Using generative artificial intelligence/ChatGPT for academic communication: Students’ perspectives. International Journal of Applied Linguistics, 34, 1437–1461. https://doi.org/10.1111/ijal.12574

26.

(2025). China’s achievements in digital education in the wake of Education Informatization 2.0 Action Plan. Science Insights Education Frontiers, 27(1), 4435–4451. https://doi.org/10.15354/sief.25.re488

27.

Mills

Mizouri

Peach

(2025). Prompting better feedback: A study of custom GPT for Formative Assessment in Undergraduate Physics. Education Sciences, 15(8), 1058. https://doi.org/10.3390/educsci15081058

28.

Premkumar

P. P.

Yatigammana

M. R. K. N.

Kannangara

(2024). Impact of generative AI on critical thinking skills in undergraduates: A systematic review. Journal of Desk Research Review and Analysis, 2(1), 199–215. https://doi.org/10.4038/jdrra.v2i1.55

29.

Rudolph

Mohamed Ismail

F. M.

Popenici

(2024). Higher education’s generative artificial intelligence paradox: The meaning of Chatbot Mania. Journal of University Teaching and Learning Practice, 21(6), 14–48. https://doi.org/10.53761/54fs5e77

30.

Samala

A. D.

Rawas

Wang

Reed

J. M.

Kim

Howard

N.-J.

Ertz

(2025). Unveiling the landscape of generative artificial intelligence in education: A comprehensive taxonomy of applications, challenges, and future prospects. Education and Information Technologies, 30, 3239–3278. https://doi.org/10.1007/s10639-024-12936-0

31.

Shen

Xia

(2025). Exploring the attitude and use of GenAI-image among art and design college students based on TAM and SDT. Interactive Learning Environments, 33(2), 1198–1215. https://doi.org/10.1080/10494820.2024.2365959

32.

Suonpää

Heikkilä

Dimkar

(2024). Students’ perceptions of generative AI usage and risks in a finnish higher education institution [Conference session]. INTED2024 Proceedings. https://doi.org/10.21125/inted.2024.0825

33.

Tantivejakul

Chantharasombat

Kongpolphrom

(2024). Voices of the future: Exploring students’ views on the use of GenAI in academic and professional PR writing. LEARN Journal: Language Education and Acquisition Research Network, 17(2), 511–537. https://doi.org/10.70730/pvtd8302

34.

Venkatesh

Morris

M. G.

Davis

G. B.

Davis

F. D.

(2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425–478. https://doi.org/10.2307/30036540

35.

Wang

Dang

Mac

(2024). Generative AI in higher education: Seeing ChatGPT through universities’ policies, resources, and guidelines. Computers and Education: Artificial Intelligence, 7, 100326. https://doi.org/10.1016/j.caeai.2024.100326

36.

Wang

K. D.

Tufts

Wieman

Salehi

Haber

(2025). Scaffold or crutch? Examining college students’ use and views of generative AI tools for STEM education [Conference session]. 2025 IEEE Global Engineering Education Conference (EDUCON). https://doi.org/10.1109/EDUCON62633.2025.11016406

37.

Yparrea

N. G.

Hernández-Rodríguez

(2024). Unveiling generative AI in higher education: Insights from engineering students and professors [Conference session]. 2024 IEEE Global Engineering Education Conference (EDUCON). https://doi.org/10.1109/EDUCON60312.2024.10578876

38.

Zhai

Wibowo

L. D.

(2024). The effects of over-reliance on AI dialogue systems on students’ cognitive abilities: A systematic review. Smart Learning Environments, 11, 28. https://doi.org/10.1186/s40561-024-00316-7

39.

Zlotnikova

Hlomani

(2025). GenAI in the context of African Universities: A crisis of Tertiary Education or its new Dawn? Digital Government, 6, 1–11. https://doi.org/10.1145/3689215