Students’ voices on generative AI: perceptions, benefits, and challenges in higher education
Open accessThe DOI matches its Crossref record; 59 of the 59 references it cites were found in a registry or an index; 1 of the 14 sources its claims cite were checked for support.
Abstract
This study explores university students' perceptions of generative AI technologies, such as ChatGPT, in higher education, focusing on familiarity, their willingness to engage, potential benefits and challenges, and effective integration.
Overview
CoChat Summary
TL;DR
This survey of 399 undergraduate and postgraduate students in Hong Kong found generally positive attitudes toward generative AI in higher education, alongside concerns about accuracy, privacy, ethics, and impacts on development, careers, and values. The authors argue that understanding student perceptions can inform responsible integration and policy.
Key findings
- Students reported generally good knowledge of GenAI limitations, with mean scores from 3.89 to 4.15; the highest was understanding limitations in handling complex tasks (M = 4.15, SD = 0.82), and the lowest was limited emotional intelligence/empathy (M = 3.89, SD = 0.97).
- Willingness to use GenAI was positive: students agreed they should learn to use it for careers (M = 4.05, SD = 0.96), it can save time (M = 4.20, SD = 0.82), and it is available 24/7 (M = 4.12, SD = 0.83). Willingness correlated with knowledge (r = 0.189, p < 0.001) and frequency of use (r = 0.326, p < 0.001).
- Concerns were moderate: the highest-rated concern was that using GenAI to complete assignments undermines the value of university education (M = 3.15, SD = 1.17 in Table 4; narrative reports M = 3.18, SD = 1.16), while over-reliance had the lowest rating (M = 2.85, SD = 1.13 in Table 4; narrative reports M = 2.89, SD = 1.13). Concerns did not significantly correlate with knowledge (r = 0.096, p > 0.05).
- Qualitative themes identified benefits in personalized/immediate learning support, writing and brainstorming, research and analysis, visual/audio multimedia, and administrative support; challenges concerned accuracy and transparency, privacy and ethics, holistic competencies, career prospects, human values, and uncertain policies.
Methods
- Online survey design with closed-ended and open-ended questions; final instrument had 26 items on a 5-point Likert scale plus 3 open-ended questions, piloted before data collection.
- Convenience sample of 399 undergraduate and postgraduate students from various disciplines across six universities in Hong Kong; 204 males (51.1%) and 195 females (48.9%), with 177 undergraduates (44.4%) and 222 postgraduates (55.6%).
- Descriptive analysis for survey data and thematic analysis for open-ended responses; two coders manually generated codes from a manageable set of responses (n = 387), discussed disagreements, and created a consensus codebook.
Limitations
- The paper states the sample size was relatively small, limiting generalizability to the broader Hong Kong student population; self-reported data may introduce social desirability or recall bias; the cross-sectional design cannot track changes over time; and the study did not explore how students were exposed to AI or GenAI’s actual impact on learning outcomes.
- Evident from the text: convenience sampling and a single-region sample further constrain generalizability, and the correlational design cannot establish causal relationships between knowledge, frequency of use, willingness, and concerns.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 1 Not valid · 10 Not checked · 71 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“suggesting that even those with a good understanding of the technology may still have reservations, similar to Dahmash et al. (2020)'s findings.”
What the sources say
- Partly supportedcites Dahmash et al. (2020) sourceExperimental — accuracy not yet measured
The Results report that students anxious about AI’s impact on radiology had higher self-perceived understanding of AI than those who were not anxious. But the Discussion says objective knowledge was low and contradicted students’ self-assessments, so the paper does not establish that students with genuinely good understanding had reservations. (LLM, conf=0.84)
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As illustrated in Table 2, participants had a generally good understanding of GenAI technologies, with mean scores ranging from 3.89 to 4.15.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The study revealed that students' knowledge of GenAI technologies and frequency of use are positively correlated.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“students also expressed worries regarding the vacuum of institutional policies on the use of GenAI.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Some participants were optimistic about AI's integration in the future.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
27 more claims
Create a free account to see all 83 claims and every source verdict.
- It may state two separate claims as one. 27
- Probably not a substantial claim. 40
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“This observation corresponds with patterns noted in the critical thinking domain, suggesting that students' perceptions of AI's usefulness and learning value do not directly lead to improved problem-solving skills via self-regulated learning mechanisms, similar to the findings of Ghotbi et al. [39] and Chan and Hu [40].”
From The Mediating Role of Generative AI Self-Regulation on Students’ Critical Thinking and Problem-Solving (2024) · cites this paper as Chan and Hu [40]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In one study, 30% of students reported using AI for generating text, while another study reported idea generation and real-time, personalised feedback to be the most common reasons for use (Chan and Hu 2023;Higher Education Policy Institute 2024).”
From Summarise, elaborate, try again: exploring the effect of feedback literacy on AI-enhanced essay writing (2026) · cites this paper as Chan and Hu 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Students and teachers worried about genAI leading to the loss of critical thinking skills, creativity, and autonomy, all findings that resonate with previous research (Chan & Hu, 2023;Encheva et al., 2025;Isiaku et al., 2024;Newell et al., 2024;Olsson, 2026;Rahman et al., 2025;Xia et al., 2024).”
From Toward Partnership Ethics for Generative Artificial Intelligence in Higher Education (2026) · cites this paper as Chan & Hu, 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent work in higher education similarly documents that students rely on AI to streamline academic writing and routine tasks yet often do so without questioning its accuracy (Chan & Hu, 2023;Pitts et al., 2025).”
From Understanding AI Literacy Among College Students: Voices from a Historically Black College and University (2026) · cites this paper as Chan & Hu, 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Past studies highlight the same ambiguity among university students, who often conflate AI-assisted writing with plagiarism but receive little clarity from faculty or policy (Chan & Hu, 2023;Pitts et al., 2025).”
From Understanding AI Literacy Among College Students: Voices from a Historically Black College and University (2026) · cites this paper as Chan & Hu, 2023
Experimental — accuracy not yet measured
26 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Concerns about generative AI technologies · p. 8
“the highest rating was for how these technologies could affect the value of university education (Mean = 3.18; SD = 1.16).”
§ Knowledge of generative AI technologies · p. 8
“| Using generative AI technologies such as ChatGPT to complete assignments undermines the value of university education | 3.15 | 1.17 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Concerns about generative AI technologies · p. 8
“They expressed the least positive opinions about if people will become over-reliant on GenAI technologies (Mean = 2.89; SD = 1.13)”
§ Knowledge of generative AI technologies · p. 8
“| I can become over-reliant on generative AI technologies | 2.85 | 1.13 |”
Citations
Cites
Perceived usefulness, perceived ease of use, and user acceptance of information technology
Davis, F. D. (1989) · MIS Quarterly · doi:10.2307/249008 · Cited by 68,483 · cited 1×
“User acceptance is key to successful uptake of technological innovations (Davis, 1989).”
BackgroundAuto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Generative adversarial networks
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2020) · Communications of the ACM · doi:10.1145/3422622 · Cited by 18,384
Generative Adversarial Networks
Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.2661 · Cited by 4,477
BAİDOO-ANU, D., & OWUSU ANSAH, L. (2023) · Journal of AI · doi:10.61969/jai.1337500 · Cited by 3,939
Teaching for Quality Learning at University: What the Student Does
BIGGS, J. B., & Tang, C. S. (1999) · Cited by 3,374
Baidoo-Anu, D., & Owusu Ansah, L. (2023) · doi:10.2139/ssrn.4337484 · Cited by 3,004
Artificial intelligence in higher education: the state of the field
Crompton, H., & Burke, D. (2023) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-023-00392-8 · Cited by 2,010 · cited 1×
“Another opportunity in which GenAI can bring benefits is learning assessment (Crompton & Burke, 2023).”
BackgroundChatGPT: five priorities for research
van Dis, E. A. M., Bollen, J., Zuidema, W., van Rooij, R., & Bockting, C. L. (2023) · Nature · doi:10.1038/d41586-023-00288-7 · Cited by 1,842 · cited 1×
“…to be useful research aids for generating ideas, synthesizing information, and summarising a vast amount of text data to help researchers analyse data and compose their writing (Berg, 2023; Chan & Zhou, 2023), contributing to efficiency in publication (Kitamura, 2023; van Dis et al., 2023).”
BackgroundA comprehensive AI policy education framework for university teaching and learning
Chan, C. K. Y. (2023) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-023-00408-3 · Cited by 1,577 · cited 2×
“Chan (2023b) suggests an AI Ecological Education Policy Framework to tackle the various implications of AI integration in university teaching and learning with three dimensions: Pedagogical, Governance, and Operational.”
BackgroundSelf-citation
Research works printed but not linked: 1
- Sumakul, D. T. Y. G., Hamied, F. A., & Sukyadi, D. (2020). Students’ perceptions of the use of AI in a writing class. Advances in Social Science, Education and Humanities Research, 624. 52–57. https://www.atlantis-press.com/article/125970061.pdf
Cited by
A comprehensive AI policy education framework for university teaching and learning
Chan, C. K. Y. (2023) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-023-00408-3 · Cited by 1,577
Self-citationChan, C. K. Y., & Lee, K. K. W. (2023) · Smart Learning Environments · doi:10.1186/s40561-023-00269-3 · Cited by 634
“…include data privacy and security concerns, potential biases in AI algorithms, and the importance of human oversight in AI-driven decision-making processes (Bisdas et al., 2021; Chan & Hu, 2023; Chan & Zhou, 2023; Chen et al., 2023; Essel et al., 2022; Gillissen et al., 2022; Jha et al., 2022).”
BackgroundSelf-citationYusuf, A., Pervin, N., & Román-González, M. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00453-6 · Cited by 554
“This aligns with the broader trend of AI advancements in natural language processing (Chan & Hu, 2023).”
BackgroundMethodResultInfluentialTPACK in the age of ChatGPT and Generative AI
Mishra, P., Warr, M., & Islam, R. (2023) · Journal of Digital Learning in Teacher Education · doi:10.1080/21532974.2023.2247480 · Cited by 484
“For example, there is some initial evidence that GenAI can help reduce teacher workload (Chan & Hu, 2023).”
Promises and challenges of generative artificial intelligence for human learning
Yan, L., Greiff, S., Teuber, Z., & Gašević, D. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-02004-5 · Cited by 499
Metadata
- Authors
- Chan, C. K. Y., & Hu, W.
- Affiliations
- University of Hong Kong (HK)
- Corresponding author
- Chan, Cecilia Ka Yuk
- ORCID
- Chan, Cecilia Ka Yuk
- Venue
- International Journal of Educational Technology in Higher Education
- Published
- 2023-07-17
- Volume
- 20
- Issue
- 1
- ISSN
- 2365-9440 (ISSN-L)
- OpenAlex
- W4384464487
- Cited by
- 2,301
- References
- 46
- Citations by year
- 2026: 967 · 2025: 878 · 2024: 423 · 2023: 53
- FWCI
- 64.06
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- International Journal of Educational Technology in Higher Education (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Social Sciences › Business, Management and Accounting › Marketing › AI in Service Interactions
- Keywords
- generative AI, ChatGPT, student perceptions, higher education, AI in teaching and learning, personalized learning, AI ethics, privacy concerns
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Students’ voices on generative AI: perceptions, benefits, and challenges in higher education. https://app.cochat.ai/papers/10.1186/s41239-023-00411-8
- The paper
- Chan, C. K. Y., & Hu, W. (2023). Students’ voices on generative AI: perceptions, benefits, and challenges in higher education. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-023-00411-8
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.