On August 26, 2026, MeiGuang SuZao announced that its proprietary AM Build process-adaptive large model has enabled dynamic, full-process control in 3D printing. In the case of blade components, support powder consumption dropped from 14.4g to 0.5g per part, with its share of total material use falling from 7.7% to 0.2%—a reduction of 97.4%. For complex structural parts, support volume decreased by 66%, material waste by 50%, and post-processing time was cut from 31 minutes 46 seconds to 16 minutes 18 seconds; support-free printed parts required almost no post-processing. The model senses the material state and heat dissipation variations of each layer in real time, dynamically optimizing parameters layer by layer, eliminating reliance on manual parameter-tuning experience. AI-driven process adaptation marks a key step in additive manufacturing's shift from experience-driven to data-driven operation, with parameter optimization capability emerging as a new competitive dimension for equipment and software. Value for powder buyers: ① AI process adaptation directly reduces metal powder waste and represents a systematic optimization opportunity on the powder utilization side; ② The significant reduction in support material usage means lower powder consumption per part, requiring adjustments to powder procurement volume forecasting models; ③ Pay close attention to the higher demands AI parameter tuning places on powder particle size consistency and batch-to-batch stability.