Crossover Talks Kaohsiung：Navigating Global Disruption with Domain Twin
Profet AI joined forces with Chunyao Digital to co-host our Crossover Talks in Kaohsiung. Themed “Navigating Global Disruption with Domain Twin”, the event explored structural challenges facing enterprises and proposed strategic AI-driven solutions amid global industrial shifts, where manufacturing is grappling with tariff uncertainties, geopolitical risks, talent mobility, and supply chain restructuring.
We invited manufacturing industry leaders, including former Innolux CIO Dr. Howard Hsieh, former Yageo CEO David Huang, ChipMOS Technology consultant Michael Wang, and Chunyao Digital General Manager Jeff Chi, who shared insights on overseas plant setup, technology transfer, organizational collaboration, and knowledge inheritance, highlighting how AI and the Domain Twin concept can help navigate disruptions and build sustainable advantages in smart manufacturing.
From Expertise to Scalable Process: A Practical Path for AI in Manufacturing
In his opening remarks, Profet AI CEO and Co-Founder Jerry Huang noted that while the potential of AI is widely recognized, few truly know how to implement it. With rising tariffs and volatile exchange rates adding pressure, manufacturers are seeking breakthroughs through AI. Yet gaps in understanding, from leadership to frontline staff, often leave projects stalled at the slogan stage. James, Executive Assistant to the CEO, added that global trade shifts and “Trump tariffs” are prompting companies to consider relocating production to the U.S. or ASEAN, but data security risks, talent shortages, and the need for rapid operational ramp-up remain major hurdles.
Many manufacturers have advanced in digitalization, ERP/MES adoption, and automation, but James emphasized that the human factor is still critical. As experienced workers become scarce due to demographic and labor shifts, AI is emerging as a key enabler of transformation. Profet AI’s five modules: the Resilience Management Framework, AutoML Platform, AILM Platform, AI Studio, and AI Thinking Workshop can help companies turn strategy into execution, accelerate knowledge transfer, and scale AI applications to optimize the entire process from R&D to production.
Innolux 4.0: The Three Pillars of Industrial Digital Transformation
Drawing on his years of frontline manufacturing experience, former Innolux CIO Dr Howard Hsieh shared insights from leading the company’s “Innolux 4.0” initiative. He emphasized that true digital transformation goes beyond technology adoption and requires reshaping both management capability and talent capability, all starting with determination. He noted that a culture of continuous improvement is the ultimate goal. For smart manufacturing, Howard outlined three pillars: Culture (breaking silos and fostering connections), Methodology (combining shop floor observation with data), and Technology. Research from the Artificial Intelligence Foundation shows that over 70% of Taiwan’s manufacturers are stuck at the third of four AI adoption stages, able to run projects but unable to scale them company-wide.
Howard believes companies do not need to wait until every condition is met before starting large-scale transformation. Instead, they should begin with an AI diagnosis. “Using AI for diagnosis is the starting point of continuous improvement,” he said. The strategy of “starting small, optimizing step by step, and then expanding” is, in his view, the key to unlocking AI’s true value.
Strategic Outlook and Time Management: Keys to Thriving Beyond Technology
David Huang, former CEO of Yageo and now Founder of Jensen Capital, reflected on past financial crises, highlighting the impact of exchange rate fluctuations on profitability and the importance of turning crises into opportunities. He recommended reserving one-third of the budget for unpredictable risks such as tariff changes or currency swings to strengthen financial resilience.
He noted that Taiwan’s manufacturing sector should leverage its strong industrial base to move AI from Preventive Maintenance to Prediction, integrating it with domain expertise. The real gap, he said, is the shortage of “Domain Agents” who can develop and transfer specialized knowledge. Facing global uncertainty, companies should anticipate multiple scenarios and plan ahead for talent and facility redeployment, ensuring readiness during supply chain shifts.
The Hidden Challenges of Overseas Factory Setup: Replicating Operations and Talent
Michael, now a consultant at ChipMOS Technology, shared the challenges he faced while leading overseas factory setups as Chief Plant Manager for Greater China at Innolux. Recalling his first relocation project over 30 years ago in China, he described hurdles such as equipment power differences, maintenance issues, and language and cultural gaps, which made SOPs hard to enforce and required hands-on training for local staff. As labor costs rose, he successfully drove automation, reducing the workforce from 6,000 to just 200 employees.
He emphasized that the push for AI and automation often encounters internal resistance, including doubts over ROI, difficulty in quantifying benefits, and reluctance to change. Michael believes that “the key to driving reform is the determination that as long as it doesn’t kill me, I will keep going.” He urged companies to focus on long-term operational efficiency and knowledge accumulation, as the value gained over time far outweighs short-term returns.
Data-Driven Supply Chain Upgrades: AI as the New Standard for Enterprises
At the forum, Jeff Chi, General Manager of Chunyao Digital, noted that enterprise management is a continuous process of transformation, requiring stronger data integration and collaboration to enhance supply chain resilience and adaptability. “AI accuracy can already reach 95%, but for multi-agent systems, enterprises demand even higher precision. Reliable AI must combine industry domain knowledge with reproducibility and controllability,” he said.
In the digital economy era, AI is already as important as, if not more critical than, traditional MES systems. With AI technology advancing rapidly, he strongly advised companies to plan their AI adoption timelines as early as possible, treating data and knowledge management as core assets. This approach not only boosts efficiency but also protects valuable expertise from being lost due to employee turnover, helping secure long-term competitiveness.
Learn how AI helps manufacturers navigate global disruption. Fill out the form to get event highlights, case studies, and transformation strategies from industry leaders.