Enterprise carbon emissions intensity (CEI) is an important measure of corporate environmental responsibility management. Existing studies exploring the net effects of single factors are insufficient to fully reveal the complex causal relationships behind them. Taking the top 40 listed manufacturing enterprises in China as examples, this study adopted the technological, organizational, and environmental (TOE) framework and selected green technology innovation, digital transformation, dual carbon leadership, financial redundancy, environmental regulatory pressure, and public environmental concern as the key antecedents influencing CEI. This study employs necessary condition analysis (NCA) and fuzzy set qualitative comparative analysis (fsQCA) methods to identify the configuration path of CEI in manufacturing enterprises and uses machine learning to rank the importance of antecedent variables. The findings reveal that no single factor alone is necessary or sufficient for determining the CEI. This study identified five different configurational pathways associated with a low CEI and three different pathways with a higher CEI. Machine learning shows that green technology innovation is the most important antecedent factor affecting CEI. These insights provide valuable guidance for manufacturing companies that adopt practices that facilitate low-carbon transformations.