Bandura’s (2001) social cognitive theory emphasizes the interaction among individual cognition, behavior, and the environment, providing an integrative framework for understanding human behavior in specific contexts. According to this theory, this study regards classroom innovation atmosphere as a key environmental factor, defines AI technology application as specific technology usage behavior, and treats intrinsic motivation and self-efficacy as motivation-related cognitive and belief-related cognitive variables, respectively. Together, they form a “environment-cognition-behavior-outcome” causal chain to explore the formation mechanism of student creativity.

In social cognitive theory, behavior is not solely influenced by the environment but also regulated and driven by individual cognition. As a supportive environment, an innovative classroom atmosphere can reduce students’ perceived risks and barriers when adopting new technologies, encouraging them to view AI as a tool for exploration and creation. This, in turn, enhances their creative performance through technological empowerment. The environment indirectly affects behavior and outcomes by activating individuals’ cognitive processes. When the classroom atmosphere encourages free exploration, students are more likely to develop intrinsic interest in learning itself, leading to sustained engagement in creative activities. From an intrinsic perspective, the adoption of AI technology is an “instrumental behavior” that relies on individuals’ self-confidence in their abilities and the supportive forces of the environment. Social cognitive theory clearly indicates that individuals with high self-efficacy are more willing to translate environmental opportunities into actual actions, demonstrating stronger adaptability and persistence when facing new technologies (Van Dinther et al., 2011). Self-efficacy not only amplifies the positive effect of an innovative classroom atmosphere on AI technology adoption but also strengthens individual persistence in innovative environments, creating a reinforcing cycle of “environment-cognition-behavior.”

Social cognitive theory was ultimately adopted as the theoretical framework for this study because its triadic reciprocal determinism model can simultaneously encompass multi-level interactions among environment, cognition, and behavior. It effectively explains the mediating role of AI technology usage between innovation atmosphere and creativity—not merely as behavioral responses to environmental stimuli, but also as instrumental practical activities driven by cognitive beliefs. This theoretical choice enables the study to transcend traditional single-path analysis and reveal the cognitive-behavioral mechanisms of student creativity cultivation within a dynamic, integrated framework.

Classroom innovation atmosphere refers to the multidimensional set of factors in the classroom environment that stimulate and support students’ creative thinking and behaviors. It emphasizes the classroom as a dynamic system, showcasing innovation-oriented characteristics in aspects such as teachers’ instructional performance, forms of student interaction, and the arrangement of teaching tasks. Its core components include supportive teacher behaviors, open communication and collaboration among students, and the design of challenging and open-ended learning tasks. Together, these elements create an environment where students feel psychologically secure, bold to experiment, and enthusiastic about exploring the unknown. Compared to the general concepts of “learning atmosphere” or “classroom environment,” classroom innovation atmosphere focuses more on the factors that directly promote innovative steps. Learning atmosphere typically covers broader academic, managerial, and emotional dimensions, whereas classroom innovation atmosphere refers to specific environmental conditions that inspire the generation of ideas, support unconventional thinking, and encourage practical exploration. Traditional classroom environments may emphasize discipline and knowledge transmission, while classroom innovation atmosphere highlights openness, exploration, and support.

In this study, we adopted and adapted the Classroom Environment Scale developed by Fraser. This scale measures multiple dimensions, such as communication atmosphere, teacher encouragement, and innovation support, allowing us to effectively quantify this construct and examine its relationship with student creativity.

Creativity is defined as the ability to generate novel and useful ideas (Amabile et al., 2018), serving as a critical foundation for individuals, teams, and organizations to achieve innovation (Shalley et al., 2004). Research on creativity primarily unfolds from three perspectives: the individual perspective, the environmental perspective, and the interaction perspective (Lee and Lee, 2023). The individual perspective focuses on the effects of cognition, personality, and motivation on creativity; the environmental perspective examines how external conditions, cultural atmosphere, and social support facilitate creative thinking; the interaction perspective investigates the dynamic relationship between individuals and their environment, demonstrating their mutual influence on creativity development.

From the individual perspective, cognition and personality traits significantly impact creativity, with openness showing a positive correlation with creative ability (McCrae and Costa, 1997). Individual motivation is also a key factor influencing creativity—just as intrinsic motivation drives students who develop an interest in a subject to explore it deeply and innovate continuously.

The environmental perspective highlights the influence of external conditions on creativity formation. Existing research indicates that people’s work environments are closely linked to their creativity (Deng et al., 2016), and a supportive work environment can stimulate individuals’ creative potential (Amabile, 1997). In education, fostering student creativity is inseparable from the “environmental perspective” of classroom innovation. When teachers cultivate an inclusive classroom atmosphere, students are more likely to express their ideas freely, thereby activating creative thinking. Teachers can employ teaching strategies such as open-ended questions, teamwork, and creative activities to facilitate group discussions and brainstorming, effectively creating a challenging learning environment that encourages students to approach problems from diverse perspectives. This nurtures their ability for creative expression and enhances problem-solving skills (Collard and Looney, 2014).

The interaction perspective emphasizes the dynamic relationship between individuals and their environment. Creativity does not develop unidirectionally but emerges through mutual influence between individuals and their surroundings—individual creativity often flourishes when supported by conducive environments (Sawyer and Henriksen, 2023). Thus, students’ creative expressions in class are closely tied to classroom atmosphere. Based on prior theoretical and empirical research, this study proposes the hypothesis:

The application scope of AI technology in education continues to expand, particularly in higher education, where artificial intelligence has been introduced into classroom activities. This not only provides students with new learning tools but also serves as a crucial enabling method for fostering student creativity. From a theoretical mechanism perspective, AI technology applications primarily enhance creativity through three pathways: cognitive offloading, cognitive augmentation, and supporting exploration in novel idea spaces. Cognitive offloading refers to AI’s ability to automate tedious tasks such as information gathering and data computation, thereby freeing students’ cognitive resources from routine, low-level mental burdens and enabling them to focus more on high-level creative planning and critical thinking. Cognitive augmentation means AI tools can provide information, perspectives, and analytical frameworks that surpass an individual’s prior knowledge base, thereby expanding students’ cognitive boundaries and creating favorable conditions for cross-disciplinary associations and knowledge integration. Supporting exploration in novel idea spaces manifests as AI’s capability to rapidly generate multiple alternative solutions and simulate outcomes under various scenarios, allowing students to conduct low-cost mental experimentation—thus encouraging bold trial-and-error and continuous optimization. These advantages are often difficult to achieve in traditional classroom settings due to time and resource constraints.

Creativity is a core element driving individual and societal progress, and high-quality educational environments form the foundation for cultivating such creativity (Csikszentmihalyi, 1996). Classrooms with a strong innovation atmosphere demonstrate significant effects in enhancing student creativity, while also encouraging active student participation and application of AI technologies. Teachers’ guidance plays an equally critical role—educators can create open learning environments that stimulate students’ thinking and experimentation, thereby facilitating easier adoption and utilization of AI technologies (Runco, 2014). For instance, teachers may employ AI tools to implement personalized instruction based on students’ interests and capabilities, which not only improves learning outcomes but also strengthens students’ creative thinking.

Students can also skillfully utilize AI technology to facilitate their creative thinking activities more smoothly. For instance, by using AI for data analysis and situational exploration, students can more swiftly and efficiently uncover the essence of complex problems. With AI’s assistance in generating multiple potential solutions, they can explore a broader idea space and ultimately develop innovative responses. AI technology further assists students in conducting self-reflection and evaluation of their learning, thereby promoting self-improvement. Effective application of AI technology not only relies on classroom innovation atmosphere but can reciprocally enhance this atmosphere’s vitality. During learning processes, AI provides real-time feedback and personalized learning suggestions, encouraging more active student engagement in creative tasks (Deci and Ryan, 2000). It transcends being merely a tool, serving instead as a powerful support for classroom innovation and creativity enhancement.

Artificial intelligence is a broad discipline encompassing multiple branches, each with unique research directions and application scenarios. In terms of its role in the creative process, AI technology in education can generally be divided into two categories: generative AI, which autonomously produces entirely new content such as text, code, and images, directly participating in the ideation phase; and instrumental AI, which primarily leverages functions like data analysis, intelligent tutoring, and process optimization to support students’ exploration and practice, facilitating the implementation phase of creativity. This study primarily focuses on instrumental AI technology—applications that serve as aids to help students more efficiently gather information, analyze data, refine solutions, and execute tasks. Examples include intelligent learning systems, data mining tools, and adaptive teaching platforms. The key role of these technologies lies in providing a foundation for deeper creative thinking and exploration through “cognitive offloading” and “cognitive extension,” which aligns more closely with the theoretical framework this study seeks to explore—leveraging environmental support to activate individuals’ intrinsic agency. Based on this analysis, we propose the following hypothesis:

Intrinsic motivation refers to the motivation generated by an individual’s interest in the learning activity itself during the learning process. Unlike extrinsic motivation, intrinsic motivation emphasizes the emotional satisfaction and sense of achievement brought by the learning activity itself. A classroom innovation environment that encourages free thinking and maintains a lively atmosphere is more likely to stimulate students’ intrinsic motivation, thereby promoting greater engagement in learning (Fan and Cai, 2022).

When students perceive learning activities as both challenging and interesting, they are more likely to invest time and effort into learning to obtain a sense of fulfillment. Teachers can employ teaching methods such as open-ended questions and cooperative learning to create an innovative classroom atmosphere, inspiring students to engage in deep thinking and discussion while experiencing the joy of learning during exploration. Meanwhile, timely encouragement from teachers is essential, as these behaviors help students recognize their continuous progress, thereby motivating them to continue exploring and seeking knowledge. Intrinsic motivation is also a key factor influencing students’ creativity levels, with existing research confirming the positive relationship between intrinsic motivation and creativity (Ryan and Deci, 2000). Zhang and Bartol’s (2010) study demonstrates that intrinsic motivation promotes creative performance by influencing individuals’ engagement in the creative process. Based on the above analysis, this study proposes the following hypothesis:

Self-efficacy refers to an individual’s belief in their own capabilities, a concept derived from Albert Bandura’s social cognitive theory (Bandura, 2000). Bandura posited that such beliefs not only influence people’s choice of behaviors but also affect their persistence levels and effort investment when facing challenges (Bandura, 2001). In educational contexts, this belief is particularly crucial as it directly impacts students’ learning motivation and strategies, ultimately affecting their academic outcomes. According to social cognitive theory, environmental influences on individual behavior and internal states do not uniformly affect everyone but are significantly modulated by individual cognitive factors. A classroom environment that fosters innovation provides opportunities and possibilities, but whether and how individuals utilize these environmental resources depends on their beliefs about their own capabilities. Treating self-efficacy as a moderating variable is key to transitioning from an “environment-driven” model to an “individual-environment interaction” model, which is essential for understanding the mechanisms of innovative classroom atmospheres.

Self-efficacy serves as a key psychological variable that moderates intrinsic motivation by influencing individuals’ cognitive processes and emotional responses (Schunk, 2003). Students with stronger self-efficacy typically demonstrate higher intrinsic motivation when approaching learning tasks, exhibit greater enthusiasm for learning, and are more willing to challenge themselves and overcome difficulties, thereby gaining more sense of achievement and satisfaction during learning (Zimmerman, 2000). This positive mindset enables them to maintain higher academic focus and persevere when encountering complex problems. Particularly when introducing emerging technologies like AI, the process itself comes with uncertainties and challenges. Students with high self-efficacy are more likely to perceive the encouragement of technological application in an innovative classroom as a challenging opportunity. They are confident in their ability to master and effectively use AI tools (Pajares, 1996), thus more actively and boldly translating environmental support into concrete exploratory actions. In contrast, students with low self-efficacy may hesitate even in an innovation-encouraging environment due to doubts about their technical competence, making it difficult for them to effectively initiate the “environment-technology application-creativity” mediating pathway. The purpose of a classroom innovation atmosphere is to stimulate students’ intrinsic motivation, but this effect also relies on students’ self-beliefs. Students with high self-efficacy, when faced with open-ended, exploratory learning tasks, are more confident in enjoying the thrill of challenging themselves. The environment more readily activates and sustains their intrinsic motivation. However, for students with low self-efficacy, the same open-ended tasks may cause them to focus more on the possibility of failure rather than the intrinsic joy of exploration, making it harder for their intrinsic motivation to be strongly ignited.

Therefore, investigating how self-efficacy moderates the influence of classroom innovation atmosphere on AI technology adoption and individual intrinsic motivation holds significant implications for cultivating university students’ creativity in the digital era. Based on the above analysis, this study proposes the following hypotheses:

In summary, we have constructed the following theoretical model (as shown in Figure 1).

This study focuses on university students in the Jiangsu-Zhejiang-Shanghai region, collecting data through questionnaire surveys. This region is one of China’s most economically developed areas with the highest concentration of educational resources. Local universities generally rank at the forefront in terms of teaching and educational reforms, as well as the integration and application of information technology, demonstrating a high level of acceptance and proficiency in innovative teaching tools like AI technology. This provides a rich context for studying classroom innovation and the application of AI. The region has always prioritized cultivating innovative talent, with diverse and representative academic programs and teaching models across its universities, reflecting the current practices and challenges of creativity cultivation in Chinese higher education. Therefore, selecting university students from this region as research subjects enhances the relevance of the study sample and the insightful value of the research findings. Electronic questionnaires were distributed to first- through fourth-year students via online platforms across various universities, with data collection conducted in two phases spaced 1 month apart. To ensure consistency between the two phases, respondents were required to use “initials + last four digits of phone number” as identifiers for subsequent screening and matching. The first phase collected students’ basic information and classroom innovation capability data, distributing 420 questionnaires and obtaining 369 valid responses. The second phase gathered data on student creativity, AI technology application, intrinsic motivation, and self-efficacy, distributing 369 questionnaires and yielding 270 valid samples after removing invalid responses, representing a valid response rate of 73.2% (Table 1).

All measurement scales used in this study were established instruments, with all variables rated on a 5-point Likert scale, as detailed in Table 2. This study selected student gender, grade level, location, and major category as control variables. Individuals of different genders may exhibit variations in cognitive styles, risk preferences, and creative performance. Controlling for gender can eliminate its potential confounding effects. The depth of education received, knowledge reserves, and university experiences vary among students across different grade levels, which may lead to changes in their creativity levels and development. Even within the Jiangsu-Zhejiang-Shanghai region, subtle differences may still exist among provinces and cities in terms of educational resources, cultural environments, and policy support. Controlling for the location variable can exclude potential regional-level influences. Different academic disciplines vary in knowledge structures, thinking training models, and evaluation criteria, imposing distinct requirements and shaping effects on students’ creativity and innovative behaviors. Controlling for major categories aims to capture variations resulting from disciplinary background differences.

Classroom Innovation Atmosphere Scale: This study employs the classroom environment scale developed by Fraser (1982). To ensure its applicability, corresponding adjustments were made based on the research context, totaling 7 items.

Student Creativity Scale: This study employs the creativity scale developed by Runco et al. (2001) with appropriate adaptations based on the research context, consisting of five items.

AI Technology Application Scale: This study references the measurement scale developed by Man Tang et al. (2022) and the challenge appraisal subscale from the challenge-threat stress appraisal scale developed by LePine et al. (2016), with appropriate modifications made for the AI context, totaling 4 items.

Intrinsic Motivation Scale: Based on Amabile’s theoretical framework and related pilot studies (Amabile, 1985). This scale’s items were designed to assess students’ level of intrinsic motivation in classroom activities, totaling 4 items.

Self-Efficacy Scale: This study employs the innovative self-efficacy scale developed by Tierney and Farmer (2002), which consists of 4 items designed to assess individuals’ confidence in their success during innovative activities.

This study is based on a regression analysis framework, utilizing the PROCESS macro program (Model 7) developed by Hayes to analyze mediating, moderating, and moderated mediation effects. The Bootstrap sampling method was employed to derive 95% confidence intervals, testing the significance of indirect and conditional effects to ensure the robustness of the analysis results.

This study used SPSS 27.0 software to analyze the reliability and validity of the variables, with the results shown in Table 2.

Although within the measurement model certain individual item factor loadings did not meet the recommended threshold of 0.7, these values still exceeded the minimum requirement of 0.5, indicating acceptable explanatory power. The composite reliability and average variance extracted for all variables met the standards, demonstrating high overall reliability and validity of the measurement model. This study concludes that the lower loadings of individual items do not compromise the reliability of the research findings.

This study conducted correlation analysis on the relationships between variables, with results shown in Table 3. The findings indicate that classroom innovation atmosphere is significantly positively correlated with student creativity (β = 0.169, p < 0.01); classroom innovation atmosphere shows significant positive correlations with both AI technology application and intrinsic motivation (β = 0.139, p < 0.05; β = 0.166, p < 0.01); while AI technology application and intrinsic motivation both demonstrate significant positive correlations with creativity (β = 0.186, p < 0.01; β = 0.585, p < 0.001). These results provide preliminary support for the research hypotheses.

This study explored the mechanism of how classroom innovation atmosphere affects university students’ creativity, yielding the following main conclusions.

This study constructs and validates a moderated mediation model, demonstrating that the classroom innovation atmosphere not only directly enhances student creativity but also operates indirectly through two pathways—promoting AI technology application and fostering intrinsic motivation—with students’ self-efficacy playing a significant positive moderating role in this process. When teachers cultivate an open and inclusive classroom environment, it effectively stimulates students’ creative thinking and enhances their innovation capacity. In such an environment, students can freely express their ideas and actively experiment with new approaches, thereby promoting the development of their creative thinking. This finding breaks through the limitations of previous research that primarily focused on singular pathways of influence. In the relationship between classroom innovation atmosphere and student creativity, the application of AI technology and intrinsic motivation play important mediating roles. This finding highlights the positive effects of classroom innovation atmosphere, which can not only directly impact students’ creativity but also facilitate the effective utilization of AI technology and enhance students’ intrinsic motivation, thereby indirectly improving their creative performance. When designing curricula, educators should consider how to effectively leverage AI technology to enhance students’ creativity. Self-efficacy plays a positive moderating role in the relationship between classroom innovation atmosphere and both AI technology application and intrinsic motivation, further revealing the significance of individual self-efficacy in fostering an innovative classroom atmosphere. When students develop confidence in their abilities, they are more likely to actively engage in innovative classroom environments and fully utilize AI technology for learning and creation. High self-efficacy not only stimulates students’ intrinsic motivation but also strengthens their practical ability to cope with challenges, thereby further promoting creativity enhancement.

This study provides crucial theoretical support and practical insights for educators in fostering students’ creativity, emphasizing the importance of integrating innovation atmosphere, AI technology application, intrinsic motivation, and self-efficacy in classroom teaching. The comprehensive theoretical framework presented analyzes the intrinsic logic of dynamic interactions among these factors, pointing the way for future educational research and practice.

From a theoretical perspective, the primary contribution of this study lies in constructing and validating a moderated multiple mediation model, which dissects the intrinsic mechanism through which classroom innovation atmosphere influences student creativity via the application of AI technology and intrinsic motivation, while also highlighting the enhancing moderating role of self-efficacy. This integrated model addresses a theoretical gap in existing literature regarding the synergistic promotion of creativity formation by “technology usage” and “psychological motivation” in classroom settings. The findings resonate with and extend the core concepts of the Technology Acceptance Model (TAM) and UTAUT. A positive classroom innovation atmosphere effectively creates facilitating conditions emphasized in TAM, lowering students’ barriers to adopting AI technology and reducing perceived risks, which aligns with the construct of perceived ease of use. Furthermore, this study reveals that in innovative educational contexts, technology adoption is not merely driven by utilitarian evaluations but also by exploratory interest and innovation-driven proactive actions. Social cognitive theory, with its emphasis on the interaction between individual agency and environmental factors, provides deeper insights into this phenomenon.

This study offers empirical support for the relationship between classroom innovative atmosphere and student creativity. The research proposes that classroom innovative atmosphere is a key factor in enhancing students’ creativity, further demonstrating the importance of environmental factors in education (Wang et al., 2023). The classroom innovative atmosphere encompasses not only teachers’ instructional methods but also student interaction and collaboration. A dynamic classroom environment encourages students to boldly experiment with new ideas and helps cultivate their creative thinking. This finding not only provides empirical evidence for educational theory but also suggests new directions for future research.

The application of AI technology and intrinsic motivation play mediating roles in the relationship between classroom innovative atmosphere and creativity. With rapid technological advancements, educators face the challenge of effectively utilizing modern AI technology to enhance students’ learning capabilities. Applying AI technology can not only enrich learning resources but also stimulate students’ intrinsic motivation, making them more proactive in engaging in creative activities (Deci and Ryan, 2000). In the process of using AI technology to promote students’ innovative abilities, educators must not overlook the importance of humanistic care. Education is not merely about imparting knowledge but also about fostering students’ holistic development. Teachers should encourage students to think boldly, thereby igniting their intrinsic motivation and promoting creativity cultivation. This further highlights that intrinsic motivation is a crucial element in driving creativity (Ballesteros et al., 2023).

This study also explores the moderating role of self-efficacy in the relationship between classroom innovation atmosphere, AI technology application, and intrinsic motivation. The results once again confirm that self-efficacy plays a crucial role in the learning process (Chiu et al., 2023). Amid the growing rise of AI in educational applications, this study clarifies the mediating role of instrumental AI technology as an enabling pathway in the relationship between innovative atmosphere and creativity. It addresses the practical challenge of how technology integrates into educational ecosystems to drive innovation by incorporating self-efficacy into the model. The research reveals self-efficacy as a key individual cognitive resource that enhances the environmental push for technology adoption while strengthening the stimulating effect of the environment on intrinsic motivation. This expands the application scope of social cognitive theory in digital educational settings. Enhancing self-efficacy not only boosts students’ learning confidence but also motivates them to actively engage in innovative practices. These findings suggest that educators should pay attention to students’ self-efficacy when designing curricula, employing practical methods to strengthen their confidence. By providing encouragement and helping students recognize their progress, educators can further facilitate the development of their creativity. This study reveals an orderly chain of effects—"environmental support, technology empowerment, motivation stimulation, and belief regulation”—establishing a more comprehensive and dynamic theoretical framework for understanding the cultivation mechanisms of student creativity. These findings not only offer theoretical guidance for educators in fostering students’ creativity but also open up new directions for future research. Educators can leverage this theoretical framework to develop more targeted teaching strategies, thereby promoting the cultivation of students’ innovative abilities and the enhancement of their comprehensive competencies.

Based on the findings of this study, the following recommendations are proposed for educational practices in higher education institutions to more effectively cultivate students’ creativity.

Educators should strive to build an innovative learning environment supported by AI technology. Teachers should incorporate data analysis tools into curriculum design and utilize intelligent literature management software to assist in research, freeing students from the cognitive burden of complex information processing. For example, “human-machine collaboration” research projects can be designed, explicitly requiring students to use these tools for tasks such as data cleaning, visualization, or literature tracing. The saved effort can then be directed toward proposing innovative research hypotheses, deeply interpreting analytical results, and explaining their theoretical and practical significance. Course objectives should clearly define the supportive role of AI, emphasizing its value in extending rather than replacing human intellect, ensuring that students hone their critical thinking and original conceptual abilities while adopting such technologies.

To ensure the healthy development of technology applications, schools should introduce concise guidelines for the use of artificial intelligence. These guidelines must first establish clear boundaries for academic integrity, prohibiting the use of AI for automated writing, data manipulation, and similar misconduct, while incorporating such violations into academic integrity policies. More importantly, they should guide students in developing responsible usage habits. When using AI for data processing tasks, students must specify the tools employed, the analytical procedures followed, and their interpretation and verification of the results in their assignments or reports, thereby enhancing scientific rigor and transparency in research. Institutions may provide standardized citation and disclosure templates to facilitate student compliance and faculty verification.

Student self-efficacy is a key factor in facilitating environmental support and technological innovation. In the teaching process, it is necessary to incorporate strategic methods to enhance technological confidence. Before introducing new AI software, teachers should provide micro-training tutorials with clear tasks and steps, rather than complex theoretical explanations, thereby reducing students’ technical anxiety at the initial stage. When assigning complex projects, the task segmentation method should be employed—breaking large projects into smaller components—and timely feedback should be given at each stage, enabling students to visibly recognize their progress and skill development. Teachers must pay special attention to openly acknowledging students’ exploratory spirit when using tools to solve practical problems. This process-oriented praise is particularly effective in building long-term, stable technological confidence.

As the trend of artificial intelligence becoming ubiquitous grows increasingly evident, the evaluation of students’ creativity must also keep pace with the times. The assessment system should shift from a singular focus on verifying outcomes to a comprehensive approach incorporating process-based portfolios. Schools should encourage students to submit assignments that include complete documentation of their AI utilization workflows—such as screenshots of key data preprocessing steps, comparisons of analytical results with different parameters and reasoning behind selections, as well as independent conclusions and innovative inferences derived from AI outputs. Evaluation criteria must explicitly establish a “depth of thinking empowered by technology” dimension, emphasizing how students leverage tools to expand cognitive horizons and what distinctive integration, critical analysis, and creative achievements they accomplish on this foundation.

The support of school administrators is the backbone for creating a sustainable innovation ecosystem. It is proposed that the school consolidate its current software resources and establish a unified AI platform for teaching tools, providing students with licensed software and stable technical assistance. Additionally, the school should initiate an “AI-empowered Teaching Innovation Seed Fund” to encourage teachers to apply for small-scale reform projects that deeply integrate specific AI tools into teaching modules, while funding the transformation of successful experiences into shareable teaching case studies. Regularly hosting “Human-Machine Collaborative Innovation Project Exhibitions” to showcase outstanding student works that utilize AI to solve real-world problems will foster a campus culture that actively embraces technology for exploration and creation.

Ultimately, educators should adopt a long-term vision to cultivate a continuously improving ecosystem of innovation. A positive, secure, and inspiring classroom environment serves as fertile ground for creativity to thrive. Thoughtful integration of AI technology can spark students’ exploratory drive and sense of competence. In teaching practices, it is crucial to emphasize the synergistic effects of intrinsic motivation, self-efficacy, and technological empowerment, helping students not only find joy in acquiring knowledge but also build self-worth through creative exploration. Creating an innovative classroom atmosphere and nurturing creativity cannot be achieved overnight—it is an evolutionary process requiring continuous evaluation and dynamic adjustment. Educators must develop keen insight and reflective capacity, conducting systematic analysis of teaching strategies, student feedback, and innovation outcomes at regular intervals. By promptly optimizing implementation pathways to align with students’ developmental needs and external environmental shifts, educators can guide students into deeper exploration and autonomous innovation. This multi-layered, sustained systematic approach is essential for cultivating highly creative talent reserves for society.

This study has the following limitations. Although the sample was drawn from universities in the Jiangsu-Zhejiang-Shanghai region and possesses certain representativeness, it may still not fully reflect the developmental landscape of student creativity nationwide. Future research could consider expanding the sample scope to include more universities from diverse regions and types, thereby enhancing the validity of the findings. The development of creativity is an ongoing process that may be influenced by multiple factors. Future studies should also examine the long-term impact of classroom innovation atmosphere. This study primarily constructs a model based on social cognitive theory. Future research could introduce and compare other theoretical frameworks, integrating the core variables of the technology acceptance model with the psychological motivation variables of this study to create a more predictive unified model. Alternatively, by leveraging the computational creativity framework, specific characteristics of student-AI collaborative creations could be analyzed, providing objective metrics for assessing creativity. Additionally, The impact mechanisms of generative AI technology on classroom innovation atmosphere and student creativity may differ from those of instrumental AI technology, the primary focus of this study, warranting further validation in future research. This study has certain limitations in the selection of control variables. Although we included common demographic variables such as gender, grade level, location, and major category as control variables, none showed significant influence in the model nor substantially improved the model’s explanatory power. Future research could consider introducing more contextually relevant control variables, such as course type or discipline-specific intrinsic constraints on creativity. These variables may more directly affect students’ behavior and cognitive processes in innovative environments, thereby facilitating a more precise analysis of the mechanisms linking classroom innovation atmosphere and student creativity. While this study provides some theoretical foundations and practical guidance regarding the influence of classroom innovation atmosphere on student creativity, subsequent research needs to improve in areas such as sample diversity and methodological richness to more deeply explore the mechanisms through which classroom innovation atmosphere affects student creativity.