In the rapidly evolving landscape of educational technology, generative artificial intelligence (AI) is emerging as a transformative force capable of reshaping how learning systems operate and how users engage with digital educational tools. A recent comprehensive meta-analysis conducted by Yan Yan and N.B. Jafri, published in BMC Psychology, delves deep into the multifaceted factors influencing the intention to utilize generative AI within educational systems. Their study synthesizes a wide array of empirical evidence to unravel the complex interplay of psychological, social, and technological determinants that drive or impede adoption. This exploration arrives at a critical juncture when educators, policymakers, and technologists seek to harness AI’s potential responsibly and effectively.
The core of this research lies in unraveling why and how intentions to adopt generative AI manifest among educators, students, and institutional decision-makers. Generative AI, distinct from traditional AI models, excels at producing novel content such as text, images, and simulations, thereby offering unique educational opportunities. However, the willingness to integrate these capabilities into learning environments is far from uniform. By aggregating data from multiple studies, the meta-analysis identifies consistent patterns and divergent trends that contribute to a nuanced understanding of adoption drivers in educational contexts.
One of the most compelling insights from Yan and Jafri’s meta-analysis is the pivotal role of perceived usefulness. This concept, deeply rooted in the Technology Acceptance Model (TAM), encapsulates users’ belief that employing generative AI will enhance their educational outcomes or processes. The analysis confirms that when users perceive clear, tangible benefits—such as personalized learning, enhanced creativity, and improved efficiency—their intention to engage with these systems significantly increases. This underscores the necessity for developers and educators to articulate and demonstrate the direct value added by generative AI tools.
Closely tied to perceived usefulness is the factor of perceived ease of use, which reflects how effortless individuals believe it is to learn and operate generative AI systems. The meta-analysis reveals that complexity and usability challenges remain substantial barriers to adoption, especially for educators who may lack technical training or resources. As a result, intuitive interfaces, robust support, and comprehensive training programs emerge as critical enablers to foster widespread acceptance. The interplay between ease of use and usefulness suggests that addressing these aspects concurrently maximizes adoption potential.
Beyond individual cognitive perceptions, social influence emerges as a significant external determinant in the intention to adopt generative AI in education. The meta-analysis highlights that endorsements from respected peers, institutional leadership, and influential thought leaders dramatically shape attitudes and behaviors. Educators and students frequently look to their professional communities and academic networks when evaluating new technologies. This social validation mechanism suggests that pilot programs, success stories, and professional development initiatives can act as catalysts for broader integration.
The psychological construct of trust also features prominently in the meta-analysis findings. Trust in the technology’s reliability, security, and ethical use profoundly impacts users’ willingness to incorporate generative AI into their educational routines. Concerns over data privacy, algorithmic biases, and potential misuse temper enthusiasm and generate skepticism. Addressing these concerns through transparent design, stringent data protection policies, and clear communication strategies is not merely prudent but essential for cultivating lasting engagement.
Importantly, the study examines demographic and contextual variables that influence adoption intentions. Factors such as age, prior experience with AI or digital tools, cultural attitudes toward technology, and institutional readiness all modulate how generative AI is perceived and embraced. For example, younger users with greater exposure to digital environments tend to exhibit higher openness toward AI integration. Conversely, institutions with limited infrastructure or conservative cultures demonstrate restrained enthusiasm. Recognizing these nuances enables tailored interventions that respect diverse learner and educator profiles.
The meta-analysis also critically assesses the educational settings where generative AI is deployed, revealing variable adoption patterns across disciplines, grade levels, and learning objectives. Subjects with creative or exploratory foci, such as art and language learning, exhibit greater receptivity to generative AI’s capabilities compared to more rigid, standardized curricula. This differential adoption hints at the need for domain-specific customization to optimize effectiveness and user satisfaction. It further suggests that one-size-fits-all approaches in AI integration are unlikely to succeed comprehensively.
On the technological front, the analysis emphasizes the impact of system features such as adaptability, interactivity, and feedback mechanisms on users’ adoption intentions. Generative AI systems that dynamically tailor content to individual needs, encourage active participation, and provide timely insights foster deeper engagement and learning. These sophisticated functionalities elevate perceived usefulness and user satisfaction, thereby reinforcing positive adoption cycles. Research and development efforts should thus prioritize these attributes to sustain momentum.
Moreover, the interplay between ethical considerations and adoption intentions surfaces as an urgent discourse within the study. Educational stakeholders increasingly demand that generative AI respects academic integrity, supports inclusivity, and avoids perpetuating inequities. Users’ concerns about plagiarism, fairness, and accessibility significantly influence their acceptance. Developers and policymakers must embed ethical frameworks into design and governance structures to align technology deployment with educational values and societal expectations.
Yan and Jafri’s meta-analysis also points to the dynamic nature of adoption intentions over time, influenced by evolving user experiences, technological advancements, and changing institutional priorities. Initial skepticism may wane as familiarity grows, or conversely, enthusiasm may diminish if unmet expectations arise. This temporal dimension calls for ongoing evaluation and adaptation in AI integration strategies, reinforcing the importance of iterative feedback loops and user-centered design in educational technology.
The study’s comprehensive methodology, employing meta-analytic techniques, provides statistically robust conclusions by aggregating results from diverse studies with varying methodologies, sample sizes, and contexts. This synthesis mitigates individual study biases and enhances generalizability, offering a valuable roadmap for stakeholders navigating the complex ecosystem of AI adoption in education. Nevertheless, the authors acknowledge limitations related to evolving AI capabilities and emerging educational paradigms that future research must address.
Crucially, the implications of this meta-analysis extend beyond academic discourse, offering actionable insights for technology developers, educators, administrators, and policymakers. Emphasizing user-centered design, transparent communication, robust training, and ethical oversight can collectively accelerate the responsible adoption of generative AI. By highlighting multifactorial influences, the study advocates for integrated approaches that consider cognitive, social, technological, and contextual dimensions simultaneously.
In conclusion, Yan Yan and N.B. Jafri’s meta-analysis is a seminal contribution illuminating the complex web of factors shaping the intention to use generative AI in educational systems. As educational landscapes continue to integrate AI-driven innovations, understanding these underlying determinants is paramount to unlocking the technology’s transformative potential. This research not only charts the current state of adoption but also lays the groundwork for informed, inclusive, and ethical advancement in educational AI applications, bearing profound implications for learners and educators worldwide.
Subject of Research: Factors influencing the intention to use generative artificial intelligence in educational systems
Article Title: Factors influencing the intention to use generative artificial intelligence in educational systems: a meta-analysis
Article References:
Yan Yan, C., Jafri, N.B. Factors influencing the intention to use generative artificial intelligence in educational systems: a meta-analysis. BMC Psychol (2026). https://doi.org/10.1186/s40359-026-03957-0
Image Credits: AI Generated
