As competition intensifies across the high-end consumer electronics supply chain, manufacturers are facing greater pressure to improve quality, delivery speed, and operational resilience across increasingly distributed global supply networks. Profet AI, a manufacturing-focused AI software company, has continued to deepen its collaboration with General Interface Solution (GIS) Holding Ltd., supporting GIS in expanding AI applications from yield, quality, and efficiency improvement on the production floor to cross-process knowledge extraction, model development, and organizational capability building.
Over roughly two years of collaboration, the two companies have completed more than 100 AI projects and built over 1,500 AI models across manufacturing scenarios including process improvement, quality analysis, and anomaly prediction. Through Profet AI’s platform and implementation methodology, GIS has moved beyond using AI to solve isolated production line issues. AI is gradually becoming a working method that supports continuous improvement and decision-making across engineering teams.
From Customer Requirements to Supply Chain Competition, AI Becomes a Key Driver of Manufacturing Upgrade
GIS Chairman and Chief Strategy Officer Chou Hsien-ying said the company’s AI adoption was driven not only by the rapid development of AI technology, but also by customer expectations around quality, delivery, and response speed. As competition increases in the high-end consumer electronics supply chain, manufacturers must continue strengthening their operational capabilities to remain competitive in an environment shaped by supply chain diversification and rising resilience requirements.
“AI adoption is not simply about following a technology trend,” Chou said. “What matters is whether AI can truly enter company operations and help us continuously improve quality, delivery, and response speed, so we can better serve our customers and supply chain partners.”
He also noted that the rapid rise of regional supply chains has made supply chain security and resilience a higher priority for both countries and enterprises. For manufacturers, the key question is no longer only where to deploy production capacity, but how to use AI to improve quality, speed, and overall competitiveness across operations.
More Than 100 AI Projects Create New Room for Process Improvement
In terms of tangible results, GIS has used AI to re-examine process data that was previously considered close to its improvement limit, uncovering new opportunities for optimization.
On one optical display film production line, for example, the defect rate related to particle contamination had previously approached 10%. With Profet AI’s support in building models, analyzing key factors, and adjusting process parameters, the defect rate was reduced to nearly zero. In another glue coating process, bubble-related defects were previously around 0.3% to 0.4%. After AI-driven analysis and improvement, the rate was reduced to approximately 0.2%, cutting defects by about half.
The significance of these results goes beyond individual metric improvements. They also helped engineering teams recognize that many processes that appear stable or already within target can still reveal new room for optimization through AI and data analysis.
From SOPs to Cross-Process Correlation, Turning Senior Engineering Experience into Replicable Capability
Manufacturing improvement has traditionally relied heavily on the experience accumulated by senior engineers, which is then translated into standard operating procedures. However, most conventional SOPs remain limited to a single station or process. When problems involve cross-process correlations, manufacturers still depend heavily on the judgment of experienced engineers.
Through its collaboration with Profet AI, GIS has further transformed process data and know-how accumulated across different stations into models. This helps engineering teams identify correlations across processes and gradually shift process improvement from individual experience toward capabilities that can be continuously accumulated, replicated, and transferred through systems.
“In the past, many improvements depended on strong engineers and senior engineers because they were the ones who held cross-process know-how,” Chou said. “The major value of AI is that it helps us identify these cross-process correlations and codify them into models and operational language. This allows process improvement to gradually move from relying on people to relying on models and systems.”
Profet AI Provides Not Just Tools, but a Method Companies Can Internalize
Speaking about why GIS chose Profet AI, Chou said GIS valued Profet AI’s manufacturing experience and its proven implementation experience across relevant industries. More importantly, Profet AI does not simply help enterprises complete individual projects. It provides a platform and methodology that helps companies build their own AI application capabilities.
Chou described the collaboration as a capability-building process. The GIS team first learned from Profet AI how AI can be introduced into production and manufacturing workflows. The two teams then jointly identified improvement topics and worked together to move from learning to implementation, with GIS’s internal teams collaborating closely with Profet AI’s consultants.
“Profet AI is more like giving us the fishing rod, not just the fish,” Chou said. “Through this process, we learned how to develop our own improvement know-how and AI application culture within the company.”
Jerry Huang, Co-founder and CEO of Profet AI, said the value of manufacturing AI lies not only in building models, but in turning those models into capabilities that can continuously operate within the enterprise.
“True manufacturing AI should not stop at isolated projects or one-time improvements,” Huang said. “It must turn frontline know-how into enterprise AI assets that can be managed, replicated, and scaled. The GIS case shows how AI can move from single-process improvement to becoming a method for continuous organizational progress.”
Through this collaboration, Profet AI and GIS demonstrate a complete path for manufacturing AI, from project implementation and process improvement to organizational capability building. As global manufacturers continue to address supply chain restructuring, knowledge transfer, and operational efficiency challenges, turning frontline experience into a Domain Twin™ that can be preserved, scaled, and replicated is becoming an important direction for enterprises seeking to move AI from deployment to long-term competitive advantage.
