The Foundations: Dr. Kang Dae-won’s Transformative Inventions
Dr. Kang Dae-won (1934-2012) stands as one of the most influential scientists in semiconductor history, yet remains largely unknown outside academic and industry circles. His pioneering work at Bell Laboratories fundamentally transformed the trajectory of global semiconductor development.
MOSFET: The Building Block of Modern Electronics
In 1960, Dr. Kang partnered with Mohamed M. Atalla at Bell Laboratories to develop the Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET). This invention revolutionized semiconductor design by enabling high integration and mass production on a single chip. The MOSFET became the foundational technology for Intel’s CPUs, Samsung’s DRAM, SK Hynix memory systems, and virtually all modern mobile processors.
What made the MOSFET truly groundbreaking was its scalability. Unlike earlier transistor designs, the MOSFET could be miniaturized and stacked in increasingly dense configurations, laying the groundwork for the exponential growth predicted by Moore’s Law.
The Floating Gate: Enabling NAND Flash Memory
Seven years after inventing the MOSFET, Dr. Kang made his second landmark contribution. In 1967, working with Simon M. Sze at Bell Labs, he developed the Floating Gate memory cell. This ingenious design placed an electrically isolated floating gate within a MOSFET structure, allowing charge to be trapped and stored without continuous power.
This principle became the architectural foundation for NAND Flash memory, which powers everything from USB drives to SSDs in modern computers. The Floating Gate’s elegance lies in its simplicity and efficiency: a sandwich structure of insulator-metal-insulator that enables long-term data retention with minimal power consumption.
Lee Kun-hee’s Visionary Gamble: Acquiring Korean Semiconductor
While Dr. Kang laid the theoretical foundations of semiconductor technology in the West, a pivotal decision in Korea would determine whether the nation could build a world-class semiconductor industry. That decision came on December 6, 1974, when Lee Kun-hee, then an executive director at Dongyang Broadcasting Company, invested $500,000 of his personal funds to acquire 50% of Korean Semiconductor.
Defying the Skeptics
The decision was audacious. Samsung’s management resisted, arguing, “We cannot even manufacture televisions properly—how can we possibly succeed in semiconductors?” Yet Lee Kun-hee saw what others could not: that semiconductor self-sufficiency was not merely Samsung’s challenge but the common imperative of South Korea’s entire electronics industry.
Lee acquired the remaining 50% stake in 1977 and renamed the company Samsung Semiconductor in 1978. More importantly, he embarked on a relentless strategy of talent acquisition, making repeated visits to Silicon Valley to recruit world-class engineers and secure technology transfer agreements. He famously negotiated with Fairchild Semiconductor to provide 30% of Samsung’s equity in exchange for critical technology licensing—a deal that would prove transformational.
Lee Jae-yong’s HBM4 Strategy: The Next Generation of AI Infrastructure
Fifty years after Lee Kun-hee’s bold 1974 decision, his successor Lee Jae-yong faced a different but equally consequential challenge. By the early 2020s, Samsung’s semiconductor dominance had begun to erode, particularly in the critical high-bandwidth memory (HBM) segment essential for artificial intelligence and large language models. Competitors, especially SK Hynix, were gaining ground.
Lee Jae-yong responded with characteristic decisiveness. Rather than accepting Samsung’s declining position, he directly visited research facilities and engaged with development teams. In late 2025, he inspected Samsung’s NRD-K (New R&D Complex for Korea), examining facilities that consolidate memory, system-on-chip, and foundry R&D under one roof.
HBM4: Industry-Leading Performance
The result of Samsung’s focused R&D strategy is impressive: HBM4 achieved 11.7 Gbps performance in internal evaluations, claiming the highest speed in the industry. The achievement represents more than raw performance metrics; it demonstrates Samsung’s ability to execute a fully integrated chip design and manufacturing process, from logic dies using 4-nanometer foundry technology to complete package assembly.
Samsung’s vertical integration advantage cannot be overstated. By controlling logic design (SF4 4nm process), memory design (D1c 10nm 6th-generation process), and packaging, Samsung eliminated the coordination overhead that plagues competitors. Samples have already shipped to major AI companies, including Nvidia, which has reportedly provided positive feedback on performance and reliability.
Lee Jae-yong framed this achievement not as catching up, but as establishing ‘fundamental technological competitiveness.’ This linguistic choice reflects his broader strategy: beyond any single product cycle, he is rebuilding Samsung’s position as a technology leader capable of sustained innovation.
Why Korea’s Semiconductor Success Matters
The Korean semiconductor story represents a remarkable convergence of scientific innovation, strategic vision, and relentless execution. Dr. Kang Dae-won discovered the fundamental principles; Lee Kun-hee recognized the strategic imperative and mobilized resources despite institutional resistance; and Lee Jae-yong is navigating the transition toward AI infrastructure dominance.
Three critical factors explain Korea’s sustained leadership:
- Visionary Leadership: Each generation has embraced technology-first thinking, committing to long-term investments over quarterly earnings.
- Vertical Integration: By controlling the entire value chain from design to manufacturing, Samsung and SK Hynix achieved flexibility and quality control unmatched by pure-play competitors.
- Talent and Infrastructure: Building world-class R&D facilities (like NRD-K) and recruiting global talent created a flywheel of continuous innovation.
As artificial intelligence reshapes global technology, HBM4 is not merely Samsung’s next product—it is the foundation upon which the next decade of AI infrastructure will be built. In that sense, Lee Jae-yong’s strategy is as consequential as Lee Kun-hee’s 1974 decision, and both stand upon the genius of Dr. Kang Dae-won’s 1960 invention.
Korea’s semiconductor journey reminds us that true technological leadership emerges not from accident but from the persistent convergence of scientific discovery, strategic vision, and organizational capability.
반도체 반도체 분인 기늤를 초래한 강대원 박사(1934-2012)는 단순 과제자가 아니라 고 살동안 세나 반도체 동캡의 귀중땜 기반을 더른 인물이다.
1960년, 미국의 범렐연구소(Bell Labs)에서 강대원 박사는 스른 모하메드 마틸 아탈라 박사와 함께 MOSFET(초센늤터 이짱 효과)를 세계 초로 개발했다. 이 기술은 단숬 몽연 기반을 널리 고집날 가능한 구조로 냈고, 지금도 인테니 CPU나 삼성렄자의 D랜으로 메모리 등 대부분의 주요 결과로 사용되동을 영보는 기술이다.