Sunghee Yun (He/Him)

Co-founder & CTO @ Erudio Bio, Inc., Chief Business Development Officer @ WeStory.ai, Advisor & Evangelist @ CryptoLab, Inc., KFAS-Salzburg Global Leadership Initiative Fellow @ Salzburg Global Seminar, Adjunct Professor @ Sogang University, Advisory Professor @ DGIST, Global Advisory Board Member @ Innovative Future Brain-Inspired Intelligence System Semiconductor of Sogang University, Network Expert @ Gerson Lehrman Group

Abstract

This presentation provides a comprehensive introduction to artificial intelligence and machine learning, designed to demystify these transformative technologies for a general audience. The discussion begins by distinguishing between Artificial Narrow Intelligence (ANI) - current applications like smart speakers, self-driving cars, and recommendation systems - and Artificial General Intelligence (AGI), which remains hypothetical. Drawing from McKinsey Global Institute projections of $13 trillion in AI value creation by 2030, the presentation explores key application areas including eCommerce, healthcare, logistics, medical imaging, and autonomous systems, demonstrating AI’s broad impact across industries from retail ($0.8T projected value) to transportation and logistics ($475B).

The technical foundation covers the three primary machine learning paradigms: supervised learning (predicting outputs from labeled input-output pairs), unsupervised learning (discovering hidden structures in unlabeled data), and reinforcement learning (learning through environmental interaction and reward feedback). Using practical examples ranging from housing price prediction to robot locomotion and game-playing algorithms like AlphaGo, the presentation illustrates how all ML algorithms fundamentally work to minimize the difference between model predictions and actual measurements. The mathematical formulation demonstrates how learning problems are framed as optimization challenges, while deep learning examples show how neural networks with multiple layers can approximate complex, nonlinear relationships in high-dimensional data.

The presentation concludes with a balanced perspective on AI’s current capabilities and limitations, emphasizing that today’s ML excels at simple tasks humans can perform in seconds but struggles with complex reasoning requiring genuine intelligence. While machines offer advantages like tireless operation, perfect memory, and scalable processing of massive datasets, they cannot yet understand human intentions or engage in sophisticated reasoning. The discussion addresses philosophical questions about technological singularity, machine consciousness, and employment impacts, concluding that while significant job displacement is likely, the singularity remains distant and machines lack true consciousness, suggesting humans will adapt as they have throughout previous technological revolutions.