<p>Manufacturing companies are increasingly adopting green transformation through government innovation subsidy policies to incentivize it, and artificial intelligence (AI) to handle specific tasks, reducing dependence on human labor. This study, for the first time, integrates the impacts of green technology implementation within a hybrid subsidy policy, based on both the input and output of green technology investment, and the adoption of AI technologies in an energy-economy-environment (3E) conscious production system. The model defines a flexible manufacturing and remanufacturing system under limited floor-space capacity and thereby formulates a constrained optimization problem within a cap-and-trade (CAT) regulation. As a result, the Karush–Kuhn–Tucker (KKT) conditions are applied after examining the cost function’s curvature, and closed-form optimal solutions are derived to minimize the manufacturer’s total annual cost. Furthermore, this study provides practical implications for authorities in setting an appropriate emission cap to ensure effective CAT regulation across different scenarios. A numerical illustration highlights that the proposed hybrid subsidy policy, combining input- and output-based incentives, achieves the best overall performance. AI efficiency in controlling production-stage emissions plays a critical role in determining both technological investment and system cost. Under the baseline numerical parameter setting, AI technology significantly reduces operational costs and energy consumption, resulting in an 18% reduction in total annual costs compared with systems without AI adoption.</p>