GeChuang Dongzhi, an industrial AI company in the TCL group, has deployed its self-developed Tianshu AI ADC visual inspection system at TCL CSOT's Suzhou display panel factory, according to a September 2025 GeChuang press release published on digi.china.com. The company says the system has replaced more than 75% of manual inspection there.
TCL CSOT (Suzhou) was awarded the "2025 Excellence-Level Intelligent Manufacturing Demonstration Factory" certification by China's MIIT for its "AI-driven new display panel intelligent factory." The 540,000-square-meter production facility features highly automated robotic arms and overhead transport rails, with the Xingzhi large model for product innovation and AI digital employees serving as the quality engineer's "eyes" and "brain."
In the critical micrometer-level defect inspection workflow, AI digital employees operate through a closed-loop "ADC+ADR" dual-system: the ADC automatic defect classification system uses AI vision technology to identify and classify over 4,000 types of micrometer-level defects at 95% accuracy; the ADR automatic defect repair system drives high-precision tools to fully automatically repair repairable defects.
The Tianshu AI platform packages deep learning work into ready-to-use tools and performs millisecond-level reclassification of defects flagged by automated optical inspection (AOI). According to GeChuang, traditional AOI equipment still misjudges more than 60% of defects and has relied on large teams of people to reclassify them, while Tianshu's AI reclassification identifies defects with accuracy above 95%.
Technical architecture spans three platforms: a training platform with built-in tens-of-millions-level defect feature library supporting low-code model iteration, an inference platform with lightweight edge computing boxes achieving millisecond-level single-image processing (5-10x faster than human eyes), and resource scheduling with dynamic GPU allocation supporting 100+ algorithm models running concurrently at TCL CSOT.
GeChuang says the Tianshu platform has been applied in more than 1,000 industrial scenarios across its customers, with inference covering visual inspection needs in more than 30 industries; for one semiconductor client it reduced labour input by 90% and reached a defect detection rate above 99%. All figures are the vendor's own.
GeChuang plans to deepen its work with the TCL Industrial Research Institute on discovering complex defect patterns and transferring knowledge across domains, as chips get smaller and inspection shifts from reacting to defects to preventing them.