The recent move of a senior Taiwan Semiconductor Manufacturing Co. (TSMC) executive to Intel has reignited debate over whether talent transfers can narrow the gap between the world's leading foundry and its American rival. A former TSMC research director, however, argues that technology alone is not the deciding factor—corporate culture is.
Speaking on Storm Media's program International"Fly to the World",Yang Kuang-lei（楊光磊）, a former research and development director at Taiwan Semiconductor Manufacturing Co. (TSMC), said Intel faces deep structural challenges in trying to replicate TSMC's foundry model.
“Even if the technology were transferred in full, the culture cannot be transferred,” Yang said, adding that Intel remains years away from matching TSMC's execution capability.
Technology Can Move—Culture Cannot
The comments follow controversy surrounding Lo Wei-jen（羅唯仁）, a former senior vice president at TSMC who allegedly accessed sensitive data related to advanced process technologies—including next-generation nodes—before leaving the company for Intel in late 2025. The incident raised concerns that proprietary knowledge could accelerate Intel's foundry ambitions.
Yang dismissed that assumption as misguided.
Drawing on his own experience as a former consultant to Intel, Yang described fundamental differences in organizational alignment between the two companies. He compared TSMC to a system in which “360 people may differ by only one degree, but all move in the same direction.” Intel, by contrast, he said, operates in a far more fragmented environment, where “360 people each point in different directions, resulting in zero net movement.”
According to Yang, this lack of alignment reflects a broader cultural problem rather than a technical one.
From Product Champion to Foundry Outsider
Yang noted that Intel once operated under a disciplined, execution-driven culture shaped during the era of its former leadership, but that cohesion has weakened over time. While current management has sought to introduce organizational reforms, including flatter decision-making structures, Yang argued that such changes cannot quickly rebuild deep-rooted R&D execution capabilities.
At the heart of Intel's struggle, Yang said, is what he described as a “competency trap.” Having dominated the industry for decades as a vertically integrated product company, Intel lacks the service-oriented mindset required for the foundry business.
“Foundry manufacturing is fundamentally a service,” Yang said. “It requires humility, customer trust, and a willingness to optimize for someone else's roadmap.”
TSMC's leadership, he argued, rests not only on process technology, but also on a deeply embedded service culture—one that prioritizes yield stability, cost control, and long-term customer relationships. On these metrics, Yang said, no competitor currently comes close.
Limited Impact From Individual Moves
Yang added that while Samsung remains the closest rival to TSMC, it still trails by a considerable margin. Intel, which is effectively rebuilding its foundry operation, is even further behind.
As a result, Yang believes the impact of Lo's move has been overstated. Even in the unlikely scenario that technical data were perfectly replicated, he said, results would still depend on the surrounding organizational system.
“Without the underlying culture, the outcomes won't be the same,” Yang said. He also expressed confidence that Intel would not engage in unethical practices to copy competitors wholesale.
In Yang's assessment, the decisive advantages in advanced semiconductor manufacturing lie not in individual hires or isolated data, but in institutional discipline, alignment, and accumulated experience—factors that cannot be transferred overnight.
