Enhancing sustainable innovation in AI companies: the role of perceived organizational support, job satisfaction, and job embeddedness.

Introduction: With the rapid advancement of the artificial intelligence (AI) industry, the demand for employee innovation performance has become increasingly critical for enterprise competitiveness. However, how organizational factors such as perceived organizational support (POS) influence innovation performance among AI employees remains underexplored. Methods: Drawing upon Social Exchange Theory and Self-Determination Theory, this study develops and empirically tests a moderated mediation model. Data were collected through a questionnaire survey of employees from AI enterprises of varying sizes (n = 536). The model examines the mediating effects of job satisfaction and job embeddedness, as well as the moderating effect of company size, on the relationship between POS and innovation performance. Results: The findings reveal that perceived organizational support indirectly enhances innovation performance through job satisfaction and job embeddedness. The direct effect of POS on innovation performance was not significant, indicating a fully mediated relationship. Moreover, company size significantly moderated the link between POS and job satisfaction, with the effect being stronger in smaller firms. Discussion: These results highlight the importance of employee attitudes as psychological mechanisms translating organizational support into innovative outcomes. The study provides theoretical insight into how POS functions differently across firm sizes and offers practical implications for tailoring HR strategies-such as enhancing perceived support and fostering job satisfaction-to strengthen innovation capacity in AI enterprises.