This study proposes an artificial intelligence (AI)-driven framework for sustainable 3D printing using recycled polymers in smart manufacturing ecosystems. Neural networks (NN) and genetic algorithms (GA) optimize process parameters—print speed, extrusion temperature, and layer height to enhance mechanical properties and energy efficiency of parts printed with recycled polylactic acid (rPLA) and polyethylene terephthalate (rPET). A simulated smart factory with five fused deposition modeling (FDM) printers was tested over 90 days. The AI-driven approach achieved 35% higher tensile strength, 30% reduced energy consumption, and 25% increased throughput compared to baseline methods. Realtime IoT monitoring and material characterization ensure robust optimization (prediction error ¡5%). The framework supports circular economy principles by minimizing waste and carbon footprints, offering a scalable model for sustainable additive manufacturing in Industry 4.0.