Unlocking research output with ChatGPT- 4 and SciSpace Ai through the mediating and moderating roles of research orientation

Abstract This study addresses a critical gap in the literature by examining the limited empirical evidence on how artificial intelligence (AI) tools interact with researchers’ cognitive orientations to influence research productivity. Specifically, it develops and tests the novel AI–Research Output (AI-RO) Model, which integrates the Technology Acceptance Model (TAM) and Socio-Technical Systems Theory to explain both the direct and conditional relationships between AI use and research output. Using an explanatory cross-sectional design, data were collected from 503 academic researchers across 20 public and private universities in Ghana. Structural Equation Modeling (SEM) was employed to examine direct, mediating, and moderating relationships among ChatGPT-4, SciSpace Ai, research orientation (RO), and research output (Rout). The findings show that ChatGPT − 4 (β = 0.387, p < 0.001) and SciSpace Ai (β = 0.182, p < 0.001) are positively associated with research output, while both tools are also significantly associated with research orientation (ChatGPT-4: β = 0.395; SciSpace Ai: β = 0.435; p < 0.001). Research orientation is positively associated with research output (β = 0.289, p < 0.001). Mediation analysis indicates that RO is significantly associated with the relationships between ChatGT- 4 (β = 0.114, p = 0.002) and SciSpace Ai (β = 0.126, p = 0.001) and research output. Moderation results reveal a differential interaction pattern, in which RO weakens the association between ChatGPT − 4 and research output (β = −0.204, p = 0.002) but strengthens the association with SciSpace AI (β = 0.130, p = 0.036). The study contributes to theory by advancing a dual mediation–moderation framework that explains how human cognitive orientation conditions AI–research relationships, extending TAM and Socio-Technical Systems perspectives. From a practical standpoint, the findings highlight the importance of aligning AI tools with researchers’ methodological capabilities through targeted training and institutional support. From a policy perspective, the results underscore the need for structured AI governance frameworks that promote responsible, ethical, and context-sensitive integration of AI in academic research. The study demonstrates that the effectiveness of AI in research is contingent not only on technological capability but also on the cognitive and institutional conditions under which it is used.