[KIST AI Special Lecture] Cross-Industry AI Innovation - Silicon Valley’s Model for AI-Driven Technology Transfer Across Industries

Abstract

In recent years, Artificial Intelligence (AI) has emerged as a universal catalyst across industries, from materials science to medicine, logistics to semiconductors. This talk examines how AI-led innovation is best realized when paired with domain expertise, quality data, and cross-disciplinary fluency. Drawing from my experience across academic, industrial, and entrepreneurial landscapes in Silicon Valley and Korea, I will explore how technological transfer and convergence can be accelerated through AI—not merely as a modeling tool, but as a medium of insight and integration.

I will share lessons from building industrial AI systems at Amazon and Gauss Labs, where deep learning was applied to large-scale manufacturing optimization, and from co-founding Erudio Bio, which combines advanced semiconductor-based multiplexing with multi-omic biomarker analysis. These efforts reveal the importance of grounding AI development in rigorous engineering, mathematical understanding, and close collaboration with domain specialists. Recent collaborations with hospitals, research institutes, and public agencies highlight how AI systems can bridge data-rich environments with meaningful real-world decisions.

The talk will also touch on my ongoing academic engagements—from seminars at KAIST, Sogang University, DGIST, POSTECH, Korea University, Seoul National University (SNU), Yonsei University, and Stanford University to AI forums that connect research with practice—and reflect on the creative and human dimensions of AI work. Rather than framing AI as an end in itself, I will advocate for a view that sees it as a powerful collaborator: one that amplifies human judgment, enables scientific exploration, and, when thoughtfully applied, helps forge new connections across traditional boundaries of industry and knowledge.