Introduction

Sustainable Development Goal (SDG) integration has been explored in business and technology sectors1,2, yet empirical research on how educational technologies and pedagogical design contribute to sustainability outcomes in teacher education remains limited. The synergistic impact of SC, generative AI tool usage (GenAITU), and knowledge-sharing practices (KSP) on perceived sustainable competencies (PSCs) is significant for preparing preservice teachers (PSTs) in the current technological landscape3,4,5,6,7,8. This study responds to calls for holistic and evidence-based educational practices that enable PSTs to develop competencies needed for long-term sustainability.

Many studies have discussed the significant role of sustainable education in producing PSCs for sustainable social and economic benefits9,10. PSCs require PSTs to put theory into practice and to become proactive and creative. They help students develop skills in (1) Critical Thinking, (2) Digital, (3) Value Creation, and (4) Opportunity Recognition competencies. These competencies prioritize long-term societal progress and mutually reinforce innovation, and vice versa, aligned with SDGs1,2,11. Recent evidence also suggests that learner engagement with educational technologies plays a critical role in fostering knowledge sharing and innovation in hybrid learning contexts12. However, limited empirical work has explored how SC, combined with GenAITU and KSP associated with PSCs in a blended learning environment.

GenAITU encompasses (i) frequency of use, (ii) application in learning for tasks like research and writing, and (iii) types of use, including content creation, text modification, and knowledge retrieval through AI-driven tools (e.g., ChatGPT, Kimi, and DouBao and others)13. However, concerns remain about potential academic integrity issues if AI tools are used improperly, highlighting the need for clear guidelines on appropriate usage. As educational institutions increasingly invest in technology-enhanced learning, it is essential to understand how these tools impact the learning process14. Some studies have discussed GenAITU as an effective approach to increase learning and improve sustainable outcomes in blended learning4,15. The present study investigates GenAITU not only as an educational input but as a mediating factor linking sustainable curricula to PSCs.

With this, the KSP is considered important for learning in the current technological landscape. KSP refers to the collaborative exchange of information, ideas, and resources through formal and informal interactions, enhancing outcomes in hybrid learning16. These practices also facilitate both online and face-to-face classes through social media platforms alongside digital devices, ensuring the smooth functioning of the learning process. The importance of KSP in higher education cannot be overlooked, as these practices significantly influence overall perceived academic performance3. Along with other factors, such as sustainable curricula and GenAITU, KSP contributes to enhancing students’ learning during this technological fusion14,17. Knowledge-sharing across other sectors supports adopting Industry 4.0 technologies in work process18. However, lack of studies have investigated the effect of sustainable curricula on PSCs in this technological landscape Despite its relevance, few studies have empirically tested the joint role of sustainable curricula, GenAITU, and KSP in promoting PSCs.

The existing literature on sustainable education and blended learning has primarily focused on isolated components, such as curriculum design or technology integration, without holistically examining the interplay between sustainable curricula, GenAITU, and KSP in relation to PSCs. Few studies have explored how these elements contribute to achieving SDGs in higher education. Prior research lacks empirical validation of the mediating roles of GenAITU and KSP in bridging sustainable curricula and PSCs. There is also limited application of frameworks like CDF, UTAUT2 and SDGs to contextualize these relationships, particularly in blended learning environments. Furthermore, the impact of GenAITU on PSCs within teacher education programs remains underexplored, leaving a gap in actionable strategies for educators and policymakers.

This study addresses these gaps by proposing and empirically testing a research model that integrates sustainable curricula, GenAITU, KSP, and PSCs, grounded in CDF, UTAUT2, KSPs and SDGs. The partial least squares structural equation modeling of data from PSTs in China demonstrates how sustainable curricula directly correlate PSCs and indirectly through GenAITU and KSP. The research validates the serial mediation effect, showing that GenAITU amplifies the sustainable curricula’s impact via KSP. By contextualizing these relationships within blended learning, this study provides evidence-based insights into how GenAITU and KSP synergize with sustainable curricula to advance SDG-aligned PSCs. This bridges theoretical and empirical voids, offering a holistic understanding of the mechanisms driving PSCs among PSTs.

This study contributes to the literature in several ways. Firstly, it outlines hypothetical relationships among sustainable curricula, GenAITU, KSP, and PSCs. Secondly, it connects this conceptual framework to empirical research, offering a practical perspective. Thirdly, the study enhances understanding of the mechanisms through which PSCs are achieved, highlighting the conditional role of sustainable curricula in fostering PSCs when GenAITU and KSP are present in HEIs. Fourth, it contributes to empirical studies grounded in CDF, UTAUT2, KSP and SDGs. This study also advances SDGs by demonstrating how sustainable curricula, GenAITU, and KSP collectively enhance PSCs. It provides policymakers and institutions with insights to improve curricula, infrastructure, and faculty training, ensuring future teachers foster sustainability through education. Finally, the authors propose directions for future research. This study addresses that gap through a model informed by the CDF, UTAUT2, and SDGs. To guide this investigation, the following research questions (RQs) were posed, with hypotheses developed to empirically test each relationship:

RQ1: Do sustainable curricula, GenAITU, and KSP influence PSCs?

RQ2: Does GenAITU, through KSP, mediate the relationship between sustainable curricula and PSCs?

The remaining paper is organized as follows: Sect. 2 consists of the literature review, including the research model and hypothesis development. Section 3 outlines the Design and sampling. Section 4 presents the results and findings. Finally, Sect. 5 discusses the findings, practical implications, limitations, and directions for future research.

Literature review

Theoretical model construction

In recent years, researchers have increasingly integrated technology-enhanced curriculum frameworks and educational technology theories to promote sustainable learning across various domains. Our research draws on four adaptive theoretical perspectives to construct the theoretical model: the CDF, an adapted UTAUT2, KSP, and the UN-SDGs. These frameworks collectively inform the conceptualization of key constructs, including sustainable curricula, generative AI tool usage, and knowledge sharing practices, positioned as interrelated drivers of PSCs among learners in blended learning environments.

Fig. 1
Fig. 1
Full size image

Theoretical model diagram.

Curriculum delivery framework in the context of sustainable curricula

The adapted CD in the context of sustainable curricula offers a structured foundation for developing PSCs through four integrated components: innovative content design, transformative teaching strategies, enhanced learning experiences, and adaptive assessment models. Innovative content design ensures learning materials are aligned with sustainability challenges and evolve dynamically over time. Transformative teaching strategies and enhanced learning experiences collectively support the engagement phase by integrating experiential and reflective learning with technological tools to deepen student involvement. Adaptive assessment models contribute to the evaluation phase by offering personalized, technology-driven feedback to foster sustainability-related skills. These components operate simultaneously to support the development of PSCs-including critical thinking, digital, value creation, and opportunity recognition competencies, preparing students to navigate and address complex sustainability challenges1,2,19. Thus, sustainable curricula, structured around these components, serve as a foundational framework for preparing pre-service teachers to meet sustainability demands through informed teaching and learning practices. The present study, therefore, centers on predicting PSCs using the sustainable curricula model.

UTAUT2 in generative AI tool usage context

UTAUT2, applied to Generative AI Tool Usage within blended learning, measures how learners adopt and utilize these technologies to advance sustainability goals. It provides critical insights through three key dimensions: Frequency of Use (how regularly learners employ AI tools for sustainability tasks), Application (how AI is used-for research, problem-solving, or creating sustainable solutions), and Types of Use (the nature of AI interactions, such as data analysis, predictive modeling, or ethical evaluation of sustainability scenarios). By analyzing these dimensions, adapted UTAUT2 uncovers behavioral patterns in AI adoption, which directly influence the development of Digital Competency and enable scalable, tech-driven, sustainable capability building8.

GenAITU refers to platforms like ChatGPT, Kimi, DouBao and others, enabling students to create content, ideas, and insights for academic/personal use, enhancing learning through diverse content generation8. These tools and the Internet serve as channels for knowledge creation and learning. China is an emerging country where approximately 1.3–1.4 billion people are mobile phone users, and Active AI Users are roughly 800 million to 1 billion people who regularly use AI applications20. GenAITU enhances learning by reducing content access barriers, enabling rapid knowledge acquisition, and fostering learning strategies21. It supports active learning through features like content generation, logical reasoning, and adaptive pedagogy, along with risks such as ethical dilemmas, cognitive laziness, addiction, and misinformation8,22. Its dual role as a facilitator of innovation and a source of potential academic challenges highlights the need for balanced integration in education. We defined GenAITU as (i) the frequency of use, indicating how often PSTs engage with AI tools in their learning process; (ii) the application in learning, reflecting how these tools support academic tasks such as searching information, problem-solving and personalized learning; and (iii) the types of use, encompassing content creation, text modification, knowledge retrieval, and skill development through AI-driven technologies. GenAITU is considered effective for learning and should be incorporated in designing curricula through university sustainable education to predict PSCs.

Knowledge sharing practices

KSP represents the collaborative engine for sustainable capability development through three core practices23,24,25: Practice 1 (Information Exchange) facilitates the sharing of sustainability data and research26. Practice 2 (Idea Exchange) enables discourse on innovative solutions to environmental/social challenges27. Practice 3 (Resources Exchange) distributes tools, frameworks, and best practices. These practices culminate in a Knowledge Co-Contribution Process, where learners collectively generate new sustainability knowledge28. Embedded within blended learning, KSP cultivates Value Creation Competency (transforming shared knowledge into actionable solutions) and Opportunity Recognition Competency (identifying sustainability opportunities through peer insights), reinforcing a culture of continuous, collaborative learning essential for long-term impact29,30.

In the context of this study, KSP refers to the processes and behaviors through which students, faculty, and peers exchange, disseminate, and collaboratively use knowledge and information within the framework of sustainable curricula12. These practices encompass formal mechanisms such as online platforms, collaborative tools, and workshops, as well as informal methods like peer discussions, mentorship, and collaborative problem-solving activities. In this study, the perceived KSP is to foster PSCs in blended learning environments12,16. The KSP refers to the (i) collaborative exchange of information, (ii) ideas, and (iii) resources through formal and informal interactions, enhancing PSCs.

Perceved sustainable competencies perspective

The PSCs construct is listed under critical thinking competency31,32, digital competency33, value creation competency34, and opportunity recognition competency, alongside constituting the target PSCs catalyzed by the synergistic framework. The curriculum equips learners with foundational knowledge and adaptive skills; UTAUT2-analyzed Generative AI usage enhances efficiency and innovation in applying these skills; and KSP fosters collective intelligence and real-world problem-solving29,30. This triad transforms blended learning into a dynamic ecosystem where competencies are not merely taught but co-created, scaled, and iteratively refined to address evolving sustainability challenges.

PSCs are relatively intuitive and frequently assumed predictors of learning outcomes35. In this study, PSCs are defined as the collective competencies of individuals to integrate quality education competencies, foster critical thinking competency, digital competency, value creation competency, and opportunity recognition competency. These competencies, developed based on SDGs, are operationalized through sustainable curricula approaches1,2, leveraging GenAITU and KSP to equip learners with PSCs essential for sustainable development. In this study, PSCs are defined as the systemic capacities to harness competencies as the central drivers for integrating and advancing digital competency33, value creation competency34, opportunity recognition competency, alongside critical thinking competency31,32, aligned with SDGs. Conceptualization of PSCs with other variables, such as sustainable curricula, GenAITU, and KSP, has received minimal systematic attention in academia. Moreover, PSCs represent the ultimate outcome of sustainable education. However, prior research has not adequately addressed this phenomenon. Therefore, this study examines the relationships among sustainable curricula, GenAITU, KSP, and PSCs, as conceptualized in the following hypothetical model.

Hypothetical model development

The proposed model highlights that sustainable curricula can bring about advancement in PSCs through GenAITU via the KSP. Sustainable curricula, GenAITU, and KSP have long been major concerns among scholars and practitioners for their role in developing PSCs. Sustainable curricula integration is being treated as a tool to achieve SDGs6, and its positive effects on PSCs have been claimed by researchers in the field of sustainable education36. The adapted CDF posits that the effective implementation of sustainable curricula plays a pivotal role in developing PSCs students based on sustainable goals2,19.

The research model was developed based on four broad theoretical perspectives: adapted CDF19, UTAUT37, SDGs1,2, and the concept of knowledge sharing. The CDF explains the role of sustainable curricula along with subscales of ICD, TTS, ELE, and AAM19. The proposed model highlights that sustainable curricula can bring about advancement in PSCs through GenAITU via the KSP. Sustainable curricula, GenAITU, and KSP have long been major concerns among scholars and practitioners for their role in developing PSCs. Sustainable curricula integration is being treated as a tool to achieve SDGs6, and its positive effects on PSCs have been claimed by researchers in the field of sustainable education36. The CDF posits that the effective implementation of PSCs plays a pivotal role in developing students’ sustainable curricula within blended learning environments2,19.

Universities can support SDGs by integrating PSCs into teaching and learning. PSCs foster critical thinking, technical skills, productivity, and opportunity recognition. PSCs have been defined as frameworks that foster critical thinking and technical skills, productivity and scalability, and opportunity recognition and value creation1,2,11. However, PSCs development should align with the design and implementation of sustainable curricula within universities, ensuring consistency with institutional sustainability goals6. Achieving SDGs through sustainable education remains challenging, but adopting inclusive strategies like blended learning and participatory pedagogy can help universities address these challenges36. Sustainable curricula should be tailored to individual student needs to enhance relevance and effectiveness. For instance, Pramesworo, Fathurrochman4 highlight that well-designed SC foster students’ PSCs by bridging theoretical knowledge with practical skills. Based on these perspectives, we propose the following hypothetical model to examine the relationships between the variables, as illustrated in Fig. 2.

Fig. 2
Fig. 2
Full size image

Hypothetical relationship model.

Sustainable curricula VS PSCs

Sustainable education fosters PSCs aligned with the SDGs. For example, Kuusalu, Laine36 used a mixed-method approach to assess sustainability themes and evaluate the ecological, social, cultural, and economic dimensions among language students at a Finnish university. Caldana, Eustachio6 investigated 274 bachelor’s students in a hybrid format through a survey questionnaire and identified that formal, non-formal, and informal learning approaches increase the development of PSCs and sustainability literacy at a Principles for Responsible Management Education (PRME) signatory business school in a Brazilian university. Pauliukevičienė, Stankevičienė2 identified SDGs 4, 8, 9, and 16 as key drivers of sustainable FinTech growth, emphasizing the role of public access to FinTech and an educated workforce in Northern Europe, including Finland, Germany, the Netherlands, and Sweden, leading in SC development. The curriculum framework proposed by Iqbal, Yi19 recommends that the relationship between indicators of CDF and competencies must be investigated further. The concept of sustainable curricula in the context of PSTs needs to be uncovered. Limited research has focused on exploring the association between sustainable curricula and PSCs. To address this gap, this study aimed to explore the positive relationship between sustainable curricula and PSCs. We developed the following hypothesis to explore this relationship:

H1: Sustainable curricula are positively associated with sustainable competencies.

Sustainable curricula VS GenAITU

Few scholars have developed theoretical concepts and models to bring out the association between curricula curriculum and learning technologies19,38. UTAUT suggests that the ease of use and perceived benefits of these technologies enhance learning37. This enables students to engage better with the tools and practices that shape their competencies39. Studies on blended learning have highlighted that effective sustainable curricula are a positive contributor to GenAITU in technology-enhanced learning environments among students37,40. In this context, it is expected that PSTs enrolled in blended learning courses in China will likely develop technological skills as they engage with GenAITU within formal blended learning environments. Such environments help students learn the effective use of generative AI tools. These arguments suggest that sustainable curricula are positively related to GenAITU, leading to the following hypothesis:

H2: Sustainable curricula positively connect with GenAITU.

Sustainable curricula VS KSP

Multiple studies on sustainable education have highlighted the positive connection between curricula and KSP in blended learning business courses3. Researchers have designed and published studies focused on the SC and its effects on KSP through actively participating in collaborative online spaces among students12,16. Researchers with a common approach claim that a hybrid learning environment allows students to engage in discussions, share resources, and work together on projects. It helps in developing their critical thinking and problem-solving skills41. According to Uyan and Şanal42, sustainable curricula have emerged as a promising approach to enhancing KSP in hybrid learning. These arguments suggest that sustainable curricula are positively related to KSP among PSTs, which leads to the following hypothesis

H3: Sustainable curricula are positively associated with KSP.

GenAITU VS KSP

Various studies have suggested that the nexus of GenAITU and KSP needs to be explored in blended learning. Eslami, Achtenhagen18 explored that advanced learning technologies connections of KSP with communication and collaboration in automative industry. Another example example is that GenAITU provides support to learning environments that can increase KSP38. It is also helpful for effectively managing KSP in organizations22. A lack of studies has reflected the connection between GenAITU and the KSP, particularly in sustainable education. Therefore, the current study addressed this gap, examining the connection between GenAITU and KSP. In line with this, the following hypothesis was proposed:

H4: GenAITU positively connects with KSP.

GenAITU VS PSCs

Educational studies have shed light on the connection between GenAITU and PSCs aligned with the SDGs43. Research has also been conducted to evaluate the impact of GenAITU on students’ PSCs, enhance the global pursuit of sustainable innovation Competencies development15. Results proved that advanced technologies have a connection with entrepreneurial Competencies, which is a component of PSCs44. Previous studies have emphasized that sustainable education should deliver better to achieve better results to develop PSCs. This may be possible through the integration of GenAITU in the teaching and learning process to achieve the SDGs. These studies enhance our understanding of the link between GenAITU and PSCs. Thus, a positive association is predicted with the following hypothesis:

H5: GenAITU positively associates with PSCs.

KSP VS PSCs

Most studies have highlighted that KSP plays a significant role in fostering PSCs. However, its implementation varies significantly across different contexts. In regions like Nepal, robust digital infrastructure provides support for collaboration and the exchange of ideas, enhancing PSCs through KSP45. Similarly, Ye, Liu46 found that knowledge sharing positively impacts employees’ innovation behavior, as evidenced by a survey of 318 frontline employees in China. Asghar, Barbera47 identified that KSP enhances leadership behaviors among PSTs in Pakistan. Moreover, there was a need to uncover how the concept of knowledge sharing in blended learning environments contributes to KSP perspectives42. The association between KSP and PSCs has received limited attention in previous studies45. Thus, we address this research gap by examining the connection between KSP and PSCs through the following hypotheses:

H6: KSP positively connects with PSCs.

Integrative mediation modeling

Several studies highlight the need to explore the relationship between sustainable curricula, GenAITU, KSP, and PSCs. GenAITU has the potential to enhance learning by bridging traditional and modern educational practices, supporting adaptive learning, personalized feedback, and collaborative problem-solving among university students43,48. Additionally, GenAITU may strengthen the role of sustainable curricula in developing PSCs. Based on these insights, this study proposes a research model examining the mediating role of GenAITU in the relationship between sustainable curricula and PSCs3,4,5,6, leading to the following hypothesis:

H7: Sustainable curricula associated with PSCs through GenAITU

Previously reviewed studies described that KSP in various forms directly affects PSCs45 through universities’ robust digital infrastructures that facilitate collaboration and the exchange of ideas on campus. Similarly, Ye, Liu46 identified that KSP positively influences innovation behavior, which is a component of PSCs. Caldana, Eustachio6 found that formal, non-formal, and informal learning approaches enhance the development of PSCs in a hybrid format in Brazil. Sharif, Tongkachok49 identified the highly constructive role of KSP that fully mediated the connection between leadership behaviors and innovativeness, which is a component of PSCs. The authors of the current research thus affirm from the findings of the aforementioned studies that KSP works as a mediator between sustainable curricula and PSCs. Limited research has focused on exploring these connections. Therefore, this study addressed this gap by measuring the mediating role of the KSP in the relationship between the sustainable curricula and PSCs. Hence, the following hypotheses are put forward:

H8: Sustainable curricula connect with PSCs through the KSP.

In addition, previous studies have highlighted the significance of SC, GenAITU, KSP, and PSCs in various educational contexts3,4,5,6. Beyond the direct and indirect hypotheses proposed above, this study investigated the serial mediation effects of sustainable curricula on PSCs through GenAITU via KSP. As described above, most prior studies have focused on the direct influence of sustainable curricula on PSCs, GenAITU, and KSP. However, few studies have examined and provided evidence for the mediating role of GenAITU via KSP in improving psychological PSCs. Moreover, little is known about the mediating role of GenAITU and KSP between sustainable curricula and PSCs. Therefore, this study aims to fill this gap in existing literature by investigating the mediation of GenAITU through KSP. Furthermore, examining this mediating influence will contribute to a more comprehensive understanding of the sustainable curricula’s impact on PSCs through the mediating associations of GenAITU via KSP. Therefore, based on the above, we propose the following hypothesis (Fig. 1).

H9: Sustainable curricula connect with PSCs through GenAITU via the KSP.

Existing literature discusses SC in relation to GenATU and KSP among university students3,4,5,19,38. Sustainable curricula support SDGs through sustainable education initiatives within universities36. Integrating sustainable curricula with ICD, TTS, ELE, and AAM in blended learning environments can enhance GenATU and KSP. Drawing on the adapted UTAUT2 framework, which emphasizes frequency of use, application in learning, and types of use, alongside the KSP perspective on collaborative exchanges of information, ideas, and resources, this study examines how sustainable curricula influence students’ PSCs by fostering GenATU and KSP4,5,19,37,38. These include critical thinking, technical skills, productivity, scalability, opportunity recognition, and value creation1,2. Furthermore, the authors assumed that effective SC enhances GenATU and KSP, which in turn helps students improve these PSCs. The proposed research model, demonstrated in Fig. 1. In blended learning environments, SC equips students with essential skills to develop PSCs. This study demonstrates that GenAITU, mediated by KSP, bridges sustainable curricula and PSCs, revealing how sustainable curricula enhance these Competencies through technology and collaboration. By clarifying the sequential pathway (sustainable curricula → GenAITU → KSP → PSCs), this research advances understanding of their interplay in emerging educational contexts.

Design and sampling

This study employed a time-lagged questionnaire survey to investigate the relationships between sustainable curricula, GenATU, and KSP among PSTs in blended learning environments. Data were collected from undergraduate, graduate, and post-graduate students enrolled in teacher education programs across six universities in Hubei, China. Participants were required to meet specific inclusion criteria: recent completion of sustainability-related courses, active use of generative AI tools in coursework, enrollment in blended learning programs, and being in their final degree years. Final-year PSTs were specifically selected because they are more likely to have completed or be actively engaged in courses that integrate sustainable curricula. At this academic stage, students possess greater exposure to both pedagogical strategies and technology-enhanced learning tools like GenAITU, making them more capable of providing informed and meaningful responses. Moreover, targeting this cohort ensures relevance and maturity in reflections on KSP and PSCs development. Participants were enrolled in blended learning courses that explicitly embedded sustainability themes and included integrated use of GenAI tools in weekly assignments, discussions, and collaborative projects. Institutional syllabi required students to demonstrate their engagement with digital platforms and AI applications as part of their final course evaluations.

Participants were selected using convenience sampling, facilitated by faculty and department heads, with surveys distributed via email and course platforms (Tencent Classroom, WeChat, and Tangce Classroom) and printed copies as well. Ethical approval was obtained from the Institutional Ethical Review Board, School of Education, Huazhong University of Science and Technology (No. IERB-SED/2205/2023, approved on 22 May 2023), and institutional permissions were secured to distribute the survey. All methods were performed in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. Participants received comprehensive information about the study before participation, and informed consent was obtained, ensuring confidentiality and voluntary participation.

Data collection was conducted in two waves (September 2023 and January 2024) to reduce common method bias. A pilot study (n = 30) was conducted to measure the reliability and validity of the instrument. Minor revisions were made based on participant feedback. In the final study, 500 students were invited to participate, yielding 446 responses (89.2% response rate). Surveys were administered using digital platforms with links and QR codes. As shown in Table 1, the gender distribution was nearly balanced, with 48.4% male and 51.6% female participants. Regarding education level, the majority of respondents were undergraduates (46.2%), followed by master’s (33.4%) and doctoral students (20.4%), indicating a diverse academic representation. The cumulative GPA data reveals that most students reported a GPA Less than 2.00 (23.3%), 2.00–3.00 (28.3),3.10–3.50 (34.5), and 3.51-4 (13.9), suggesting an overall moderate to high academic performance level. This distribution supports the credibility of the sample for analyzing academic and behavioral constructs in a higher education context.

Table 1 Demographic characteristics of survey participants (N = 446).
Full size table

Implementation framework for key constructs

Our study integrated four key approaches to operationalize sustainable curricula within teacher education programs in China: ICD, TTS, ELE, and AAM. ICD was aligned with the United Nations Sustainable Development Goals (SDGs) and implemented through modules such as Climate Change, Social Justice, Education for All, Digital Literacy and AI in Education, Rural Education and Equity, Gender, Ethics and Inclusive Pedagogy, and Mental Health and Well-being in Schools. These modules emphasized interdisciplinary, equity-oriented, and globally relevant content to foster sustainability competencies among students enrolled in teacher education programs.

TTS included the use of AI tools like ChatGPT, DouBao, Kimi, or others to simulate lesson plans and explore ethical dilemmas in sustainability education. Students engaged with these tools both asynchronously and during blended learning to develop AI-integrated pedagogy. ELE activities incorporated real-world problem-solving tasks, such as environmental policy debates and inclusive curriculum planning, using generative AI tools to deepen engagement with SDG-related topics. AAM featured AI-based formative assessments, interactive quizzes, and automated feedback tools that supported students’ PSCs. GenAITU was captured through three lenses: frequency (regular use of AI tools for coursework), application (e.g., assignment drafting, quiz prep), and ethical use (adherence to AI responsibility guidelines). KSP emerged through group co-authorship projects, peer editing using AI, and informal exchanges via platforms such as WeChat, Tencent, and others (see Table 2).

Table 2 Implementation of sustainable curricula, genaitu, and KSP.
Full size table

Questionnaire design

A structured 50-item questionnaire, consisting of three main sections, was adapted and developed. The first section outlines the purpose of the study, defines key terms, and includes statements regarding privacy and confidentiality. The second section collected demographic information from the respondents. The third section focuses on the primary variables: sustainable curricula, GenAITU, KSP, and PSCs. A seven-point Likert scale was used (1 = strongly disagree to 7 = strongly agree). Most scale items were developed and adapted from previous research, while a few were newly developed to ensure that the scales were contextually valid. Before the final data collection, the reliability and validity of the questionnaire were tested through a pilot study involving 30 respondents with characteristics similar to those of the target population. Their feedback helped refine the questionnaire items. Five academic experts were consulted to ensure face and content validity. Based on feedback from the pilot participants and experts, minor adjustments were made to enhance the validity of the questionnaire. The finalized instrument was used to collect the data.

Measures

Innovative content design

The five items were adapted and developed from the work of United-Nations1, Pauliukevičienė, Stankevičienė2, Iqbal, Yi19, Iqbal, Asghar44. Sample item included: “The curriculum content provides a comprehensive foundation for understanding sustainable principles.” The Cronbach’s alpha coefficient for this scale was 0.729, indicating acceptable reliability.

Transformative teaching strategies

Seven items were adapted and developed from the work of Iqbal, Asghar, et al. (2022); Iqbal, Yi, et al. (2022); Pauliukevičienė et al. (2024); United-Nations (2015). Example item included: “The teaching strategies used effectively engaged me in sustainable learning.” A Cronbach’s alpha of 0.869 indicated strong reliability.

Enhanced learning experiences

Six items were adapted and developed from the work of Iqbal, Asghar, et al. (2022); Iqbal, Yi, et al. (2022); Pauliukevičienė et al. (2024); United-Nations (2015). Examples of items included: “My academic experience has motivated and engaged me in sustainable Competencies.” A Cronbach’s alpha of 0.773 indicated acceptable reliability.

Adaptive assessment model

Six items adapted and developed from the work of Iqbal, Asghar, et al. (2022); Iqbal, Yi, et al. (2022); Pauliukevičienė et al. (2024); United-Nations (2015). The sample item “The feedback provided on my work was constructive and helped improve my sustainable Competencies.” A Cronbach’s alpha of 0.879 indicated that the scale was reliable.

Generative AI tool usage

Eight items adapted and developed from existing research8. Sample item included “I use Generative AI tools to generate new ideas and solutions for sustainability challenges.” A Cronbach’s alpha of 0.863 reflected strong reliability.

Knowledge-sharing practices

Six items related to KSP were adapted and developed from the studies of Zhang and Mangmeechai12, Asghar, Barbera47. Sample item included, “I regularly share my knowledge with peers to support sustainable innovation tasks.” Cronbach’s alpha, 0.805, showed acceptable reliability.

Perceived sustainable competencies

Twelve items were adapted and developed from the work of United-Nations1, Pramesworo, Fathurrochman4, Kuusalu, Laine36, Iqbal, Asghar44. Sample item included: “I possess the skills to solve complex problems and apply knowledge to promote sustainable practices.” A Cronbach’s alpha of 0.884 indicated high reliability.

Data analysis procedure

Statistical analyses were performed using SPSS 26 and SmartPLS 4 using descriptive and inferential statistics. First, the demographics of the participants were analyzed. Second, we analyzed measurement models using partial least squares structural equation modeling (PLS-SEM). This included factor loading, Cronbach’s alpha (CA), composite reliability (roh_A and roh_C), convergent validity (Average Variance Extracted, AVE), multicollinearity (Variance Inflation Factor, VIF), and discriminant validity (heterotrait-monotrait, HTMT) of the scales. Finally, we employed PLS-SEM to explore the direct and indirect relationships between variables used in the research model, using second-order constructs to capture higher-level relationships between first-order latent variables for a deeper understanding of this model. The analysis includes collinearity, R-squared, F-square, model fit, and structural relationship measurements. The data screening ensured accuracy by addressing missing values (none, as all questions were mandatory), identifying outliers (none exceeded ± 3.29 Z-scores), and confirming acceptable normality. Skewness and kurtosis values for all items fell within the acceptable range of ± 3, except for ICD3 and ELE3, which showed elevated values indicating non-normality; however, given the large sample size (N = 446), the data are considered robust for SEM analysis, as SmartPLS does not require normally distributed data due to its variance-based approach. To address potential common method bias and improve the validity of causal inferences, a time-lagged questionnaire design was adopted. This design allowed participants to reflect more accurately on their experiences at each stage, while minimizing the effects of social desirability and common method variance. Common method bias was assessed using procedural remedies for wave data and Harman’s single-factor test, which explained 31.693% of the variance50,51.

Results

Measurement modeling

The study used PLS-SEM to measure the connections among variables defined in the research model. SmartPLS-SEM was chosen as it is more statistically efficient and relatively less sensitive to sample size compared with other statistical software used for covariance-based SEM50. This study measured the connections between sustainable curricula, GenAITU, KSP, and PSCs among PSTs. Thus, prior to testing the hypothesized impact, we examined the validity and reliability of scales in the model.

We measured the factor loadings for each factor. The threshold value for factor loadings is 0.60 if the AVE is above 0.50. Items exhibiting low factor loadings or negatively impacting the validity and reliability of the scale were eliminated through dimensionality reduction procedures in accordance with established psychometric guidelines. Results indicated that all items were above the threshold, confirming the scale’s validity and reliability. Cronbach’s alpha and composite reliability (rho_A and rho_C) have a cut-off value of 0.70. The results indicated that Cronbach’s alpha and composite reliability (rho_A and rho_C) confirmed the scale’s reliability (Table 3). Multicollinearity was assessed using VIF, with a threshold value of < 5. Table 2 indicates that multicollinearity was not an issue. We also assessed the convergent validity using the AVE values with a threshold of 0.5019,50. The AVE values for all scales met the standard, confirming the acceptable validity of each construct (Table 3).

Table 3 Measuring reliability, validity and multicollinearity.
Full size table

The HTMT ratio was applied to assess discriminant validity. It is considered better than the Fornell and Larcker52 approach. The HTMT allows us to evaluate the correlation between different constructs and within-item relationships in the same factor50. Henseler, Ringle53 recommended a cutoff value of 0.90, with values exceeding this threshold indicating insufficient discriminant validity. In our study, all HTMT values fell below 0.90, confirming adequate discriminant validity. Specifically, the correlation between PSCs and sustainable curricula was 0.601, between PSCs and GenAITU was 0.609, and between PSCs and KSP was 0.787. Additionally, the correlation between sustainable curricula and GenAITU was 0.791, and between sustainable curricula and KSP was 0.529, while the correlation between GenAITU and KSP was 0.592. These results affirm the discriminant validity of the reflective constructs, as the correlations among constructs remain below the critical threshold of multicollinearity, and each construct retains conceptual distinction.

The model fit indices demonstrate an acceptable model fit, with SRMR values of 0.058 and 0.059 for the saturated and estimated models, respectively, both below the recommended threshold of 0.08. The d_ULS values (1.385 and 1.403) and d_G values (0.444 and 0.445) were relatively low, further indicating a good model fit. Additionally, the NFI values of 0.804 and 0.803 are slightly below the commonly accepted benchmark of 0.90 but still suggest a reasonable model approximation. The Chi-square values (1124.048 and 1126.905) also support the overall adequacy of the model50, indicating that the proposed model structure fits the observed data well.

Furthermore, we measured R² square using the following criteria: 0.75, substantial explanatory power, 0.50 indicates moderate explanatory power; and 0.25, weak or below-average explanatory power50. Based on these criteria, the results indicate that PSCs have relatively strong explanatory power, with an R² value of 0.518 (adjusted to 0.524), nearing the threshold for substantial explanatory power. The explanatory power for GenAITU was moderate, with an R² value of 0.457 (adjusted to 0.456). Meanwhile, KSP exhibited a weaker, yet still moderate, level of explanatory power, with an R² value of 0.266 (adjusted to 0.263).

Cohen (1988) proposed the f2 metric to determine the influence of an exogenous construct on a dependent one. An f2 value exceeding 0.35 signifies a substantial effect, 0.15 indicates a moderate effect, and anything below 0.02 suggests a weak influence50. The sustainable curricula slightly affect PSCs, with an f² of 0.040. In contrast, its impact on Generative AI Tool Usage (GenAITU) is very strong, as reflected by an f² of 0.842. The curriculum also exhibits a small effect on Knowledge Sharing Practices (KSP), with an f² of 0.025. GenAITU shows a small to moderate influence on both PSCs (f² = 0.025) and KSP (f² = 0.104). Notably, KSP has a moderate effect on PSCs, demonstrated by an f² of 0.400.

Structural modeling

We measured the direct and indirect connections through SmartPLS-SEM (bootstrap mechanism 20000) proposed by Hair, Risher50. The results indicated that the sustainable curricula significantly correlate with students’ PSCs (β = 0.190, p <.05), validating their educational value in promoting long-term competencies aligned with sustainability goals. It also shows a positive and significant direct connection with GenAITU (β = 0.676, p <.05) and KSP (β = 0.183, p <.05), which confirms H1, H2, and H3, respectively. Similarly, GenAITU has a positive and significant direct connection with KSP (β = 0.374, p <.05) and PSCs (β = 0.156, p <.05), which supports H4 and H5, respectively. These findings suggest that integrating generative AI tools into the curriculum strengthens students’ knowledge-sharing practices and their ability to act sustainably. Additionally, KSP has a positive and significant direct connection with PSCs (β = 0.510, p <.05), which supports H6. This highlights that when students actively share and exchange knowledge, it significantly boosts their capacity to make informed and sustainable decisions. Moreover, our research assessed two control variables, gender and background; neither had a significant effect on PSCs (β = 0.021, p >.05; β = −0.027, p >.05) (Fig. 3), indicating that the curriculum’s effectiveness in building PSCs is consistent across diverse student demographics (see Table 4).

Table 4 Direct relationships between constructs and control variables.
Full size table
Fig. 3
Fig. 3
Full size image

Structural model illustrating the connections between sustainable curricula, generative ai tool usage, and knowledge sharing practices on perceived sustainable competencies.

Mediating effect

We also measured indirect connections through structural modeling. The results indicated that sustainable curricula have a positive and significant indirect connection with PSCs through GenAITU (β = 0.105, p <.05), which confirmed H7. The results also show that sustainable curricula have a positive and significant indirect connection with PSCs through KSP (β = 0.093, p <.05), which supports H8. Finally, findings indicated that sustainable curricula have a positive and significant indirect connection with PSCs through GenAITU via KSP (β = 0.129, p <.05), which confirmed H9 (Table 5). These mediated pathways demonstrate that the curriculum not only delivers direct learning outcomes but also activates digital and collaborative mechanisms that enhance students’ ability to engage in sustainable innovation.

Table 5 Indirect connection between variables.
Full size table

Discussion

This study established meaningful relationships using a synthesized theoretical framework. Previous research in this domain has predominantly focused on Western contexts2,36, with limited attention paid to emerging nations such as China. Moreover, existing studies in these contexts have largely overlooked the role of sustainable curricula in blended learning environments, emphasizing the efficient use of academic resources. To the best of our knowledge, this study is the first to investigate the impact of sustainable curricula on PSCs within blended higher education, particularly by positioning GenAITU and KSP as mediating constructs. Our study advances our understanding of how sustainable curricula and technological tools (GenAITU) are not just theoretically effective but are also perceived by students as impactful components of their learning experience.

The results indicate that sustainable curricula have a positive, direct, and significant association with PSCs, confirming H1. The findings were consistent with the outcomes of the previous study, which highlighted the pivotal role of integrating sustainable curricula in teacher education programs to promote SDGs, emphasizing the development of PSCs among language students at a Finnish university36. Similarly, Iqbal, Yi19 found that CDF has a positive connection with students’ skills, which is a component of PSCs. Similarly, it was identified that sustainable curricula develop critical thinking, adaptability, and creative problem-solving, which are components of PSCs among higher education students39,40. Overall, these findings highlight the importance of integrated sustainable curricula in fostering PSCs. Students evaluated the curriculum as both relevant and personally transformative, particularly in developing their reflective thinking and values-creation competencies aligned with sustainability goals.

Likewise, the outcomes showed that sustainable curricula have a positive, direct, and significant connection with GenAITU, supporting H2. These results are aligned with previous findings, emphasizing that sustainable curricula are a positive contributor to GenAITU in sustainable education within technology-enhanced learning environments40. The results are also aligned with the framework proposed by Iqbal, Yi19 that effective CDF contributes to ICT Competencies among PSTs. Similarly, Shimizu, Kasai48 highlighted the significant connection between curricula and GenAITU in medical education. Overall, the findings highlight the significant role of sustainable curricula in enhancing GenAITU in sustainable education. Students reported that the use of GenAITU tools improved their ability to generate and refine ideas, facilitating practical applications of theoretical knowledge in blended classroom settings.

Similarly, the findings indicated that sustainable curricula have a positive, direct, and significant association with KSP, which supports H3. The results were consistent with the outcomes reported that learning environments enable KSP in a hybrid approach17. Zhao and Breslow54 concluded that educational environments are predictors of KSP in blended learning. Similarly, the boundaryless classroom allows KSP from a larger distance than the institution17,55. It was reported that sustainable curricula have an important role in fostering KSP42. Students’ feedback suggests that knowledge-sharing activities embedded in the curriculum encouraged peer learning, co-creation, and application of sustainable practices beyond the virtual classroom.

Moreover, the results showed that GenAITU had a positive, direct, and significant relationship with PSCs, which confirmed H4. The outcomes were consistent with the results reported in previous studies that GenAITU fosters collaboration among stakeholders and enhances sustainable innovation Competencies, which is a component of PSCs15,43. Moreover, advanced technologies play a pivotal role in developing critical student Competencies, such as creativity, critical thinking, problem-solving, and adaptability44. The findings highlight the transformative role of GenAITU in advancing PSCs that are aligned with the SDGs. Students viewed GenAITU as a catalyst for enhancing their innovation Competencies and as a supportive mechanism for independently exploring sustainability challenges.

Additionally, outcomes indicated that GenAITU has a positive, direct, and significant connection with PSCs, which supported H5. The results were consistent with the findings of previous studies that indicated advanced learning technologies such as GenAITU create immersive and interactive learning environments, strengthening communication and collaboration18,38. Moreover, GenAITU integration offers effective management of KSP in contemporary organizations22. These findings emphasize the crucial role of GenAITU in fostering effective KSP practices that enable technological advancement in university teacher education settings. GenAITU helps the global pursuit of such PSCs development15. This result suggests that GenAITU enhances PSCs and can be aligned with modern educational needs among PSTs in blended learning environments. Based on student experiences, GenAITU not only facilitated content creation and information synthesis but also stimulated collaborative dialogue, reinforcing shared responsibility for learning.

Correspondingly, the results showed that KSP has a positive, direct, and significant relationship with PSCs, which validated H6. The outcomes of this study were consistent with the results of a previous study that robust digital infrastructures facilitate collaboration and idea exchange, enhancing sustainable curricula in developing nations like Nepal45. Similarly, Ye, Liu46 found that KSP positively impacts innovation behavior among employees in China. Research has indicated that collaborative culture, knowledge sharing, and technical innovation Competencies, which are components of PSCs, are connected to the Small and Medium Enterprise environment in Pakistan56. This suggested the significance of KSP in the development of PSCs among PSTs in blended learning.

Furthermore, the results showed that sustainable curricula have an indirect and significant positive connection with PSCs through GenAITU, supporting H7. The results were consistent with findings of the previous study reported by Shimizu, Kasai48, which emphasized that GenAITU enhances the effectiveness of the educational curricula to foster PSCs. Similarly, Iqbal, Asghar44 reported that the curriculum indirectly enhances entrepreneurship through technology environments. Likewise, research has identified that advanced learning technologies can amplify the potential of sustainable curricula in fostering PSCs3,4,5,6. The results highlighted that GenAITU serves as a vital mediator in the connection between sustainable curricula and PSCs among PSTs in blended learning environments. These findings resonate with student narratives where digital tools were seen as enablers of active exploration, supporting meaningful learning journeys toward sustainability.

Similarly, the outcomes indicated that sustainable curricula have an indirect and significant positive relationship with PSCs in the presence of KSP, supporting H8. The outcomes were aligned with previous research results showing that the KSP enhanced competencies within a hybrid learning environment16,17. Similar results were reported by Iqbal, Yi19, who found that effective CDF indirectly led to enhanced entrepreneurial Competencies. Furthermore, Caldana, Eustachio6 found that formal, informal, and informal education can be helpful for the development of PSCs in hybrid learning in Brazil. The reason behind such results is that the effective integration of sustainable curricula can predict PSCs by facilitating effective KSP among PSTs. Students acknowledged that peer-led discussions and shared projects played a formative role in shaping their ability to critically engage with sustainability issues.

Likewise, the outcomes showed that sustainable curricula have an indirect and significant positive association with PSCs through GenAITU via KSP, which validated H9.The outcomes aligned with previous research on the relationship between educational technologies, curriculum design, and PSCs44. The same findings were given by Iqbal, Yi19 that ICT knowledge of information communication technologies strengthens the relationship between SC and entrepreneurial Competencies. In short, sustainable curricula and PSCs had a significant, positive, direct, and indirect association in the presence of GenAITU via KSP. This integrated pathway reflects the multi-layered nature of students’ learning experiences, where curriculum, technology, and collaboration converge to support applied, real-world learning.

Conclusions

The study was designed based on the CDF, UTAUT2, knowledge-sharing perspectives, and the SDGs. It examined the associations between sustainable curricula, GenAITU, knowledge-sharing practices (KSP), and PSCs in an emerging national context (China). The results showed a positive and significant association between sustainable curricula and PSCs. The study found that sustainable curricula were positively associated with GenAITU. Results also indicated a positive and direct relationship between sustainable curricula and KSP. Moreover, GenAITU was found to be positively associated with PSCs. Findings also revealed a positive and significant association between GenAITU and KSP. Additionally, KSP was positively associated with PSCs. The study identified that sustainable curricula had an indirect positive association with PSCs through GenAITU. Similarly, sustainable curricula showed an indirect positive association with PSCs through KSP. Furthermore, the study found that sustainable curricula had both direct and indirect positive associations with PSCs, with serial mediation via GenAITU and KSP.

The conclusions of this study are based on empirical findings and discussions, statistically validated using SmartPLS-SEM. They can be interpreted as follows:First, an effective sustainable curriculum may serve as a significant explanatory variable for PSCs. Second, sustainable curricula are positively associated with GenAITU. Third, sustainable curricula may also function as an explanatory variable for KSP. Fourth, GenAITU appears to be positively associated with PSCs. Fifth, GenAITU is linked to KSP. Sixth, KSP may act as an explanatory variable for PSCs. Seventh, GenAITU may serve as an intervening factor in the association between sustainable curricula and PSCs. Eighth, KSP may also mediate that association. Finally, GenAITU, through KSP, may serve as a serial mediator linking sustainable curricula to PSCs.

Theoretical contributions

Our study theoretically contributes to four key ways: First, it enhances understanding of the relationships between sustainable curricula, GenAITU, KSP, and PSCs, adapting CDF, UTAUT Model, knowledge sharing perspectives and SDGs as guiding research frameworks. Second, the study bridges a gap in the literature regarding how sustainable curricula influence PSCs, not only directly but also through the mediating roles of advanced technologies and collaborative KSP, by introducing a novel serial mediation model. Third, it expands existing theoretical perspectives by highlighting the strategic role of GenAITU in developing PSCs within higher education. Finally, this study adds to the theoretical discourse on the impact of sustainable curricula in the context of emerging technologies, providing a comprehensive framework that connects education, technology, and PSCs in blended learning.

Practical implications

The current study provides some valuable and practical implications for teachers, curriculum policymakers, and university administration. The results indicated that sustainable curricula, GenAITU, and KSP enhanced PSCs among students. Practitioners can use these results to design and implement a sustainable curriculum that integrates GenAITU and KSP. Faculty should adopt strategies to make more effective, sustainable curricula with the integration of GenAITU to encourage students to actively participate in KSP, which can increase PSCs. Curriculum policymakers can leverage these findings to develop better policies that promote the integration of sustainable curricula and GenAITU in universities’curricula. Policymakers can develop policies that advance curricula to create a balance between traditional and AI-driven frameworks that support PSCs’development among PSTs. Universities should invest in the professional development of their faculty to integrate GenAITU into their teaching strategies. Universities’ management should focus on building infrastructure to support GenAITU, such as labs, equipment, and software, to improve KSP, which ultimately increases the effectiveness of sustainable curricula to enhance PSCs. Students benefit from the adoption of sustainable curricula that actively engage them in GenAITU and knowledge-sharing platforms. They can use GenAITU in sustainable curricula to equip themselves with the required PSCs to tackle complex sustainable challenges. For curriculum designers, this study recommends embedding AI-integrated, sustainability-oriented modules that encourage project-based learning, peer collaboration, and interdisciplinary thinking. These strategies support the development of critical, digital, and sustainability competencies in an applied context. For teacher education programs, structured training on the ethical and pedagogical use of Generative AI tools should be incorporated into practicum and coursework. Scenario-based learning that simulates real classroom challenges involving sustainability issues can equip PSTs to make informed, responsible decisions using AI. For policymakers, the findings suggest the need to align national teacher education standards with SDGs and AI literacy benchmarks. Policymakers should support funding mechanisms and institutional guidelines that promote the use of emerging technologies for sustainability-oriented teacher preparation.

Limitations and future research

Our study has a few limitations. This study was conducted in a specific region of China. This may limit the generalizability of the findings. Futu re research should replicate this study across different countries and cultural contexts. This validates the results and provides broader insights. We collected data only from students from educational programs, excluding other majors such as business, engineering, and medical programs. Future research should include these programs to explore this phenomenon across different academic disciplines. The use of a time-lag survey limits the ability to establish causal relationships between variables. Longitudinal studies should be conducted to track changes in PSCs. The long-term effects of sustainable curricula, GenAITU, and KSP should also be examined. This study did not explore the full range of GenAITU, such as blockchain and augmented reality. Future studies should investigate how different technologies can influence PSCs and KSP in higher education. Moreover, we adapted the UTAUT model; future studies could use the full-scale UTAUT2 model to explore this phenomenon. While our study highlights the potential benefits of GenAITU, we did not suggest that its use could lead to issues such as laziness, which has been underplayed or overlooked. Future research should address these concerns and explore the negative impact of GenAITU on students’learning processes. The overrepresentation of male respondents may limit the generalizability of the findings to the broader population of preservice teachers. Future studies should aim for more gender-balanced samples to improve external validity. Lastly, given the survey and self-report nature of the data, along with the non-experimental design, causal conclusions cannot be drawn. Future research employing longitudinal or experimental methods is recommended to validate the directionality and causality of these associations more robustly.