Available via license: CC BY 4.0
Content may be subject to copyright.
What is Ethical: AIHED Driving Humans or Human-Driven AIHED? A
Conceptual Framework enabling the ‘Ethos’ of AI-driven Higher education
Prashant Mahajan
R. C. Patel Institute of Technology, Shirpur, India, registrar@rcpit.ac.in, ORCID ID: 0000-0002-5761-5757
Abstract
Artificial Intelligence (AI) is transforming higher education (HE) by enabling personalized learning,
automating administrative processes, and enhancing decision-making. However, AI adoption
presents significant ethical and institutional challenges, including algorithmic bias, data privacy
concerns, and governance inconsistencies. This study introduces the Human-Driven AI in Higher
Education (HD-AIHED) Framework, an adaptive and structured model designed to integrate human
intelligence (HI) into every phase of the AI lifecycle—adoption, design, deployment, evaluation,
and exploration. Unlike conventional AI models that prioritize automation, HD-AIHED emphasizes
human-centered governance, ethical compliance, and participatory decision-making to ensure that
AI enhances rather than replaces human agency in HE.
The framework aligns AI applications with both institutional and student needs, fostering trust,
adaptability, and transparency. Its dual-layered approach integrates AI across both the AI lifecycle
and the student lifecycle, ensuring context-sensitive, equitable, and goal-oriented AI
implementation. A key contribution of this study is its regionally adaptable approach, recognizing
variations in technological infrastructure and policy landscapes. Additionally, the integration of
SWOC (Strengths, Weaknesses, Opportunities, and Challenges) analysis during the adoption phase
allows HE institutions to evaluate AI readiness, mitigate risks, and refine governance structures,
while the exploration phase ensures long-term adaptability, scalability, sustainability, and AI
promotion through continuous research, innovation, and interdisciplinary collaboration.
To ensure AI remains a responsible enabler in HE, this study advocates for the establishment of
University/HE Institutional AI Ethical Review Boards, alignment with global regulatory
frameworks (e.g., UNESCO AI Ethics Guidelines, GDPR, Sustainable Development Goal 4), and
the promotion of inclusive and transparent AI adoption policies.
Key insights from the HD-AIHED model highlight its role in bridging AI research gaps and
overcoming real-time challenges in global HE institutions. The framework offers tailored strategies
for diverse educational contexts, including developed and emerging countries, ensuring that AI
implementation is contextually relevant and ethically sound. By emphasizing interdisciplinary
collaboration among policymakers, educators, industry leaders, and students, this study envisions
AI as an ethical and equitable force for innovation in HE.
Ultimately, the HD-AIHED model serves as a catalyst for AI inclusivity rather than exclusion and
a driver of educational equity rather than disparity by embedding the ethos of AIHED into higher
education systems. Future research should focus on the empirical validation of HD-AIHED through
institutional case studies, AI bias audits, and longitudinal assessments to ensure ethical integrity,
transparency, and sustainability in AI deployment.
Keywords: Artificial Intelligence; Higher Education; AI in HE; Human Intelligence; AI Lifecycle;
Ethics; AI Adoption; Scalability and Sustainability.
1 of 63
1. Introduction
Artificial Intelligence (AI) is no longer just an emerging technology; it is fundamentally
transforming higher education (HE). AI-driven applications such as adaptive learning platforms,
predictive analytics, and automated grading systems are reshaping the way institutions function.
These tools offer personalized learning pathways, improve administrative efficiency, and enable
data-driven decision-making in student retention and academic performance tracking [1–3]
The evolution of Artificial Intelligence in Education (AIED) has predominantly focused on K-12
education (primary and secondary), driving advancements in adaptive learning, automated
assessments, and intelligent tutoring systems [4,5]. However, these innovations serve as a
foundation for AIHED, which extends AI applications to complexity of HE [6]. While AIED
research is rooted in general education, its principles apply to HE with necessary contextual
adaptations [7].
The rapid integration of AI in HE (AIHED) raises critical ethical and institutional concerns,
including algorithmic bias, student data privacy risks, and regulatory inconsistencies [5,8]. AI's
potential to amplify existing educational inequities has been widely documented, particularly
regarding biased datasets that disadvantage marginalized student groups [9]. Furthermore, AI
adoption remains fragmented, with developing nations facing infrastructure constraints and digital
divide issues, exacerbating educational disparities [10].
HE, long revered as a beacon of enlightenment, stands at a crossroads in this AI-driven
transformation. AIHED offers personalized learning, optimized institutional processes, and
expanded access for underserved populations [6,11]. Adaptive learning platforms, predictive
analytics, and immersive virtual environments redefine student success by enabling individualized
learning trajectories [10,12,13]. Empirical research demonstrates that AI-driven learning systems
outperform traditional methods, providing real-time feedback, personalized content, and adaptive
learning pathways [4,5,14]. AIHED not only enhances learning outcomes but frees educators to
focus on mentorship, creativity, and moral guidance [7]. Additionally, AI fosters global
collaboration, breaking barriers of geography, age, and accessibility, pushing HE into a realm of
limitless learning and discovery [6,15]. As an extension of human intellect, AIHED is not merely
transformative—it is transcendent [3].
By 2030, AI is projected to contribute approximately $15.7 trillion to the global economy, with
higher education emerging as a key beneficiary [16–18]. While this projection underscores AI’s
potential to democratize access to education, empirical studies indicate that current AI
implementations in HE often fail to meet student expectations, primarily due to fairness concerns,
regulatory gaps, and algorithmic biases [19–21]. Research has shown that AI-driven assessments
and decision-making systems sometimes exhibit discriminatory tendencies, disproportionately
affecting marginalized student populations and raising ethical concerns [9]. Additionally, the lack
of standardized AI governance models has led to inconsistencies in AI adoption across institutions,
impacting access and equity in educational opportunities [10].
Empirical evidence suggests that structured AI governance frameworks—integrating algorithmic
fairness reviews, human oversight, and participatory governance—can significantly enhance AI’s
effectiveness in HE [22]. To ensure responsible and sustainable AI deployment, HE institutions
must prioritize ethical compliance, transparency, and stakeholder engagement, mitigating biases
and fostering equitable educational outcomes.
This study proposes the Human-Driven AI in Higher Education (HD-AIHED) Framework—a
structured and ethical approach that incorporates human oversight into AI adoption across HE
institutions. Unlike conventional AI models that prioritize automation, HD-AIHED ensures that AI
remains a tool for empowerment rather than exclusion. The framework is context-sensitive,
addressing the distinct needs of diverse educational settings, while aligning AI applications with
institutional priorities, global ethical standards, and student-centric learning models.
2 of 63
1.1 Problem Statement: Reporting And Analysis
The integration of AI into higher education (HE) is not merely a technological advancement but an
ethical reckoning that challenges the foundational principles of HE [23,24]. While AI offers
transformative opportunities—such as personalized learning, enhanced administrative efficiency,
and expanded accessibility—it also introduces a duality that must be critically examined.
The 2024 Global AI Student Survey highlights this dual nature of AI in HE, revealing both its
potential for transformation and its pressing challenges. On the positive side, 86% of students
worldwide report incorporating AI tools into their studies, showcasing AI’s growing role in HE.
However, 60% of respondents express concerns about fairness, citing risks of algorithmic bias in
AI-driven decision-making [19,22].
Additionally, privacy emerges as a key issue, with students questioning the transparency and
security of their data, raising concerns about institutional accountability. Despite AI’s promise, 80%
of students believe that institutional AI implementations fail to meet expectations, highlighting gaps
between AI innovation and its real-world application in HE. These findings underscore AI’s double-
edged impact—driving innovation while reinforcing the urgent need for ethical, inclusive, and
transparent governance frameworks in HE [19].
Similarly, a McKinsey & Company report found that approximately 85% of AI practitioners
identified bias in algorithms as a key challenge for implementation in HE, emphasizing the necessity
of robust frameworks for transparency, fairness, and responsible AI adoption (Singh, 2024).
Addressing these concerns requires multi-stakeholder collaboration, rigorous policy frameworks,
and adaptive AI governance models that align AI advancements with educational equity, privacy
protection, and institutional integrity in HE [9,20].
In the global context, institutions exemplify both the successes and challenges of AIHED. This dual
impact showcases transformative achievements alongside critical shortcomings. For instance, AI-
powered grading tools like Gradescope have streamlined grading processes, reducing grading time
by 50% at UC Berkeley [25]. However, similar AI-driven systems, such as the UK’s algorithmic
grading during the COVID-19 pandemic, revealed systemic biases that disproportionately
disadvantaged underprivileged students [26].
Additionally, predictive analytics at Georgia State University improved graduation rates by 22% by
identifying and supporting at-risk students [27]. Yet, similar tools like Ellucian Analytics
disproportionately flagged minority students as “at-risk”, exacerbating existing inequities [25,28].
Furthermore, platforms like Coursera have expanded access to high-quality education for over 100
million learners globally, particularly in developing regions [29]. However, their limited
accessibility in under-resourced areas has deepened the digital divide [30]. Similarly, Squirrel AI
in China has demonstrated success by personalizing learning for millions in underserved areas,
improving educational outcomes [31], yet its invasive tracking of student behaviors without consent
has raised significant privacy and ethical concerns [32].
The widespread integration of AIHED challenges fundamental social and human principles,
reshaping the dynamic between humanity and technology [2,33,34]. As [35] notes, every
technological innovation brings both empowerment and unintended consequences. While AIHED
has democratized access to knowledge and fostered inclusivity, it also introduces ethical dilemmas
such as algorithmic bias, data privacy violations, and the erosion of interpersonal connections, as
revealed in UNESCO’s report [36,37].
The focus must shift from merely questioning AI’s future influence to thoughtfully directing its
integration to align with human values. In HE—a domain that fosters intellectual growth and
societal advancement—this balance between technological progress and purposeful implementation
is essential [38,39].
3 of 63
Leveraging AI’s transformative potential while addressing its unintended consequences will help
preserve HE as a human-centered domain that nurtures empathy, creativity, and diversity. To
achieve this, human oversight must ensure AI aligns with pedagogical goals and the ethical
imperatives of HE [6,37].
However, one of the key challenges in AI adoption is ensuring that governance structures
incorporate student agency and participatory decision-making. Existing AI ethics discussions in HE
tend to focus primarily on educators, policymakers, and institutions, with limited involvement of
students in shaping AI policy frameworks [10,22].
This study proposes an inclusive approach where students actively contribute to AI governance,
ensuring that educational AI applications remain human-centered and aligned with user needs [40]
[19,41].
As [42,43] posits, technology is never neutral, as it inherently reshapes human interactions and
values, illuminating certain possibilities while obscuring others. In this vein, [40] raise a
fundamental question about AIHED: Is AI merely performing ethical actions, or is it processing
ethically? [36] argue that AI tools are not neutral but embedded with the values and biases of their
developers.
Thus, without careful human oversight, AI risks reducing HE to algorithmic decision-making,
stripping it of the moral and emotional depth essential for meaningful learning [44].
1.2 Framing the Study: Aspirations, Purpose, and Objectives
The integration of AI into HE remains in its nascent stages, underscoring the pressing need for
comprehensive research into its adoption, application, and broader impact [37]. This study seeks to
address the duality of AIHED by proposing a conceptual framework that emphasizes ethical,
inclusive, and human-centered AI integration.
The primary objective is to harmonize the role of AI as both a transformative driver in HE and a
humanized collaborative tool anchored in core human values. By achieving this balance, the study
aims to design a strategic roadmap that enhances learning outcomes, fosters equity, and safeguards
the ethical and emotional dimensions vital to meaningful HE [6,8].
Central to this study is embedding accountability, adaptability, and empathy into every phase of the
AI lifecycle, from adoption to deployment. By treating AI as an augmentation of human intelligence
rather than a replacement, the framework aligns with arguments that transparency and inclusivity
must be prioritized to empower rather than exclude [26]. Additionally, AI tools are not neutral; they
reflect the values and biases of their creators [36]. By emphasizing human oversight, the framework
seeks to mitigate potential harm and ensure that AI complements rather than diminishes the human
essence of HE. To preserve creativity, empathy, and interpersonal connections, the study proposes
measures that ensure AI applications align with human-centric values.
Furthermore, as [6] highlight, the lack of stakeholder’s involvement in AI development risks
undermining pedagogical integrity. This framework prioritizes active educator participation in AI
implementation to align AI capabilities with institutional goals and pedagogical principles.
Similarly, [45] underscore the importance of addressing the social dimensions of AI, including
cultural and societal diversity. By incorporating these factors, the proposed framework ensures
inclusivity, scalability, and responsiveness to the diverse needs of learners across various contexts.
Progress in AIHED, as envisioned in this study, is measured not only by learning outcomes but also
by the ethical processes and intentionality guiding its adoption. By aligning AI’s transformative
capabilities with the moral foundations of HE, the study aspires to create a future where innovation
enhances humanity while preserving HE's core values [8].
4 of 63
To achieve these objectives, the study proposes the Human-Driven AIHED framework, a
conceptual model emphasizing equity, transparency, and accountability. This framework bridges
the gap between technological advancements and ethical integrity, ensuring that AI serves as a
collaborative tool that upholds ethical and institutional values rather than undermining them. In
doing so, it redefines AI’s role in HE as a driver of innovation that supports humanity, enhancing
HE while safeguarding its foundational principles.
Accordingly, the study’s objectives are:
To identify and analyze gaps in existing AIHED integration frameworks, particularly in
terms of ethics, inclusivity, and human agency, and propose effective strategies to address
them.
To investigate the dual role of AIHED in shaping regional AI adoption patterns and examine
the ethical challenges and opportunities it presents for higher education (HE) institutions
worldwide.
To develop and empirically validate the Human-Driven AIHED framework, ensuring that it
supports ethical, inclusive, and human-centered AI integration within HE systems.
To examine real-time global challenges that hinder AI integration in HE, and assess the
operability, adaptability, and effectiveness of the Human-Driven AIHED model.
To explore the role of interdisciplinary collaboration and participatory co-design in
promoting the ethical, responsible, and sustainable deployment of AIHED in HE
institutions.
To formulate actionable policy and implementation recommendations for key stakeholders,
including educators and policymakers, to ensure the ethical, effective, and context-sensitive
integration of AIHED in HE.
By addressing these objectives, the study provides a structured pathway for ethically integrating AI
in HE, ensuring that technological innovation aligns with humanity’s highest ideals and educational
values. To achieve the research aims and objectives of promoting a Human-Driven AIHED
framework, this study focuses on addressing the following research questions:
1. What gaps exist in current AIHED integration frameworks regarding ethics, inclusivity, and
human agency, and how can they be effectively addressed?
2. How does the dual role of AIHED influence regional AI adoption patterns, and what ethical
challenges and opportunities does it present for HE institutions worldwide
3. How can the Human-Driven AIHED framework be developed and empirically validated to
ensure ethical, inclusive, and human-centered AI integration in HE systems?
4. What real-time global challenges hinder the integration of AI in HE, and how can the
operability of the Human-Driven AIHED model be evaluated?
5. How can interdisciplinary collaboration and participatory co-design enhance the ethical and
responsible deployment of AIHED?
6. What actionable recommendations can be provided to key stakeholders—including
educators, and policymakers—to ensure the ethical and effective integration of AI in HE?
2.0 Research Methodology
Figure 1 details the structured research methodology used in this study. It outlines the sequential
steps, starting with research methods, literature review, and theoretical framework, followed by data
extraction, synthesis, and conceptual model development, culminating in model validation to ensure
applicability and reliability.
5 of 63
Figure 1: Research Methodology for developing a Conceptual Framework (Own Source)
2.1 Nature of Study, Method and Purpose
This study employs a qualitative meta-synthesis approach, integrating findings from multiple
qualitative and quantitative studies to identify patterns, contradictions, and gaps in AI adoption
within HE [46,47]. The meta-synthesis method is particularly effective for conceptual research
[10,48], as it reinterprets existing datasets through theoretical and ethical lenses to develop new
governance frameworks for AI in HE [49].
2.2 Data Sources and Selection
Secondary data available in English was synthesized from four key sources. Academic literature
provided critical insights into the benefits, successes, challenges, and ethical considerations
surrounding AI applications in education [8]. Secondary data for this study was sourced from four
primary domains:
Academic Literature – This included peer-reviewed journals and conference proceedings providing
empirical evidence on AI applications, ethical considerations, and institutional challenges in HE
[10].
Global Policy Reports – Documents from UNESCO, OECD, and IEEE were examined to
understand AI’s ethical and regulatory landscape [31,50,51].
Institutional Case Studies – Real-world applications of AI in HE were analyzed to extract both best
practices and challenges in AI implementation [38,39].
Technology Adoption Theories – To contextualize AI integration within established educational
paradigms, this study reviewed key technology adoption and learning theories. These include the
Technology Acceptance Model (TAM) frameworks [52], the Diffusion of Innovation (DoI) Theory
[53], the Unified Theory of Acceptance and Use of Technology (UTAUT) [54], and Constructivist
Learning Theory (CLT). These models provide insights into user acceptance, innovation diffusion,
and pedagogical adaptability in AI-enhanced learning environments.
6 of 63
Additionally, to understand broader socio-educational implications, the study incorporates Trow’s
Massification Model [55], Bourdieu’s Social Capital Theory [56], and Barnett’s Higher Education
Theory [56]. These theories help frame AI integration within higher education’s structural
evolution, social dynamics, and institutional governance.
To ensure a comprehensive and high-quality literature review, sources were gathered from reputable
academic databases, including: Web of Science, Scopus, ERIC (Education Resources Information
Center), EBSCOHost, IEEE Xplore, ScienceDirect, ACM Digital Library, SpringerNature and
Google Scholar [10].
The search strategy prioritized studies focusing on ethics, inclusivity, and institutional alignment in
AI integration within education or higher education. Research exclusively centered on technical
implementations of AI, without pedagogical or ethical considerations, was excluded to maintain the
study’s education-focused scope [57].
2.3 Data Synthesis: Extraction and Categorization
The data synthesis was conducted through two complementary approaches: data-driven descriptive
themes and theory-driven analytical themes. The descriptive themes emerged directly from the
literature, focusing on observable patterns and recurring insights, providing a practical
understanding of AI’s current role in HE [40]. Meanwhile, the analytical themes were developed
through a conceptual lens, employing established theoretical frameworks to interpret and analyze
the data. By integrating these approaches, this study synthesizes practical observations with
theoretical insights to develop the HD-AIHED framework, ensuring that AI adoption in higher
education remains equitable, scalable, and ethically aligned
2.3.1 Thematic Approach
Thematic analysis was used to identify and categorize recurring themes in the literature [58,59].
The structured process included:
Evaluation of Foundational Theories – AI adoption models were critically examined, revealing
gaps in addressing institutional alignment and stakeholder concerns [7,54].
Highlighting AI’s Duality – AI’s potential for efficiency, personalization, and scalability was
contrasted against ethical concerns, algorithmic bias, data privacy risks, and inclusivity challenges
[26,60–62].
Comparing Global Frameworks – UNESCO’s AI Ethics Guidelines, OECD’s AI Principles, and
the European AI Act were analyzed for their applicability in HE, highlighting the need for local
adaptation while preserving human integrity [31,50,63].
Human-Centered Integrated System Approach – AI integration was examined in the context of
human-centered systems, ensuring that technology remains a facilitator of human intelligence rather
than a replacement [36,39].
2.3.2 Analytical Approach
The analytical approach introduced explicit mechanisms to extend primary study content into a
theoretical framework. Analytical themes were carefully designed to address gaps identified in the
descriptive findings, integrating ethical, operational, and human-centric perspectives to ensure a
holistic understanding of AI in HE [31,64]. The process involved:
Mapping Global Challenges – Identifying algorithmic bias, digital divide, regulatory fragmentation,
and ethical dilemmas in AI adoption [8,26].
Integrating Ethical Considerations – Aligning AI adoption with beneficence, non-maleficence,
autonomy, justice, and explicability to ensure ethical AI deployment in HE [8,40].
7 of 63
Stakeholder-Centric AI Adoption – Ensuring AI integration incorporates educator participation,
student agency, and regulatory oversight to balance technological progress with human-centered
HE [13,38].
Decision-Making and Feedback as Human Intelligence – AI adoption was assessed through
decision-making processes, emphasizing the role of human oversight in guiding AI-generated
outcomes [6,39].
AI Life Cycle as Phased Human Intelligence – The AI integration cycle was examined as a phased
approach, incorporating human intelligence at various stages of adoption, design, deployment,
analysis, and exploration [13,36]. This approach ensures that AI integration in HE is innovative,
ethically sound, and adaptable to the diverse and evolving needs of HE contexts throughout the AI
lifecycle [10,49].
SWOC Analysis for Adoption and Exploration – A structured SWOC (Strengths, Weaknesses,
Opportunities, and Challenges) framework was utilized to evaluate AI adoption strategies and
potential areas for future growth [38,40].
External Intelligence – AI systems were analyzed in the context of external intelligence,
incorporating interdisciplinary perspectives to enhance adaptability and scalability [8,20].
This double powered approach provided a comprehensive framework that aligns AI integration with
institutional priorities and ethical imperatives, ensuring that AI adoption remains inclusive, scalable,
and accountable [31,50].
2.4 Conceptualization of the Framework
The conceptual framework of this study integrates insights from the literature review, theoretical
models, and the researcher’s intelligence [39]. It emphasizes the synergy between technological
innovation and humanized ethics, promoting accountability, transparency, and scalability. The
framework establishes modeling relationships between theories, synthesizing complex theoretical
data into thematic and analytical while linking existing research with practical implementation
strategies [65].
By prioritizing human oversight, inclusivity, and adaptability, the framework ensures ethically
grounded AI applications in HE, aligning innovation with core human values and institutional
priorities [31,40].
The framework is designed to distinguish AI's roles across key higher education (HE) functions—
teaching, assessment, and administration—ensuring that ethical guidelines are tailored to each
domain [11,66]. Additionally, a structured governance roadmap is integrated to define the
responsibilities of institutions, policymakers, and AI developers (industry) in promoting responsible
AI deployment [10,31,40].
2.5 Ethical Considerations and Framework Validation
The proposed framework underwent validation through comparative analysis with global AI ethics
frameworks and institutional case studies [31,50]. Validation involved:
Testing Real-World AI Applications – Evaluating each phase of the HD-AIHED framework
(adoption, design, deployment, analysis, and exploration) using real AI systems to assess feasibility
and ethical alignment [8].
Analyzing Case Studies of Real-World Challenges– Reviewing institutional experiences with AI to
determine the framework’s adaptability across diverse HE settings [38].
Guidelines for Quantitative Research with Empirical Data for Validation – Incorporating
quantitative research methodologies, including statistical analysis, surveys, and experimental
8 of 63
designs, to empirically validate AI implementation outcomes in HE. This ensures that AI models
are assessed based on measurable effectiveness, equity, and ethical considerations [3] [37].
These validation efforts ensured that the framework remains applicable, scalable, and aligned with
global ethical standards while addressing localized challenges [31,51].
2.6 Empirical Validation Strategy
While this study develops a conceptual framework, future empirical testing is required to evaluate
the HD-AIHED model’s applicability by adopting following strategies.
Institutional Case Studies: A comparative analysis of HE institutions applying the
framework in diverse contexts to assess effectiveness [11,38,67].
AI Bias Auditing: Implementing a structured evaluation method to identify and mitigate
algorithmic biases in AI-driven educational tools [20,68,69].
Human-AI Interaction Metrics: Developing key performance indicators (KPIs) for
inclusivity, transparency, and student engagement to assess AI integration success [10,70].
Future research should focus on mixed-method approaches, incorporating both qualitative
and quantitative analyses to assess AI's real-world impact on HE [10,11].
2.7 Limitations
While this study provides a strong conceptual foundation, its reliance on secondary data and meta-
synthesis introduces limitations regarding empirical validation. Future research should incorporate
longitudinal studies, primary data collection, and comparative regional analyses to enhance the
framework’s practical applicability [10,37]. The study also acknowledges the challenges posed by
rapid advancements in AI, requiring continuous updates to the framework to ensure relevance in
evolving HE landscapes.
Another limitation is the variability in institutional AI adoption rates, which may influence the
applicability of the proposed framework across different HE contexts. Additionally, the integration
of external intelligence and human-centered AI approaches requires further empirical validation
through experimental and real-world deployment studies [13,20]. Despite these limitations, this
study makes a significant contribution to the ethical and human-centric adoption of AI in HE.
3. Literature Review and Theoretical Frameworks
The field of Artificial Intelligence in Education (AIED), particularly within Higher Education (HE),
has undergone significant growth since 2018, driven by rapid technological advancements and the
global shift toward digital learning, particularly accelerated by the COVID-19 pandemic. This
expansion has led to innovations in personalized learning, predictive analytics, and administrative
automation. However, the ethical implications of AI deployment in education remain a crucial area
of academic investigation. Research indicates that China is a leading contributor to AIED
scholarship, followed by the United States and Spain, with AIED consistently being a dominant
research priority [37].
Aligned with the principles of conceptual framework development [39], this literature review
critically assesses the evolution, global trends, and applications of AIED in HE, alongside the ethical
challenges it presents. The review interrogates the intersection of human agency and AI,
emphasizing how AI’s integration in HE is reshaping learning models and institutional governance.
By synthesizing findings from empirical studies and theoretical perspectives, the review highlights
research gaps and proposes a strategic framework to balance AI’s transformative potential with
inclusivity, accountability, and equity.
3.1 An Overview of AIED
3.1.1 Defining Artificial Intelligence
9 of 63
Artificial intelligence (AI) has been defined and conceptualized through diverse lenses across
academic and industrial domains, reflecting its multifaceted and transformative nature. Broadly, AI
is characterized as a system capable of simulating human intelligence to solve problems, analyze
complex data, and support decision-making in challenging environments [3]. This perspective
underscores AI’s transformative potential, particularly its capacity to process and analyse vast
datasets, revolutionizing decision-making processes across industries, including HE.
An alternative definition by [71], frames AI as systems capable of achieving or exceeding human-
level performance in various contexts. This view emphasizes AI’s adaptability and self-learning
capabilities, which extend beyond static, predefined knowledge. Similarly, [7] describe AI as a
cognitive catalyst that solves complex problems through pattern recognition and predictive
analytics. This characterization highlights AI’s ability to enhance human cognition, particularly in
domains that demand sophisticated analytical and decision-making capabilities.
In line with this, [72] offer a comprehensive understanding of AI, emphasizing its ability to perform
intelligent behaviors, including reasoning, learning, and planning. Their foundational work
establishes a theoretical basis for understanding AI's applications in various sectors. Additionally,
[73] argue that AI encompasses any activity performed by machines that was previously thought to
require human intelligence, further broadening the field’s scope.
Despite the nuanced distinctions among these definitions, they converge on the central idea of AI
as a powerful enabler that transcends human cognitive limitations. The applications of AI span
diverse fields, including HE, where its impact is both profound and transformative. AI’s evolution,
marked by rapid advancements in machine learning, natural language processing, and autonomous
systems, has transitioned from an evolutionary phase to a revolutionary era, representing a paradigm
shift in technology and innovation [74].
3.1.2 Emergence of AIHED
Although the theoretical foundations of AI trace back over seven decades, its integration into HE
has gained significant traction only in recent years [75,76]. The COVID-19 pandemic served as a
critical inflection point, driving the rapid adoption of AI tools and catalyzing double-digit growth
in AI technologies [77]. This unprecedented global crisis accelerated the deployment of AI-driven
solutions in teaching, learning, and administrative processes, fundamentally reshaping HE
ecosystems [39,78].
The shift toward AI in HE has also been documented by [79], who explore the role of AI in
designing personalized learning experiences. They highlight AI’s potential to foster deeper learning
by analysing student behaviours and tailoring interventions. Similarly, [5] provide evidence of AI’s
role in improving student engagement and retention through adaptive learning systems.
Once considered an enigmatic “black box” technology, AI has evolved into a transparent and
integral component of educational ecosystems. Notably, the EDUCAUSE Horizon Report: 2019
Higher Education Edition anticipated the swift proliferation of AI applications in HE, predicting
exponential growth and transformative integration, surpassing adoption rates in many other
industries [41]. Subsequent trends validated these forecasts, with the AI Index 2024 Report
documenting a threefold increase in AIED offerings since 2017, underscoring the sector’s
accelerating pace of AI adoption [80].
3.1.3 Market Growth and Global Adoption of AIED
The global AIED market is projected to grow at a compound annual growth rate (CAGR) of 31.2%
between 2025 and 2030, signifying substantial investments in AI-driven educational transformation
[81]. ‘The Pearson 2024 End of Year AI Report for HE’ corroborates these trends, reporting
significant increases in student engagement and academic participation between Spring and Fall
2024 [82]. Additionally, [31] underscores AI’s potential to bridge educational disparities,
10 of 63
particularly in underserved regions. However, challenges such as digital access inequalities and AI-
induced biases persist, necessitating robust governance frameworks [45].
3.2 AIED and AIHED: Convergence and Divergence
The integration of AIED has significantly influenced the development of AIHED, as both domains
share foundational principles in personalized learning, student support, and adaptive assessments
[6]. While AIED research offers valuable insights into student engagement, automated assessments,
and adaptive learning models, AI in HE extends beyond classroom applications to impact
institutional governance, faculty recruitment, research integrity, and accreditation processes [26].
AI applications in HE often originate from AIED-driven innovations, particularly in adaptive
learning, student support systems, and assessment automation [10,11]. For example, AI-powered
platforms such as Khan Academy, Carnegie Learning, and Duolingo have set the foundation for
university-level MOOCs like Coursera, edX, and FutureLearn, which provide personalized learning
pathways and AI-based skill certification programs [26]. Similarly, AI chatbots such as Georgia
State University’s Pounce AI, initially developed for K-12 student support, have evolved into
predictive analytics systems that enhance university student retention and success [6]. AI-driven
grading and plagiarism detection, originally designed for school-level academic integrity
enforcement, now play a crucial role in peer review automation, faculty evaluations, and research
assessments in HE [8].
Despite these overlaps, AI in HE introduces distinct challenges that necessitate specialized
governance frameworks addressing faculty governance, student job placements, research integrity,
institutional decision-making and autonomy, external linkages, academic regulatory bodies,
industry, accreditations and alumni networks [6,37,83]. While [84] outline the broad potential of AI
in education, AI’s integration in HE presents specific governance challenges that require structured
regulatory frameworks [38].
While early AI research in education [85] emphasized adaptive learning, recent studies in higher
education (HE) have shifted focus toward AI’s role in academic autonomy, decision-making,
stakeholder integration, and research integrity management. AI is increasingly being integrated into
institutional policies to streamline decision-making and governance structures [24,38]. Universities
are exploring AI-driven strategies to enhance administrative efficiency while maintaining alignment
with pedagogical and ethical values [86].
The integration of AI also extends to internal and external stakeholder collaboration, ensuring
participatory governance in HEIs. Institutional policies now emphasize structured AI adoption that
includes faculty, students, policymakers, and industry partners [22,87]. Additionally, generative AI
and learning analytics are being leveraged to create more inclusive, personalized learning
environments while addressing ethical concerns [11].
A critical aspect of AI adoption in HE is research integrity management, with scholars highlighting
the need for fairness, accountability, and transparency in AI applications [20]. AI-driven tools are
being evaluated to ensure that they uphold academic integrity, mitigate bias, and comply with
ethical standards [37,88]. As HE institutions navigate this transformation, it is crucial to maintain a
balance between AI-driven efficiency and core human-centric values, ensuring responsible and
sustainable AI integration.
However, concerns about algorithmic bias, data privacy risks, and AI over-reliance, observed in AI-
driven K-12 assessments, continue to persist in HE as well [8]. This is why this study also considers
the fundamental approaches of AIED within AIHED to some extent, ensuring that AI adoption in
HE remains ethically and strategically aligned with academic goals.
However, given the study’s focus on the structured integration of AI in HE, it also underscores the
necessity of clear oversight to ensure that universities and HE institutions retain their critical role
11 of 63
in knowledge production, intellectual discourse and the preservation of fundamental principles of
HE [31].
3.3 Theoretical Frameworks Enabling Global Adoption
Artificial Intelligence in Higher Education (AIHED) is influenced by various theoretical models
that shape its adoption, governance, and impact. Table 1 presents a global perspective on AIHED
theoretical frameworks, highlighting their relevance, applications, and associated challenges in
different educational contexts.
Table 1: Global Perspective on AIHED Theoretical Framework
Theory &
Founder(s)
Focus
Domain &
Country
Core Idea & AI
Relevance
Impact on AI in
Higher Education
Applications in
AI-Driven
Education
Limitations & Risks
of AI in HE
Trow’s
Massification
Model (1973) [55]
Higher
Education
(World-
wide)
Mass education,
scalability, MOOCs,
AI tutors
AI democratization,
access expansion,
scalable models
Coursera, edX,
AI tutors
Standardization risk,
reduced engagement
[6]
Bourdieu’s Social
Capital Theory
(1986) [56]
Higher
Education
(World-
wide)
Social capital,
networked learning,
mentorships
AI-powered
networking, peer
support, knowledge
sharing
LinkedIn
Learning,
ResearchGate
Digital literacy gaps,
academic inequality
[83]
Barnett’s Higher
Education Theory
(1990) [89]
Higher
Education
(World-
wide)
Knowledge
production, student
engagement,
research
collaboration
AI-driven learning
paths, lifelong
learning,
curriculum
enhancement
AI-driven
advising, research
assistants
Overuse of AI, critical
thinking reduction
[26]
Technology
Acceptance Model
(TAM) (1989) [52]
Higher
Education
(United
States)
User-friendliness, AI
adoption, student
engagement
Learning efficiency,
LMS adoption
AI-enhanced
LMS, adaptive
learning
Cost, rural-urban
divide, privacy
concerns [27,80]
Unified Theory of
Acceptance and
Use of Technology
(UTAUT) (2003)
[54]
Higher
Education
(European
Union)
AI adoption, ethical
deployment,
performance metrics
AI governance,
structured policies,
analytics
AI-driven
performance
tracking, ethical
AI
Infrastructure
disparity, regulatory
delays [27,90]
Diffusion of
Innovations (DoI)
Theory (1962) [91]
Higher
Education
(China)
AI adoption patterns,
innovation scaling,
government support
Large-scale AI
implementation,
government
backing
State-led AI
platforms,
research
investment
Urban-rural gap, data
privacy concerns [90]
Constructivist
Learning Theory
(CLT) by Jean
Piaget and Lev
Vygotsky [92,93]
Higher
Education
(India)
Experiential
learning, equity-
focused AI, diverse
learning needs
AI-personalized
education, access
expansion
AI literacy
programs, mobile
AI tools
Infrastructure limits,
training gaps,
scalability [80]
The global adoption of AI in higher education (AIHED) is influenced by diverse theoretical
frameworks, each shaping AI's role in educational governance and pedagogy [27,80,94]. In the
United States, AI integration follows TAM, CLR, and Trow’s Massification Model, promoting
scalable access but facing challenges related to high costs and privacy concerns. The European
Union relies on UTAUT and Barnett’s Higher Education Theory, focusing on ethical AI governance
and performance metrics, yet struggles with regulatory delays and infrastructural disparities.
China’s AIHED adoption, guided by DoI and Trow’s Model, benefits from strong government
12 of 63
support for rapid AI scaling but encounters urban-rural resource gaps and ethical issues. In India,
CLR and Bourdieu’s Social Capital Theory emphasize equity-focused AI applications and peer
networks, though infrastructure limitations and digital inequalities persist. At the global level, AI
adoption varies, shaped by governance models like Barnett’s Higher Education Theory, but remains
hindered by ethical bias, digital disparities, and lack of universal regulation. Thus, these adoption
theories explain AI trends in HE but fail to address ethical AI governance and stakeholder-driven
decision-making. The proposed framework extends these models by incorporating human oversight,
participatory governance, and real-time AI monitoring, ensuring a more ethical and sustainable AI
integration strategy.
3.4 AIHED Applications and Perspectives: A Global Transformation
Advancements in AIED are revolutionizing HE worldwide, merging technology, pedagogy, and
innovation to create inclusive, personalized, and effective learning ecosystems [76]. AI-driven tools
are reshaping educational practices by introducing adaptive learning platforms, predictive analytics,
and real-time engagement systems that cater to diverse learner needs and institutional priorities
[6,67,83,95,96]. However, these innovations must balance technological potential with ethical
considerations to ensure equitable and sustainable adoption [83,95]. The integration of AIHED
extends beyond functionality—it is a force that redefines human cognition, decision-making, and
creativity [22]
3.4.1 Teaching and Learning: A Paradigm Shift
AIED transforms teaching and learning by introducing personalized learning platforms, predictive
analytics, and real-time engagement tools, which collectively create student-centric and adaptive
environments.
Personalized and Adaptive Learning:
AI-powered platforms like Carnegie Learning and Khanmigo—intelligent tutoring systems
(ITS)—dynamically assess student performance and personalize content delivery, ensuring
targeted support for individual learning needs [7,12]. These AI-driven adaptive learning
tools leverage real-time analytics to enhance student engagement, foster individualized
instruction, and address learning gaps [17,25]. Empirical studies further indicate that human-
centered AI models integrated with ITS platforms improve learning outcomes and
accessibility by providing automated, tailored feedback [22,97].
Global platforms like Coursera and edX extend these capabilities to lifelong learning,
aligning educational content with learners' goals to foster career-ready outcomes [11,84].
Platforms like Coursera and edX connect educators and students worldwide, promoting
cross-cultural exchanges and aligning with Sustainable Development Goal 4 for equitable
access to education [98].
Global platforms like Coursera and edX extend AI-driven personalized learning to lifelong
education, aligning educational content with learners' goals to foster career-ready outcomes
[11,84]. These platforms leverage AI-powered analytics and adaptive learning strategies to
support individualized skill development and professional growth [25,97]. Furthermore,
Coursera and edX serve as global learning hubs, connecting educators and students
worldwide, fostering cross-cultural exchanges, and advancing Sustainable Development
Goal 4 by promoting equitable access to quality education [98,99].
Adaptive AI tutors, such as Carnegie Learning’s Math Tutor, have demonstrated superior
effectiveness over traditional teaching methods by providing real-time feedback and
individualized learning paths [79]. Similarly, AI-driven tools like Pounce, a chatbot at
Georgia State University, reduce enrollment attrition by offering real-time, personalized
support, highlighting AI’s potential to enhance student retention and success [11].
By integrating real-time analytics and predictive modeling, adaptive learning systems
continuously refine content delivery, improving accessibility, engagement, and learning
outcomes [6,10].
13 of 63
Predictive Analytics for Academic Success:
Predictive tools in HE empower institutions to anticipate and address challenges by
analyzing student behavior and performance data. For example, the University of Guayaquil
identifies at-risk students using predictive analytics, enabling timely resource allocation to
improve retention [100,101] [37]. In Africa, similar tools address institutional inefficiencies
and improve learning outcomes [102].
Educational Data Mining (EDM) and predictive analytics play a crucial role in identifying
at-risk students early, facilitating timely interventions that improve retention and
engagement [14]. Additionally, predictive analytics enhance resource management by
forecasting enrollment trends and operational needs, ensuring institutional strategies align
with student success goals [10]. Predictive analyse and optimize resource allocation,
enabling institutions like the University of Melbourne to achieve significant efficiency gains
[100]. Squirrel AI has tailored education to millions in China, improving students’ outcomes
for underserved populations [99].
Real-Time Assistance and Engagement:
AI-powered chatbots and gamified platforms create interactive, personalized learning
experiences that foster engagement and improve outcomes and skills development.
Gamification promotes intrinsic motivation and long-term retention, while immersive tools
like VR and AR bridge theoretical concepts and practical applications [10,12,37,103,104].
These global studies suggest that AIED has fundamentally transformed traditional teaching and
learning practices into dynamic, adaptive environments that emphasize personalization, inclusivity,
and engagement. By leveraging adaptive learning platforms, predictive analytics, and immersive
technologies, higher education institutions have shifted from standardized methods to tailored,
student-centric approaches that address diverse needs and foster better academic outcomes. AI-
powered tools not only enable real-time interventions and personalized support but also optimize
resource allocation and institutional efficiency. The integration of technologies like gamification
and VR/AR bridges theoretical knowledge with practical application, enhancing engagement and
preparing students for the challenges of a rapidly evolving global workforce. Ultimately, these
advancements mark a paradigm shift in higher education, demonstrating the potential of AIED to
create equitable, effective, and impactful learning experiences. As these technologies continue to
evolve, they promise to redefine the future of education, ensuring that it is adaptive, inclusive, and
aligned with the demands of a globalized and digital society.
3.4.2 Administrative Processes
AI is redefining administrative operations in HE, automating repetitive tasks, enhancing resource
management, and enabling strategic innovation. AI automates grading, attendance tracking, and
scheduling, reducing workloads and improving efficiency. Platforms like Absorb LMS streamline
operations, freeing educators to focus on pedagogy [105].
Automated Administrative Tasks: AI automates grading, attendance tracking, and
scheduling, reducing workloads and improving efficiency. Platforms like Absorb LMS
streamline operations, freeing educators to focus on pedagogy [103,106,107].
Predictive Resource Management: Predictive analytics forecast enrollment trends and
optimize resource allocation, enabling institutions like the University of Melbourne to
achieve significant efficiency gains [108,109] This data-driven approach ensures alignment
with institutional goals and sustainability.
Strategic Innovation: AI supports curriculum design, faculty development, and
interdisciplinary collaboration by aligning strategic priorities with data insights [10,104].
By enabling timely interventions and operational resilience, AI empowers institutions to
innovate and thrive in an ever-changing educational landscape.
These global studies revealed that AI is transforming administrative operations in higher education
by automating routine tasks, enhancing resource management, and fostering strategic innovation.
14 of 63
Automation tools streamline grading, scheduling, and attendance tracking, reducing workloads and
allowing educators to prioritize pedagogy. Predictive analytics further optimize resource allocation
and enrollment management, as seen in institutions like the University of Melbourne, ensuring
efficiency and sustainability. Moreover, AI supports strategic goals such as curriculum design and
faculty development, enabling institutions to align their priorities with data-driven insights. By
fostering operational resilience and adaptability, AI empowers higher education institutions to
navigate an ever-changing landscape, achieving greater efficiency, innovation, and institutional
growth.
3.4.3 Inclusivity and Accessibility
Insights from global research demonstrate that AI plays a pivotal role in fostering inclusivity by
addressing barriers faced by marginalized groups and students with disabilities. AI-powered
assistive technologies provide meaningful support for learners with cognitive and physical
challenges, while multilingual AI tools break down language barriers, extending access to quality
education in diverse cultural and linguistic contexts [24,110].
By adopting scalable and innovative AI solutions, institutions are actively reducing educational
disparities, ensuring that all learners—regardless of their background or abilities—have equal
opportunities to succeed [97]. Empirical studies highlight that AI-powered adaptive learning
platforms and personalized learning analytics enhance accessibility, fostering greater inclusivity in
digital education [99]. Additionally, human-centered AI models in higher education are
transforming digital equity, reinforcing sustainable, accessible, and equitable learning environments
on a global scale
Key AI-Driven strategies for inclusive HE includes;
Assistive Technologies: Text-to-speech tools, adaptive content delivery systems, and speech
recognition software enable students with cognitive and physical impairments to engage in
meaningful learning experiences [111,112].
Multilingual Support: AI-powered translation tools bridge language barriers, extending
access to quality education in diverse linguistic and cultural contexts. These tools are
particularly impactful in resource-constrained regions, reducing disparities and promoting
equity [104,113].
3.4.4 Long-Term Benefits
AIED transcends traditional education boundaries, fostering global collaboration, sustainability,
and workforce readiness.
Global Collaboration: Platforms like Coursera and edX connect educators and students
worldwide, promoting cross-cultural exchanges and aligning with Sustainable Development
Goal 4 for equitable access to education [114,115].
Preparing Future-Ready Students: AI fosters skills such as computational thinking and
digital literacy, equipping students for success in an AI-driven economy [104,116].
Sustainability and Innovation: AI-driven educational systems address disparities in
resource-constrained regions, ensuring scalable, high-quality learning opportunities [5,10].
Worldwide studies showed that AIED transcends traditional educational boundaries and becoming
a catalyst for the future of education by fostering global collaboration, equipping students with
future-ready skills, and driving sustainability in education. Platforms like Coursera and edX enable
cross-cultural exchanges, aligning with global goals for equitable access to education. Furthermore,
AI empowers learners with essential skills such as computational thinking and digital literacy,
preparing them for success in an AI-driven economy. By addressing disparities and providing
scalable, high-quality learning opportunities in resource-constrained regions, AIED establishes
15 of 63
itself as a cornerstone for societal progress, global educational equity, and the sustainable
transformation of education systems.
3.4.5 Global Dynamics of AIHED Adoption
HE institutions worldwide are leveraging artificial intelligence to foster innovation, equity, and
inclusion. The pace and scope of AI adoption are shaped by regional priorities, economic resources,
and policy frameworks. As highlighted in the AI Index Reports [27,80,90], AI is redefining the
higher educational landscape on a global scale as illustrated in Table 2. AI adoption in HE varies
by region, reflecting unique priorities, as discussed below:
Asia: Countries like China and India leverage AI for scalability and personalized learning,
with platforms like Squirrel AI and Byju’s addressing urban-rural divides [80,117].
Europe: Initiatives like Finland’s Elements of AI democratize AI literacy, while Germany’s
AI Campus integrates vocational training with AI technologies [84].
North America: Institutions like Stanford lead in deploying predictive analytics and adaptive
platforms, emphasizing personalization and workforce readiness [27,80].
Africa: AI-powered tools bridge the digital divide, extending access to rural areas through
mobile learning [7].
Oceania and the Middle East: Universities integrate AI with cultural and regional
knowledge systems to promote inclusivity and address societal challenges [11,80].
Table 2: Global Dynamics of AIHED Adoption
Region
AIHED Application
Area of Focus
Literature
Support
Asia
Scalability and Personalization; Adaptive
technologies (Squirrel AI, Byju's, Digital
Textbook); Multilingual tools (Duolingo,
Google Assistant); Lifelong learning
platforms (Coursera, edX).
Addresses diverse educational
needs; Promotes cross-cultural
learning; Reskilling and upskilling
opportunities in AI fields.
[23,117–120]
Europe
Equity and Digital Competence; National
strategies for inclusivity and digital
competence; AI literacy programs (Elements
of AI); Intelligent tutoring systems (AI
Campus).
Ensures equitable learning
outcomes; Promotes inclusivity;
Advances vocational and applied
learning.
[7,10,11,84]
North
America
Personalization and Workforce Readiness;
Adaptive learning platforms (Coursera,
ALEKS); Predictive analytics for learning
outcomes; Indigenous knowledge integration.
Enhances individualized learning;
Prepares workforce for AI-driven
economy; Fosters inclusivity with
indigenous knowledge.
[27,70,80,121,122]
South
America
Overcoming Constraints; Adaptive platforms
for personalized courses; Predictive analytics
to identify at-risk students; International
collaborations.
Improves equity and accessibility;
Enables proactive student
interventions; Strengthens regional
collaborations.
[80,123,124]
Africa
Bridging the Digital Divide; Mobile learning
applications with offline capabilities;
Multilingual AI tools for inclusivity; Pan-
African AI literacy initiatives.
Expands rural education access;
Breaks language barriers; Promotes
educator inclusivity.
[24,38,111]
Oceania
Cultural Sensitivity and Advanced Analytics;
Integration of AI with indigenous knowledge
systems; Optimized resource allocation;
Lifelong learning platforms.
Optimizes learning through cultural
relevance; Promotes equity and
personalized education.
[11,39,80]
Middle
East
Smart Campuses and Problem-Solving;
Attendance monitoring, engagement analysis;
AI for retention improvement; Research on
water scarcity and renewable energy.
Improves student engagement and
retention; Addresses region-specific
challenges; Supports innovation
through research.
[7,17,99,121]
16 of 63
Above Table 2 illustrates how regional efforts collectively highlight AI's capacity to transform
traditional educational paradigms into dynamic, student-centric, and adaptive ecosystems. By
addressing unique socio-economic contexts, AI drives a global movement toward more equitable,
efficient, and future-ready higher education systems, establishing a foundation for long-term
societal progress and innovation.
3.5 Global Regulatory Frameworks for Ethical AIHED
Ethical AIED requires robust regulatory frameworks to ensure fairness, transparency,
accountability, and inclusivity. Below is an overview of key existing frameworks and guidelines
developed by international organizations, national bodies, and regional initiatives, supported with
proper references.
3.5.1 UNESCO and OECD’s AI Ethics Guidelines
The UNESCO Recommendation on the Ethics of Artificial Intelligence (2021) emphasizes fairness,
inclusivity, accountability, and respect for human rights in AI implementation. It advocates for
capacity building, data governance, and monitoring mechanisms, particularly within HE, to ensure
ethical AI adoption [24,31,125]. However, a lack of enforceability at the national level has led to
inconsistent adoption across countries, limiting the global impact of these principles.
Similarly, the OECD’s AI Principles (2019) promote human-centered values, transparency, and
robustness in AI systems. Adopted by 42 countries, these principles serve as a global benchmark,
influencing regional policies such as the EU’s AI framework [50]. Despite their significance, these
guidelines lack actionable mechanisms for sector-specific challenges, particularly those unique to
higher education, where AI governance must balance ethical compliance with academic autonomy
and institutional diversity.
3.5.2 NASSCOM and NITI Aayog’s AI Initiatives (India)
In India, NASSCOM and NITI Aayog have developed initiatives to regulate AI ethically:
NASSCOM’s Responsible AI Hub promotes fairness, accountability, and inclusivity in AI
systems, with a focus on higher education [127].
NITI Aayog’s National Strategy for AI (2018) emphasizes AI as a tool for social
empowerment, addressing sectors such as higher education, health, and agriculture. It
stresses transparency, data privacy, and bias mitigation while advocating for public-private
collaborations [128].
While these initiatives drive innovation, they often lack robust regulatory mechanisms to monitor
compliance, and their implementation faces challenges in rural and resource-constrained areas.
3.5.3 US Department of Commerce and NIST AI Risk Management Framework
In the United States, the NIST AI Risk Management Framework (2023) provides risk assessment
strategies aligned with ethical principles such as fairness, transparency, and accountability [63]. It
encourages participatory processes to ensure diverse stakeholder needs are addressed within higher
education, supporting ethical AI deployment [9,20]. Additionally, federal bodies such as the U.S.
Department of Higher Education (2021) have issued guidelines to safeguard student data and
privacy, reinforcing the importance of institutional compliance and governance [129].
However, the NIST framework lacks specificity in integrating diverse stakeholder inputs and
struggles to address the complexities of global data governance and rapidly evolving AI
technologies. Empirical studies highlight that AI governance models in HE must integrate adaptive
risk management strategies to mitigate bias, enhance algorithmic fairness, and align with national
and international AI ethics guidelines [130]. Addressing these limitations requires sector-specific
policies tailored to the unique ethical and operational challenges of AI deployment in HE.
17 of 63
3.5.4 European AI Ethics and Regulatory Frameworks
The European Commission’s Ethics Guidelines for Trustworthy AI (2019) and the proposed EU
Artificial Intelligence Act (2021) focus on regulating high-risk AI systems in higher education.
These frameworks emphasize human agency, technical robustness, privacy, transparency, diversity,
environmental sustainability, and accountability [131,132].
Despite their rigorous standards, the high compliance costs and tailored focus on well-resourced
EU member states pose challenges for smaller institutions and those in lower-income regions.
3.5.5 Other Global Ethical Standards for AI
Various global bodies, such as the Global Partnership on AI (GPAI) and IEEE’s Global Initiative
on Ethics of Autonomous and Intelligent Systems, have established principles to guide ethical AI
deployment.
GPAI promotes collaboration between governments, academia, and private sectors to ensure
inclusivity, transparency, and accountability in AI governance [133].
IEEE’s Ethical Guidelines prioritize accountability, transparency, and human rights
protection, offering recommendations for preserving dignity and agency in higher education
AI systems [51].
Both frameworks face challenges in enforceability and accessibility, with GPAI’s collaborative
approach encountering difficulties in aligning diverse stakeholder priorities, and IEEE’s technical
focus limiting its applicability for non-technical educators and policymakers.
All available global frameworks provide a strong foundation for ethical AIED by emphasizing
fairness, transparency, and accountability. However, challenges such as enforceability, inclusivity,
and regional disparities limit their practical impact as reflected in Table 3. Effective implementation
will require more actionable mechanisms, broader stakeholder engagement, and tailored solutions
to meet the diverse needs of global higher education systems.
Table 3: Global Ethical and Regulatory Frameworks – Focus, Strengths and Limitations
Global Ethical and
Regulatory Frameworks
Focus and Strengths
Limitations
UNESCO Recommendation
on the Ethics of AI (2021)
Fairness, inclusivity, accountability,
human rights; capacity building and
monitoring mechanisms.
No enforceability at the national level;
inconsistent adoption across countries.
OECD AI Principles (2019)
Human-centered values,
transparency, robustness; a global
benchmark for ethical AI adoption.
Lacks detailed mechanisms for sector-
specific challenges, such as those
unique to higher education.
NASSCOM Responsible AI
Hub (India)
Fairness, accountability, inclusivity
in AI systems with a focus on higher
education.
Limited regulatory mechanisms and
accountability; challenges in rural and
under-resourced settings.
NITI Aayog National Strategy
for AI (India, 2018)
Social empowerment through
transparency, data privacy, and bias
mitigation in key sectors like higher
education.
Focuses heavily on public-private
collaboration but overlooks diverse
stakeholder interests and lacks
monitoring compliance.
NIST AI Risk Management
Framework (USA)
Risk assessment strategies,
participatory processes, and
safeguarding student data in
education technologies.
Lacks specificity for integrating
diverse stakeholder inputs; limited
adaptability for global data governance.
European Commission Ethics
Guidelines for Trustworthy AI
(2019)
Seven requirements including human
agency, technical robustness,
transparency, and bias mitigation for
high-risk AI systems.
High compliance costs; less accessible
for smaller institutions and low-income
regions.
18 of 63
Global Partnership on AI
(GPAI)
Collaboration between governments,
academia, and private sectors for
ethical AI governance.
Enforceability challenges; delays due
to diverse stakeholder priorities.
IEEE Global Initiative on
Ethics of Autonomous and
Intelligent Systems
Accountability, transparency, and
human rights protection for
preserving dignity and agency in AI
systems.
Highly technical focus; limited
accessibility for non-technical
educators and policymakers.
3.6 Global Ethical Challenges and Barriers
Despite the availability of ethical and regulatory frameworks at the global level, the integration of
AI in higher education continues to present complex ethical challenges. These issues have been
extensively highlighted in numerous studies [44] and several reviews in global context [6,37,104]
[10,36]. These issues arise as institutions adopt advanced technologies, with challenges shaped by
regional disparities in technological progress, regulatory frameworks, and socio-economic factors.
[36] stress the importance of reflective and ethical considerations in AI adoption, advocating for
approaches that balance innovation with equity and inclusivity.
The integration of AI in HE brings significant ethical challenges influenced by regional disparities,
regulatory frameworks, and socio-economic contexts. As highlighted by [44] and global reviews
[6,10,36,37,104], these challenges underscore the need for reflective approaches that balance
technological innovation with equity and inclusivity. Institutions must navigate these complexities
to ensure AI adoption aligns with ethical and societal values.
The ethical governance of AI in HE institutions requires structured AI governance frameworks to
regulate its use special areas of HE such as in faculty hiring, research integrity, plagiarism detection,
and accreditation processes [8].
3.6.1 Basic Ethical Instincts - Privacy, Bias, and Fairness
AI systems in HE are subject to ethical imperatives of fairness, accountability, transparency, and
ethics (FATE), particularly regarding data privacy, algorithmic bias, and equitable implementation
[9,97]. Key challenges include privacy breaches, surveillance risks, and biased algorithms
influencing admissions, grading, and resource allocation. While frameworks like GDPR enforce
data protection, systemic inequities in AI-driven education persist due to biased training datasets
and algorithmic opacity [6,26,134,135]. Empirical studies highlight that bias in AI models
disproportionately affects marginalized student groups, reinforcing educational inequalities despite
regulatory safeguards [20,130].
Addressing these challenges requires enhanced algorithmic transparency, fairness audits, and
diverse training data to mitigate discriminatory outcomes in AI-powered learning environments [9].
To ensure accountability and inclusivity, robust governance frameworks must be established,
integrating human-centered evaluations [36]. Empirical studies further emphasize that
interdisciplinary collaboration, as advocated by UNESCO guidelines and [136], is essential for
fostering ethical AI deployment in education. Additionally, contextual and cultural sensitivities
should inform policy development to ensure AI technologies align with diverse societal needs [83].
However, a lack of technical expertise within educational institutions can exacerbate ethical risks,
including data insecurity and algorithmic biases [60,68,123]. Moreover, the increasing adoption of
surveillance technologies, such as facial recognition, raises ethical concerns about privacy and its
impact on pedagogical practices [137]. Addressing these challenges requires embedding ethical
standards in AI design and ensuring transparency in deployment. Professional development for
educators and continuous oversight are equally crucial for fostering equity and fairness.
19 of 63
The following Table 4 provides a summary of literature review on global scale exploring the ethical
challenges and barriers associated with AI-driven higher education (HE), along with the key
consequences identified in these studies.
Table 4: Literature Review: Global Ethical Challenges and Barriers
Ethical
Challenge
Focus of the Study
Impacted
Country/Region
Consequences
Identified
Literature Support
Bias in AI
Systems
AI models in grading and
predictive systems
perpetuate biases when
training data is
unrepresentative.
Global – Adaptive
Learning and
Predictive
Platforms
Disadvantages
marginalized groups and
perpetuates systemic
inequities in educational
outcomes.
[7,10,20,30,36,90,138]
Privacy and
Data
Concerns
AI platforms collect vast
student data for learning
analytics, raising risks of
breaches and misuse.
Global – Learning
Analytics
Platforms
Undermines trust in AI
systems, discouraging
their adoption in
educational contexts.
[25,26,90,139–141]
Scalability
and Equity
Issues
AI-enabled tools like
MOOCs provide scalable
education but are often
inaccessible in rural or
underfunded areas.
Developing
Countries – Rural
Institutions
Worsens educational
inequalities, leaving
disadvantaged
communities unable to
access innovative AI
resources.
[7,19,26,30,45,83]
AI for
Personalized
Learning
Adaptive platforms like
Knewton personalize
learning but risk isolating
students and reducing peer
interaction.
Global – AI Tools
like ALEKS,
Knewton
Enhances individual
learning efficiency but
overlooks collaborative
and social learning
dimensions.
[7,11,13,70,90,142,143]
Ethical
Challenges in
AI Use
Lack of transparency and
ethical frameworks in AI use
creates accountability and
trust issues.
Global – AI-driven
EdTech platforms
Raises ethical dilemmas
and undermines trust
among students and
educators.
[8,10,26,121,141,144]
Teacher
Readiness for
AI
Many educators lack skills
to effectively integrate AI
into their teaching and
feedback practices.
Global –
Professional
Development
Programs
Reduces the potential
impact of AI
technologies on
improving teaching and
learning processes.
[12,140,145–147]
Human-AI
Collaboration
AI tools like automated
writing assistants and
grading systems impact
teacher and student
autonomy.
Global – AI-
Assisted Academic
Writing Tools
Reduces human
engagement and may
foster over-reliance on
AI for academic tasks.
[17,20,26,27,70]
Impact on
Mental
Wellbeing
Overuse of AI in work and
learning environments leads
to isolation and reduced
social connections.
Global – AI in
Higher Education
and Workplaces
Increases loneliness and
stress, negatively
affecting mental health.
[27,39,122,140,148]
Student
Engagement
Challenges
AI fails to address the
emotional and motivational
needs of learners in virtual
learning environments.
Global – Platforms
like Coursera,
Kahoot
Reduces retention and
long-term learning
outcomes, limiting
holistic educational
experiences.
[10,17,26,139,142]
Language
and Cultural
Barriers
AI systems fail to adapt to
diverse linguistic and
cultural contexts,
marginalizing non-dominant
groups.
India, China – AI
in Language
Learning
Limits inclusivity and
prevents equal access to
AI-driven learning tools
for diverse learners.
[13,24,147,149]
20 of 63
Bias in
Predictive
Analytics
Predictive systems flagging
“at-risk” students
disproportionately target
specific demographic
groups.
United States –
Student Retention
Platforms
Reinforces systemic
inequalities, limiting
equitable access to
educational
opportunities.
[25,32,38,103,138]
Ethical
Concerns in
Mentoring
AI-supported mentoring
systems, such as career
advising, risk providing
biased or harmful
recommendations.
Global – AI
Mentoring
Platforms
Creates ethical dilemmas
and reinforces inequities
in personalized advice
for students.
[8,10,28,109,150]
3.6.2 Sustaining Humanized Ethics
AI has transformed HE processes, such as grading and administrative workflows, but it also
introduces challenges, including job displacement and diminished human engagement. Rather than
eliminating jobs, AI often redefines roles. [151] underscores the importance of reskilling to address
this transition, while [152] advocate for human-centric AI designs to preserve mentorship and
creativity. Ethical frameworks that incorporate societal and cultural nuances are essential, as (Bond
et al., 2024) point out, to ensure inclusive workforce adaptation. Proactive measures, such as
stakeholder engagement and transparency, foster equitable transitions [6,36].
Another critical concern is over-reliance on AI, which risks diminishing essential human skills like
critical thinking and problem-solving. [11] caution against passive intellectual behavior encouraged
by AI dominance, and [84] call for frameworks that balance efficiency with active human
involvement. [37] highlight the importance of preserving the emotional and cultural dimensions of
learning, ensuring AI does not dehumanize education. Embedding ethical norms and meaningful
human oversight is essential to maintaining alignment between AI systems and educational values.
For future generations, the benefits of enhanced accessibility and efficiency must be weighed
against potential drawbacks such as “intellectual atrophy"; intellectual stagnation and reduced
intergenerational knowledge transfer [148,153]. Collaborative frameworks and ethical design, as
emphasized by [10] and [142], are key to ensuring AI systems foster creativity and resilience.
Equally important is the role of transparency and trust in preserving motivation and engagement for
both students and educators [154].
Finally, while AI enhances operational efficiency, excessive automation can erode human
connections critical to holistic education. Over-automation risks isolating students, undermining
communication and collaboration skills [11]. Embedding emotional resilience and inclusivity in AI-
driven systems ensures these tools enhance, rather than replace, the relational aspects of education
[33] (Pedro et al., 2019).
3.7 Research Gaps and Missing Links:
AIHE has made significant advancements, with frameworks striving to enhance learning
experiences, boost operational efficiency, and meet diverse educational needs. However, these
developments frequently fail to address critical dimensions essential for holistic and effective AI
integration. Drawing on global literature, including journal articles, case studies, and reports, this
section identifies the key gaps hindering the successful adoption and implementation of AIED
systems. It underscores the necessity for ethical grounding, practical operability, and alignment with
human-centric principles to create robust, adaptive, and equitable frameworks for AI in education.
Inclusive and Collaborative AI: Bridging Stakeholder Gaps in Participatory Design
AIHED often overlooks ethical challenges such as algorithmic bias, data privacy, and
transparency, leading to real-world failures like Amazon’s AI hiring tool perpetuating
gender bias [6]. Current frameworks tend to focus narrowly on fairness and privacy while
neglecting broader socio-political concerns, such as democratic oversight [135].
21 of 63
Additionally, limited stakeholder involvement—including educators, students, and
policymakers—results in AI systems misaligned with diverse needs due to a lack of
participatory co-design and feedback mechanisms [37,45,155]. Moreover, AIHED
frameworks frequently treat AI as a standalone solution, disregarding the potential of
human–AI collaboration. Systems like Georgia Tech’s "Jill Watson" AI assistant
demonstrate efficiency in handling routine tasks but lack human empathy and contextual
adaptability [153]. Research is needed to develop collaborative AI frameworks that integrate
AI efficiency with human intuition, creativity, and relational engagement [85]. Addressing
these gaps will enable inclusive, participatory, and ethically robust AI systems that better
serve diverse educational communities.
Bridging Theory and Practice: Operational Gaps in AI Adoption
Many frameworks fail to translate theoretical advancements into actionable solutions. For
instance, the UK government’s AI grading system during COVID-19 faced public backlash
due to inadequate phased testing [7]. Policies often overlook practical guidance for
implementation, as noted by [76], leaving institutions without effective workflows to bridge
theoretical benefits and practical applications.
Dynamic Feedback: The Foundation for Continuous Improvement
A persistent limitation in AIED frameworks is the reliance on static feedback mechanisms,
which prevent adaptation to evolving needs. Adaptive learning systems often fail to
incorporate real-time feedback from students and educators, resulting in stagnant outcomes
[7]. Dynamic feedback loops, as emphasized by [140], are essential for continuous
refinement and responsiveness, enabling systems to align with institutional goals and learner
requirements.
Adapting to Diversity: Aligning AI with Institutional Goals
Current AI frameworks in higher education (HE) often adopt a one-size-fits-all approach,
neglecting institutional and cultural diversity [38,96]. Research is needed on adaptive AI
models that accommodate varying infrastructure and resources, particularly for institutions
with limited AI capacity [66,111]. The use of structured assessments, such as SWOC
(Strengths, Weaknesses, Opportunities, Challenges), remains underexplored in guiding AI
integration and sustainability across diverse HE settings [10,156]. Additionally, static AI
systems struggle in culturally diverse environments, highlighting the need for context-
sensitive frameworks that align with regional needs [7,12]. Finally, ensuring AI alignment
with institutional missions, inclusivity, and scalability remains a critical challenge, requiring
further research on responsive and equitable AI implementation [80].
Automation to Augmentation: Preserving Human Values
AIHED frameworks often overemphasize scalability and efficiency, side-lining educational
values like human creativity, mentorship, and critical thinking [6,36]. This techno-centric
focus risks dehumanizing education, as noted by [8]. Balancing technological advancements
with human-centric principles is crucial to preserving equity, intellectual growth, and
holistic development.
Missing Links in AI: Resilience and Sustainability for the Future
Most frameworks prioritize short-term gains over long-term adaptability, ignoring evolving
societal needs and regulatory changes. Over-reliance on AI risks fostering "intellectual
atrophy," where human creativity and critical thinking diminish educators [153]. [148]
emphasize the need for future-proof strategies, including scenario planning and continuous
research, to ensure AIED systems remain adaptable, resilient, and aligned with broader
educational values.
Phased Human Intelligence: Filling Gaps in the AI Lifecycle
A critical gap in AIED frameworks is the lack of integration between the AI lifecycle
(adoption, design, deployment, evaluation and exploration) and phased human intelligence.
Human oversight is rarely embedded across these phases, leading to misaligned algorithms
and unchecked biases [6,135]. Effective frameworks must involve interdisciplinary teams
of educators, AI developers, and policymakers to guide each phase, ensuring alignment with
22 of 63
ethical and pedagogical goals [37,140]. Regular audits and dynamic feedback loops can
detect inefficiencies and biases, fostering systems that are both transparent and contextually
relevant [12,85].
Promotion and Benchmarking AIED: The Need for External Intelligence
Most frameworks fail to emphasize multi-stakeholder collaborations, limiting adaptability
and scalability. External partnerships with industry, governments, and academia can provide
funding, emerging technologies, and diverse perspectives essential for robust AI
implementation [6,45]. Furthermore, there is limited documentation of real-world successes
and failures, such as the chatbot "Pounce" at Georgia State University [153]. Transparent
reporting can foster innovation, build trust, and prevent repeated mistakes.
Need for University / Institutional AI Ethics Committees
AIHED lacks standardized ethical governance models, raising concerns about algorithmic
bias, academic autonomy, and decision transparency. While institutions like the University
of Amsterdam have established AI Ethics Committees [51], many universities lack
structured oversight for AI-driven faculty evaluations, admissions, and tenure reviews [26].
The literature reveals critical research gaps in terms of the absence of AI Ethics Review
Boards / Committees in universities and HE institutions.
Unaddressed Challenges in AIHED
Existing research on AI in Education (AIED) focuses on adaptive learning and student
support, while AI in Higher Education (AIHED) requires structured governance, policy
frameworks, and institutional oversight [6,26]. AI applications in faculty governance,
research integrity, and accreditation remain underexplored [37]. Current studies lack
insights into participatory AI governance, academic autonomy, and standardized regulatory
frameworks to address bias, transparency, and ethical compliance [20]. This study bridges
these gaps by proposing a structured AI governance model that ensures AI adoption aligns
with HE's core values and institutional integrity (UNESCO, 2021).
Vision without Action: A Call for Practical and Constructive Solutions
Global frameworks, such as those by [31] and [50], often emphasize fairness and
accountability but lack localized adaptability and participatory design. Initiatives like
NASSCOM and the EU AI Act prioritize compliance but fail to integrate relational
engagement and co-design critical for higher education [37]. Practical, actionable strategies
are needed to bridge this gap and facilitate safe, ethical AI integration into diverse HE
contexts.
Addressing these research gaps is critical to bridging the divide between theoretical aspirations and
practical applications of AIHED. By embedding human-centric ethics, dynamic feedback systems,
participatory and collaborative frameworks, and long-term strategies, AIED frameworks can align
technological advancements with human values, fostering adaptive, equitable, and sustainable
systems for HE.
4. Global Reality and Experiences: Ethical or Exploitation?
This section critically examines the global realities and experiences of AIHED, assessing whether
ethical principles such as fairness, accountability, inclusivity, and human-centric values are
genuinely upheld in practice. While AIHED has made significant strides, achieving remarkable
progress, its implementation has also revealed a duality of success and setbacks. This analysis
underscores the urgent need for a Human-Driven AIHED Framework to ensure the integration of
AIHED is meaningful, equitable, and aligned with ethical standards.
4.1 Duality in AIHED: Successes and Setbacks
On one hand, AI has empowered higher education institutions to personalize learning, enhance
operational efficiency, and democratize access to education. On the other hand, it has introduced
significant ethical challenges, including algorithmic bias, privacy violations, and the erosion of
23 of 63
interpersonal connections. This duality raises a critical question: Is AIHED merely “doing ethical
things right,” or is it truly “doing things ethically”? AIHED exemplifies a duality of transformative
successes and critical setbacks across global contexts as exemplified below.
Personalized Learning and Accessibility:
Platforms like Squirrel AI have tailored education to millions in China, improving outcomes
for underserved populations [29,31]. However, invasive tracking of student behaviors
without consent has raised concerns about privacy and trust [32]. Similarly, Coursera has
democratized access to quality education for over 100 million learners, particularly in
developing regions like India and sub-Saharan Africa [19,29]. Yet, infrastructural inequities
in under-resourced areas exacerbate the digital divide [30,157].
Efficiency in Grading and Assessments:
Tools like Gradescope and Turnitin have streamlined grading and improved writing
outcomes at institutions like UC Berkeley and Stanford [25] (Bond et al., 2024). However,
the UK’s algorithmic grading system during COVID-19 exposed systemic biases,
disproportionately disadvantaging students from underprivileged backgrounds [9,26].
Predictive Analytics and Student Success:
Predictive analytics at Georgia State University increased graduation rates by 22%,
supporting at-risk students through data-driven interventions [27]. Conversely, tools like
Ellucian Analytics flagged minority students disproportionately as “at-risk,” reinforcing
existing inequities [25].
AI and Relational Learning:
While tools like Cognitive Tutor have improved STEM learning outcomes by 40% [12],
their overuse has diminished meaningful teacher-student interactions, leading to isolation
and disengagement [148,153]. Automated systems, like those at Deakin University, have
reduced opportunities for faculty engagement, weakening relational aspects of education
[13].
Privacy and Surveillance Concerns:
Proctoring tools such as Proctorio have raised significant concerns due to invasive
surveillance practices that have led to racial bias and trust erosion among marginalized
communities [10,141]
These examples highlight the complex landscape of AIHED, illustrating its capacity to transform
HE while revealing its inherent risks. They also emphasize that while AIHED often achieves
efficiency and scalability, it frequently undermines fundamental educational values, such as equity,
trust, and interpersonal connection.
4.2 A Call for Action: Emergence of Human-Driven AIHED
The future of HE lies not in the raw power of AI but in the intentionality and ethical rigor with
which it is employed. A Human-Driven AIHED Framework is essential to align AI systems with
human values, ensuring fairness, inclusivity, and empathy remain at the heart of education.
Following are the key considerations for a Human-Driven AIHED Approach. A Human-Driven
AIHED Framework is essential to ensure AI adoption prioritizes human values, ethical oversight,
and equity in education.
To overcome Ethical and Governance Challenges in AIHED
AI-driven decision-making systems can reinforce biases, compromise student privacy, and
erode institutional accountability if not designed with ethical safeguards. AI-based grading,
predictive analytics, and faculty assessments must be monitored to prevent discriminatory
outcomes and ensure transparency, explainability, and fairness [20]. The absence of
structured AI governance frameworks in HE leads to disparities in adoption, ethical
inconsistencies, and regulatory gaps, making a human-driven approach critical [26].
Ensuring Equity and Inclusivity in AIHED
AI must support, not replace, human expertise. AI-driven learning platforms should align
with institutional values, cultural diversity, and equitable access to resources [45]. Without
24 of 63
human oversight, AI risks exacerbating digital divides, disproportionately affecting
underrepresented student populations [7]. Ensuring co-governance with educators, students,
and policymakers fosters trust, inclusivity, and balanced AI adoption [19].
The Role of Human Oversight in AI Adoption
Embedding human oversight across the AI lifecycle—from design to deployment and
evaluation—ensures ethical compliance, institutional accountability, and adaptability [29].
Universities must implement co-design strategies, integrating faculty, administrators,
students, and industry experts into AI policymaking to reflect academic and societal needs
[157].
Human-Centric AI for Sustainable HE Transformation
A Human-Centric framework provides a structured roadmap for responsible AI adoption,
balancing technological efficiency with empathy, fairness, and long-term sustainability
[148,158]. Ethical AI integration must incorporate dynamic feedback loops, algorithmic
fairness audits, and participatory governance models to uphold academic integrity and
educational equity.
5. Conceptualizing Human-Driven AIHED: Enabling ‘Ethos’ in AIHED
As [36] emphasize, the adoption of digital technologies in HE should not be seen as an unchecked
celebration of progress but rather as an opportunity for critical reflection, ensuring inclusivity,
fairness, and sustainability. The HD-AIHED framework provides a structured approach to
addressing research gaps and real-time challenges in AI adoption for higher education. By
integrating ethical principles and structural components, it ensures AI deployment aligns with
institutional values, academic integrity, and equitable governance [159].
Universities and HE institutions face key challenges, including policy gaps, algorithmic biases,
governance inconsistencies, and scalability constraints, requiring human oversight, participatory
decision-making, and adaptive AI strategies [9,87]. They also face complex challenges regarding
special services as discussed in literature review section. The HD-AIHED framework addresses
these gaps by embedding ethical AI governance, fairness reviews, transparency mechanisms, and
continuous feedback loops to optimize AI integration in diverse educational settings [20].
5.1 Structural Components of Proposed Framework and Potential Solutions: Addressing
Research Gaps and Global Real-time Challenges:
The figure 2 presents key structural components and potential solutions for addressing global AI
challenges in higher education through the Human-Driven framework. It systematically maps real-
time challenges, such as algorithmic bias, data privacy concerns, faculty governance gaps, and
digital divide, to corresponding research gaps and structural solutions [8,26]. The structural
components of the proposed AIHED framework—such as Institutional AI Ethical Review Boards,
AI fairness audits, SWOC analysis, participatory governance, and phased AI adoption strategies—
are integrated as possible solutions to enhance AI transparency, inclusivity, and adaptability [20,31].
By incorporating internal (faculty, students, IT teams, auditors) and external stakeholders (industry,
policymakers, global HE institutions, regulators), the framework ensures a balanced AI adoption
process that is ethically responsible, scalable, and human-centered [6,50]. The figure serves as a
guiding tool for developing final framework for HE institutions to align AI advancements with
sustainable, participatory, and equitable educational policies [7].
25 of 63
Figure 2: Structural Components of Framework and Potential Solutions for Addressing Research Gaps and Global
Real-time Challenges
26 of 63
5.2 Integrated System: Ethical Essence and Essential Structural Components
The proposed framework operates on distinctive system principles, encompassing inputs,
processing, and outputs facilitated through seamless integration of AI and unified human
intelligence, as depicted in the accompanying Figure 2. The effectiveness of AI in higher education
depends on an integrated system where AI, human intelligence, and external intelligence function
as a unified structure, fostering continuous interaction between automated processes and human
oversight. Unlike isolated AI applications that operate independently, this framework embeds AI
within a structured ecosystem that emphasizes ethical integrity, transparency, and accountability
[8,26]. By interlinking AI operations with human decision-making and continuous feedback
mechanisms, the system mitigates automation risks such as bias, ethical lapses, and the erosion of
contextual adaptability [8,36].
This integration ensures that AI does not function as an autonomous entity but rather as an adaptive,
human-aligned tool that evolves alongside institutional needs [85]. A structured flow from input to
processing and output is maintained through unified human intelligence, ensuring that AI-driven
insights remain relevant, ethical, and operationally effective [29,31]. The incorporation of external
intelligence further enhances this model by introducing interdisciplinary perspectives, regulatory
compliance, and benchmarking against global best practices, making AI implementations more
adaptable and responsive to dynamic institutional and societal expectations [19,30].
The necessity of an integrated system lies in its ability to harmonize AI’s computational efficiency
with human judgment, ensuring that decision-making remains aligned with the core values of
education [33,70]. Integrated system approach (refer Figure 3) fosters trust by maintaining
transparency in AI-driven decision-making and reinforcing accountability through structured
oversight [26,80]. By embedding AI within a holistic and adaptive system, the framework ensures
that technology functions as an enabler of education rather than a disruptor, preserving equity,
inclusivity, and institutional integrity while optimizing learning and administrative processes
[39,157].
Figure 3: HD-AIHED as an Integrated System
5.2.1 AI System Input
At the foundation of the system is the system input, which consists of structured institutional data,
including student records, administrative policies, resource allocation, and regulatory frameworks.
This phase ensures that human oversight is embedded in data selection, curation, and pre-processing
27 of 63
to uphold ethical compliance, mitigate biases, and ensure fairness in AI operations [6,26]. By
integrating human intelligence at the input stage, study’s framework safeguard AI from ingesting
flawed, outdated, or contextually irrelevant data, thereby enhancing the accuracy and integrity of
AI-driven decisions [11,36]
Institutions leveraging AI for decision-making must ensure that the data feeding AI systems is
diverse, current, and free from historical biases that could reinforce systemic inequities [30,135].
Without human intervention in the input stage, AI models risk perpetuating inaccurate patterns that
can disproportionately affect marginalized groups, compromise institutional inclusivity, and hinder
equitable student outcomes [8]. The study’s framework emphasizes the importance of incorporating
human intelligence during the data input phase to mitigate potential risks. By involving human
oversight, it ensures contextual validation of data, aligns AI operations with institutional policies,
and adapts processes to meet ethical and regulatory standards effectively. This integration provides
a critical layer of accountability, fostering the responsible and compliant use of AI systems. [11].
Additionally, AI systems in HE rely on data from multiple institutional repositories, such as learning
management systems, assessment records, and faculty evaluations. However, disparities in data
accessibility, inconsistencies in reporting structures, and variations in institutional policies can
create challenges in ensuring uniformity across AI-driven systems [70]. Human oversight in the
input stage helps reconcile these discrepancies, fostering a structured and standardized approach to
AI-driven decision-making while maintaining the integrity of academic governance [33,80]. By
embedding human expertise at this foundational level, institutions not only enhance AI accuracy
and adaptability but also reinforce their commitment to ethical, transparent, and inclusive AI
adoption in HE [19,31].
5.2.2 AI Processing
The AI Processing phase represents the analytical core of the integrated system, where raw input
data is transformed into actionable insights using machine learning algorithms, predictive analytics,
and decision-making models. This stage plays a pivotal role in enhancing institutional and
pedagogical efficiency by optimizing operations across the student lifecycle, including admissions,
personalized learning pathways, early identification of at-risk students, and streamlining
administrative processes [11,160]. Key AI applications, such as predictive modeling, automated
grading, and resource allocation, facilitate data-driven decisions, improving both educational
outcomes and operational effectiveness [6,121].
Despite its transformative potential, AI processing alone cannot guarantee ethical compliance,
fairness, or contextual sensitivity. The proposed framework addresses these limitations by
embedding Unified Human Intelligence as an integral regulatory mechanism, ensuring that AI-
driven outputs align with institutional goals, ethical principles, and pedagogical objectives [25,26].
Human oversight mitigates risks such as automation bias, algorithmic opacity, and the perpetuation
of inequities by ensuring that AI systems adhere to fairness, transparency, and accountability
standards [8].
Dynamic feedback loops are integrated to enable real-time refinements of AI models, fostering
continuous learning and adaptability to meet changing pedagogical demands and regulatory
requirements [12,33]. These iterative mechanisms prevent stagnation in AI functionality and ensure
the recalibration of algorithms to reflect contextual shifts, thereby addressing the challenges of
algorithmic misinterpretation and static decision-making processes [140].
By embedding ethical oversight and dynamic feedback within the AI Processing phase, the
framework mitigates risks associated with biased decision-making and algorithmic opacity,
ensuring that AI operates as a complementary tool rather than an autonomous determinant of
educational outcomes [6,85]. This structured interaction between AI-driven analytics and human
intelligence reinforces trust, accountability, and adaptability, positioning AI processing as a
cornerstone for the ethical and effective integration
28 of 63
5.2.3 AI System Output
The "System Output" component of the proposed framework represents the culmination of
integrated AI processes, producing actionable insights and tools that enhance decision-making,
operational efficiency, and personalized learning experiences in HE. This segment of the ecosystem
delivers outputs such as personalized recommendations for students, administrative insights for
resource optimization, and accessibility tools that cater to diverse learner needs. These outputs are
presented in user-friendly formats, including dashboards, reports, alerts, transcriptions, and
interactive interfaces such as chatbots and virtual assistants, enabling seamless interpretation and
application of data [6,11].
A key strength of the framework lies in embedding human intelligence throughout the output
process. Human oversight is integral to ensuring the accuracy, ethical compliance, and contextual
relevance of AI-driven outputs. This involves validating data accuracy, refining algorithms based
on feedback, and interpreting insights to align outputs with institutional values and objectives. For
instance, personalized learning recommendations are tailored to meet individual student needs while
avoiding algorithmic bias, and administrative outputs support equitable resource allocation across
departments [26,160].
Furthermore, the framework ensures that outputs lead to actionable outcomes. These include
intervention strategies for at-risk students, adjustments to institutional curricula, and data-informed
policy updates that drive improvements in teaching, learning, and administration. By fostering
inclusivity and addressing the diverse needs of stakeholders, the framework transforms raw AI-
generated outputs into meaningful, institutionally aligned actions [33].
The integration of dynamic feedback loops within the output process ensures continuous refinement
and adaptability of AI systems. Feedback from educators, students, and administrators helps fine-
tune the outputs, ensuring they remain responsive to changing institutional priorities and societal
demands. For example, real-time updates to predictive analytics can guide faculty in providing
timely interventions, while adaptive dashboards support decision-makers in tracking progress and
identifying areas for improvement [11,140].
By emphasizing transparency and inclusivity in the generation and application of outputs, the
framework mitigates risks such as bias, misinterpretation, and inefficiency. It reinforces institutional
trust and accountability by aligning AI-driven outcomes with ethical standards and stakeholder
expectations. This ensures that AI outputs not only optimize operational processes but also advance
the broader mission of equity, accessibility, and excellence in higher education [6,45].
The System Output component, therefore, represents a pivotal stage in the framework, translating
AI's computational capabilities into actionable, human-centered insights that drive meaningful and
sustainable progress in HE.
5.2.4 SWOC Analysis
In line with [100], the proposed framework adopts a predictive AI preparedness model, reinforced
by [130]’s roadmap for institutional competitiveness. Through SWOC analysis, the framework
ensures AI adoption aligns with strategic goals while maintaining scalability and long-term
adaptability in the AI exploration phase [38,40].
AI Adoption Phase: The SWOC analysis framework integrates AI adoption with an institution’s
strategic goals, resources, and socio-cultural context [38,40]. During the AI adoption phase,
institutional strengths such as faculty expertise, existing infrastructure, and organizational readiness
facilitate smooth AI integration [104]. Weaknesses, including gaps in technological infrastructure,
limited funding, and lack of staff training, are mitigated through capacity-building initiatives and
interdisciplinary collaboration [6,85]. Opportunities arise from expanding global reach, improving
student outcomes, and innovating pedagogy through partnerships with NGOs, funding agencies,
29 of 63
and industry experts [26,31]. Challenges such as resistance to change, governance complexities,
and regulatory compliance are addressed through proactive feedback loops and inclusive decision-
making [156].
AI Exploration Phase: At the AI exploration stage, institutions benefit from strengths such as
accumulated AI expertise, expanding research collaborations, and enhanced digital infrastructure.
Weaknesses like regulatory uncertainties and integration complexities are tackled through policy
framing, iterative governance, and benchmarking [31,161]. Opportunities include academic
expansion, international collaboration, and innovation in AI-driven pedagogy. Challenges focus on
ensuring scalability and long-term adaptability while aligning with sustainable development goals.
By embedding SWOC analysis at both phases, institutions can strategically plan AI implementation
while fostering resilience, scalability, inclusivity, and equity [26,38].
5.2.5 Human Intelligence
AI governance in HE is designed to be inclusive and participatory, integrating faculty, students,
administrators, and external stakeholders into AI decision-making processes. A participatory and
collaborative approach ensures AI governance remains adaptive to regional contexts, facilitating
cost-effective implementation across diverse educational landscapes [8,30,85].
Internal Stakeholders:
Internal stakeholders play a direct role in AI implementation, ensuring AI aligns with educational
objectives and ethical considerations [66].
Management, Educators, Researchers, IT Technicians and Administrators: University / HE
institutional top management—including presidents, vice-chancellors, provosts, and governing
board members—plays a strategic role in AI integration, ensuring alignment with institutional
vision, regulatory frameworks, and global educational trends. They are responsible for driving the
long-term AI strategy, ensuring AI integration aligns with institutional goals and global best
practices [162]. They oversee institutional AI governance structures, ensuring compliance with
national and international regulatory frameworks such as UNESCO AI Ethics Guidelines, GDPR,
and OECD AI Principles [31,103]. Senior leadership allocates resources for AI infrastructure
development, ensuring scalability, sustainability, and ethical AI adoption [38]. They fosters cross-
institutional collaborations with universities, AI research institutes, and industry partners, enabling
access to cutting-edge AI technologies and research funding [11].
Educators (faculty, deans, and academic experts) serve as primary users of AI-driven tools for
teaching and assessment, ensuring AI supports rather than dictates teaching methodologies [85].
Faculty also integrate AI into curriculum design while upholding academic integrity and ethical
compliance [163]. Administrators oversee AI policy formulation and institutional strategies,
ensuring AI aligns with institutional goals, regulatory compliance, and financial considerations [38].
They facilitate AI-driven decision-making and feedback for academic and administrative
improvements, including student retention, operational efficiency, and institutional governance
[130].
The research team advances AI innovation by conducting empirical studies, bias assessments, and
ethical evaluations [20]. They collaborate with faculty, students, and AI governance boards to test
AI models, refine policies, and ensure transparent, ethical AI deployment [8]. Research teams also
secure AI-related grants and partnerships, keeping institutions at the forefront of ethical and
responsible AI adoption [162].
The IT team is responsible for AI system deployment, cyber security, data governance, and
infrastructure optimization to align AI tools with institutional policies [26]. They ensure secure AI
integration, prevent data breaches, and maintain compliance with GDPR and UNESCO AI Ethics
Guidelines. IT specialists facilitate AI-driven automation, interoperability, and student analytics,
ensuring seamless AI adoption in learning management and administrative systems [6].
30 of 63
Students: Students, as the primary recipients of AI-driven services, provide direct experiential
feedback, shaping AI’s effectiveness and highlighting potential biases [164,165]. Their inclusion as
a partner in AI ethical review boards and as AI auditors is critical to ensuring equity and
transparency in AI governance. Actively involving students in AI decision-making and feedback
ensures AI-driven solutions remain user-centric, fair, and aligned with diverse educational needs.
Institutional AI Ethical Review Boards And Auditors: Institutional AI Ethical Review Boards
govern responsible AI use, mitigating automation bias, data privacy concerns, and academic equity
issues [23,166]. Research teams refine AI applications for HE, ensuring transparency, adaptability,
and fairness [11,167]. Institutional auditors enhance AI oversight by monitoring effectiveness,
identifying biases, and reinforcing ethical compliance in AI-driven decision-making [168,169]. The
Institutional AI Ethical Review Board comprises advisory panels, compliance bodies, faculty
experts, legal professionals, and student representatives. It approves AI adoption strategies, ensures
ethical compliance, and aligns AI policies with global regulatory frameworks, including UNESCO
AI Ethics, GDPR, and OECD AI Principles [31,103]. The board ensures institutions uphold global
ethical standards while safeguarding student data privacy [8,26].
Institutional AI Ethical Review Boards ensure responsible AI use, addressing automation bias, data
privacy concerns, and academic equity [23,166]. Research teams refine AI applications for
transparency, adaptability, and fairness [11,167]. Institutional auditors strengthen AI oversight by
evaluating effectiveness, identifying biases, and ensuring ethical compliance [168,169].
Comprising advisory panels, compliance bodies, faculty experts, legal professionals, and student
representatives, the Institutional AI Ethical Review Board approves AI adoption strategies, enforces
ethical compliance, and aligns AI policies with global regulatory frameworks such as UNESCO AI
Ethics, GDPR, and OECD AI Principles [31,103]. It also safeguards student data privacy while
ensuring institutions uphold global ethical standards [8,26].
Institutional auditors monitor AI integrity, reinforcing transparency, fairness, and regulatory
compliance [168]. Regular AI bias audits assess fairness, supported by clear policies, oversight
mechanisms, and external reviews. Ethical AI auditing and bias mitigation strategies ensure AI-
driven assessments remain accountable and impartial, with compliance officers conducting ongoing
evaluations. To mitigate data misuse and systemic inequities, the framework integrates validation
mechanisms and dynamic oversight, ensuring AI models continuously evolve to meet emerging
challenges [6,121].
External Stakeholders:
The External Intelligence component is crucial for aligning AI integration with global trends,
technological advancements, and societal expectations. By engaging industry experts, academic
professionals, global institutions, ethical and regulatory bodies, government agencies, and society,
the framework strengthens institutional capacity, enhances AI governance, and fosters trust among
stakeholders [130,170]. The framework facilitates industry and academic collaborations to provide
technological expertise, allowing AI tools to be tailored to institutional needs while enhancing
operational efficiency and accessibility [171,172].
Ethical AI Regulatory Bodies / Boards: International organizations and ethical regulatory bodies
establish AI governance benchmarks, ensuring responsible AI deployment in higher education.
Organizations such as UNESCO, OECD, and the European Union set AI adoption principles
emphasizing transparency, fairness, and accountability [130,173]. These regulatory bodies provide
ethical guidelines that align AI governance policies with global human rights frameworks, ensuring
compliance with data privacy laws such as GDPR and AI ethics principles [26,31]. By defining
ethical AI standards, these institutions support universities in implementing AI systems that are fair,
unbiased, and accountable.
Alumni: Alumni play a key role in AI integration and governance by leveraging their industry
expertise, professional networks, and funding potential. Their contributions foster AI-driven
31 of 63
mentorship programs, research funding, and industry partnerships. By serving as advisory members
on institutional AI governance boards, they ensure AI applications align with real-world industry
needs. Their engagement strengthens global institutional networks, enhancing higher education’s
adaptability to AI advancements [170].
Government Agencies: Government agencies and funding organizations help shape AI policies and
provide financial assistance for AI integration. Investments in AI infrastructure and technology
empower institutions, particularly in under-resourced regions, to implement AI effectively and
ensure equitable access to AI-driven higher education [174].
Global Academic Institutions: International organizations establish AI governance benchmarks,
ensuring ethical AI deployment in higher education. UNESCO, OECD, and similar bodies set AI
adoption principles that emphasize transparency, fairness, and accountability [130,173]. Their
policies help align AI governance frameworks with global ethical and regulatory standards,
ensuring responsible AI integration [26].
Industry Partnerships: Public-private partnerships advance AI solutions that address digital
inequalities and drive AI-powered innovation. Collaboration with industry leaders fosters
technological advancements, ensuring AI solutions remain relevant, adaptable, and impactful within
diverse educational settings [11,170].
Societal Engagement: Engaging society through transparent communication and public discourse
fosters institutional credibility and trust in AI governance. Knowledge-sharing initiatives and
benchmarking efforts further refine AI strategies, strengthening global collaboration and continuous
improvements in AI adoption [6,121]. Encouraging interdisciplinary cooperation ensures AI
policies remain inclusive, equitable, and aligned with societal expectations [26,30].
The synergy of human intelligence (internal and external stakeholders) ensures AI in HE remains
ethical, transparent, and aligned with institutional values. By combining internal and external
intelligence, AI governance becomes collaborative, adaptive, and inclusive. Unified human
intelligence bridges technological advancements with ethical governance, ensuring AI serves as a
tool for enhancement rather than automation-driven disruption. Sustainable AI governance relies
on continuous stakeholder engagement, regulatory alignment, and participatory oversight,
reinforcing trust, fairness, and accountability in AI adoption
Collaborative Human Intelligence: Synergy Between Internal and External Stakeholders:
The integration of internal and external intelligence ensures a country’s technological readiness and
policy maturity. This framework enables institutions in developed countries to emphasize advanced
AI research, regulatory compliance, and industry collaboration, while institutions in
underdeveloped regions focus on scalable, cost-effective AI solutions that align with local
infrastructure and resource availability [26,45]. The interplay between internal and external
stakeholders ensures AI governance remains both institutionally responsive and globally aligned.
Internal stakeholders, including faculty, administrators, and students, oversee AI’s implementation
and impact within educational institutions. Meanwhile, external stakeholders—such as
policymakers, industry leaders, and regulatory bodies—provide oversight, technological
advancements, and ethical benchmarks. This collaborative structure prevents ethical lapses while
promoting innovation and institutional adaptability.
5.2.6 Unified Human Intelligence: Integrating Decision-making and Feedback Mechanism for
AI Governance
The Unified Human Intelligence component of the proposed framework establishes a structured
approach to AI governance in HE, ensuring that AI integration remains ethical, accountable, and
strategically aligned with institutional objectives. This system functions through a continuous
interaction between decision-making and feedback loops, creating an adaptive mechanism that
32 of 63
refines AI-driven processes while maintaining institutional and ethical integrity (refer Figure 4).
Decision-making directs institutions toward their strategic goals, while feedback mechanisms assess
the impact of those decisions, ensuring AI remains under human oversight [130]. As illustrated in
Figure 3, these interdependent pillars prevent AI from operating in isolation, reinforcing human
intelligence to build self-awareness, trust, collaboration, accountability, and ethical compliance.
This synergy safeguards against automation bias, ethical lapses, and misalignment with educational
values, ensuring AI remains a transparent and human-centered tool [88,163,175].
Figure 4: Unified Human Intelligence: Collaborative Decision-making and Feedback Mechanism for AI Governance
Decision-Making as the Driver of Ethical AI Implementation
Empirical studies highlight the need for structured AI governance, addressing gaps in AI
curriculum implementation and ethics compliance across universities / HEIs [22,98]. The
European Commission’s Ethics Guidelines for Trustworthy AI (2019) reinforce HD-
AIHED’s emphasis on human agency, bias mitigation, and regulatory compliance, ensuring
responsible and sustainable AI adoption in HE [37,38].
To address these challenges, this study’s framework, the HD-AIHED model, adopts a multi-
tiered decision-making approach—collaborative, directive, and participatory—to ensure
ethical AI governance in HE. By fostering fairness, transparency, and accountability, this
model aligns with global AI ethics standards, promoting the responsible deployment and
long-term sustainability of AI in HEIs [9,20].
Collaborative decision-making involves internal and external stakeholders, including
educators, students, industry experts, regulatory bodies, and policymakers, ensuring AI
systems address diverse educational needs while mitigating algorithmic biases [176].
Directive decision-making assigns clear governance responsibilities to institutional
leadership, reinforcing structured execution, transparency, and accountability [166].
Participatory decision-making fosters direct engagement between faculty, administrators,
and students in the co-development of AI tools, ensuring AI serves as an enhancement to
pedagogy and administration rather than a directive force [167,177]. By integrating
structured decision-making, the framework ensures AI remains aligned with human values,
reinforcing educational integrity while fostering innovation [169].
Feedback Loops for AI Optimization and Refinement
33 of 63
The framework embeds dynamic feedback loops to enhance AI’s adaptability, accuracy, and
alignment with institutional goals. These mechanisms facilitate real-time refinements,
enabling institutions to detect algorithmic inconsistencies, recalibrate AI strategies, and
mitigate emerging biases [6]. By ensuring AI governance remains responsive to regulatory
changes, technological advancements, and shifting demographics, these feedback processes
reinforce long-term AI sustainability [176]. This self-correcting AI ecosystem fosters
responsible evolution, maintaining alignment with institutional, ethical, and regulatory
principles [88,166,167]. Furthermore, [176] explore decision-making models in AI,
reinforcing the necessity of adaptive feedback mechanisms to support ethical governance
and strategic oversight.
Synergy Between Decision-Making and Feedback
[99] emphasize that upholding institutional credibility and ethical integrity in HE requires a
harmonized approach to AI decision-making and feedback, ensuring alignment with
regulatory and ethical standards. This harmony ensures that institutional AI strategies
remain dynamic, transparent, and adaptable. While decision-making establishes strategic
direction and governance frameworks, feedback mechanisms continuously refine AI-driven
outputs, fostering a cycle of learning, optimization, and trust-building [6]. This
interdependent process enhances AI’s accuracy, institutional credibility, and stakeholder
confidence, positioning AI as an adaptive, ethical, and human-centered educational tool
[176]. As a result, the framework enables a strategically governed, human-aligned
transformation that upholds the core values of higher education [169,175].
5.2.7 Phased Human Intelligence Corresponding to AI Lifecycle
Synchronizing Phased Human Intelligence Across the AI Lifecycle
Ethical AI implementation in higher education requires systematic and structured
synchronization and integration of human intelligence across every stage of the AI lifecycle
to ensure accountability, transparency, and value alignment [161]. A phased approach to AI
governance is essential for ethical compliance and sustainable impact in HE. [22] advocate
for a structured AI-driven model with continuous governance, while [110] highlight AI’s
role in advancing Sustainable Development Goals (SDGs) through phased implementation.
Expanding on this, [6] emphasize the necessity of continuous monitoring and adaptive
feedback loops to refine AI deployment and mitigate risks. Aligning with these insights, the
HD-AIHED model incorporates a self-correcting governance framework, ensuring real-time
adjustments while upholding academic integrity. Building on these principles, the proposed
framework introduces phased Human Intelligence, strategically mapped across the AI
lifecycle—adoption, design, deployment, evaluation, and exploration—to embed structured
oversight, institutional adaptability, and ethical safeguards at every stage as illustrated in
Figure 5 [31].
The AI lifecycle, as shown in the Figure 4, consists of five interconnected phases: adoption,
design, deployment, analysis, and exploration. Corresponding to these, the framework
integrates Phased Human Intelligence components, ensuring strategic governance, ethical
compliance, and institutional trust at each stage. As depicted in the Figure 4, this framework
integrates decision-making and feedback loops, fostering ethical, transparent, and
accountable AI implementation in higher education.
34 of 63
Figure 5: Phased Human Intelligence corresponding to AI Lifecycle
Phased Unified Human Intelligence for Addressing AI Lifecycle Challenges
Table 7 presents a structured mapping of Phased Human Intelligence within the AI lifecycle,
ensuring ethical, inclusive, and transparent AI governance from feasibility assessment to
long-term sustainability. The validation criteria establish governance benchmarks,
reinforcing AI’s accountability, fairness, and effectiveness in higher education.
The framework structures AI governance into five key phases, integrating human and
technological aspects to align AI implementation with institutional goals. As illustrated in
Figure 4, each AI lifecycle stage is paired with a specific governance mechanism, enabling
continuous evaluation, decision-making, and feedback integration. This adaptive model
mitigates biases and ethical risks while fostering trust, scalability, and institutional resilience
in AI-driven education.
In the adoption phase, Critical Evaluation focuses on ethical decision-making and capacity
building to assess AI readiness. During design, Resource Selection emphasizes bias
minimization, participatory design, and pilot testing to validate AI’s role. The deployment
phase involves Overcoming Challenges, ensuring ethical compliance, privacy protection,
and institutional trust. In analysis, Evaluation of Outcomes measures data-driven decision-
making, institutional adaptability, and performance impact to refine AI implementation.
Finally, Future Scope in the exploration phase emphasizes long-term AI innovation,
scalability, and global benchmarking for sustainable governance.
Throughout all five phases, stakeholder participation is crucial to mitigate biases and
reinforce human oversight, ensuring AI remains transparent, ethical, and aligned with higher
education values.
Table 7: Phased Unified Human Intelligence for Addressing AI Lifecycle Challenges
Phase
Phased Unified Human Intelligence
AI Lifecycle
Literature
Support
Phase
Focus
Human Aspect
Technological Aspects
Phase
Outcome
Validation Criteria
35 of 63
Phase
1
Critical
Evaluation
Stakeholder's
participation,
Collaborative
&
Participative
decision-
making,
Ethical &
Inclusive AI
decision-
making,
Real-time,
Iterative &
Continuous
Feed-back,
Training &
Re-skilling,
Capacity
building,
Stakeholder's
confident,
trust and
satisfaction
SWOC Analysis,
Institutional readiness,
External Links, Risk
Mitigation, Compliance &
Funding Insights
Adoption
Feasibility Validation:
Usefulness, Innovativeness,
Resource Capabilities,
Personalization,
Performance and Effort
Expectancy.
[6,23,24,178–180]
Phase
2
Resources
Selection
AI Tool selection,
Resource selection,
Participatory design, Bias
Minimization, Pilot testing,
Processing
Design
Application & Benefits
Validation: Teaching-
Learning & Research,
Predictive Analytics,
Administrative
Optimization, Admissions
and Enrollment, Academic
Integrity, Inclusivity and
Accessibility, Performance
Assessment.
[10,26,33,36,38]
Phase
3
Overcoming
Challenges
and Barriers
Barrier identification,
Ethical compliance,
Privacy protection,
Institutional trust-building,
Ethical AI acceptance,
Transparency in AI
Decision-making
Deployment
Ethical & Regulatory
Validation:
AI Governance Adherence,
Data Security, Algorithmic
Bias, Resources &
Infrastructure Gaps, Human
Balance, Equity &
Inclusion.
[11,31,45,50,145,
181]
Phase
4
Evaluation
of
Outcomes
and
Benefits
Data-driven Decision
Making, Accountability &
Accuracy, Impact
assessment, Outcome
Performance matrix,
Institutional adaptability,
Performance &
Institutional Trust Metrics
Evaluation
Outcome Validation:
Lifelong Learning,
Operational Efficiency,
Inclusivity of Diversity,
Improved Student
Engagement, Strengthened
Academic Integrity,
Improved Academic Life.
[6,8,17,39,140]
Phase
5
Future
Scope
SWOC Analysis, External
Links, Sustainability &
Scalability, Long-term AI
Innovation, Global
Collaboration &
benchmarking, Global
Ethical AI Partnership,
Adaptive AI Regulations,
Future AI Governance
Standards, AI promotion
for broader adoption.
Exploration
Future Orientation
Validation: Scalability,
Sustainability & Long-term
Adaptability, Competitive
Advantage, Benchmarking,
Academic Expansion,
Global Collaboration,
Long-Term Benefits,
AIHED Promotion.
[8,11,22,31,99]
5.3 Establishing and Executing Human-Driven AIHED
This study aims to develop a conceptual framework for the ethical, inclusive, and sustainable
integration of AI in higher education (AIHED) while ensuring alignment with institutional goals,
societal values, and ethical safeguards. To achieve this, a unified human synergy in decision-making
and feedback processes is essential across all stages of AIHED, from adoption to deployment and
long-term sustainability [6].
This unified synergy is designed to:
Facilitate Adaptability – Feedback mechanisms enable AI systems to evolve in response
to stakeholder needs and contextual changes [85].
Enhance Accountability – Decision-making checkpoints ensure transparency and provide
structured interventions when AI outcomes deviate from intended goals [8].
36 of 63
Promote Inclusivity – Active stakeholder involvement incorporates diverse perspectives
into strategic AI decision-making [175].
Support Ethical Compliance – Regular audits and human oversight address data privacy,
fairness, and responsible AI deployment [182].
Through a thematic and analytical approach, this study proposes a conceptual framework for the
ethical, inclusive, and sustainable integration of AI in higher education (AIHED) while ensuring
alignment with institutional goals, societal values, and ethical safeguards. The framework is
structured into five distinct phases, representing Unified Human Intelligence corresponding to the
AI lifecycle. These phases, spanning from AI adoption to exploration and long-term sustainability,
are depicted as interconnected stages illustrating a sequential and systematic AI governance
approach [26].
Human and external intelligence are represented as ovals, with five human connections assigned to
each phase and a single external connection integrated into the system. Human decision-making
points (DM), symbolized by green arrows, link strategic human roles to AI governance. Decision-
making functions in two key modes [183]:
Collaborative and participative – Ensuring stakeholder engagement in AI decision-making.
Operative and directive – Providing structured implementation and strategic oversight.
Additionally, dotted red arrows form a dynamic feedback loop (FB) to enable continuous system
reviews, refinements, and iterative enhancements.
To maintain efficiency, accountability, and adaptability, the framework integrates real-time
feedback loops for continuous system improvements, while decision-making checkpoints ensure
validation and governance oversight. These interconnected processes enhance precision, alignment,
and AI optimization, reinforcing accountability, efficiency, and long-term adaptability [6].
The resulting model, termed Human-Driven AIHED (HD-AIHED), is illustrated in Figure 6. It
represents a structured workflow where human intelligence, decision-making, and feedback
mechanisms seamlessly integrate into AI-driven higher education, ensuring ethical AI governance,
institutional sustainability, and adaptive transformation.
37 of 63
Figure 6: Human-Driven AIHED Model
5.3.1 Phase 1: Critical Evaluation (AI Adoption)
Phase 1 of the AI adoption evaluation is spearheaded by Human Intelligence 1 (HI1), comprising
the institution’s top management and governing body. This phase is designed to critically assess the
feasibility of AI integration while ensuring strategic alignment with institutional objectives and
long-term sustainability [85].
38 of 63
To establish a robust foundation, HI1 engages in external intelligence gathering through External
Human Intelligence (EHI), facilitated by Feedback Loop 1 (FB1). This continuous input is sourced
from regulatory bodies, industry experts, and societal trends, providing valuable insights into
emerging opportunities, potential challenges, and best practices in AI adoption [38].
A key focus of this phase is ethical compliance, ensuring that AI adoption adheres to internationally
recognized frameworks, such as UNESCO’s AI governance principles [31]. Additionally, socio-
cultural adaptability is rigorously evaluated to assess AI’s compatibility with institutional values,
resource capabilities, and contextual applicability [6].
The evaluation framework incorporates an iterative feedback mechanism, wherein HI1 receives
structured insights from Human Intelligence 2 (HI2)—a group comprising senior educators,
researchers, and academic leaders. These insights facilitate an in-depth assessment of the
institution’s internal strengths and weaknesses concerning AI implementation.
In cases where unresolved issues persist, an additional iterative feedback loop (FB3) enables HI1
to reassess strategies, integrate newly acquired intelligence, and refine AI adoption approaches
before advancing to the next phase. This dynamic and adaptive process ensures a high degree of
transparency and responsiveness to emerging challenges.
A comprehensive SWOC (Strengths, Weaknesses, Opportunities, and Challenges) analysis is
conducted to evaluate both internal institutional capabilities and external market dynamics. This
structured assessment ensures that AI adoption decisions are data-driven, risk-aware, and aligned
with institutional priorities [6].
Decision-making in this phase is twofold:
Analytical decision-making (DM 1a) is employed to ensure a rigorous, data-driven
evaluation of AI adoption, focusing on usefulness, innovativeness, resource efficiency,
personalization, performance expectations, and socio-cultural adaptability.
Directive decision-making (DM 1b) establishes a clear framework for operational execution,
outlining the roles and responsibilities of HI2 to ensure a seamless transition to Phase 2.
The culmination of this phase is a validated AI adoption strategy that aligns with institutional
objectives, ethical standards, and socio-cultural considerations. By leveraging continuous feedback
loops, systematic evaluations, and strategic decision-making, Phase 1 ensures that AI adoption is
methodically structured, ethically compliant, and institutionally sustainable before progressing to
the next phase of implementation [85].
5.3.2 Phase 2: Selection of AI Tools and Resources, Pilot Testing and Processing (AI Design)
Building on the critical evaluation in Phase 1, Phase 2 advances toward the selection, design, and
integration of AI tools, ensuring their alignment with institutional objectives [66]. This phase is led
by Human Intelligence 2 (HI2), comprising senior educators, deans, academicians, and researchers,
and ethical review board who are responsible for assessing AI tools, managing their
implementation, and designing AI frameworks tailored to institutional needs [11].
HI2 gathers external intelligence to evaluate AI tools and industry suppliers based on the AI
applications recommended by HI1. This ensures that AI solutions are selected with a strong
foundation of technological feasibility, ethical compliance, and institutional relevance.
The selection and design process is governed by strategic decision-making (DM2a), which is based
on DM1a from Phase 1. This approach leverages the data-driven, analytical evaluation framework
of DM1a, ensuring continuity in decision-making. DM2a is collaborative and participative, ensuring
that AI solutions align with:
39 of 63
Teaching-learning processes
Predictive analytics and academic integrity
Administrative optimization (admissions, enrollment, operations)
Inclusivity and accessibility
AI design is a critical component of this phase, ensuring that AI solutions are contextually relevant,
ethically responsible, and aligned with institutional goals. AI frameworks are developed with a
focus on:
Personalization to enhance user engagement
Interoperability for seamless integration into institutional systems
Scalability to support long-term adoption
Compliance with regulatory and ethical standards
Upon the selection and design of AI tools, operative and directive decision-making (DM2b) ensures
appropriate resource allocation, establishes clear roles and responsibilities within Human
Intelligence 3 (HI3), and prepares for the transition to Phase 3.
Following AI selection and design, HI2 integrates AI tools into an institutional repository to
facilitate pilot testing. This ensures that the AI system operates within a controlled environment,
allowing for initial validation before full-scale deployment.
Pilot testing is continuously assessed through real-time iterative feedback (FB4) from HI3 (middle
management, IT staff, and ethical review boards), confirming the effectiveness of AI
implementation. Additionally, Feedback Loop 3 (FB3) from HI2 to HI1 ensures compliance with
institutional AI adoption policies and strategic objectives.
Refinements to AI design are made dynamically based on feedback, ensuring the system’s
adaptability, accuracy, and efficiency before moving forward.
Once pilot testing is successfully completed, HI2 activates final processing, confirming that AI
design and integration meet institutional standards. At this stage, HI2 facilitates operative, directive,
and tactical decision-making to ensure a seamless transition of responsibilities to HI3 for full-scale
implementation in Phase 3.
This decision-making framework promotes swift and effective AI selection, design, and
implementation, while fostering a collaborative ecosystem involving both internal and external
stakeholders.
Phase 2 transitions into tactical execution, addressing medium-term objectives such as repository
integration, AI design validation, and pilot testing. HI2 oversees human role assignments, manages
pilot programs, and monitors AI system functionality, ensuring that feedback continuously refines
both selection and testing processes.
Upon the successful completion of pilot testing, Feedback Loop 3 (FB3) informs HI1 that the system
aligns with AI adoption practices and institutional objectives, establishing a solid foundation for
full-scale deployment in Phase 3.
5.2.3 Phase 3: Overcoming Challenges and Barriers (AI Deployment)
Following the selection and pilot testing of AI tools in Phase 2, Phase 3 focuses on overcoming
challenges and barriers to ensure a seamless AI deployment process [50]. This phase is led by
Human Intelligence 3 (HI3), comprising middle management, IT teams, educators, students (end
users), and auditors. Their role is to address ethical, regulatory, technical, and human-centered
challenges through a structured decision-making framework.
40 of 63
HI3 adopts a collaborative and participatory approach to resolve barriers effectively. The decision-
making process is structured into two key components:
Analytical decision-making (DM3a) ensures a thorough evaluation of ethical, regulatory,
and operational challenges in compliance with governing bodies.
Operative and directive decision-making (DM3b) facilitates cross-functional collaboration,
ensuring that issues are addressed efficiently, with a clear task allocation for stakeholders to
implement the framework [184].
Tactical and operational decision-making plays a pivotal role in implementing solution-driven
strategies, allowing HI3 to manage day-to-day operations while ensuring AI deployment remains
aligned with institutional objectives.
HI3 is responsible for overcoming AI deployment challenges through DM3a, which includes:
Ethical and regulatory compliance: Addressing data security, algorithmic bias,
infrastructural limitations, and resource constraints to ensure AI operates within ethical and
legal standards.
Ensuring a human-centered approach: Maintaining a balance between technological
advancement and human values, focusing on:
o Emotional intelligence and creativity
o Job retention and creation
o Engagement, reskilling, and training
o User satisfaction and adaptability
Real-time and iterative feedback (FB4) plays a crucial role in refining AI processes and tools while
emphasizing ethical considerations. This continuous feedback mechanism ensures proactive issue
resolution and iterative improvement in AI system deployment [7].
HI3 facilitates effective communication and engagement among stakeholders, ensuring timely
problem resolution and continuous system refinement. Ethical compliance and inclusivity are key
performance indicators guiding AI deployment adjustments.
Through directive and operative decision-making (DM3b), HI3 allocates tasks and responsibilities
to HI4, ensuring a seamless transition to Phase 4, where AI deployment moves towards long-term
sustainability, institutional integration, and strategic scaling.
5.3.4 Phase 4: Analysing Outcomes and Benefits (AI Evaluation)
In Phase 4, Human Intelligence (HI4)—comprising senior faculty, administrators, auditors and
students—plays a pivotal role in evaluating the effectiveness of the AIHED system. This assessment
is based on critical metrics, including operational efficiency, ethical compliance, inclusivity,
institutional trust, and performance benchmarks [10]. Analytical decision-making facilitates a data-
driven evaluation of these outcomes, ensuring that AI adoption leads to meaningful educational and
institutional benefits.
Strategic Decision-Making (DM4a) operates through a collaborative and participative approach,
engaging HI4 to determine whether AI-driven outcomes justify enhanced operational efficiency,
diversity and inclusivity, improved student engagement, academic integrity, and lifelong learning
benefits.
A structured feedback mechanism supports continuous refinement:
41 of 63
Continuous Feedback (FB6) collects audit reports, user performance data, and insights into
institutional reputation for HI2, ensuring transparency and iterative system enhancements
[45]. Further passed to HI1 through FB7.
Iterative Feedback (FB5) helps identify gaps in performance metrics, enabling HI3 to refine
strategies, resolve unresolved issues, and drive system improvements.
Operative and Directive Decision-Making (DM4b) ensures cross-stakeholder collaboration,
integrating diverse perspectives for a comprehensive and balanced evaluation.
Tactical decision-making focuses on implementing recommendations at the operational
level.
Directive decision-making assigns specific tasks and responsibilities to ensure effective
execution.
By leveraging research and innovative methodologies, DM4b ensures that collective input leads to
meaningful advancements. This multi-faceted and integrated framework optimizes the AIHED
system and prepares it for a seamless transition to Phase 5
5.3.5 Phase 5: Future Scope (AI Exploration)
Phase 5 of the HD-AIHED Model corresponds to future-proof strategies, emphasizing scenario
planning and continuous research to ensure the long-term adaptability of AI systems to evolving
societal and educational needs, as recommended by [99]. In the final phase, Human Intelligence
(HI5), consisting of researchers and strategists, collaborates with HI2 and External Intelligence to
evaluate emerging needs, technological advancements, and opportunities for enhancing global
competitiveness through FB8 [26]. The SWOC analysis done by HI5 on future opportunity is
communicated to HI2 then passed to HI1 through FB10 and FB11 respectively. After collaborative
and participative discussion held at HI1 and HI2, FB8 provides feedback on AI exploration.
Continuous feedback (FB8) from HI2 informs HI5 of areas requiring improvement, addressing
evolving institutional needs, policy reforms, and newly established goals. Feedback loops (FB10)
guide HI2 in exploring future partnerships and driving innovation to ensure adaptability in a
dynamic landscape. Conceptual decision-making (DM5a) prioritizes scalability and long-term
sustainability, ensuring AI remains an evolving enabler in higher education [80].
Feedback loops (FB9) provide iterative insights into potential opportunities and risks, enabling HI4
to refine strategies and prepare for the future. This dual feedback mechanism fosters transparent
communication, adaptability, and alignment with long-term institutional objectives. It supports life-
long learning opportunities and ensures that institutions remain agile and competitive. By
integrating these iterative insights, institutions can proactively address future challenges, capitalize
on advancements in AI, and align with global standards, ensuring sustained relevance and impact.
Lastly, after successful implementation and achieving satisfactory benefits and outcomes, Human
Intelligence 1 (HI1) takes responsibility for communicating (FB12) success stories and with
External Human Intelligence. This involves the publication of success stories and strategies to
overcome obstacles associated with AIHED through various media channels to foster social appeal
and demonstrate the institution's accomplishments. This communication reinforces transparency,
enhances institutional reputation, and inspires broader engagement with the institution’s
advancements in AI adoption.
The HD-AIHED operational flow (Refer Figure 5) ensures AI adoption, design, deployment,
evaluation, and exploration remain ethically sound, institutionally integrated, and dynamically
evolving. By embedding human oversight and external intelligence at every stage, this framework
provides continuous adaptability through structured feedback loops, accountability and
transparency via decision-making checkpoints, and scalability aligned with long-term educational
needs. This integrated methodology strengthens institutional trust, fosters inclusive AI ecosystems,
42 of 63
and ensures that AI remains an enabler of human intelligence rather than a replacement, preserving
ethical, academic, and pedagogical integrity in higher education.
6. Key Insights from the HD-AIHED Model and Interpretation
6.1 Significance and Strengths of the HD-AIHED Model
The HD-AIHED framework envisions AI as an ethical force for innovation, inclusivity, and
sustainability in higher education (HE), ensuring that AI enhances rather than replaces human
intelligence [44,185]. It serves as a comprehensive and human-centric model that prioritizes
responsible AI adoption, ensuring that AI-driven transformation in HE remains aligned with
institutional goals, pedagogical values, and stakeholder needs [38,39].
By emphasizing strategic decision-making, dynamic feedback mechanisms, and phased
implementation, the HD-AIHED Model provides a structured and ethical pathway for AI
integration, addressing key challenges such as algorithmic bias, privacy concerns, and student
agency in AI governance [9,20].
Comprehensive Sequential Flow: The HD-AIHED Model encompasses the entire AI
lifecycle, from critical evaluation during adoption to analyzing outcomes and planning for
future scalability. This phased approach ensures that no essential aspect of AI
implementation is overlooked, creating a holistic framework for integrating AI into higher
education
Human-Centric Approach: The model emphasizes human intelligence; oversight,
integrating multiple feedback and strategic decision-making connections (e.g., HI1-HI5)
throughout all phases. This human-centric approach ensures ethical compliance, alignment
with institutional goals, and cultural sensitivity, fostering trust and inclusivity in AI systems
through participatory design.
Dynamic Feedback Mechanisms: Continuous and iterative feedback loops (FB1-12)
embedded at every phase ensure iterative improvements, adaptability, and accountability in
AI implementation. These mechanisms align with institutional objectives while addressing
real-time inefficiencies and stakeholder concerns.
Pilot Testing for Validation: The inclusion of pilot testing in Phase 2 validates AI tools
against functional and institutional requirements before full deployment. This reduces risks
associated with misalignment, errors, and inefficiencies, ensuring a smoother integration
process.
Addressing Ethical and Operational Barriers: Phase 3 focuses on overcoming barriers such
as algorithmic bias, data security, and institutional resource gaps. By addressing these
common challenges proactively, the framework ensures ethical and practical readiness for
AI deployment through capacity building.
Rigorous Outcome Evaluation and Scalability: Phases 4 and 5 are dedicated to evaluating
AI outcomes, analyzing long-term impacts, and identifying opportunities and risks for
growth and scalability. This future-oriented focus ensures the AI system remains relevant,
effective, and aligned with technological advancements.
Support Across the Student Lifecycle: The framework supports AI applications throughout
the student lifecycle, from admissions and personalized learning to graduation. This ensures
equitable opportunities, personalized engagement, and improved student outcomes,
reflecting a commitment to inclusivity and student success.
Benchmarking and Global Competitiveness: In Phase 5, the model incorporates
benchmarking to assess global standards, ensuring competitiveness, scalability and
sustainability. By aligning with emerging technologies and ethical principles, it prepares
institutions for future advancements while maintaining compliance with regulatory norms.
Knowledge Repository for Institutional Learning: The integration of lessons learned
through feedback loops creates a valuable knowledge base, enabling institutions to refine
strategies for future AI implementations. This facilitates organizational learning and
innovation, building a robust foundation for continuous improvement.
43 of 63
Ethical Oversight and Trust-Building: The model’s emphasis on ethical oversight ensures
compliance with principles such as beneficence, autonomy, and justice. Transparent
processes foster trust among stakeholders by demonstrating accountability and fairness,
promoting confidence in AI-driven systems.
Global Integration: The Model underscores the significance of external connections to
enhance its scalability, relevance, and impact with HI1 & HI2). By adhering to regulatory
compliance, such as GDPR and global AI governance frameworks, it ensures ethical
alignment and legal accountability. The model leverages technological advancements and
market analyses to adapt AI-driven tools to emerging trends and innovations, while global
benchmarking fosters competitiveness by aligning with international best practices.
Additionally, it emphasizes the importance of cross-sector collaborations, partnerships with
industries, and engagement with global stakeholders. A key strength lies in its commitment
to communicating success stories and failures, enabling institutions to share insights, foster
trust, and refine strategies. This transparent approach supports sustainable growth,
innovation, and credibility within the AI-driven higher education ecosystem
6.2 HD-AIHED Model in Action: A Strategic Mapping and Solution for Overcoming Real-
Time Challenges and Barriers
Table 8 serves as an analytical repository, mapping AI applications across global universities and
assessing their adoption challenges, solutions, and institutional impact. It provides an evidence-
based evaluation of how AI systems perform in diverse educational settings, identifying best
practices, gaps, and opportunities for expansion.
By applying the HD-AIHED Model, this strategic mapping offers a validated framework to navigate
AI adoption complexities, ensuring that institutions overcome governance barriers, mitigate ethical
risks, and drive sustainable AI transformation in higher education. The table systematically analyzes
AI applications across universities, aligning them with institutional goals and innovations while
assessing their real-world effectiveness.
Each AI application is evaluated through five key phases—Adoption, Design, Deployment,
Evaluation, and Exploration—providing a structured approach to overcoming institutional barriers
and ethical challenges. This phased framework ensures AI integration remains strategic, adaptive,
and aligned with human-driven governance, ultimately optimizing AI’s role in shaping the future
of higher education.
Table 8: HD-AIHED Model: A Strategic Mapping and Solution for Overcoming Challenges and Barriers
Region/
University/
Institution
AI
Application &
AI Tool
Repository
Input
Phase 1 -
Adoption
Phase 2 - Design
Phase 3 -
Deployment
Phase 4 -
Evaluation
Phase 5 -
Exploration
Critical
Evaluation
Selection of
Resources,
Testing & Final
Processing
Overcoming
Challenges
Analysing
Outcomes
Future Scope -
Opportunities
& Risks
Arizona
State
University
(USA)
Personalized
Learning
Pathways (AI
platforms like
Squirrel AI,
ALEKS)
Student
performance
data, learning
styles, and
curriculum
objectives
Usefulness,
resource
capabilities, and
personalization
validated
through SWOC
analysis
Pilot testing of
personalized
learning paths
ensuring
curriculum
alignment
Address
regulatory
barriers and
institutional
compatibility
Analyze
outcomes such
as learning
pace,
engagement,
and curriculum
fit
Expand to
support hybrid
learning models
and gamified
experiences for
broader
implementation
University of
Toronto
(Canada)
AI-Powered
Research
Assistance
(Tools like
Research
Rabbit, Iris.ai)
Institutional
research
outputs,
journal
databases, and
researcher
profiles
Evaluate the
relevance and
usefulness of AI
tools for
interdisciplinary
research
Selection of tools
based on literature
gap identification
and AI-driven
recommendations
Address
challenges like
data
accessibility, AI
recommendatio
n accuracy, and
inclusivity
Assess research
outcomes,
interdisciplinary
connections,
and user
satisfaction
metrics
Scale tools for
multilingual
research
synthesis and
integrate with
global research
databases
44 of 63
University of
Melbourne
(Australia)
AI-Powered
Chatbots for
24/7 Support
(Chatbots like
IBM Watson
Assistant, Ada)
FAQs,
institutional
policies,
student
queries, and
service logs
Determine
chatbot
usefulness for
resolving
routine queries
and escalating
complex cases
Pilot testing
chatbot
workflows within
small departments
before scaling up
Overcome
challenges
related to
response
accuracy,
escalation
processes, and
language
support
Analyze
outcomes like
improved
service
efficiency,
query resolution
rates, and user
feedback
Expand chatbot
capabilities to
multilingual
functionality
and
alumni/commun
ity engagement
University of
California
(USA)
Proctoring and
Exam Integrity
(Proctoring
tools like
ProctorU,
Honorlock)
Exam
question sets,
student
credentials,
live webcam
and keystroke
monitoring
data
Validate AI's
effectiveness in
proctoring
against ethical
and privacy
considerations
Implementation of
AI-based
proctoring tools
with faculty
feedback and
gradual
integration
Address ethical
concerns, false
positives, and
system accuracy
challenges
Analyze
outcomes such
as proctoring
effectiveness,
fairness, and
user satisfaction
rates
Enhance AI-
driven post-
exam analytics
and adaptive
questioning
systems
Stanford
University
(USA)
AI in Research
& Patent
Ethics (AI-
driven
plagiarism
detection &
research
integrity tools)
Research
papers, patent
filings, and
institutional
policies
Evaluate AI's
role in ensuring
research
integrity and
detecting
plagiarism
AI-driven analysis
of patent
applications and
automated
research
verification
Address patent
ownership
disputes and
legal ambiguity
in AI-generated
research
Assess AI's
effectiveness in
maintaining
ethical research
standards
Implement AI
governance
frameworks to
manage AI-
generated
patents and
research ethics
Georgia
Tech (USA)
AI-Driven
Recruitment
Analytics
(Recruitment
tools like
Eightfold.ai,
Entelo)
Historical
hiring data,
skills
databases, and
institutional
workforce
needs
Evaluate tools
for predictive
hiring, skill gap
analysis, and
workforce
planning
Selection of AI
tools aligned with
institutional
workforce needs
and recruitment
priorities
Address
challenges
related to
fairness,
diversity, and
predictive
analytics
outcomes
Analyze
recruitment
efficiency,
diversity
outcomes, and
alignment with
institutional
goals
Scale AI to
support long-
term workforce
planning and
global talent
acquisition
National
University of
Singapore
(Singapore)
AI-Powered
Accessibility
for
Neurodiverse
Students
(Tools like
CogniFit,
Glean)
Cognitive
ability
assessments,
behavioral
data, and
neurodiverse
student
profiles
Assess
inclusivity,
usability, and
alignment with
institutional
accessibility
goals
Implementation of
AI tools tailored
to neurodiverse
needs through
pilot testing
Overcome
accessibility
barriers and
validate
inclusivity
benchmarks
Measure
learning
outcomes,
usability
feedback, and
cognitive
adaptability
Expand with
AR/VR tools for
broader
neurodiverse
learning
enhancements
Harvard
University
(USA)
AI in
Accreditation
(AI-based
compliance
tools)
Faculty
performance
metrics,
accreditation
frameworks
Validate AI for
automating
compliance
tracking and
performance
evaluation
Implement AI-
driven
accreditation
monitoring
systems
Address bias in
AI-based
faculty
performance
assessments
Assess
improvements
in accreditation
transparency
and compliance
rates
Develop
adaptive AI
credentialing
frameworks for
global
accreditation
MIT (USA)
Virtual Reality
(VR)
Immersive
Learning (VR
tools like
Oculus for
Higher
Education,
zSpace)
Lesson plans,
3D models,
and student
progress data
Validate
alignment of
VR learning
tools with
curriculum
goals and
resource
capabilities
Selection and
pilot testing of
VR tools for
specific
curriculum
modules
Address barriers
such as cost,
resource
limitations, and
teacher training
needs
Evaluate
learning
engagement,
retention rates,
and
performance
outcomes
Broaden VR
integration
across subjects
and develop
AR-based real-
world learning
systems
IIT Bombay
(India)
AI in Campus
Placements
(AI-driven job
recommendatio
n systems)
Student
resumes,
employer
requirements,
industry
hiring trends
Assess AI's
potential in
optimizing job
matching and
placement
success
Implement AI-
powered resume
analysis and
employer-student
matchmaking
Address AI bias
in resume
shortlisting and
candidate
ranking
Evaluate job
placement
success rates
and employer
satisfaction
Develop
human-AI
hybrid hiring
models for
fairer
recruitment
processes
45 of 63
Oxford
University
(UK)
AI-Powered
Emotional
Intelligence
Analysis
(Sentiment
analysis tools
like Affectiva)
Video/audio
data from
classrooms,
student survey
responses, and
interaction
logs
Validate
effectiveness in
detecting
emotions and
engagement
while ensuring
privacy
safeguards
Implementation of
sentiment analysis
tools and targeted
emotional support
programs
Overcome
challenges
related to
emotional data
interpretation
and ethical
concerns
Compare
emotional well-
being metrics
pre- and post-
intervention for
outcome
validation
Expand into AI-
driven
emotional
development
and teacher
training
programs
Cambridge
University
(UK)
AI-Powered
Campus
Energy
Optimization
(Smart campus
tools like
BuildingIQ,
Enel X)
Real-time
energy
consumption
data, weather
patterns, and
occupancy
rates
Evaluate the
cost-
effectiveness
and eco-friendly
potential of
energy
optimization
tools
Selection of tools
for
implementation
with pilot energy
audits and
sustainability
benchmarks
Overcome
integration
challenges with
existing
infrastructure
and energy
compliance
standards
Assess energy
savings, carbon
footprint
reduction, and
user satisfaction
metrics
Integrate
renewable
energy sources
and improve
sustainability
across campuses
Carnegie
Mellon
University
(USA)
AI in Industry
Collaboration
(AI tools for
corporate-
academic
partnerships)
Research
collaboration
agreements,
industry-
funded
projects
Assess AI’s role
in strengthening
university-
industry
partnerships
Implement AI-
driven
collaboration
platforms for
project
matchmaking
Address data
ownership
conflicts
between
universities and
private firms
Evaluate
research impact
and industry
engagement
success rates
Establish
contractual AI
governance
models to
ensure
institutional
data control
University of
Cambridge
(UK)
AI-Driven
Alumni
Engagement
(Engagement
tools like
Salesforce
Einstein)
Alumni
profiles, event
participation
data, and
donation
history
Evaluate tools
for fostering
meaningful
alumni
relationships
and event
participation
Pilot testing AI
recommendations
for mentorship,
event planning,
and fundraising
efforts
Address
challenges like
alumni data
quality and
maintaining
personalization
at scale
Measure
success rates in
alumni
engagement,
participation,
and fundraising
Scale AI-driven
predictive
models for long-
term alumni
collaboration
and donor
engagement
7. Discussions
7.1 Implications and Recommendations
The AIED-HDMFB model's step-by-step approach integrates critical evaluation, tool selection,
ethical compliance, and scalability into the adoption of AI in higher education. Its feedback loops
ensure adaptability and inclusivity, aligning with the broader objectives of the Human-driven
AIHED framework [36]. For instance, during the tool selection phase, human decision-makers
validate AI tools for compliance with institutional missions and stakeholder needs, reflecting the
principles of dynamic feedback and ethical alignment [10]. The framework proposed in this study
is pivotal for ensuring the sustainable and ethical adoption of Artificial Intelligence (AI) in higher
education (HE). It establishes a foundation for continuous improvement by aligning AI applications
with institutional goals, addressing limitations, and balancing AI's capabilities with human roles.
Beyond enhancing efficiency, the framework emphasizes AI as a tool for assistance rather than
replacement, fostering innovation through collaboration, lifelong learning, and emerging
technologies [155].
Towards a new notion, Human-Driven AIHED: A Paradigm Shift
The model—HD-AIHED, conceptualized from this study, transitions the notion of "AIHED
driving Humans" to "Human-driven AIHED”, placing human agency at the core of AI
integration. This approach highlights the collaborative role of both internal and external
stakeholders—educators, students, policymakers, and technology providers in ensuring AI
adoption aligns with ethical, cultural, and institutional values. Educators play a crucial role
in co-designing AI systems to meet pedagogical goals, while institutions must establish
frameworks where human oversight remains central to AI applications. Governments can
facilitate this transition by developing policies that incentivize human-centric AI initiatives
that uphold inclusivity and [10,36].
Human-Driven AIHED: Empowering Not Replacing
46 of 63
The HD-AIHED framework positions AI as an assistive tool that automates routine tasks,
enabling educators to focus on mentorship, creativity, and strategic problem-solving. The
integration of feedback loops and decision making ensures that AI evolves continuously to
meet institutional and student needs through empowering human intelligence. To make this
possible, professional development programs are essential for reskilling educators in AI-
enabled methodologies, while students must be encouraged to engage actively with AI tools
to supplement critical thinking and creativity. Ethical bodies need to establish
interdisciplinary guidelines to ensure AI deployment is aligned with inclusivity and
transparency, preserving education’s human-centric essence [7,186]
The Ethics of Equity: A Path to Inclusive HE
By integrating adaptive learning, multilingual support, and accessibility features, HD-
AIHED aims to address cultural and physical disparities in education. Feedback mechanisms
embedded in the framework help detect and mitigate biases, ensuring that AI-driven
solutions provide equitable access for diverse learners. Governments can expand AI-driven
initiatives to underserved populations through strategic funding and infrastructure
investments, while institutions must embed cultural and contextual needs into their AI tools.
Society at large should advocate for community-driven efforts that reduce digital divides
and promote inclusivity in AI-powered education [6,10].
Integrating Innovation and Institutional Values in a Digital World
The HD-AIHED framework incorporates innovative technologies such as blockchain,
AR/VR, and advanced analytics to enhance transparency, security, and efficiency, while
maintaining institutional values. These innovations, when supported by feedback
mechanisms, allow institutions to adapt technologies responsibly and ensure alignment with
evolving stakeholder needs. Industry partners are integral in co-developing AI systems
tailored to educational requirements, while institutions must adopt structured AI life-cycle
phases that balance innovation with core values. Regular audits by ethical bodies can ensure
these technologies comply with equity and ethical standards, sustaining a mission-driven
approach [10,148].
A Shared Vision: Toward a Unified and Agile Future
Aligning with NEP 2020 of India, Higher Education 4.0, and Industry 4.0, HD-AIHED
bridges learner-centric education with advanced technologies. The framework’s scalability
and adaptability prepare students for AI-augmented workplaces, while dynamic feedback
loops ensure its responsiveness to workforce demands. Institutions should embrace global
benchmarking practices and share success stories to enhance competitiveness, while
policymakers must design regulations that promote scalable and responsible AI adoption.
Students, in turn, can benefit from AI-powered career tools that offer personalized skill
development and prepare them for global opportunities [8].
Inspiring Global Transformation Through Collaborative Practices
The HD-AIHED framework emphasizes international collaboration to share best practices,
address global challenges, and foster inclusivity. Benchmarking efforts and alignment with
diverse cultural and educational contexts amplify its global impact, making it a
transformative tool for inclusive HE. Institutions must actively engage in global forums to
co-develop solutions and promote trust in AI’s transformative potential. Public dialogue
should be encouraged to address societal concerns about equity and transparency, while
ethical bodies facilitate the development of harmonized global practices, creating an
interconnected HE ecosystem [109,155].
Catalysts of Transformation: Role and Responsibilities
Governments, institutions, educators, and industry stakeholders play critical roles in the
implementation and promotion of HD-AIHED by addressing access gaps, fostering
reskilling initiatives, and ensuring ethical AI integration. Governments can transform
education by investing in infrastructure, fostering AI literacy, supporting curricula aligned
with Industry 4.0, and constructing ethical regulations to ensure inclusivity and transparency
[148]. Institutions must leverage AI to enhance operational efficiency and deliver
personalized learning experiences that attract diverse learners while maintaining socio-
cultural equity. By sharing success stories and adopting strategic enrollment approaches,
institutions can build trust and forge global collaborations [10]
47 of 63
Educators, as ethical stewards, must embrace reskilling to adapt to AI-enhanced roles,
focusing on mentorship, creativity, and innovation as AI automates routine tasks. Industry
stakeholders can collaborate with academia to design scalable, ethical AI tools tailored to
institutional needs, while creating jobs such as AI trainers, ethical auditors, and data analysts
to support workforce transformation [7]. Students benefit from AI-driven personalization
throughout their educational journey, from admissions to career preparation, gaining critical
skills for AI-augmented industries.
By combining their efforts, these stakeholders ensure HD-AIHED is adopted inclusively,
ethically, and innovatively, driving the creation of a sustainable and globally impactful AI-
powered education ecosystem.
HD-AIHED: A Universal Model for Borderless HE Future
The HD-AIHED framework serves as a universal model to address global challenges in
higher education by harmonizing innovation with equity, transparency, and lifelong
learning. It aligns with international initiatives such as SDG 4 and UNESCO’s guidelines
on AI ethics, promoting ethical AI adoption while ensuring inclusivity and accessibility for
diverse populations. In developing nations, the framework bridges gaps in infrastructure and
resources by leveraging affordable, culturally sensitive technologies that empower
underserved communities with lifelong learning opportunities and skill development [8].
Advanced regions benefit from HD-AIHED’s mechanisms for refining AI tools to enhance
personalization, address algorithmic bias, and uphold data privacy while positioning
themselves as innovation hubs for global research.
For low-economic regions, the framework provides a cost-effective pathway for higher
education reform, emphasizing ethical compliance and partnerships with governments and
NGOs to tackle infrastructural challenges. Cross-border collaboration further strengthens its
global impact, enabling institutions to share knowledge, co-develop scalable AI solutions,
and benchmark best practices. These efforts align with the EU’s AI Act and UNESCO’s
ethical AI guidelines, ensuring standardized, transparent, and equitable AI integration across
diverse contexts [36].
Through international partnerships and cultural sensitivity, the HD-AIHED framework
transforms HE into an interconnected ecosystem, fostering innovation while preparing
institutions and learners for a dynamic and equitable future. Institutions that actively adopt
this framework and engage in global collaborations can position themselves as leaders in
advancing ethical and inclusive AI in education.
Long-term Adaptability: Awareness, Promotion and Training
Institutions should actively leverage social media platforms, research publications, and
academic collaborations to drive awareness and engage stakeholders in AIHED adoption
[187]. Social media can be a powerful tool for sharing success stories, challenges, and best
practices, fostering a global knowledge-sharing ecosystem. Engaging with policymakers,
faculty, students, and industry leaders through these platforms can help build trust and
encourage cross-border collaborations [6,11].
Regular publications in academic journals, magazines, and institutional reports can further
reinforce the credibility of HD-AIHED, providing universities with frameworks, case
studies, and real-world insights [30]. Hosting webinars, podcasts, and panel discussions will
create an open dialogue about AI’s evolving role in HE, fostering a community of
practitioners, researchers, and decision-makers [38]. International conferences and
institutional partnerships will enhance AI governance models, ensuring that HD-AIHED
remains an evolving, globally adopted initiative [31].
For responsible AI integration, faculty and students must be trained in AI ethics and
governance [8]. Universities should implement AI ethics literacy programs that emphasize
bias detection, data privacy, fairness, and responsible AI usage [20]. Educators must be
equipped with knowledge to critically engage with AI-driven decision-making, ensuring that
AI complements human expertise rather than replacing it [7]. Students should be trained in
AI governance policies to enable their participation in institutional AI decision-making,
ethical review boards, and AI auditing [26].
Transformative Perspectives: AI as Critical Infrastructure
48 of 63
The HD-AIHED framework reimagines AI not just as a tool but as critical infrastructure
that fosters inclusive, ethical, and transformative progress in higher education. By
emphasizing human agency, creativity, and ethical accountability, it positions AI as a
collaborative partner that complements rather than replaces human intelligence. The
framework incorporates dynamic feedback loops to ensure the continuous evolution of AI
systems, enabling them to address the ever-changing challenges of higher education.
Through the integration of SWOC analyses and the alignment of human and external
intelligence across the phases of the AI life cycle, the framework bridges equity, inclusivity,
and data ethics gaps.
It promotes practical strategies to enhance lifelong learning and sustainable innovation,
making higher education more accessible and future-focused. Policymakers have a critical
role in establishing metrics to evaluate AI's long-term societal and educational impacts,
while ethical bodies advocate for iterative refinements to ensure AI aligns with institutional
missions and broader societal values. This transformative approach ensures AI’s adoption
as a sustainable, equitable, and dynamic infrastructure that drives innovation and progress
in HE globally [188,189].
AI Bias Mitigation and Ethical AI Auditing
The HD-AIHED framework requires robust bias mitigation strategies to prevent algorithmic
discrimination in grading, admissions, and learning analytics [68]. AI systems, if not
carefully managed, can reinforce existing social inequalities and lead to unintended biases
[60]. Ensuring fairness in AI decision-making is essential to maintaining trust, transparency,
and inclusivity in higher education.
To enhance fairness and accountability, following strategies should be adopted:
Bias Detection Frameworks: Institutions should conduct periodic AI audits to assess fairness
and eliminate biases in student performance predictions. This includes using algorithmic
fairness metrics and AI transparency evaluations to ensure responsible AI deployment [39].
The AI Risk Management Framework developed by [63] offers an approach to assessing
and mitigating AI risks, which could be applied in educational settings.
Explainable AI (XAI): AI tools used in HE should incorporate explainable AI models that
provide clear, interpretable decision-making processes [187]. Transparent AI models allow
educators, administrators, and students to understand how AI-based decisions are made and
intervene when necessary [26].
Interdisciplinary Ethics Committees: Universities/institutions should establish AI Ethics
Boards, composed of faculty, students, AI developers, and policymakers, to oversee AI
deployment and governance in education. These committees would evaluate potential
biases, risks, and ethical concerns related to AI systems in HE [8]. Several global
frameworks, such as the UNESCO Recommendation on the Ethics of AI (2021) and the
OECD Principles on Artificial Intelligence (2019), emphasize the importance of ethics-
based AI governance in education [31,50].
By integrating these mechanisms, HE institutions can ensure that AI enhances educational
equity without reinforcing systemic biases. Adopting these strategies will not only improve
AI fairness and transparency but also foster trust and inclusivity among students and
educators in AI-driven learning environments.
7.2 Future Research Directions
7.2.1 Quantitative Survey for Validating Study’s Framework
To ensure the HD-AIHED framework is empirically validated, a quantitative survey combined with
Structural Equation Modeling (SEM) is recommended. SEM has been widely recognized in
educational research for assessing complex relationships between latent constructs, ensuring robust
theoretical validation [190,191]. Given the increasing role of AI in higher education, it is essential
to establish a data-driven approach that quantifies AI's impact across its adoption, design,
deployment, performance, and sustainability phases [6]. Following Table 8 showcases future
research directions. Table 8 presents a structured future research agenda aimed at empirically testing
the HD-AIHED model through quantitative survey-based methodologies.
49 of 63
Table 8: Future Research Outline: Testing HD-AIHED Model With Quantitative Survey
Steps
Description
1. Define the Research
Framework
Develop hypotheses based on the five-phase AI lifecycle (adoption, design,
deployment, analysis, and exploration). Identify key indicators such as ethical
scalability, inclusivity, and human intelligence integration. Utilize technology
adoption models like TAM, UTAUT, or DoI to examine AI’s impact on educational
institutions.
2. Design a Quantitative
Survey
Create a structured questionnaire targeting higher education stakeholders,
including faculty, administrators, and students. Incorporate Likert-scale questions
to assess alignment with institutional ethics, operational effectiveness, stakeholder
acceptance, and ethical compliance in AI integration.
3. Conduct Data
Collection
Implement a large-scale survey across universities using random or stratified
sampling methods. Collect data in either a cross-sectional or longitudinal format to
ensure diverse representation and measure framework effectiveness over time.
4. Perform Statistical
Validation
Use Structural Equation Modeling (SEM) to examine relationships between AI
lifecycle phases and ethical outcomes. Apply factor analysis to identify key
dimensions influencing AI implementation, and conduct regression analysis to
predict AI’s impact on learning outcomes and institutional efficiency.
5. Assess Practical
Implications
Validate the scalability and adaptability of the HD-AIHED framework by
comparing results across different geographical and institutional contexts. Provide
actionable insights for policymakers and educators to enhance AI integration in
higher education.
6. Recommend Future
Research Directions
Integrate AI-driven data analytics to refine survey insights, combine quantitative
and qualitative methods for deeper validation, and benchmark findings against
global AI policy frameworks to assess ethical AI integration in higher education.
A structured survey-based methodology should be employed, gathering responses from key
stakeholders, including faculty, students, and administrators. Survey indicators should be designed
based on well-established theoretical models such as the Technology Acceptance Model (TAM)
[52] and the Unified Theory of Acceptance and Use of Technology (UTAUT) [54], ensuring
reliability and construct validity. Likert-scale items measuring AI readiness, human-centered
design, stakeholder-driven deployment, ethical considerations, and long-term sustainability should
be incorporated [192].
To analyze these relationships, SEM should be applied, as it enables simultaneous testing of
multiple hypotheses and constructs within a hierarchical framework [193]. This method will allow
researchers to determine the extent to which AI adoption, ethical AI design, and stakeholder
involvement contribute to institutional scalability and sustainability [194]. Furthermore,
incorporating covariance relationships among first-order constructs will provide a comprehensive
understanding of interdependencies in AI integration [195].
By validating the HD-AIHED framework through quantitative analysis and SEM, this study will
offer generalizable insights into the operational, ethical, and pedagogical implications of AI in
education. This approach will align AI deployment with institutional policies, ethical standards, and
global governance frameworks, ensuring a scalable and human-centric AI transformation in higher
education [8].
Table 9 exhibits key constructs and hypotheses aligned with the five phases of the HD-AIHED
model, offering a structured approach for its empirical validation. By applying Structural Equation
Modeling (SEM), this research proposes to test the relationships between constructs associated with
various phases of the AI lifecycle. The table outlines relevant constructs, proposed hypotheses, and
expected correlations, ensuring a data-driven assessment of the HD-AIHED model’s effectiveness
in AI-driven higher education.
Table 9: Validating HD-AIHED Model through Structural Equation Model (SEM)
50 of 63
Higher-Order
Construct
Phase
Sub-Constructs
(Latent
Variables)
Description
Hypothesis
Connection
Description
Human-
Driven AI in
Higher
Education
(HD-AIHED)
1. Adoption
Phase
Institutional
Readiness
The level of infrastructure,
policies, and technical support
for AI adoption.
H1: The successful
adoption of AI in
higher education is
positively associated
with institutional
readiness, faculty &
students acceptance,
and ethical policy
frameworks.
Institutions must
have the
necessary
infrastructure,
policies, and
stakeholder
engagement for
AI to be adopted
effectively.
Faculty and
Student
Acceptance
The willingness of educators
and learners to integrate AI into
educational practices.
Ethical and
Regulatory
Compliance
The adherence to AI governance
policies and ethical standards
during adoption.
2. Design Phase
Human-
Centered AI
Design
The extent to which AI systems
integrate human intelligence,
ethical considerations, and
pedagogical principles.
H2: AI system design
that integrates human
intelligence, ethical
safeguards, and
institutional values
enhances user trust
and engagement
among students and
faculty.
AI should be
designed with
transparency,
ethical
considerations,
and human
intelligence
integration to
ensure responsible
use.
Transparency
and Trust
The clarity of AI decision-
making processes and users’
confidence in AI tools.
Usability and
Accessibility
The ease of use and inclusivity
of AI-driven learning systems.
3. Deployment
Phase
Stakeholder
Involvement
The engagement of faculty,
administrators, and students in
AI implementation.
H3: The effectiveness
of AI deployment in
higher education is
significantly
influenced by
stakeholder
involvement,
transparency in
implementation, and
compliance with
ethical guidelines.
Effective AI
implementation
requires active
involvement and
ethical decision-
making from
faculty, students,
and
administrators.
Implementation
Transparency
The degree of openness in AI
integration, including clear
communication about AI’s role.
Ethical AI
Practices
The extent to which AI
deployment aligns with fairness,
accountability, and privacy
standards.
4. Analysis
Phase
AI-Driven
Decision-Making
The effectiveness of AI in
enhancing learning analytics,
student performance tracking,
and administrative decision-
making.
H4: AI-driven
analytics that
incorporate real-time
feedback and ethical
oversight improve
decision-making,
learning outcomes,
and institutional
efficiency.
AI's success in
HE depends on its
ability to enhance
learning
outcomes,
efficiency, and
fairness while
ensuring
compliance with
ethical standards.
Feedback
Integration
The responsiveness of AI
systems to human feedback for
continuous improvement.
Institutional
Efficiency
The impact of AI on
streamlining administrative and
academic operations.
5. Exploration
Phase
Long-Term
Sustainability
The adaptability and continuous
evolution of AI applications in
education.
H5: The long-term
sustainability and
scalability of AI in
higher education
depend on continuous
evaluation, policy
evolution, and
adaptive integration
of emerging
technologies.
The future of AI
in education relies
on continuous
evaluation, policy
evolution, and
adaptability to
emerging
technologies.
Policy Evolution
and Scalability
The ability of institutions to
update policies and expand AI
solutions.
Emerging
Technology
Integration
The readiness of institutions to
adopt next-generation AI
innovations.
7.2.2 Other Emerging Research Priorities:
While this study develops a conceptual framework for ethical AI adoption in HE, its empirical
validation remains a critical next step. Future research should focus on longitudinal assessment,
51 of 63
ethical compliance, comparative analysis, and iterative refinement to ensure AI’s effectiveness,
adaptability, and scalability across diverse institutional settings.
Longitudinal Studies on AI’s Institutional Impact: Long-term impact studies should track AI
adoption’s influence on student outcomes, institutional governance, and educational equity over a
multi-year period. By comparing pre-adoption and post-adoption performance metrics, these studies
will provide insights into AI’s role in academic innovation, faculty efficiency, and student support.
Empirical Testing of the HD-AIHED Framework: Pilot studies across diverse university settings
should measure the effectiveness of the HD-AIHED framework in AI governance, faculty decision-
making, and student engagement. Testing AI applications in different cultural, regulatory, and
economic contexts will ensure their adaptability and scalability while refining ethical oversight
mechanisms.
Development of AI Bias Auditing Tools: AI-powered decision-making systems in grading,
admissions, and academic feedback must be regularly assessed for bias detection and mitigation.
Research should focus on developing AI fairness auditing tools that ensure transparency,
accountability, and ethical compliance with standards such as UNESCO AI Principles and GDPR
regulations.
Exploring AI’s Role in Faculty Governance and Research Integrity: AI’s integration into peer
review, faculty recruitment, and academic fraud detection requires further study to ensure human
oversight and ethical accountability. Research should evaluate how AI can enhance academic
quality assurance while preventing unintended biases or manipulation in research evaluation and
faculty assessments.
Advancing AI-Powered Student Support Systems: Future studies should examine AI’s role in student
well-being, including AI-driven academic advising, mental health chatbots, and personalized
learning platforms. Evaluating AI’s impact on student retention, emotional intelligence, and
inclusivity will ensure AI complements human educators rather than replacing critical human
interactions.
Cross-Institutional Comparative Analysis and AI Refinement: AI adoption patterns across different
academic, geographical, and socio-economic contexts should be analyzed to identify best practices
and governance models for optimizing AI’s role in HE worldwide. A real-time feedback-driven
refinement model will enable AI governance frameworks to evolve iteratively, ensuring
interdisciplinary collaboration, institutional adaptability, and strategic AI expansion.
Through empirical, data-driven research, these future directions will strengthen AI integration in
HE, ensuring alignment with institutional goals, ethical governance, and the evolving academic
landscape while fostering a human-centered, transparent, and sustainable AI ecosystem.
7.3 AIHED Research Milestones And Accomplishments
This study successfully meets its research objectives by conducting a comprehensive review of
literature, case studies, and global reports, critically analyzing existing research to identify key gaps
in AIED frameworks, particularly in the domains of ethics, inclusivity, and human agency. This
foundational analysis supports the development of a comprehensive and human-centered
framework, ensuring AI’s responsible integration into higher education.
This directly addresses Research Question 1 (RQ1) by identifying deficiencies in existing
AIHED integration frameworks and proposing evidence-based strategies to address these
gaps. Through a comprehensive review of literature, case studies, and global reports, the
study evaluates key shortcomings in ethics, inclusivity, and human agency, laying the
foundation for a human-centered AI framework that ensures responsible AI integration into
higher education.
52 of 63
Proposing Solutions for Ethical and Inclusive AI: The research presents strategies for
responsible AI adoption, emphasizing transparency, fairness, and inclusivity in AI-driven
educational environments.
Laying the Foundation for a Human-Centered AI Framework: The findings support the
development of a comprehensive AIHED model that prioritizes ethical governance, human
collaboration, and institutional alignment.
Addressing AI’s Dual Role in Higher Education: The research examines AI as both a
transformative driver and a collaborative tool, highlighting global opportunities and
challenges associated with AI deployment. This analysis directly addresses Research
Question 2 (RQ2) by balancing technological advancements with human-centric values,
ensuring that AI enhances, rather than replaces, human agency in educational settings.
Development of the Human-Driven AIHED (HD-AIHED) Framework: Integrating core
principles of accountability, adaptability, and empathy, this framework ensures AI adoption
remains ethical and human-centric. This aligns with Research Question 3 (RQ3) by fostering
a balanced, inclusive, and responsible AI implementation strategy in higher education.
Evaluating Framework Operability in Real-Time Global Contexts: The study assesses AI
deployment across diverse educational settings, directly responding to Research Question 4
(RQ4). By incorporating feedback mechanisms and adaptive design, the framework remains
scalable, relevant, and accessible, even in resource-constrained regions, promoting equitable
AI integration.
Promoting Interdisciplinary Collaboration and Participatory Co-Design: The research
provides actionable insights for ethical AI adoption, directly contributing to Research
Question 5 (RQ5). These insights foster trust, inclusivity, and alignment with institutional
and educational values, ensuring AI applications are developed and implemented with a
strong ethical foundation.
Strategic Recommendations for Key Stakeholders: Finally, the study meets its objective of
offering strategic recommendations for educators, policymakers, and technology
developers, addressing Research Question 6 (RQ6). These recommendations ensure
effective AI integration while mitigating challenges related to equity, accountability, and
cultural sensitivity, fostering a more sustainable and ethical AI-driven educational
ecosystem.
8. Conclusions
This study provides a critical analysis of Artificial Intelligence in Higher Education (AIHED),
highlighting its dual role as both a transformative driver and a collaborative tool. The proposed
Human-Driven AIHED (HD-AIHED) framework establishes a structured and ethical roadmap for
AI adoption, ensuring that AI integration in higher education remains human-centered, ethically
sound, and strategically aligned with institutional objectives. By positioning AI as a facilitator rather
than a replacement for human intelligence, the framework preserves the core values of
accountability, adaptability, and inclusivity, reinforcing ethical integrity, transparency, and
accessibility. The model underscores that the future of higher education does not rest in AI’s
capabilities alone but in human intelligence guiding AI towards its highest potential.
The HD-AIHED framework's applicability spans both the AI life cycle and the student life cycle,
ensuring a dual-layered synergy that integrates Human Intelligence at each phase of AI development
and deployment. Through its phased approach—adoption, design, deployment, and exploration—
the framework ensures AI remains context-sensitive, adaptable, and responsive to evolving higher
education landscapes. By mapping global real-time AI challenges, the HD-AIHED model provides
actionable insights on how institutions can leverage AI-driven models without compromising
human-centric values, thus ensuring scalability, sustainability, and cross-border collaboration in AI
governance.
The HD-AIHED framework serves as a strategic enabler, leveraging AI’s potential to enhance
personalized learning, optimize administrative processes, and strengthen institutional decision-
making while embedding a strong human-centered ethical foundation. In developed nations, the
53 of 63
framework supports institutional competitiveness, workforce readiness, and advanced research
capabilities by integrating governance models, ethical safeguards, and continuous feedback
mechanisms into AI implementation. In developing regions, the framework plays a democratizing
role, fostering scalable, cost-effective AI solutions that bridge digital and infrastructural gaps,
ensuring equitable access to AI-driven education. Recognizing the risks associated with AI
adoption, the framework proactively addresses bias, data privacy, algorithmic transparency, and
student agency, reinforcing human oversight and ethical compliance as fundamental pillars of
responsible AI integration.
To strengthen ethical AI governance, institutions must establish dedicated AI governance boards to
oversee bias mitigation, privacy protection, and regulatory compliance. Aligning AI policies with
UNESCO AI ethics guidelines, GDPR, and Sustainable Development Goal 4 (SDG 4) ensures that
AI adoption remains globally relevant, ethically responsible, and institutionally aligned. The HD-
AIHED framework emphasizes that AI applications must be customized to institutional goals,
ensuring AI enhances adaptive learning, predictive analytics, and administrative efficiency, while
placing students at the center of AI integration.
Stakeholder engagement plays a crucial role in AI governance, requiring participation from internal
and external stakeholders, including faculty, students, policymakers, industry partners, and
regulatory bodies. A participatory decision-making approach combined with dynamic feedback
mechanisms fosters AI systems that are responsive, inclusive, and aligned with institutional values.
To ensure strategic alignment, ethical compliance, and institutional readiness, the implementation
of SWOC (Strengths, Weaknesses, Opportunities, and Challenges) analysis at both the adoption
phase and the AI exploration phase is critical. This structured evaluation allows institutions to
identify risks, leverage opportunities, and establish a sustainable AI adoption strategy that is
scalable, adaptable, and future-ready. Furthermore, AI tools must be culturally and linguistically
adaptable, ensuring inclusivity and diversity in educational environments.
In addition, the promotion of AIHED is imperative for institutions to learn from both success stories
and failures, enabling them to refine AI strategies, mitigate risks, and enhance best practices. By
fostering knowledge-sharing and institutional collaboration, AI integration can be continuously
improved, ensuring its long-term impact and ethical alignment in higher education.
Bridging the AI readiness gap in higher education necessitates significant investment in AI literacy
programs, equipping faculty, administrators, and students with the skills necessary to navigate AI’s
evolving role. Institutions in low-resource settings should develop collaborative partnerships to
build AI infrastructure and implement capacity-building initiatives that foster sustainable AI
adoption. Additionally, real-time feedback mechanisms must be embedded in AI models, ensuring
continuous assessment, refinement, and alignment with institutional and policy shifts. The iterative
testing and improvement of AI applications will help institutions remain responsive to emerging
technological advancements and evolving educational needs.
Future research should prioritize the empirical validation of the HD-AIHED framework through
pilot implementations, longitudinal assessments, and cross-institutional comparative studies.
Testing the framework in real-world educational settings will provide insights into its scalability,
effectiveness, and governance impact. Multi-institutional pilot programs should be initiated to
evaluate AI’s influence on student engagement, institutional decision-making, and ethical AI
adoption.
The development of a standardized AIHED governance roadmap is essential to define best
practices, compliance mechanisms, and long-term monitoring strategies for responsible AI
integration in higher education. Research must also focus on designing AI bias auditing tools to
detect and mitigate algorithmic bias, ensuring fairness, accountability, and inclusivity in AI-driven
decision-making.
54 of 63
Additionally, exploring regional AIHED adaptation strategies is crucial for understanding how
policy frameworks, cultural dynamics, and infrastructure disparities shape AI adoption across
diverse educational landscapes. Identifying context-sensitive governance models will help
institutions develop adaptive AI policies that align with global ethical standards while addressing
local challenges.
By adopting the HD-AIHED model proposed in this study, universities and higher education
institutions can harness AI’s transformative power not just as a ‘tool’, but as a core ‘infrastructure’
that upholds ethical integrity and human-centered progress. This Human-Driven AIHED
Framework ensures that AI serves as a catalyst for empowerment, rather than exclusion, fostering
a future where technology and human expertise collaborate to advance equitable and sustainable
education.
Ultimately, the HD-AIHED framework envisions AI as an inclusive, ethical, and transformative
force in higher education rather than a disruptive replacement for human intelligence. By fostering
collaboration between governments, institutions, industry stakeholders, and students, the
framework ensures that AI integration remains human-driven, ethically aligned, and strategically
positioned to enhance global education. As AI continues to redefine higher education, its success
will depend on our collective ability to balance technological progress with ethical imperatives,
ensuring that AI in HE remains a beacon of knowledge, accessibility, and integrity for future
generations.
55 of 63
References
1. Tegmark M. Life 3.0: Being human in the age of artificial intelligence. Vintage; 2018.
2. Tegmark M. Being human in the age of artificial intelligence. 2019.
3. Brynjolfsson BYE, McAfee A. Artificial intelligence for real. Harv Bus Rev. 2017;July: 1–31. Available:
https://starlab-alliance.com/wp-content/uploads/2017/09/AI-Article.pdf
4. Luckin R, Holmes W. Intelligence Unleashed: An argument for AI in Education. UCL Knowledge Lab:
London, UK. . 2016. Available: https://www.pearson.com/content/dam/corporate/global/pearson-dot-
com/files/innovation/Intelligence-Unleashed-Publication.pdf
5. Holmes W, Bialik M, Fadel C. Artificial intelligence in education promises and implications for teaching and
learning. Center for Curriculum Redesign; 2019.
6. Zawacki-Richter O, Marín VI, Bond M, Gouverneur F. Systematic review of research on artificial intelligence
applications in higher education–where are the educators? Int J Educ Technol High Educ. 2019;16: 1–27.
7. Pedro F, Subosa M, Rivas A, Valverde P. Artificial intelligence in education: Challenges and opportunities
for sustainable development. 2019.
8. Floridi L, Cowls J. A unified framework of five principles for AI in society. Mach Learn city Appl Archit
urban Des. 2022; 535–545.
9. Wu C, Zhang H, Carroll JM. AI Governance in Higher Education: Case Studies of Guidance at Big Ten
Universities. Futur Internet. 2024;16. doi:10.3390/fi16100354
10. Bond M, Khosravi H, De Laat M, Bergdahl N, Negrea V, Oxley E, et al. A meta systematic review of
artificial intelligence in higher education: a call for increased ethics, collaboration, and rigour. Int J Educ
Technol High Educ. 2024;21. doi:10.1186/s41239-023-00436-z
11. Crompton H, Burke D. Artificial intelligence in higher education: the state of the field. Int J Educ Technol
High Educ. 2023;20. doi:10.1186/s41239-023-00392-8
12. Lin CC, Huang AYQ, Lu OHT. Artificial intelligence in intelligent tutoring systems toward sustainable
education: a systematic review. Smart Learn Environ. 2023;10. doi:10.1186/s40561-023-00260-y
13. Oyebola Olusola Ayeni, Nancy Mohd Al Hamad, Onyebuchi Nneamaka Chisom, Blessing Osawaru, Ololade
Elizabeth Adewusi. AI in education: A review of personalized learning and educational technology. GSC Adv
Res Rev. 2024;18: 261–271. doi:10.30574/gscarr.2024.18.2.0062
14. Baker RS. Challenges for the future of educational data mining: The Baker learning analytics prizes. J Educ
data Min. 2019;11: 1–17.
15. Wang T, Lund BD, Marengo A, Pagano A, Mannuru NR, Teel ZA, et al. Exploring the Potential Impact of
Artificial Intelligence (AI) on International Students in Higher Education: Generative AI, Chatbots, Analytics,
and International Student Success. Appl Sci. 2023;13. doi:10.3390/app13116716
16. PwC. Sizing the prize: What’s the real value of AI for your business and how can you capitalise? PwC; 2019.
Available: https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf
17. Alotaibi NS. The Impact of AI and LMS Integration on the Future of Higher Education: Opportunities,
Challenges, and Strategies for Transformation. Sustain. 2024;16. doi:10.3390/su162310357
18. Lytras MD, Serban AC, Alkhaldi A, Aldosemani T, Malik S. What’s Next in Higher Education: The AI
Revolution 2030. In: Lytras MD, Serban AC, Alkhaldi A, Malik S, Aldosemani T, editors. Digital
Transformation in Higher Education, Part A. Emerald Publishing Limited; 2024. pp. 155–172.
doi:10.1108/978-1-83549-480-620241008
19. Digital Education Council. Digital Education Council Global AI Student Survey 2024 AI or Not AI : What
Students Want. 2024. Available: https://www.digitaleducationcouncil.com/post/digital-education-council-
global-ai-student-survey-2024
20. Memarian B, Doleck T. Fairness, Accountability, Transparency, and Ethics (FATE) in Artificial Intelligence
(AI) and higher education: A systematic review. Comput Educ Artif Intell. 2023;5: 100152.
doi:10.1016/j.caeai.2023.100152
21. Khatri BB, Karki PD. Artificial Intelligence (AI) in Higher Education: Growing Academic Integrity and
Ethical Concerns. Nepal J Dev Rural Stud. 2023;20: 1–7. doi:10.3126/njdrs.v20i01.64134
22. Batista J, Mesquita A, Carnaz G. Generative AI and Higher Education: Trends, Challenges, and Future
Directions from a Systematic Literature Review. Inf. 2024;15. doi:10.3390/info15110676
23. Airaj M. Ethical artificial intelligence for teaching-learning in higher education. Educ Inf Technol. 2024;29:
17145–17167. doi:10.1007/s10639-024-12545-x
24. Al-Zahrani AM, Alasmari TM. Exploring the impact of artificial intelligence on higher education: The
dynamics of ethical, social, and educational implications. Humanit Soc Sci Commun. 2024;11: 1–12.
doi:10.1057/s41599-024-03432-4
56 of 63
25. Alfredo R, Echeverria V, Jin Y, Yan L, Swiecki Z, Gašević D, et al. Human-centred learning analytics and AI
in education: A systematic literature review. Comput Educ Artif Intell. 2024;6.
doi:10.1016/j.caeai.2024.100215
26. Holmes W, Miao F. Guidance for generative AI in education and research. UNESCO Publishing; 2023.
Available: https://unesdoc.unesco.org/ark:/48223/pf0000386693.locale=en
27. Maslej N, Fattorini L, Brynjolfsson E, Etchemendy J, Ligett K, Lyons T, et al. Artificial intelligence index
report 2023. arXiv Prepr arXiv231003715. 2023.
28. Hamid OH. Impact of Artificial Intelligence (AI) in Addressing Students at-Risk Challenges in Higher
Education (HE). The Evolution of Artificial Intelligence in Higher Education: Challenges, Risks, and Ethical
Considerations. Emerald Publishing Limited; 2024. pp. 183–194.
29. Stanford University. AI Index Report 2023: Measuring trends in Artificial Intelligence. AI Index Rep. 2023.
Available: https://aiindex.stanford.edu/report/
30. Marengo A, Pagano A, Pange J, Soomro KA. The educational value of artificial intelligence in higher
education: a 10-year systematic literature review. Interact Technol Smart Educ. 2024.
31. UNESCO. Recommendation on the Ethics of Artificial Intelligence. Paris: UNESCO; 2021.
32. Sambasivan N, Kapania S, Highfill H, Akrong D, Paritosh P, Aroyo LM. “Everyone wants to do the model
work, not the data work”: Data Cascades in High-Stakes AI. Proceedings of the 2021 CHI Conference on
Human Factors in Computing Systems. New York, NY, USA: Association for Computing Machinery; 2021.
doi:10.1145/3411764.3445518
33. Ozmen Garibay O, Winslow B, Andolina S, Antona M, Bodenschatz A, Coursaris C, et al. Six Human-
Centered Artificial Intelligence Grand Challenges. Int J Hum Comput Interact. 2023;39: 391–437.
doi:10.1080/10447318.2022.2153320
34. Rakowski R, Polak P, Kowalikova P. Ethical Aspects of the Impact of AI: the Status of Humans in the Era of
Artificial Intelligence. Society. 2021;58: 196–203. doi:10.1007/s12115-021-00586-8
35. Postman N. Technopoly: The surrender of culture to technology. Vintage; 2011.
36. Castañeda L, Selwyn N. More than tools? Making sense of the ongoing digitizations of higher education. Int J
Educ Technol High Educ. 2018;15. doi:10.1186/s41239-018-0109-y
37. Mustafa MY, Tlili A, Lampropoulos G, Huang R, Jandrić P, Zhao J, et al. A systematic review of literature
reviews on artificial intelligence in education (AIED): a roadmap to a future research agenda. Smart Learning
Environments. Springer Nature Singapore; 2024. doi:10.1186/s40561-024-00350-5
38. Alqahtani N, Wafula Z. Artificial Intelligence Integration: Pedagogical Strategies and Policies at Leading
Universities. Innov High Educ. 2024. doi:10.1007/s10755-024-09749-x
39. Crawford J, Allen KA, Pani B, Cowling M. When artificial intelligence substitutes humans in higher
education: the cost of loneliness, student success, and retention. Stud High Educ. 2024;49: 883–897.
doi:10.1080/03075079.2024.2326956
40. Holmes W, Porayska-Pomsta K, Holstein K, Sutherland E, Baker T, Shum SB, et al. Ethics of AI in
Education: Towards a Community-Wide Framework. Int J Artif Intell Educ. 2022;32: 504–526.
doi:10.1007/s40593-021-00239-1
41. Alexander B, Ashford-Rowe K, Barajas-Murphy N, Dobbin G, Knott J, McCormack M, et al. Educause
Horizon report: 2019 Higher Education edition. 2019.
42. Heidegger M. The question concerning technology. 1977.
43. Heidegger M. The question concerning technology. Readings Philos Technol. 2009; 9–24.
44. Kamali J, Alpat MF, Bozkurt A. AI ethics as a complex and multifaceted challenge: decoding educators’ AI
ethics alignment through the lens of activity theory. Int J Educ Technol High Educ. 2024;21: 1–20.
45. Alshahrani BT, Pileggi SF, Karimi F. A Social Perspective on AI in the Higher Education System: A
Semisystematic Literature Review. Electron. 2024;13: 1–27. doi:10.3390/electronics13081572
46. Walsh D, Downe S. Meta‐synthesis method for qualitative research: a literature review. J Adv Nurs. 2005;50:
204–211.
47. Finfgeld-Connett D. Introduction To Theorygenerating Meta- Synthesis Research. A Guid to Qual Meta-
Synthesis. 2018; 1–12. doi:10.4324/9781351212793-1
48. Heaton J. Secondary analysis of qualitative data: An overview. Hist Soc Res Sozialforsch. 2008; 33–45.
49. Zimmer L. Qualitative meta‐synthesis: a question of dialoguing with texts. J Adv Nurs. 2006;53: 311–318.
50. OECD. OECD principles on artificial intelligence. OECD Paris, France; 2019.
51. IEEE. Ethics of Autonomous and Intelligent Systems. New York: IEEE; 2020.
52. Davis FD, Bagozzi RP, Warshaw PR. Technology acceptance model. J Manag Sci. 1989;35: 982–1003.
53. Rogers EM. Diffusion of innovations, 5th edn London. UK Free Press. 2003.
57 of 63
54. Venkatesh V, Bala H. Technology acceptance model 3 and a research agenda on interventions. Decis Sci.
2008;39: 273–315.
55. Trow M. Reflections on the transition from elite to mass to universal access: Forms and phases of higher
education in modern societies since WWII. International handbook of higher education. Springer; 2007. pp.
243–280.
56. Bourdieu P. The forms of capital.(1986). Cult theory An Anthol. 2011;1: 949.
57. Sandelowski M, Barroso J. Handbook for synthesizing qualitative research. springer publishing company;
2006.
58. Noblit GW. Meta-ethnography: Synthesizing qualitative studies. Stage Publ. 1988.
59. Hannes K, Lockwood C. Synthesizing qualitative research: choosing the right approach. John Wiley & Sons;
2011.
60. Kordzadeh N, Ghasemaghaei M. Algorithmic bias: review, synthesis, and future research directions. Eur J Inf
Syst. 2022;31: 388–409.
61. Artyukhov A, Wołowiec T, Artyukhova N, Bogacki S, Vasylieva T. SDG 4, Academic Integrity and Artificial
Intelligence: Clash or Win-Win Cooperation? Sustainability. 2024;16: 8483.
62. Heintz F. ARTIFICIAL INTELLIGENCE DUALITY. Comput Think Educ K-12 Artif Intell Lit Phys
Comput. 2022;143.
63. NIST. AI Risk Management Framework. Gaithersburg: National Institute of Standards and Technology;
2023.
64. Thomas J, Harden A. Methods for the thematic synthesis of qualitative research in systematic reviews. BMC
Med Res Methodol. 2008;8: 1–10.
65. Leshem S, Trafford V. Overlooking the conceptual framework. Innov Educ Teach Int. 2007;44: 93–105.
66. Ahmad SF, Alam MM, Rahmat MK, Mubarik MS, Hyder SI. Academic and Administrative Role of Artificial
Intelligence in Education. Sustain. 2022;14: 1–11. doi:10.3390/su14031101
67. Hu S, Ke F, Vyortkina D, Hu P, Luby S, O’Shea J. Artificial Intelligence in Higher Education: Applications,
Challenges, and Policy Development and Further Considerations BT - Higher Education: Handbook of
Theory and Research: Volume 40. In: Perna LW, editor. Cham: Springer Nature Switzerland; 2025. pp. 1–52.
doi:10.1007/978-3-031-51930-7_13-1
68. Baker RS, Hawn A. Algorithmic bias in education. Int J Artif Intell Educ. 2022; 1–41.
69. Singh N. Keeping artificial intelligence real: The importance of AI governance. McKinsey & Company;
2024. Available: https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/keeping-artificial-
intelligence-real
70. Nguyen A, Hong Y, Dang B, Huang X. Human-AI collaboration patterns in AI-assisted academic writing.
Stud High Educ. 2024;49: 847–864. doi:10.1080/03075079.2024.2323593
71. Dobrev D. A Definition of Artificial Intelligence. 2012; 1–7. Available: http://arxiv.org/abs/1210.1568
72. Russell SJ, Norvig P. Artificial intelligence: a modern approach. Pearson; 2016.
73. Kaplan A, Haenlein M. Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations,
illustrations, and implications of artificial intelligence. Bus Horiz. 2019;62: 15–25.
doi:10.1016/j.bushor.2018.08.004
74. Roll I, Wylie R. Evolution and Revolution in Artificial Intelligence in Education. Int J Artif Intell Educ.
2016;26: 582–599. doi:10.1007/s40593-016-0110-3
75. Chen X, Xie H, Zou D, Hwang GJ. Application and theory gaps during the rise of Artificial Intelligence in
Education. Comput Educ Artif Intell. 2020;1: 100002. doi:10.1016/j.caeai.2020.100002
76. Mohammed AT, Velander J, Milrad M. A Retrospective Analysis of Artificial Intelligence in Education (
AIEd ) Studies : Perspectives , Learning Theories , Challenges , and Emerging Opportunities. Springer Nature
Singapore; 2024. doi:10.1007/978-981-97-8638-1
77. Claudio S, Sirois J, Owen K, Marya V, Krishnan C, Brasca C. How technology is shaping learning in higher
education. McKinsey Co. 2022; 1–9. Available: https://www.mckinsey.com/industries/education/our-
insights/how-technology-is-shaping-learning-in-higher-education
78. Agarwal P, Swami S, Malhotra SK. Artificial Intelligence Adoption in the Post COVID-19 New-Normal and
Role of Smart Technologies in Transforming Business: a Review. J Sci Technol Policy Manag. 2024;15:
506–529. doi:10.1108/JSTPM-08-2021-0122
79. Luckin R, Holmes W. Intelligence unleashed: An argument for AI in education. 2016.
80. Perrault R, Clark J. Artificial Intelligence Index Report 2024. 2024.
81. Grand View Research. Artificial Intelligence (AI) in Education Market Report. 2025. Available:
https://www.grandviewresearch.com/industry-analysis/artificial-intelligence-ai-education-market-report
58 of 63
82. Anon. Pearson 2024 End of Year AI Report for Higher Education. Pearson; 2024.
83. Marques-Cobeta N. Artificial Intelligence in Education: Unveiling Opportunities and Challenges BT -
Innovation and Technologies for the Digital Transformation of Education: European and Latin American
Perspectives. In: García-Peñalvo FJ, Sein-Echaluce ML, Fidalgo-Blanco Á, editors. Singapore: Springer
Nature Singapore; 2024. pp. 33–42. doi:10.1007/978-981-97-2468-0_4
84. Luckin R, Holmes W, Griffiths M, Forcier LB. Artificial intelligence in education: Opportunities and
challenges for a new era. Educ Technol. 2021;61: 16–22.
85. Holstein K, Aleven V, Rummel N. A conceptual framework for human–AI hybrid adaptivity in education.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture
Notes in Bioinformatics). Springer International Publishing; 2020. doi:10.1007/978-3-030-52237-7_20
86. George B, Wooden O. Managing the Strategic Transformation of Higher Education through Artificial
Intelligence. Adm Sci. 2023;13. doi:10.3390/admsci13090196
87. Spivakovsky O V., Omelchuk SA, Kobets V V., Valko N V., Malchykova DS. Institutional Policies on
Artificial Intelligence in University Learning, Teaching and Research. Inf Technol Learn Tools. 2023;97:
181–202. doi:10.33407/itlt.v97i5.5395
88. Chan CKY. A comprehensive AI policy education framework for university teaching and learning. Int J Educ
Technol High Educ. 2023;20: 38.
89. Barnett R. Being a university. Routledge; 2010.
90. Zhang D, Mishra S, Brynjolfsson E, Etchemendy J, Ganguli D, Grosz B, et al. 2021 AI Index Report. AI
Index Steer Committee, Stanford Univ Human-Centered Artif Intell Institute, Stanford Univ. 2021; 1–222.
Available: https://aiindex.stanford.edu/report/
91. Rogers EM, Singhal A, Quinlan MM. Diffusion of innovations. An integrated approach to communication
theory and research. Routledge; 2014. pp. 432–448.
92. Chapman M. Constructive evolution: Origins and development of Piaget’s thought. Cambridge University
Press; 1988.
93. Bruner J. Humon Development Celebrating Divergence: Piaget and Vygotsky1. Hum Dev. 1997;40: 63–73.
94. Zhang D, Mishra S, Brynjolfsson E, Etchemendy J, Ganguli D, Grosz B, et al. The AI index 2021 annual
report. arXiv Prepr arXiv210306312. 2021.
95. Urmeneta A, Romero M. Creative Applications of Artificial Intelligence in Education. Springer Nature; 2024.
96. Arsen’eva N V, Putyatina LM, Tarasova N V, Tikhonov G V. Advantages and Disadvantages of Using
Artificial Intelligence in Higher Education. Russ Eng Res. 2024;44: 1687–1690.
doi:10.3103/S1068798X24702794
97. Osorio C, Fuster N, Chen W, Men Y, Juan AA. Enhancing accessibility to analytics courses in higher
education through AI, simulation, and e-collaborative tools. Information. 2024;15: 430.
98. Simeonov S, Feradov F, Marinov A, Abu-Alam T. Integration of AI Training in the Field of Higher
Education in the Republic of Bulgaria: An Overview. Educ Sci. 2024;14. doi:10.3390/educsci14101063
99. Boustani NM, Sidani D, Boustany Z. Leveraging ICT and Generative AI in Higher Education for Sustainable
Development: The Case of a Lebanese Private University. Adm Sci. 2024;14. doi:10.3390/admsci14100251
100. Pacheco-Mendoza S, Guevara C, Mayorga-Albán A, Fernández-Escobar J. Artificial Intelligence in Higher
Education: A Predictive Model for Academic Performance. Educ Sci. 2023;13. doi:10.3390/educsci13100990
101. Sinha RK, Deb Roy A, Kumar N, Mondal H. Applicability of ChatGPT in Assisting to Solve Higher Order
Problems in Pathology. Cureus. 2023;15. doi:10.7759/cureus.35237
102. Ade-Ibijola A, Okonkwo C. Artificial Intelligence in Africa: Emerging Challenges. Responsible AI in Africa:
Challenges and Opportunities. Springer; 2023. pp. 101–117. Available:
https://link.springer.com/chapter/10.1007/978-3-031-08215-3_5#citeas
103. Wang S, Wang F, Zhu Z, Wang J, Tran T, Du Z. Artificial intelligence in education: A systematic literature
review. Expert Syst Appl. 2024;252: 124167. doi:https://doi.org/10.1016/j.eswa.2024.124167
104. Kerimbayev N, Adamova K, Shadiev R, Altinay Z. Intelligent educational technologies in individual
learning: a systematic literature review. Smart Learn Environ. 2025;12: 1–30. doi:10.1186/s40561-024-
00360-3
105. Sova R, Tudor C, Tartavulea CV, Dieaconescu RI. Artificial Intelligence Tool Adoption in Higher Education:
A Structural Equation Modeling Approach to Understanding Impact Factors among Economics Students.
Electronics. 2024;13: 3632.
106. Mahmood NZ, Ahmed SR, Al-Hayaly AF, Algburi S, Rasheed J. The Evolution of Administrative
Information Systems: Assessing the Revolutionary Impact of Artificial Intelligence. 2023 7th International
Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT). 2023. pp. 1–7.
doi:10.1109/ISMSIT58785.2023.10304973
59 of 63
107. Automation Anywhere. University of Melbourne saves 10,000 hours annually with Automation Anywhere.
2024. Available: https://www.automationanywhere.com/resources/customer-stories/university-of-melbourne-
aau
108. Zong Z, Guan Y. AI-Driven Intelligent Data Analytics and Predictive Analysis in Industry 4.0: Transforming
Knowledge, Innovation, and Efficiency. J Knowl Econ. 2024; 1–40.
109. Mariani MM, Machado I, Magrelli V, Dwivedi YK. Artificial intelligence in innovation research: A
systematic review, conceptual framework, and future research directions. Technovation. 2023;122: 102623.
doi:10.1016/j.technovation.2022.102623
110. Ferk Savec V, Jedrinović S. The Role of AI Implementation in Higher Education in Achieving the
Sustainable Development Goals: A Case Study from Slovenia. Sustain. 2025;17: 1–22.
doi:10.3390/su17010183
111. Funda V, Piderit R. A review of the application of artificial intelligence in South African Higher Education.
2024 Conf Inf Commun Technol Soc ICTAS 2024 - Proc. 2024; 44–50.
doi:10.1109/ICTAS59620.2024.10507113
112. Roscoe RD, Salehi S, Nixon N, Worsley M, Piech C, Luckin R. Inclusion and equity as a paradigm shift for
artificial intelligence in education. Artificial Intelligence in STEM Education. CRC Press; 2022. pp. 359–374.
113. Gemiharto I, CMS S. Inclusivity and Accessibility in Digital Communication Tools: Case Study of AI-
Enhanced Platforms in INDONESIA. J Pewarta Indones. 2024;6: 78–88. doi:10.25008/jpi.v6i1.154
114. Schwab K. The fourth industrial revolution. Crown Currency; 2017.
115. Sajja R, Sermet Y, Cikmaz M, Cwiertny D, Demir I. Artificial Intelligence-Enabled Intelligent Assistant for
Personalized and Adaptive Learning in Higher Education. 2023; 1–23. doi:10.3390/info15100596
116. Mohamed Hashim MA, Tlemsani I, Matthews R. Higher education strategy in digital transformation. Educ
Inf Technol. 2022;27: 3171–3195. doi:10.1007/s10639-021-10739-1
117. Knox J. Artificial intelligence and education in China. Learn Media Technol. 2020;45: 298–311.
doi:10.1080/17439884.2020.1754236
118. Kalyanakrishnan S, Panicker RA, Natarajan S, Rao S. Opportunities and Challenges for Artificial Intelligence
in India. Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society. New York, NY, USA:
Association for Computing Machinery; 2018. pp. 164–170. doi:10.1145/3278721.3278738
119. Sharma S, Singh G, Sharma CS, Kapoor S. Artificial intelligence in Indian higher education institutions: a
quantitative study on adoption and perceptions. Int J Syst Assur Eng Manag. 2024. doi:10.1007/s13198-023-
02193-8
120. World Economic Forum. State of the education report for India, 2022: artificial intelligence in education;
here, there and everywhere. World Economic Forum; 2022. Available:
https://unesdoc.unesco.org/ark:/48223/pf0000382661
121. Mumtaz S, Carmichael J, Weiss M, Nimon-Peters A. Ethical use of artificial intelligence based tools in higher
education: are future business leaders ready? Educ Inf Technol. 2024. doi:10.1007/s10639-024-13099-8
122. O’Dea X. Generative AI: is it a paradigm shift for higher education? Stud High Educ. 2024;49: 811–816.
doi:10.1080/03075079.2024.2332944
123. Salas-Pilco SZ, Yang Y. Artificial intelligence applications in Latin American higher education: a systematic
review. Int J Educ Technol High Educ. 2022;19: 21. doi:10.1186/s41239-022-00326-w
124. Rios-Campos C, Zambrano EOG, Cantos MFM, Anchundia-Gómez O, León MEC, Moya GEM, et al.
Universities and Artificial Intelligence. South Florida J Dev. 2024;5: e4016. doi:10.46932/sfjdv5n6-010
125. Thagard P. The Ethics of Artificial Intelligence. Bots Beasts. 2021; 225–248.
doi:10.7551/mitpress/14102.003.0010
126. UNESCO. Recommendation on the Ethics of AI. Unesco. 2021; 1–21. Available:
https://unesdoc.unesco.org/ark:/48223/pf0000380455
127. NASSCOM. Responsible AI: Principles and Practices. New Delhi: National Association of Software and
Service Companies; 2021.
128. Aayog N. National Strategy for Artificial Intelligence. New Delhi: NITI Aayog; 2018.
129. U.S. Department of Education. Guidelines for AI in Education. Washington, D.C.: Department of Education;
2021.
130. Katsamakas E, Pavlov O V, Saklad R. Artificial intelligence and the transformation of higher education
institutions: A systems approach. Sustainability. 2024;16: 6118.
131. European Commission. Ethics Guidelines for Trustworthy AI. Brussels: European Commission; 2019.
132. European Commission. EU Artificial Intelligence Act. Brussels: European Commission; 2021.
133. GPAI. Ethical AI in Practice. Montreal: Global Partnership on AI; 2020.
60 of 63
134. Silberg J, Manyika J. Notes from the AI frontier : Tackling bias in AI ( and in humans ). Mckinsey Glob Inst.
2019; 1–8. Available: https://www.mckinsey.com/~/media/mckinsey/featured insights/artificial
intelligence/tackling bias in artificial intelligence and in humans/mgi-tackling-bias-in-ai-june-2019.pdf
135. Prem E. From ethical AI frameworks to tools: a review of approaches. AI Ethics. 2023;3: 699–716.
doi:10.1007/s43681-023-00258-9
136. Pedro F, Subosa M, Rivas A, Valverde P. Artificial intelligence in education: Challenges and opportunities
for sustainable development. 2019.
137. Kaplan A, Haenlein M. Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations,
illustrations, and implications of artificial intelligence. Bus Horiz. 2019;62: 15–25.
doi:10.1016/j.bushor.2018.08.004
138. Essien A, Bukoye OT, O’Dea X, Kremantzis M. The influence of AI text generators on critical thinking skills
in UK business schools. Stud High Educ. 2024;49: 865–882. doi:10.1080/03075079.2024.2316881
139. Pinto M, Leite C. Digital technologies in support of students learning in higher education: Literature review.
Digit Educ Rev. 2020; 343–360. doi:10.1344/DER.2020.37.343-360
140. Hunt P, Leijen Ä, van der Schaaf M. Automated feedback is nice and human presence makes it better:
Teachers’ perceptions of feedback by means of an e-portfolio enhanced with learning analytics. Educ Sci.
2021;11. doi:10.3390/educsci11060278
141. UNESCO. Foundation models such as ChatGPT through the prism of the UNESCO Recommendation on the
Ethics of Artificial Intelligence. 2023; 13.
142. Escotet MÁ. The optimistic future of Artificial Intelligence in higher education. Prospects. 2023.
doi:10.1007/s11125-023-09642-z
143. Gligorea I, Cioca M, Oancea R, Gorski A-T, Gorski H, Tudorache P. Adaptive learning using artificial
intelligence in e-learning: a literature review. Educ Sci. 2023;13: 1216.
144. Stanford University. Introduction to the AI Index Report 2024. 2024.
145. Alordiah CO. Proliferation of Artificial Intelligence Tools: Adaptation Strategies in the Higher Education
Sector. Propellers J Educ. 2023;2: 53–65. Available: https://ijvocter.com/pjed/article/view/68
146. Correa-Peralta M, Vinueza-Martínez J, Castillo-Heredia L. Evolution, topics and relevant research
methodologies in Business Intelligence and Data Analysis in the Academic Management of Higher Education
Institutions. A literature review. Results Eng. 2024; 103782. doi:https://doi.org/10.1016/j.rineng.2024.103782
147. Sharma S, Singh G. Adoption of artificial intelligence in higher education: an empirical study of the UTAUT
model in Indian universities. Int J Syst Assur Eng Manag. 2024. doi:10.1007/s13198-024-02558-7
148. Tang PM, Koopman J, Mai KM, De Cremer D, Zhang JH, Reynders P, et al. No person is an island:
Unpacking the work and after-work consequences of interacting with artificial intelligence. J Appl Psychol.
2023.
149. MA W, MA W, HU Y, BI X. The who, why, and how of ai-based chatbots for learning and teaching in higher
education: A systematic review. Educ Inf Technol. 2024. doi:10.1007/s10639-024-13128-6
150. Köbis L, Mehner C. Ethical Questions Raised by AI-Supported Mentoring in Higher Education. Front Artif
Intell. 2021;4: 1–9. doi:10.3389/frai.2021.624050
151. George AS. Artificial Intelligence and the Future of Work: Job Shifting Not Job Loss. Partners Univers Innov
Res Publ. 2024;2: 17–37.
152. Lopez M, Gonzalez I. Artificial Intelligence Is Not Human: The Legal Determination of Inventorship and Co-
Inventorship, the Intellectual Property of AI Inventions, and the Development of Risk Management
Guidelines. J Pat Trademark Off Soc’y. 2024;104: 135.
153. Cremer D De, Koopman J. Using AI at Work Makes Us Lonelier and Less Healthy. In: Havard Business
Review [Internet]. 2024 [cited 26 Dec 2024]. Available: https://hbr.org/2024/06/research-using-ai-at-work-
makes-us-lonelier-and-less-healthy
154. Danaher J, Nyholm S. Automation, work and the achievement gap. AI Ethics. 2021;1: 227–237.
doi:10.1007/s43681-020-00028-x
155. Nemorin S, Vlachidis A, Ayerakwa HM, Andriotis P. AI hyped? A horizon scan of discourse on artificial
intelligence in education (AIED) and development. Learn Media Technol. 2023;48: 38–51.
doi:10.1080/17439884.2022.2095568
156. Palomares I, Martínez-Cámara E, Montes R, García-Moral P, Chiachio M, Chiachio J, et al. A panoramic
view and swot analysis of artificial intelligence for achieving the sustainable development goals by 2030:
progress and prospects. Applied Intelligence. 2021. doi:10.1007/s10489-021-02264-y
157. UNESCO. Challenges and Opportunities for Sustainable Development Education Sector. UNESCO Work
Pap Educ Policy, No 7 Fr Pedró (Ed) . 2019; 1–48. Available: https://en.unesco.org/themes/education-policy-
158. van de Poel I. Embedding Values in Artificial Intelligence (AI) Systems. Minds Mach. 2020;30: 385–409.
61 of 63
doi:10.1007/s11023-020-09537-4
159. Crawford LM. Conceptual and theoretical frameworks in research. Research Design and Methods: An
Applied Guide for the Scholar-Practitioner. 2019. pp. 35–48. Available:
https://uk.sagepub.com/sites/default/files/upm-binaries/105274_ch03_14.pdf
160. Yang SJH, Ogata H, Matsui T, Chen NS. Human-centered artificial intelligence in education: Seeing the
invisible through the visible. Comput Educ Artif Intell. 2021;2: 100008. doi:10.1016/j.caeai.2021.100008
161. Bond RR, Mulvenna M, Wang H. Human centered artificial intelligence: Weaving UX into algorithmic
decision making. RoCHI 2019 Int Conf Human-Computer Interact. 2019; 2–9. Available:
https://pure.ulster.ac.uk/en/publications/human-centered-artificial-intelligence-weaving-ux-into-algorithmi
162. Pedro F, Subosa M, Rivas A, Valverde P. Artificial Intelligence in Education: Challenges and Opportunities
for Sustainable Development Education Sector United Nations Educational, Scientific and Cultural
Organization. Minist Educ. 2019; 1–46. Available: https://en.unesco.org/themes/education-policy-
163. Cowls J, Floridi L. Prolegomena to a white paper on an ethical framework for a good AI society. 2018.
164. Guilbault M. Students as customers in higher education: The (controversial) debate needs to end. J Retail
Consum Serv. 2016; 0–1. doi:10.1016/j.jretconser.2017.03.006
165. Yelkur R. Customer Satisfaction and the Services Marketing Mix. J Prof Serv Mark. 2000;21: 105–115.
doi:10.1300/J090v21n01
166. Boddington P. Towards a Code of Ethics for Artificial Intelligence. Springer; 2017.
167. Cope B, Kalantzis M, Searsmith D. Artificial intelligence for education: Knowledge and its assessment in AI-
enabled learning ecologies. Educ Philos Theory. 2021;53: 1229–1245. doi:10.1080/00131857.2020.1728732
168. Ojha M, kumar Mishra A, Kandpal V, Singh A. The Ethical Dimensions of AI Development in the Future of
Higher Education: Balancing Innovation with Responsibility. Ethical Dimensions of AI Development. IGI
Global; 2025. pp. 401–436.
169. Seldon A, Abidoye O, Metcalf T. The Fourth Education Revolution Reconsidered: Will Artificial Intelligence
Enrich Or Diminish Humanity? Legend Press Ltd; 2020.
170. Bucea-Manea-țoniş R, Kuleto V, Gudei SCD, Lianu C, Lianu C, Ilić MP, et al. Artificial Intelligence
Potential in Higher Education Institutions Enhanced Learning Environment in Romania and Serbia. Sustain.
2022;14: 1–18. doi:10.3390/su14105842
171. Slimi Z, Villarejo-Carballido B. Unveiling the Potential: Experts’ Perspectives on Artificial Intelligence
Integration in Higher Education. Eur J Educ Res. 2024;13.
172. Murdan AP, Halkhoree R. Integration of Artificial Intelligence for educational excellence and innovation in
higher education institutions. 2024 1st International Conference on Smart Energy Systems and Artificial
Intelligence (SESAI). IEEE; 2024. pp. 1–6.
173. Leoste J, Jõgi L, Õun T, Pastor L, San Martín López J, Grauberg I. Perceptions about the future of integrating
emerging technologies into higher education—the case of robotics with artificial intelligence. Computers.
2021;10. doi:10.3390/computers10090110
174. Lukichyov PM, Chekmarev OP. Artificial intelligence in higher education. Russ J Innov Econ. 2023;13: 485–
502.
175. Luckin R. Machine Learning and Human Intelligence. The future of education for the 21st century. UCL
institute of education press; 2018.
176. Holstein K, McLaren BM, Aleven V. Designing for complementarity: Teacher and student needs for
orchestration support in AI-enhanced classrooms. Artificial Intelligence in Education: 20th International
Conference, AIED 2019, Chicago, IL, USA, June 25-29, 2019, Proceedings, Part I 20. Springer; 2019. pp.
157–171.
177. Ivanov S. The dark side of artificial intelligence in higher education. Serv Ind J. 2023;43: 1055–1082.
178. Helmiatin, Hidayat A, Kahar MR. Investigating the adoption of AI in higher education: a study of public
universities in Indonesia. Cogent Educ. 2024;11: 2380175.
179. Luckin R. Machine Learning and Human Intelligence: The Future of Education for the 21st Century. UCL
IOE Press. 2018. Available: https://eric.ed.gov/?id=ED584840
180. Dwivedi YK, Hughes L, Ismagilova E, Aarts G, Coombs C, Crick T, et al. Artificial Intelligence (AI):
Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and
policy. Int J Inf Manage. 2021;57. doi:10.1016/j.ijinfomgt.2019.08.002
181. Bhatnagar H. Artificial Intelligence-New Horizon in Indian Higher Education. J Learn Teach Digit Age.
2020;2020: 30–34.
182. Xu W, Member IEEE S, Gao Z, Dainoff M. An HCAI Methodological Framework: Putting It Into Action to
Enable Human-Centered AI. arXiv Prepr arXiv231116027. 2023. Available:
https://arxiv.org/abs/2311.16027v3
62 of 63
183. Dwivedi YK, Hughes L, Ismagilova E, Aarts G, Coombs C, Crick T, et al. Artificial Intelligence (AI):
Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and
policy. Int J Inf Manage. 2021;57: 101994.
184. Floridi L, Cowls J. A unified framework of five principles for AI in society. Mach Learn City Appl Archit
Urban Des. 2022; 535–545. doi:10.1002/9781119815075.ch45
185. Kurtz G, Amzalag M, Shaked N, Zaguri Y, Kohen-Vacs D, Gal E, et al. Strategies for Integrating Generative
AI into Higher Education: Navigating Challenges and Leveraging Opportunities. Educ Sci. 2024;14.
doi:10.3390/educsci14050503
186. Xu W, Ouyang F. A systematic review of AI role in the educational system based on a proposed conceptual
framework. Educ Inf Technol. 2022;27: 4195–4223. doi:10.1007/s10639-021-10774-y
187. Rai A. Explainable AI: from black box to glass box. J Acad Mark Sci. 2020;48: 137–141.
doi:10.1007/s11747-019-00710-5
188. Laplante P, Amaba B. Artificial Intelligence in Critical Infrastructure Systems. Computer (Long Beach Calif).
2021;54: 14–24. doi:10.1109/MC.2021.3055892
189. Chaudhry J, Pathan ASK, Rehmani MH, Bashir AK. Threats to critical infrastructure from AI and human
intelligence. J Supercomput. 2018;74: 4865–4866. doi:10.1007/s11227-018-2614-0
190. Hair JF. Multivariate data analysis. 2009.
191. Green T. A methodological review of structural equation modelling in higher education research. Stud High
Educ. 2016;41: 2125–2155. doi:10.1080/03075079.2015.1021670
192. Teo T. Handbook of quantitative methods for educational research. Springer Science & Business Media;
2014.
193. Byrne BM. Structural Equation Modeling With AMOS. Structural Equation Modeling With AMOS. 2013.
doi:10.4324/9781410600219
194. Kline RB. Principles and practice of structural equation modelling (4th ed.). Methodol Soc Sci. 2015; 1–554.
195. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus
new alternatives. Struct Equ Model a Multidiscip J. 1999;6: 1–55.
63 of 63
