Artificial Intelligence in Higher Education: A Bibliometric Analysis of Trends, Gaps, and Future Directions (1986–2025)

Artificial Intelligence in Higher Education: A Bibliometric Analysis of Trends, Gaps, and Future Directions (1986–2025) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Artificial Intelligence in Higher Education: A Bibliometric Analysis of Trends, Gaps, and Future Directions (1986–2025) Khalifa AHSINA, Mohamed AKHLAFFOU, Said LOUKIL, Adraa JAAFAR, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7187231/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Since its conceptualization by McCarthy (1956), artificial intelligence (AI) has evolved significantly, offering multiple applications in higher education (HE), such as personalized learning, automated assessment, and student success prediction. However, despite the proliferation of research, studies remain fragmented and lack a global vision, particularly after 2022. This bibliometric study analyzes 1,780 articles published between 1986 and 2025, extracted from the Scopus database. The PRISMA approach was applied to ensure a rigorous and reproducible selection of publications. The VOSviewer tool was used to explore research trends, scientific collaborations, and priority themes concerning AI in HE. The analysis reveals three main phases: (1) a latency phase (1986–2013) focused on expert systems; (2) a phase of moderate growth (2014–2019) with the emergence of chatbots and adaptive learning; and (3) a post-2020 explosion due to the rise of generative models such as ChatGPT. Most publications originate from the United States and China, with limited interdisciplinary collaborations. Themes related to accessibility, ethics, and frugal AI are scarcely addressed. Current research shows geographical and thematic concentration, with a lack of diversity and interdisciplinarity. The absence of longitudinal studies and methodological standardization limits the comprehensive understanding of AI's impact on HE. To bridge these gaps, it is recommended to foster international collaborations, particularly North-South, and explore neglected themes such as accessibility, ethics, frugal AI, and pedagogical integration for a better understanding of AI's impact on HE. Artificial Intelligence (AI) Higher Education (HE) Bibliometric Analysis Research Trends Generative Models Interdisciplinary Collaboration Ethics and Accessibility Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Artificial intelligence (AI) has become a transformative force across sectors, including higher education (HE). Since McCarthy (1956), AI has evolved into systems enabling adaptive learning and complex data processing, fostering innovative applications in HE. These technologies support personalized learning, automated assessment, and student success prediction, particularly in languages, engineering, and medicine. Despite extensive research, a gap remains: studies are fragmented, discipline-specific, and lack a recent global synthesis, especially post-2022. This limits understanding AI’s potential and challenges in HE, underscoring the need for an updated bibliometric study. AI has revolutionized HE by enhancing pedagogical adaptation, automated evaluation, and personalized learning. While studies highlight AI’s benefits, they remain fragmented, overlooking key aspects. Notably, no recent comprehensive review examines global research trends, collaborations, or disciplinary impacts. This lack of synthesis hinders understanding research dynamics and future directions. A central research question arises: How has AI research in HE evolved in terms of publication volume, application domains, and scientific collaboration patterns? This study addresses this gap through a bibliometric analysis of AI research in HE (1986–2025), aiming to: Map the geographic and temporal evolution of AI research in HE. Identify key actors (affiliations, collaborations) and priority disciplines. Examine AI application domains and beneficiaries in HE. Identify research trends and gaps to guide future studies. The study follows a structured approach: the theoretical framework defines AI and reviews key studies on its educational applications. The methodology details the bibliometric approach, data sources, and analysis criteria. The results section presents key findings, followed by a discussion and conclusion highlighting limitations and future research perspectives. 1. Literature Review on AI in Education AI seeks to replicate human cognitive processes through computational systems, including expert systems, neural networks, fuzzy logic, genetic algorithms, and intelligent agents (Rainer et al., 2016). AI falls into two categories: Strong AI: Capable of reasoning and emotional responses (Wells, 2023). Weak AI: Task-specific applications like fraud detection. HE, being cognitively intensive, is an ideal AI application domain, enhancing educational practices across disciplines (Perrotta & Selwyn, 2020). AI applications include: Content development and dissemination. Interaction and collaboration. Performance evaluation (Chassignol et al., 2018; Perrotta & Selwyn, 2020). Recent literature reviews cover general AI applications (Chassignol et al., 2018; Chen et al., 2020; Chiu et al., 2023) and specific areas: Educational chatbots (Okonkwo & Ade-Ibijola, 2021). Precision education (Luan & Chin-Chung, 2021). STEM (Xu & Ouyang, 2022). Student assessment (González-Calatayud et al., 2021). AI enhances learning through personalized experiences, automated assessments, and accessibility improvements but also raises concerns about credibility, plagiarism, and academic integrity (Rudolph et al., 2023). Challenges include: Data security risks due to extensive student data collection. Algorithmic bias reinforcing inequalities. Teachers’ autonomy concerns (Zhang et al., 2022; Nguyen et al., 2022). Despite AI’s potential, its integration in HE must balance automation and human interaction to preserve learning’s social dimension (Popenici & Kerr, 2017). A comprehensive review is needed to synthesize empirical research and assess AI’s evolving impact in HE. 2. Research Methodology This study uses Scopus, a widely accepted database for systematic reviews. Following Goksel & Bozkurt (2019), a query in March 2025 identified English-language publications featuring “artificial intelligence” and “higher education” in titles, abstracts, or keywords, retrieving 1,780 articles from 1986–2025. The study follows the PRISMA framework for systematic reviews and employs VOSviewer for bibliometric analysis, ensuring rigorous article selection and a comprehensive exploration of AI research trends in HE. a. Data Source The articles were extracted from Scopus, a leading database covering indexed academic journals. b. Search Strategy The following query was used to filter relevant articles: TITLE-ABS-KEY ("artificial intelligence" AND "Higher education") AND (LIMIT-TO (DOCTYPE, "ar")) AND (LIMIT-TO (LANGUAGE, "English")) AND (LIMIT-TO (SRCTYPE, "j")) This query identified 1,780 articles published within the studied period. c. Selection Criteria The selected studies meet the following criteria: Published in peer-reviewed journals. Written in English. Focused on AI in higher education, including empirical studies and systematic reviews. Publications in other languages or those not explicitly addressing the impact of AI on higher education were excluded. 3. Results 3.1. Descriptive Analysis of Bibliometric Data Artificial intelligence (AI) is reshaping higher education (HE) by enabling personalized learning, automated assessments, and predictive analytics. Since McCarthy (1956), AI has evolved into adaptive systems applied across disciplines like languages, engineering, and medicine. Despite extensive research, studies remain fragmented, lacking a comprehensive synthesis post-2022, limiting understanding of AI’s potential and challenges in HE. This study conducts a bibliometric analysis (1986–2025) to map AI research trends, key actors, and application domains. AI enhances learning but raises concerns about data security, bias, and teacher autonomy (Zhang et al., 2022). Using Scopus and PRISMA methodology, we analyze 1,780 publications, employing VOSviewer for research mapping. Findings will highlight gaps, collaborations, and future research directions, ensuring a balanced integration of AI in HE (Popenici & Kerr, 2017). 3.2 The Most Influential Journals on AI in Higher Education Journal rankings reveal a concentration of AI in education research in a few key publications. The top five journals are Education and Information Technologies (57 articles), Computers and Education Artificial Intelligence (47), Sustainability Switzerland (44), Education Sciences (43), and the International Journal of Educational Technology in Higher Education (27). These journals explore AI integration, educational technologies, and pedagogical innovations. Over 120 journals have published on this topic, reflecting interdisciplinarity. Notable contributions come from IEEE Access (22), Frontiers in Psychology (12), and Scientific Reports (10). The rise in AI-related publications aligns with digital transformation, AI advancements, and the shift to online learning post-COVID-19. The presence of high-impact journals confirms AI’s growing role as a transformative force in higher education. 3.3 The Most Prolific Authors in the Field of AI in Higher Education AI in education research is concentrated in a few key journals. The top five are Education and Information Technologies (57 articles), Computers and Education AI (47), Sustainability Switzerland (44), Education Sciences (43), and the International Journal of Educational Technology in Higher Education (27). These journals cover AI integration, educational technologies, and pedagogy. Over 120 journals have contributed, highlighting interdisciplinarity. Notable ones include IEEE Access (22), Frontiers in Psychology (12), and Scientific Reports (10). The surge in AI publications reflects digital transformation, AI advancements, and post-COVID-19 online learning. High-impact journals confirm AI’s growing influence in higher education. 3.4 Keyword Co-Occurrence Analysis The keyword co-occurrence analysis of AI and higher education using VOSviewer highlights four main research themes. Based on Scopus data (2022–2024), this mapping relies on three elements: keyword frequency (nodes), their connections (co-occurrences), and thematic clustering (colors). The first cluster focuses on generative technologies and educational challenges (e.g., ChatGPT, Generative AI, Academic Integrity), addressing the impact of LLMs on fraud and their pedagogical integration. The second cluster covers active learning and education stakeholders (e.g., active learning, teachers, students), emphasizing the adaptation of teaching methods to AI. The third cluster explores advanced machine learning (e.g., contrastive learning, algorithmic biases), while the fourth examines AI applications in specific contexts (STEM, virtual reality, pandemic). Trends reveal a shift from COVID-19 concerns (2022) to generative AI (2023–2024), with rising tensions over academic integrity and automation. However, gaps remain, particularly in accessibility and student data governance. These findings underscore the need for ethical frameworks and a redefined role for teachers in an increasingly automated environment. 3.5. Research Gaps in AI and Higher Education This analysis highlights critical gaps in AI applications in higher education, particularly in ethics, pedagogy, cultural diversity, methodology, technology, and social aspects. Ethically, there is a lack of frameworks for detecting biases, managing student data, and addressing legal accountability for algorithmic errors. Pedagogically, few studies explore AI’s impact on metacognition, academic disengagement, or teacher adaptation, including prompt training and critical AI analysis. Culturally, nearly 90% of research focuses on Western contexts, overlooking minority languages and collectivist pedagogies. Methodologically, the absence of longitudinal studies and interdisciplinary collaboration limits AI’s educational impact assessment. Technologically, research on explainable AI (XAI) and low-resource solutions remains scarce. Socially, disparities in AI access and teacher acceptance remain overlooked. Addressing these issues requires ethical guidelines, interdisciplinary studies, AI literacy training, and increased funding. 3.6. Analysis of Authors Publishing in the Field of AI and Higher Education The co-authorship network mapping in AI and higher education, using VOSviewer, reveals collaboration patterns among researchers in 2023. Some authors, like Ruddjbh J., hold central positions, forming star-shaped networks, while others, such as Tajlj S., appear more isolated, possibly due to specialization, emerging status, or incomplete data. Crawford J. and Cowling M. form a connected subgroup, indicating close collaboration. The analyzed period (March–June 2023) coincides with the rise of ChatGPT, suggesting intensified academic production. However, limitations exist: restricted author visibility, potential name recognition errors, and the absence of quantitative indicators like node size or link thickness. The short timeframe also hinders long-term trend analysis. Compared to Zhang et al.'s Chinese network, this Western network appears less dense and covers a different period. Future research should refine author identification, extend the study period, and cross-reference with publication keywords for deeper insights. 3.7. Analysis of Countries in the Field of AI and Higher Education his analysis of the geographical distribution of authors in AI applied to higher education (2022–2024) reveals significant concentration and disparities. The U.S. dominates research on techno-pedagogy, followed by China, reflecting its massive AI education investments. India, Malaysia, and Indonesia also emerge as key Asian players, while Gulf countries, notably the UAE and Saudi Arabia, strategically invest in Education 4.0. However, Sub-Saharan Africa is nearly absent, with Nigeria barely mentioned. Latin America is also underrepresented, except for Brazil. Two distinct periods emerge: 2022–2023 saw the rise of Vietnam, Sri Lanka, and Middle Eastern contributors like Kuwait and Oman. In 2023–2024, U.S. and Chinese hubs consolidated, alongside Eastern Europe’s emergence, notably Ukraine and Hungary. Regional clusters show thematic specializations: the Middle East focuses on smart universities (e.g., NEOM), Southeast Asia on AI for local languages, and Europe/North America on ethics and pedagogy. Yet, biases persist. Data errors (e.g., "Migera" for Nigeria) and duplication (e.g., Sweden) reinforce English-speaking dominance, sidelining Africa and Latin America. Expanding inclusion by integrating the Global South and linking data to AI education funding is essential. Strengthening North-South collaborations and conducting comparative studies could bridge the AI divide, ensuring more equitable representation in higher education research. Discussion The analysis of the results, informed by the literature review, highlights key trends in the evolution of artificial intelligence (AI) in education, its interdisciplinary nature, collaboration networks, addressed themes, and identified gaps. Its chronological progression shows a shift from expert systems and neural networks to specific applications such as STEM education and language learning. Bibliometric data confirm this trajectory in three phases: from 1986 to 2013, a latency period characterized by "weak" AI and limited tutorial systems; between 2014 and 2019, moderate growth marked by chatbots and adaptive learning (Goksel & Bozkurt, 2019); and from 2020 to 2025, an explosion of research driven by post-COVID disruptions and the rise of generative models such as ChatGPT. This trend validates the hypothesis of Perrotta & Selwyn (2020) on AI's key role in educational transformation. Despite recommendations from Chassignol et al. (2018), interdisciplinarity remains limited. Techno-centric journals such as Education and Information Technologies dominate, although a gradual opening toward psychology and sustainability is observable (Sustainability, Frontiers in Psychology). However, the lack of journals in education sciences highlights weak integration of pedagogical theories. Regarding collaboration networks, Chiu T.K.F. and his co-authors dominate, reflecting a strong interest in generative AI, absent before 2019. Furthermore, findings reveal a significant geographical bias, with an overrepresentation of the United States and China, while Africa and Latin America remain marginalized, confirming a North-South imbalance (Xu & Ouyang, 2022). Transdisciplinary collaborations between educators and engineers remain rare. The thematic analysis of keywords confirms these trends. Generative AI, particularly ChatGPT, occupies a central place but also raises concerns about academic integrity (González-Calatayud et al., 2021). Critical thematic gaps persist, such as the lack of research on disability or data governance, despite their crucial role in contemporary ethical debates (Wells, 2023). The limited exploration of pedagogical theoretical frameworks indicates a lack of theoretical coherence. The identified research gaps align with findings from the literature review. Geographically, 90% of studies originate from English-speaking or Western countries, reflecting a lack of diversity. Methodologically, few longitudinal studies exist, making it difficult to assess AI's long-term effects in education. Additionally, the absence of standardized evaluation methods for AI tools, particularly chatbots, limits result comparability. Thematically, blind spots persist, such as operational ethics (bias auditing, user consent) and frugal AI, which could provide solutions for less connected regions. A shift toward more qualitative research, especially ethical and pedagogical issues, appears to have emerged after 2023. To address these gaps, several recommendations can be made. Methodologically, mixed-method studies combining learning analytics and ethnography would provide a more comprehensive assessment of generative AI's impact on metacognition. In terms of international collaborations, it is crucial to promote North-South projects on frugal AI, particularly through strategic partnerships between the European Union and Africa. From an editorial perspective, encouraging education science journals to publish more on AI, including special issues on ethics and inclusion, would be relevant. Finally, at the institutional level, training teachers in AI literacy and targeted funding for research in underrepresented regions are essential. Conclusion This study highlights that AI research in higher education is primarily structured around post-ChatGPT generative AI but does not fully integrate essential human dimensions such as equity and critical pedagogy. The observed systemic biases, both geographically and methodologically, call for an integrated approach that combines bibliometric data and qualitative studies to move beyond mere publication quantification. In conclusion, this research underscores the rapid evolution of AI in higher education, particularly due to advances in generative AI. While scientific production has grown significantly, gaps remain in interdisciplinarity, geographical diversity, and theoretical grounding. A more integrative approach combining bibliometric analyses with qualitative studies would refine the understanding of AI's transformative role in education. Declarations Author contributions K A conceived and led the study. M A and S L contributed to data collection and preliminary analysis. I M conducted the advanced data analysis and prepared the visualizations. A J and K A wrote the main manuscript text. All authors reviewed and approved the final version of the manuscript. Funding The authors did not receive support or funding from any organization for the submitted work. All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript. Data availability The data supporting the findings of this study were retrieved from the Scopus database. Access to these data is subject to license restrictions. Data can be made available by the corresponding author upon reasonable request. Clinical trial number: Not applicable. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Acknowledgements Not applicable. References McCarthy, J., Minsky, M. L., Rochester, N., & Shannon, C. E. (1956). Artificial intelligence (AI) coined at dartmouth. Retrieved October, 28, 2021. Rainer, K., Prince, B., Splettstoesser-Hogeterp, I., & SanchezRodriguez, C. (2016). Introduction to Information Systems ((4th ed.).). John Wiley & Sons Inc. Wells, R.E. (2023). Strong AI vs. weak AI: What’s the difference? Strong AI can do anything a human can do, while weak AI is limited to a specific task. LifeWire. Retrieved August 22, Chassignol, M., Khoroshavin, A., Klimova, A., & Bilyatdinova, A. (2018). Artificial intelligence trends in education: A narrative overview. Procedia Computer Science 136, 16–24. Chen, L., Chen, P., & Lin, Z. (2020). Artificial intelligence in education: A review. IEEE Access, 8, 75264–––75278. Chiu, T. K. F., Xia, Q., Zhou, X., Chai, C. S., & Cheng, M. (2023). Systematic literature review on opportunities, challenges, and future research recommendations of artificial intelligence in education. Computers and Education: Artificial Intelligence, 4, Article 100118. https://doi.org/10.1016/j.caeai.2022.100118 Okonkwo, C. W., & Ade-Ibijola, A. (2021). Chatbots applications in education: A systematic review. Computers and Education: Artificial Intelligence, 2, Article 100033. https://doi.org/10.1016/j.caeai.2021.100033 Luan, H., & Chin-Chung, T. (2021). A Review of using machine learning approaches for precision education. Journal of Educational Technology & Society, 24, 250–266. Hwang, G.-J., & Tu, Y.-F. (2021). Roles and research trends of artificial intelligence in mathematics education: A bibliometric mapping analysis and systematic review. Mathematics, 9, 594. https://doi.org/10.3390/math9060584 Xu, W., & Ouyang, F. (2022). The application of AI technologies in STEM education: A systematic review from 2011 to 2021. International Journal of STEM Education, 9, 1–20. https://doi.org/10.1186/s40594-022-00377-5 Gonz´alez-Calatayud, V., Prendes-Espinosa, P., & Roig-Vila, R. (2021). Artificial intelligence for student assessment: A systematic review. Applied Sciences, 11(12), 5467. https://doi.org/10.3390/app11125467 Goksel, N., & Bozkurt, A. (2019). Artificial intelligence in education: Current insights and future perspectives In S. Sisman-Ugur & G. Kurubacak (Eds.), Handbook of Research on Learning in the Age of Transhumanism (pp. 224-236). IGI Global. doi: 10.4018/978-1-5225-8431-5.ch014. Rudolph, J., Tan, S., & Tan, S. (2023). ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching, 6(1). https://doi.org/10.37074/jalt.2023.6.1.9 Liu, Z., Kong, X., Liu, S., Yang, Z., & Zhang, C. (2022). Automated detection of emotional and cognitive engagement in MOOC discussions to predict learning achievement. Computers & Education, 181, Article 104461. https://doi.org/10.1016/j.compedu.2022.104461 Nguyen, A., Ngo, H. N., Hong, Y., Dang, B., & Nguyen, B. P. T. (2022). Ethical principles for artificial intelligence in education. Education and Information Technologies, 1-21. Popenici, S. A., & Kerr, S. (2017). Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning, 12, 1–13. https://doi.org/10.1186/S41039-017-0062-8 Tahiru, F. (2021). AI in education: A systematic literature review. Journal of Cases on Information Technology (JCIT), 23(1), 1-20. Chiu, T. K., Xia, Q., Zhou, X., Chai, C. S., & Cheng, M. (2023). Systematic literature review on opportunities, challenges, and future research recommendations of artificial intelligence in education. Computers and Education: Artificial Intelligence, 4, 100118. Rudolph, J., Tan, S., & Tan, S. (2023). War of the chatbots: Bard, Bing Chat, ChatGPT, Ernie and beyond. The new AI gold rush and its impact on higher education. Journal of Applied Learning and Teaching, 6(1), 364-389. Crawford, J., Cowling, M., & Allen, K. A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). Journal of University Teaching and Learning Practice, 20(3), 1-19. Bannister, P., Urbieta, A. S., & Peñalver, E. A. (2023). A systematic review of generative AI and (English medium instruction) higher education. Aula Abierta, 52(4), 401-409. Bozkurt, A., Junhong, X., Lambert, S., Pazurek, A., Crompton, H., Koseoglu, S., Farrow, R., Bond, M., Nerantzi, C., Honeychurch, S., Bali, M., Dron, J., Mir, K., Stewart, B., Costello, E., Mason, J., Stracke, C. M., & Romero-Hall, E. (2023). Speculative Futures on ChatGPT and Generative Artificial Intelligence (AI): A Collective Reflection from the Educational Landscape. Asian Journal of Distance Education, 18(1), 53-130. Article 1. https://doi.org/10.5281/zenodo.7636568 Crawford, J., Cowling, M., & Allen, K.-A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). Journal of University Teaching and Learning Practice, 20(3), 1–19. https://search.informit.org/doi/10.3316/ Perkins, M. (2023). Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching and Learning Practice, 20(2), 1–24. https://search.informit.org/doi/10.3316/informit.T2024111300009591751711095 Acosta-Enriquez, B.G., Arbulú Ballesteros, M.A., Huamaní Jordan, O. et al. Analysis of college students' attitudes toward the use of ChatGPT in their academic activities: effect of intent to use, verification of information and responsible use. BMC Psychol 12, 255 (2024). https://doi.org/10.1186/s40359-024-01764-z Romero-Rodríguez, J. M., Ramírez-Montoya, M. S., Buenestado-Fernández, M., & Lara-Lara, F. (2023). Use of ChatGPT at university as a tool for complex thinking: Students’ perceived usefulness. Journal of New Approaches in Educational Research, 12(2), 323-339. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7187231","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":502795513,"identity":"83ccea6a-3bc0-4ad4-8639-dcda0644cd1c","order_by":0,"name":"Khalifa AHSINA","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1ElEQVRIie3OoQvCQBTH8d8YzHJgvWOgf4FwIpzJ+a/sWLCaTIaTgWta/TMEwTwYaBmsLhgUwbwkCyKezeTNZrhPe/C+vAdY1h+igAdwoA1H/Zgw9Vui8bTpY0y5t0s9PXUHZRRXmAfoJYbYhzccEH7r70upOA4RRB5+TzqA8MEz552E8FKI1Ji07qzm2Xi3kSrFUyfF2fQYEZTwTG6pXChnqZPScIXFZObrJNrklxhyFRFRGq7QY7Jn9SMbrZPJtaruQUcUhitwPwe9TAz7lmVZVhMvO0hAX2R0agEAAAAASUVORK5CYII=","orcid":"","institution":"Ibn Zohr University","correspondingAuthor":true,"prefix":"","firstName":"Khalifa","middleName":"","lastName":"AHSINA","suffix":""},{"id":502795514,"identity":"4d4f6280-811c-47e1-bac8-923b17522c29","order_by":1,"name":"Mohamed AKHLAFFOU","email":"","orcid":"","institution":"National School of Business and Management of El Jadida, Chouaib Doukkali University","correspondingAuthor":false,"prefix":"","firstName":"Mohamed","middleName":"","lastName":"AKHLAFFOU","suffix":""},{"id":502795515,"identity":"e220ba84-6dd0-4add-a923-3d28419f9396","order_by":2,"name":"Said LOUKIL","email":"","orcid":"","institution":"National School of Business and Management of Tangier, Abdelmalek Essaadi University","correspondingAuthor":false,"prefix":"","firstName":"Said","middleName":"","lastName":"LOUKIL","suffix":""},{"id":502795516,"identity":"c414be68-8e4e-4287-9188-3efb665b5d81","order_by":3,"name":"Adraa JAAFAR","email":"","orcid":"","institution":"National School of Business and Management of Tangier, Abdelmalek Essaadi University","correspondingAuthor":false,"prefix":"","firstName":"Adraa","middleName":"","lastName":"JAAFAR","suffix":""},{"id":502795517,"identity":"021916e6-ac57-4703-af4a-06c77e7028ab","order_by":4,"name":"Idriss MOKSIT","email":"","orcid":"","institution":"Ibn Zohr University","correspondingAuthor":false,"prefix":"","firstName":"Idriss","middleName":"","lastName":"MOKSIT","suffix":""}],"badges":[],"createdAt":"2025-07-22 12:53:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7187231/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7187231/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89651329,"identity":"59c1bd0c-8fe9-4672-b8fa-5f8434477732","added_by":"auto","created_at":"2025-08-22 09:47:03","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":43882,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of publications over time\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/7621d9d596664ae3bf581074.jpg"},{"id":89650948,"identity":"c7e12b07-38c4-4932-8286-6ee8a538de65","added_by":"auto","created_at":"2025-08-22 09:39:05","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":68165,"visible":true,"origin":"","legend":"\u003cp\u003eMost influential journals in AI and Higher Education\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/238b7f252a55b594e6c30765.jpg"},{"id":89650828,"identity":"bae79bf9-1f98-4962-bbc5-b527d103b9ba","added_by":"auto","created_at":"2025-08-22 09:39:02","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":77516,"visible":true,"origin":"","legend":"\u003cp\u003eThe most prolific authors in the field of AI in higher education\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/f771f1f9a2ddfaf6329e52b2.jpg"},{"id":89650954,"identity":"a457f5fd-de85-416d-895f-995348e838d1","added_by":"auto","created_at":"2025-08-22 09:39:06","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":184427,"visible":true,"origin":"","legend":"\u003cp\u003eClustering of Keywords Related to AI Adoption in Higher Education\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/1dd0c7d60d3822e25150a504.jpg"},{"id":89650936,"identity":"859495f5-f79d-4e65-ad33-7b56483a43e5","added_by":"auto","created_at":"2025-08-22 09:39:05","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":31704,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCo-authorship Network Map in AI and Higher Education\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/669680e2d0bbcf66c6a22635.jpg"},{"id":89650950,"identity":"d41a911a-12ba-4ba2-94d3-33b15855fcc1","added_by":"auto","created_at":"2025-08-22 09:39:06","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":126784,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCountry Co-authorship in AI and Higher Education\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSource: Developed by the authors\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/ec5b19b91145f0b1705f5c28.jpg"},{"id":103405462,"identity":"80e46644-ae01-4bef-bcad-421998649318","added_by":"auto","created_at":"2026-02-25 09:58:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1171542,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7187231/v1/89b03d78-c499-4632-be24-e648c2cb6bb0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Artificial Intelligence in Higher Education: A Bibliometric Analysis of Trends, Gaps, and Future Directions (1986–2025)","fulltext":[{"header":"Introduction","content":"\u003cp\u003eArtificial intelligence (AI) has become a transformative force across sectors, including higher education (HE). Since McCarthy (1956), AI has evolved into systems enabling adaptive learning and complex data processing, fostering innovative applications in HE. These technologies support personalized learning, automated assessment, and student success prediction, particularly in languages, engineering, and medicine.\u003c/p\u003e\u003cp\u003eDespite extensive research, a gap remains: studies are fragmented, discipline-specific, and lack a recent global synthesis, especially post-2022. This limits understanding AI\u0026rsquo;s potential and challenges in HE, underscoring the need for an updated bibliometric study.\u003c/p\u003e\u003cp\u003eAI has revolutionized HE by enhancing pedagogical adaptation, automated evaluation, and personalized learning. While studies highlight AI\u0026rsquo;s benefits, they remain fragmented, overlooking key aspects. Notably, no recent comprehensive review examines global research trends, collaborations, or disciplinary impacts. This lack of synthesis hinders understanding research dynamics and future directions.\u003c/p\u003e\u003cp\u003eA central research question arises: How has AI research in HE evolved in terms of publication volume, application domains, and scientific collaboration patterns? This study addresses this gap through a bibliometric analysis of AI research in HE (1986\u0026ndash;2025), aiming to:\u003c/p\u003e\u003cp\u003eMap the geographic and temporal evolution of AI research in HE.\u003c/p\u003e\u003cp\u003eIdentify key actors (affiliations, collaborations) and priority disciplines.\u003c/p\u003e\u003cp\u003eExamine AI application domains and beneficiaries in HE.\u003c/p\u003e\u003cp\u003eIdentify research trends and gaps to guide future studies.\u003c/p\u003e\u003cp\u003eThe study follows a structured approach: the theoretical framework defines AI and reviews key studies on its educational applications. The methodology details the bibliometric approach, data sources, and analysis criteria. The results section presents key findings, followed by a discussion and conclusion highlighting limitations and future research perspectives.\u003c/p\u003e"},{"header":"1. Literature Review on AI in Education","content":"\u003cp\u003eAI seeks to replicate human cognitive processes through computational systems, including expert systems, neural networks, fuzzy logic, genetic algorithms, and intelligent agents (Rainer et al., 2016). AI falls into two categories:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eStrong AI: Capable of reasoning and emotional responses (Wells, 2023).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWeak AI: Task-specific applications like fraud detection.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eHE, being cognitively intensive, is an ideal AI application domain, enhancing educational practices across disciplines (Perrotta \u0026amp; Selwyn, 2020). AI applications include:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eContent development and dissemination.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eInteraction and collaboration.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePerformance evaluation (Chassignol et al., 2018; Perrotta \u0026amp; Selwyn, 2020).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eRecent literature reviews cover general AI applications (Chassignol et al., 2018; Chen et al., 2020; Chiu et al., 2023) and specific areas:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eEducational chatbots (Okonkwo \u0026amp; Ade-Ibijola, 2021).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePrecision education (Luan \u0026amp; Chin-Chung, 2021).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSTEM (Xu \u0026amp; Ouyang, 2022).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eStudent assessment (Gonz\u0026aacute;lez-Calatayud et al., 2021).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eAI enhances learning through personalized experiences, automated assessments, and accessibility improvements but also raises concerns about credibility, plagiarism, and academic integrity (Rudolph et al., 2023). Challenges include:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eData security risks due to extensive student data collection.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eAlgorithmic bias reinforcing inequalities.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eTeachers\u0026rsquo; autonomy concerns (Zhang et al., 2022; Nguyen et al., 2022).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eDespite AI\u0026rsquo;s potential, its integration in HE must balance automation and human interaction to preserve learning\u0026rsquo;s social dimension (Popenici \u0026amp; Kerr, 2017). A comprehensive review is needed to synthesize empirical research and assess AI\u0026rsquo;s evolving impact in HE.\u003c/p\u003e"},{"header":"2. Research Methodology","content":"\u003cp\u003eThis study uses Scopus, a widely accepted database for systematic reviews. Following Goksel \u0026amp; Bozkurt (2019), a query in March 2025 identified English-language publications featuring \u0026ldquo;artificial intelligence\u0026rdquo; and \u0026ldquo;higher education\u0026rdquo; in titles, abstracts, or keywords, retrieving 1,780 articles from 1986\u0026ndash;2025.\u003c/p\u003e\u003cp\u003eThe study follows the PRISMA framework for systematic reviews and employs VOSviewer for bibliometric analysis, ensuring rigorous article selection and a comprehensive exploration of AI research trends in HE.\u003c/p\u003e\u003cp\u003e\u003cb\u003ea. Data Source\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe articles were extracted from Scopus, a leading database covering indexed academic journals.\u003c/p\u003e\u003cp\u003e\u003cb\u003eb. Search Strategy\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe following query was used to filter relevant articles:\u003c/p\u003e\u003cp\u003eTITLE-ABS-KEY (\"artificial intelligence\" AND \"Higher education\") AND (LIMIT-TO (DOCTYPE, \"ar\")) AND (LIMIT-TO (LANGUAGE, \"English\")) AND (LIMIT-TO (SRCTYPE, \"j\"))\u003c/p\u003e\u003cp\u003eThis query identified 1,780 articles published within the studied period.\u003c/p\u003e\u003cp\u003e\u003cb\u003ec. Selection Criteria\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe selected studies meet the following criteria:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003ePublished in peer-reviewed journals.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWritten in English.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eFocused on AI in higher education, including empirical studies and systematic reviews.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePublications in other languages or those not explicitly addressing the impact of AI on higher education were excluded.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Descriptive Analysis of Bibliometric Data\u003c/h2\u003e\u003cp\u003eArtificial intelligence (AI) is reshaping higher education (HE) by enabling personalized learning, automated assessments, and predictive analytics. Since McCarthy (1956), AI has evolved into adaptive systems applied across disciplines like languages, engineering, and medicine. Despite extensive research, studies remain fragmented, lacking a comprehensive synthesis post-2022, limiting understanding of AI\u0026rsquo;s potential and challenges in HE.\u003c/p\u003e\u003cp\u003eThis study conducts a bibliometric analysis (1986\u0026ndash;2025) to map AI research trends, key actors, and application domains. AI enhances learning but raises concerns about data security, bias, and teacher autonomy (Zhang et al., 2022). Using Scopus and PRISMA methodology, we analyze 1,780 publications, employing VOSviewer for research mapping. Findings will highlight gaps, collaborations, and future research directions, ensuring a balanced integration of AI in HE (Popenici \u0026amp; Kerr, 2017).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e3.2 The Most Influential Journals on AI in Higher Education\u003c/h2\u003e\u003cp\u003eJournal rankings reveal a concentration of AI in education research in a few key publications. The top five journals are Education and Information Technologies (57 articles), Computers and Education Artificial Intelligence (47), Sustainability Switzerland (44), Education Sciences (43), and the International Journal of Educational Technology in Higher Education (27). These journals explore AI integration, educational technologies, and pedagogical innovations.\u003c/p\u003e\u003cp\u003eOver 120 journals have published on this topic, reflecting interdisciplinarity. Notable contributions come from IEEE Access (22), Frontiers in Psychology (12), and Scientific Reports (10). The rise in AI-related publications aligns with digital transformation, AI advancements, and the shift to online learning post-COVID-19. The presence of high-impact journals confirms AI\u0026rsquo;s growing role as a transformative force in higher education.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e3.3 The Most Prolific Authors in the Field of AI in Higher Education\u003c/h2\u003e\u003cp\u003eAI in education research is concentrated in a few key journals. The top five are Education and Information Technologies (57 articles), Computers and Education AI (47), Sustainability Switzerland (44), Education Sciences (43), and the International Journal of Educational Technology in Higher Education (27). These journals cover AI integration, educational technologies, and pedagogy.\u003c/p\u003e\u003cp\u003eOver 120 journals have contributed, highlighting interdisciplinarity. Notable ones include IEEE Access (22), Frontiers in Psychology (12), and Scientific Reports (10). The surge in AI publications reflects digital transformation, AI advancements, and post-COVID-19 online learning. High-impact journals confirm AI\u0026rsquo;s growing influence in higher education.\u003c/p\u003e\u003cp\u003e\u003cb\u003e3.4 Keyword Co-Occurrence Analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe keyword co-occurrence analysis of AI and higher education using VOSviewer highlights four main research themes. Based on Scopus data (2022\u0026ndash;2024), this mapping relies on three elements: keyword frequency (nodes), their connections (co-occurrences), and thematic clustering (colors).\u003c/p\u003e\u003cp\u003eThe first cluster focuses on generative technologies and educational challenges (e.g., ChatGPT, Generative AI, Academic Integrity), addressing the impact of LLMs on fraud and their pedagogical integration. The second cluster covers active learning and education stakeholders (e.g., active learning, teachers, students), emphasizing the adaptation of teaching methods to AI.\u003c/p\u003e\u003cp\u003eThe third cluster explores advanced machine learning (e.g., contrastive learning, algorithmic biases), while the fourth examines AI applications in specific contexts (STEM, virtual reality, pandemic).\u003c/p\u003e\u003cp\u003eTrends reveal a shift from COVID-19 concerns (2022) to generative AI (2023\u0026ndash;2024), with rising tensions over academic integrity and automation. However, gaps remain, particularly in accessibility and student data governance. These findings underscore the need for ethical frameworks and a redefined role for teachers in an increasingly automated environment.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.5. Research Gaps in AI and Higher Education\u003c/h2\u003e\u003cp\u003eThis analysis highlights critical gaps in AI applications in higher education, particularly in ethics, pedagogy, cultural diversity, methodology, technology, and social aspects.\u003c/p\u003e\u003cp\u003eEthically, there is a lack of frameworks for detecting biases, managing student data, and addressing legal accountability for algorithmic errors. Pedagogically, few studies explore AI\u0026rsquo;s impact on metacognition, academic disengagement, or teacher adaptation, including prompt training and critical AI analysis.\u003c/p\u003e\u003cp\u003eCulturally, nearly 90% of research focuses on Western contexts, overlooking minority languages and collectivist pedagogies. Methodologically, the absence of longitudinal studies and interdisciplinary collaboration limits AI\u0026rsquo;s educational impact assessment.\u003c/p\u003e\u003cp\u003eTechnologically, research on explainable AI (XAI) and low-resource solutions remains scarce. Socially, disparities in AI access and teacher acceptance remain overlooked. Addressing these issues requires ethical guidelines, interdisciplinary studies, AI literacy training, and increased funding.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.6. Analysis of Authors Publishing in the Field of AI and Higher Education\u003c/h2\u003e\u003cp\u003eThe co-authorship network mapping in AI and higher education, using VOSviewer, reveals collaboration patterns among researchers in 2023. Some authors, like Ruddjbh J., hold central positions, forming star-shaped networks, while others, such as Tajlj S., appear more isolated, possibly due to specialization, emerging status, or incomplete data. Crawford J. and Cowling M. form a connected subgroup, indicating close collaboration.\u003c/p\u003e\u003cp\u003eThe analyzed period (March\u0026ndash;June 2023) coincides with the rise of ChatGPT, suggesting intensified academic production. However, limitations exist: restricted author visibility, potential name recognition errors, and the absence of quantitative indicators like node size or link thickness. The short timeframe also hinders long-term trend analysis.\u003c/p\u003e\u003cp\u003eCompared to Zhang et al.'s Chinese network, this Western network appears less dense and covers a different period. Future research should refine author identification, extend the study period, and cross-reference with publication keywords for deeper insights.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.7. Analysis of Countries in the Field of AI and Higher Education\u003c/h2\u003e\u003cp\u003ehis analysis of the geographical distribution of authors in AI applied to higher education (2022\u0026ndash;2024) reveals significant concentration and disparities. The U.S. dominates research on techno-pedagogy, followed by China, reflecting its massive AI education investments. India, Malaysia, and Indonesia also emerge as key Asian players, while Gulf countries, notably the UAE and Saudi Arabia, strategically invest in Education 4.0. However, Sub-Saharan Africa is nearly absent, with Nigeria barely mentioned. Latin America is also underrepresented, except for Brazil.\u003c/p\u003e\u003cp\u003eTwo distinct periods emerge: 2022\u0026ndash;2023 saw the rise of Vietnam, Sri Lanka, and Middle Eastern contributors like Kuwait and Oman. In 2023\u0026ndash;2024, U.S. and Chinese hubs consolidated, alongside Eastern Europe\u0026rsquo;s emergence, notably Ukraine and Hungary. Regional clusters show thematic specializations: the Middle East focuses on smart universities (e.g., NEOM), Southeast Asia on AI for local languages, and Europe/North America on ethics and pedagogy.\u003c/p\u003e\u003cp\u003eYet, biases persist. Data errors (e.g., \"Migera\" for Nigeria) and duplication (e.g., Sweden) reinforce English-speaking dominance, sidelining Africa and Latin America. Expanding inclusion by integrating the Global South and linking data to AI education funding is essential. Strengthening North-South collaborations and conducting comparative studies could bridge the AI divide, ensuring more equitable representation in higher education research.\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe analysis of the results, informed by the literature review, highlights key trends in the evolution of artificial intelligence (AI) in education, its interdisciplinary nature, collaboration networks, addressed themes, and identified gaps. Its chronological progression shows a shift from expert systems and neural networks to specific applications such as STEM education and language learning. Bibliometric data confirm this trajectory in three phases: from 1986 to 2013, a latency period characterized by \"weak\" AI and limited tutorial systems; between 2014 and 2019, moderate growth marked by chatbots and adaptive learning (Goksel \u0026amp; Bozkurt, 2019); and from 2020 to 2025, an explosion of research driven by post-COVID disruptions and the rise of generative models such as ChatGPT. This trend validates the hypothesis of Perrotta \u0026amp; Selwyn (2020) on AI's key role in educational transformation.\u003c/p\u003e\u003cp\u003eDespite recommendations from Chassignol et al. (2018), interdisciplinarity remains limited. Techno-centric journals such as Education and Information Technologies dominate, although a gradual opening toward psychology and sustainability is observable (Sustainability, Frontiers in Psychology). However, the lack of journals in education sciences highlights weak integration of pedagogical theories.\u003c/p\u003e\u003cp\u003eRegarding collaboration networks, Chiu T.K.F. and his co-authors dominate, reflecting a strong interest in generative AI, absent before 2019. Furthermore, findings reveal a significant geographical bias, with an overrepresentation of the United States and China, while Africa and Latin America remain marginalized, confirming a North-South imbalance (Xu \u0026amp; Ouyang, 2022). Transdisciplinary collaborations between educators and engineers remain rare.\u003c/p\u003e\u003cp\u003eThe thematic analysis of keywords confirms these trends. Generative AI, particularly ChatGPT, occupies a central place but also raises concerns about academic integrity (Gonz\u0026aacute;lez-Calatayud et al., 2021). Critical thematic gaps persist, such as the lack of research on disability or data governance, despite their crucial role in contemporary ethical debates (Wells, 2023). The limited exploration of pedagogical theoretical frameworks indicates a lack of theoretical coherence.\u003c/p\u003e\u003cp\u003eThe identified research gaps align with findings from the literature review. Geographically, 90% of studies originate from English-speaking or Western countries, reflecting a lack of diversity. Methodologically, few longitudinal studies exist, making it difficult to assess AI's long-term effects in education. Additionally, the absence of standardized evaluation methods for AI tools, particularly chatbots, limits result comparability. Thematically, blind spots persist, such as operational ethics (bias auditing, user consent) and frugal AI, which could provide solutions for less connected regions. A shift toward more qualitative research, especially ethical and pedagogical issues, appears to have emerged after 2023.\u003c/p\u003e\u003cp\u003eTo address these gaps, several recommendations can be made. Methodologically, mixed-method studies combining learning analytics and ethnography would provide a more comprehensive assessment of generative AI's impact on metacognition. In terms of international collaborations, it is crucial to promote North-South projects on frugal AI, particularly through strategic partnerships between the European Union and Africa.\u003c/p\u003e\u003cp\u003eFrom an editorial perspective, encouraging education science journals to publish more on AI, including special issues on ethics and inclusion, would be relevant. Finally, at the institutional level, training teachers in AI literacy and targeted funding for research in underrepresented regions are essential.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study highlights that AI research in higher education is primarily structured around post-ChatGPT generative AI but does not fully integrate essential human dimensions such as equity and critical pedagogy. The observed systemic biases, both geographically and methodologically, call for an integrated approach that combines bibliometric data and qualitative studies to move beyond mere publication quantification.\u003c/p\u003e\u003cp\u003eIn conclusion, this research underscores the rapid evolution of AI in higher education, particularly due to advances in generative AI. While scientific production has grown significantly, gaps remain in interdisciplinarity, geographical diversity, and theoretical grounding. A more integrative approach combining bibliometric analyses with qualitative studies would refine the understanding of AI's transformative role in education.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eK A conceived and led the study. M A and S L contributed to data collection and preliminary analysis. I M conducted the advanced data analysis and prepared the visualizations. A J and K A wrote the main manuscript text. All authors reviewed and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors did not receive support or funding from any organization for the submitted work. All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data supporting the findings of this study were retrieved from the Scopus database. Access to these data is subject to license restrictions. Data can be made available by the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMcCarthy, J., Minsky, M. L., Rochester, N., \u0026amp; Shannon, C. E. (1956). Artificial intelligence (AI) coined at dartmouth. Retrieved October, 28, 2021.\u003c/li\u003e\n\u003cli\u003eRainer, K., Prince, B., Splettstoesser-Hogeterp, I., \u0026amp; SanchezRodriguez, C. (2016). Introduction to Information Systems ((4th ed.).). John Wiley \u0026amp; Sons Inc.\u003c/li\u003e\n\u003cli\u003eWells, R.E. (2023). Strong AI vs. weak AI: What\u0026rsquo;s the difference? Strong AI can do anything a human can do, while weak AI is limited to a specific task. LifeWire. Retrieved August 22,\u003c/li\u003e\n\u003cli\u003eChassignol, M., Khoroshavin, A., Klimova, A., \u0026amp; Bilyatdinova, A. (2018). Artificial intelligence trends in education: A narrative overview. Procedia Computer Science 136, 16\u0026ndash;24.\u003c/li\u003e\n\u003cli\u003eChen, L., Chen, P., \u0026amp; Lin, Z. (2020). Artificial intelligence in education: A review. IEEE Access, 8, 75264\u0026ndash;\u0026ndash;\u0026ndash;75278.\u003c/li\u003e\n\u003cli\u003eChiu, T. K. F., Xia, Q., Zhou, X., Chai, C. S., \u0026amp; Cheng, M. (2023). Systematic literature\u003c/li\u003e\n\u003cli\u003ereview on opportunities, challenges, and future research recommendations of artificial intelligence in education. Computers and Education: Artificial Intelligence, 4, Article 100118. https://doi.org/10.1016/j.caeai.2022.100118\u003c/li\u003e\n\u003cli\u003eOkonkwo, C. W., \u0026amp; Ade-Ibijola, A. (2021). Chatbots applications in education: A systematic review. Computers and Education: Artificial Intelligence, 2, Article 100033. https://doi.org/10.1016/j.caeai.2021.100033\u003c/li\u003e\n\u003cli\u003eLuan, H., \u0026amp; Chin-Chung, T. (2021). A Review of using machine learning approaches for precision education. Journal of Educational Technology \u0026amp; Society, 24, 250\u0026ndash;266.\u003c/li\u003e\n\u003cli\u003eHwang, G.-J., \u0026amp; Tu, Y.-F. (2021). Roles and research trends of artificial intelligence in mathematics education: A bibliometric mapping analysis and systematic review. Mathematics, 9, 594. https://doi.org/10.3390/math9060584\u003c/li\u003e\n\u003cli\u003eXu, W., \u0026amp; Ouyang, F. (2022). The application of AI technologies in STEM education: A systematic review from 2011 to 2021. International Journal of STEM Education, 9, 1\u0026ndash;20. https://doi.org/10.1186/s40594-022-00377-5\u003c/li\u003e\n\u003cli\u003eGonz\u0026acute;alez-Calatayud, V., Prendes-Espinosa, P., \u0026amp; Roig-Vila, R. (2021). Artificial intelligence for student assessment: A systematic review. Applied Sciences, 11(12), 5467. https://doi.org/10.3390/app11125467\u003c/li\u003e\n\u003cli\u003eGoksel, N., \u0026amp; Bozkurt, A. (2019). Artificial intelligence in education: Current insights and future perspectives In S. Sisman-Ugur \u0026amp; G. Kurubacak (Eds.), Handbook of Research on Learning in the Age of Transhumanism (pp. 224-236). IGI Global. doi: 10.4018/978-1-5225-8431-5.ch014.\u003c/li\u003e\n\u003cli\u003eRudolph, J., Tan, S., \u0026amp; Tan, S. (2023). ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching, 6(1). https://doi.org/10.37074/jalt.2023.6.1.9\u003c/li\u003e\n\u003cli\u003eLiu, Z., Kong, X., Liu, S., Yang, Z., \u0026amp; Zhang, C. (2022). Automated detection of emotional and cognitive engagement in MOOC discussions to predict learning achievement. Computers \u0026amp; Education, 181, Article 104461. https://doi.org/10.1016/j.compedu.2022.104461\u003c/li\u003e\n\u003cli\u003eNguyen, A., Ngo, H. N., Hong, Y., Dang, B., \u0026amp; Nguyen, B. P. T. (2022). Ethical principles for artificial intelligence in education. Education and Information Technologies, 1-21.\u003c/li\u003e\n\u003cli\u003ePopenici, S. A., \u0026amp; Kerr, S. (2017). Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning, 12, 1\u0026ndash;13. https://doi.org/10.1186/S41039-017-0062-8\u003c/li\u003e\n\u003cli\u003eTahiru, F. (2021). AI in education: A systematic literature review. Journal of Cases on Information Technology (JCIT), 23(1), 1-20.\u003c/li\u003e\n\u003cli\u003eChiu, T. K., Xia, Q., Zhou, X., Chai, C. S., \u0026amp; Cheng, M. (2023). Systematic literature review on opportunities, challenges, and future research recommendations of artificial intelligence in education. Computers and Education: Artificial Intelligence, 4, 100118.\u003c/li\u003e\n\u003cli\u003eRudolph, J., Tan, S., \u0026amp; Tan, S. (2023). War of the chatbots: Bard, Bing Chat, ChatGPT, Ernie and beyond. The new AI gold rush and its impact on higher education. Journal of Applied Learning and Teaching, 6(1), 364-389.\u003c/li\u003e\n\u003cli\u003eCrawford, J., Cowling, M., \u0026amp; Allen, K. A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). Journal of University Teaching and Learning Practice, 20(3), 1-19.\u003c/li\u003e\n\u003cli\u003eBannister, P., Urbieta, A. S., \u0026amp; Pe\u0026ntilde;alver, E. A. (2023). A systematic review of generative AI and (English medium instruction) higher education. Aula Abierta, 52(4), 401-409.\u003c/li\u003e\n\u003cli\u003eBozkurt, A., Junhong, X., Lambert, S., Pazurek, A., Crompton, H., Koseoglu, S., Farrow, R., Bond, M., Nerantzi, C., Honeychurch, S., Bali, M., Dron, J., Mir, K., Stewart, B., Costello, E., Mason, J., Stracke, C. M., \u0026amp; Romero-Hall, E. (2023). Speculative Futures on ChatGPT and Generative Artificial Intelligence (AI): A Collective Reflection from the Educational Landscape. Asian Journal of Distance Education, 18(1), 53-130. Article 1. https://doi.org/10.5281/zenodo.7636568\u003c/li\u003e\n\u003cli\u003eCrawford, J., Cowling, M., \u0026amp; Allen, K.-A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). Journal of University Teaching and Learning Practice, 20(3), 1\u0026ndash;19. https://search.informit.org/doi/10.3316/\u003c/li\u003e\n\u003cli\u003ePerkins, M. (2023). Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching and Learning Practice, 20(2), 1\u0026ndash;24. https://search.informit.org/doi/10.3316/informit.T2024111300009591751711095\u003c/li\u003e\n\u003cli\u003eAcosta-Enriquez, B.G., Arbul\u0026uacute; Ballesteros, M.A., Huaman\u0026iacute; Jordan, O. et al. Analysis of college students\u0026apos; attitudes toward the use of ChatGPT in their academic activities: effect of intent to use, verification of information and responsible use. BMC Psychol 12, 255 (2024). https://doi.org/10.1186/s40359-024-01764-z\u003c/li\u003e\n\u003cli\u003eRomero-Rodr\u0026iacute;guez, J. M., Ram\u0026iacute;rez-Montoya, M. S., Buenestado-Fern\u0026aacute;ndez, M., \u0026amp; Lara-Lara, F. (2023). Use of ChatGPT at university as a tool for complex thinking: Students\u0026rsquo; perceived usefulness. Journal of New Approaches in Educational Research, 12(2), 323-339.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Artificial Intelligence (AI), Higher Education (HE), Bibliometric Analysis, Research Trends, Generative Models, Interdisciplinary Collaboration, Ethics and Accessibility","lastPublishedDoi":"10.21203/rs.3.rs-7187231/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7187231/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSince its conceptualization by McCarthy (1956), artificial intelligence (AI) has evolved significantly, offering multiple applications in higher education (HE), such as personalized learning, automated assessment, and student success prediction. However, despite the proliferation of research, studies remain fragmented and lack a global vision, particularly after 2022.\u003c/p\u003e\u003cp\u003eThis bibliometric study analyzes 1,780 articles published between 1986 and 2025, extracted from the Scopus database. The PRISMA approach was applied to ensure a rigorous and reproducible selection of publications. The VOSviewer tool was used to explore research trends, scientific collaborations, and priority themes concerning AI in HE.\u003c/p\u003e\u003cp\u003eThe analysis reveals three main phases: (1) a latency phase (1986\u0026ndash;2013) focused on expert systems; (2) a phase of moderate growth (2014\u0026ndash;2019) with the emergence of chatbots and adaptive learning; and (3) a post-2020 explosion due to the rise of generative models such as ChatGPT. Most publications originate from the United States and China, with limited interdisciplinary collaborations. Themes related to accessibility, ethics, and frugal AI are scarcely addressed.\u003c/p\u003e\u003cp\u003eCurrent research shows geographical and thematic concentration, with a lack of diversity and interdisciplinarity. The absence of longitudinal studies and methodological standardization limits the comprehensive understanding of AI's impact on HE.\u003c/p\u003e\u003cp\u003eTo bridge these gaps, it is recommended to foster international collaborations, particularly North-South, and explore neglected themes such as accessibility, ethics, frugal AI, and pedagogical integration for a better understanding of AI's impact on HE.\u003c/p\u003e","manuscriptTitle":"Artificial Intelligence in Higher Education: A Bibliometric Analysis of Trends, Gaps, and Future Directions (1986–2025)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-22 09:38:29","doi":"10.21203/rs.3.rs-7187231/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3ca429d6-701c-4cc8-8dad-50922bf041ac","owner":[],"postedDate":"August 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-25T09:57:06+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-22 09:38:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7187231","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7187231","identity":"rs-7187231","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}