Abstract: Innovation performance is an important metric for evaluating the contributions of researchers. As a strategic technology propelling the new wave of technological transformation, artificial intelligence (AI) is reshaping research paradigms across disciplines. However, prior studies have primarily focused on the direct impact of AI adoption on research activities, neglecting the role of individual differences among researchers in shaping Innovation performance. This study adopts the Stimulus-Organism-Response (SOR) theoretical model to explore the relationship between AI adoption and researchers' Innovation performance. The SOR framework posits that external stimuli influence internal psychological states, leading to specific outcomes. In this context, AI adoption acts as a stimulus that influences researchers’ psychological perceptions, which in turn affect their innovation performance. Drawing on Self-Determination Theory (SDT), the study incorporates three key psychological factors—research autonomy, research competence, and research relatedness—as mediators to capture individual differences in how researchers experience AI adoption and to provide a deeper understanding of its effects on innovation. Based on empirical data from 252 researchers, this study provides three principal findings. First, AI adoption positively influences the Innovation performance of researchers. Second, the positive relationship between AI adoption and innovation performance is mediated by research autonomy, competence, and relatedness, demonstrating the critical role of these psychological factors. Third, the organizational innovation climate moderates the impact of AI adoption on research competence and relatedness, underscoring the importance of a supportive environment for maximizing the benefits of AI in research. This study makes three key theoretical contributions. First, it extends the theoretical boundaries and application scope of the SOR model, revealing the underlying mechanisms—previously a “black box”—through which AI adoption influences researchers’ Innovation performance. Second, it offers an explanation for the divergences in research on AI technology and Innovation performance at the micro level. Third, it elucidates the mediating mechanisms and boundary conditions between AI adoption and the Innovation performance of researchers. This study provides three practical insights for exploring AI-driven research management. First, it emphasizes actively embracing AI technologies in research while adhering to research ethics and integrity standards. Second, it highlights the importance of fostering an innovative organizational climate that encourages originality and exploration, with particular attention to researchers’ autonomy, competence, and relatedness. Third, it recommends promoting the development of researchers' AI-related skills to apply them in major scientific studies and technological innovations. Future work could examine the mechanisms through which different organizational structures and modes impact Innovation performance. Additionally, the relevance of AI technologies varies across research fields, it could focus on specific subfields to provide deeper insights.