This international study on pharmacy students’ perspectives regarding GenAI comprehensively examines technology adoption across nine countries, utilising the Extended UTAUT framework. The research included a cross-sectional survey targeting pharmacy students in various geographical regions, specifically Asia, the Middle East, Africa, and Europe, encompassing Egypt, Türkiye, Indonesia, Pakistan, Iraq, Nigeria, Malaysia, Saudi Arabia, and the United Arab Emirates. This study represents one of the largest investigations focused on accepting and using GenAI among pharmacy students. Employing the extended UTAUT framework delivers an in-depth analysis of the acceptance and usage domains based on eight key constructs of the model. This thorough assessment across different contexts aims to guide future initiatives and strategies for fostering responsible GenAI use among pharmacy students.

Through a theory-driven large-scale assessment, this study addresses an important topic, providing detailed findings on the acceptance and use of GenAI globally and highlighting differences across UTAUT domains based on country, gender, level of study and academic performance. This may help enhance the understanding of GenAI adoption in pharmacy education. However, this study has several limitations, including its cross-sectional design that limits causal inference, potential selection bias in participant recruitment, geographical concentration in specific regions, and self-reported data that might introduce response bias. Finally, the analysis of the associated factors with GenAI was informative but not extensive enough to uncover all potential factors that might impact adoption. These factors necessitate a careful interpretation and suggest the need for future research to thoroughly examine GenAI’s evolving role in pharmacy education.

The research uncovered several significant findings regarding using GenAI among pharmacy students. Expectedly, the patterns of frequency and preferred tools highlighted commonly used resources such as ChatGPT, followed by academic tools for content writing like Quillbot. This aligns with previous research that underpinned these academic writing tools, which are widely common in higher education settings [11, 19]. In contrast, specific single-purpose academic tools like Gamma and Tome experienced minimal adoption. Additionally, many respondents mentioned other emerging tools beyond the commonly referenced GenAI options. Earlier research highlighted that the choices between different tools are continuously changing and impacted by perceived efficiency, interaction, and intention [20], making multi-purpose tools appealing options to satisfy several needs through one platform. The interest in emerging GenAI tools indicates that educational institutions should enhance awareness and provide continuously updated guidance on these resources.

The research provided valuable insights into how pharmacy students engage with various GenAI tools across academic tasks. The wide array of applications for GenAI tools demonstrates their adaptability and integration into multiple facets of academic work. The high percentages of students utilising these tools for explanation, research, and improvement indicate that they primarily view them as supportive resources to enhance their learning and output quality. This wide range of uses is impacted by opportunities offered by GenAI to streamline learning, research, and assessment processes while making it a personalised and engaging experience [21]. The comparatively lower percentage of students relying on GenAI to complete assignments entirely is a positive sign, suggesting that most students are not overly dependent on these tools. However, this also highlights the need to reinforce the ethical aspects of interacting with these tools and areas where additional support or guidelines may be essential [22]. This data mainly benefits educators and institutions, highlighting the need to effectively integrate GenAI tools into the curriculum and guide students in their usage.

The analysis of the Extended UTAUT model revealed insightful findings at the construct level. Performance and effort expectancy were among the top-scoring constructs, indicating that students primarily perceive generative AI tools as valuable, accessible, and efficient for completing academic tasks. Previous studies highlighted that performance expectancy, effort expectancy, and social influence significantly influence the intention to use GenAI, but only performance expectancy and social influence directly impact academic performance [23]. Conversely, habit followed by price value constructs received the lowest scores, suggesting that adopting AI tools has not yet become a deeply established practice among students, while affordability continues to be a significant concern for individual users. This may raise the concern of equitable access to these tools, which requires a clear organisational perspective on a structured and targeted strategy to integrate GenAI [24].

The exploratory factor analysis (EFA) revealed a three-factor solution that partially aligns with the original UTAUT framework while introducing novel interactions among constructs. Most notably, Habit (HT) and Price Value (PV) loaded together under Factor 2 (Affordability and Habitual Integration). The coupling of Habit and Price Value suggests that students’ habitual use of GenAI tools is closely tied to affordability barriers, particularly in lower-income regions. For instance, tools requiring paid licenses were underutilised (Fig. 1). Research indicates that technology adoption in developing economies is significantly influenced by cost-effectiveness, even during the initial stages of habitual usage [25]. In such contexts, habitual use may only emerge if tools are perceived as financially accessible, creating a feedback loop where affordability reinforces routine engagement [26].

Moreover, the standalone Social Influence (SI) factor (Factor 3) aligns with theoretical emphasis on peer and mentor pressures. Existing interpretations of social influence in technology adoption suggest that while social influence may align with UTAUT’s emphasis on peer pressures, its independence from constructs like Performance Expectancy could vary across cultures [27]. In Southeast Asian societies, social norms significantly impact technology adoption. In contrast, social influence may be less tied to utility-driven adoption in contexts with weaker institutional support. These findings highlight the necessity of contextualising UTAUT extensions for emerging technologies such as GenAI. Our research shows that cost and cultural factors can influence theoretical relationships, especially in diverse, cross-national samples.

Curriculum-related findings revealed significant gaps in formal GenAI education. An overwhelming 60% of participants reported no exposure to GenAI-related events within their pharmacy curriculum. Merely 10% highlighted legal and ethical considerations. This disparity highlights the need for comprehensive GenAI integration strategies in pharmacy education. A relatively small-scale international study recruited 387 pharmacy students and highlighted a positive attitude towards this technology, indicating a need for relevant education and training [28]. This raises important questions about whether pharmacy educators are equipped to lead by example and upskill students’ skills in this area. Beyond the broad applications of GenAI use for generating study aids, brainstorming ideas, and offering practice opportunities for clinical problems, educators have started to develop successful examples of integrating this technology into the pharmacy curriculum in a subject-specific manner [29, 30].

Compared to previous work conducted among pharmacy students, the current study highlighted country-specific variations in extended UTAUT constructs and provided additional depth to the analysis. Malaysia had the highest ranking in performance expectancy, while Egypt and Iraq had the lowest scores for effort expectancy. Egypt ranked lowest in facilitating conditions and behavioural intention to use GenAI tools. Türkiye and Malaysia scored highest in social influence. Pakistan and Egypt recorded the lowest price values, while Indonesia had the highest. These variations highlight the complex landscape of GenAI acceptance across different educational and cultural contexts, informing the need for a context-specific approach to promoting responsible GenAI integration in pharmacy education [23].

Finally, our analysis revealed notable gender differences in the UTAUT constructs, with male students demonstrating higher acceptance. This is consistent with an earlier study that showed better perceptions and a higher pattern of use for broader applications among males compared to females, who were more specific and critically evaluating the usefulness of adopting these tools [31]. On the other hand, a recent study based on the technology acceptance model reported no significant gender-based differences in the perceived effectiveness of GenAI writing tools [32]. In a small study among second- and third-year US pharmacy students to investigate perceptions on utilising ChatGPT for clinical presentations, third-year students were more familiar and confident [33], consistent with our data that showed that third-year students exhibited the highest performance and effort expectancy. In the present study, academic performance was found to influence only Effort Expectancy and Facilitating Conditions, while higher-achieving students reported superior median scores compared to their average peers. Previous studies have shown mixed results regarding the relationship between academic performance and attitudes toward adopting technology. Some research indicates that students with higher academic achievement tend to have a more positive attitude toward technology adoption [34], while other studies have not found a significant impact of academic performance on this attitude [35]. The findings suggest that acceptance of GENAI tools in pharmacy education may be influenced more significantly by demographic factors, such as gender and educational level, than by academic performance. This highlights the importance of considering these factors in the development of future initiatives.

This study reveals implications for GenAI in pharmacy education. Three policy priorities are identified for responsible GenAI utilisation. First, establishing clear ethical standards and policies is crucial to maintaining academic integrity while maximising GenAI’s potential [36]. Comprehensive ethical guidelines must be developed to mitigate concerns regarding excessive reliance on GenAI for academic tasks [22]. Such guidelines are vital for preserving academic integrity and enhancing critical thinking skills. Second, curriculum Integration is essential to incorporate mandatory GenAI literacy modules in pharmacy programs, focusing on ethical usage and skill development (e.g., critical evaluation of AI outputs) while exploring future directions. Building capacity among pharmacy educators and developing structured strategies for GenAI integration into the curriculum is imperative to uphold quality standards and improve efficiency [37]. Third, context-specific Training that should utilise cultural strengths and social influences through peer-led training initiatives while ensuring equitable access to specific GenAI tools via institution-sponsored programs.