Abstract: In developing countries such as the Philippines, the demand for low-cost housing continues to increase alongside the need to reduce material-related carbon emissions. This study developed an artificial neural network (ANN)-based decisionsupport framework integrated with Pareto optimization to evaluate the mechanical performance and embodied carbon of lateritic soil-stabilized blocks (LSSBs). A total of 216 specimens were tested using different stabilizer types, stabilizer dosages, compaction pressures, and curing durations. Compressive strength was determined through laboratory testing, while embodied carbon was quantified using a cradle-to-gate life cycle assessment approach. ANN models were used to predict compressive strength and embodied carbon, followed by sensitivity analysis and multi-objective optimization to identify strength-carbon trade-offs. The results showed that compaction pressure and stabilizer dosage primarily influenced compressive strength, while stabilizer type and dosage governed embodied carbon. Pareto-optimal solutions indicated that structurally adequate and lower-carbon LSSBs can be achieved by improving production control rather than relying on excessive stabilizer use. The proposed framework provides practical mix-selection guidance for sustainable low-cost housing applications.