[AK View] 10 Years Since the 'Lee Sedol vs AlphaGo Moment'

A Decade Since Go Proved AI's Power
The Era of Human vs. AI Competition Is Over
Those Who Master AI Become Unbeatable Champions

"The Genius Makers: The Mavericks Who Brought AI to Google, Facebook, and the World," written by Cade Metz, a technology reporter for The New York Times, chronicles the fierce battle for artificial intelligence (AI) supremacy over the past 60 years. The book features the untold stories of more than 400 pioneers who led the AI revolution, including Geoffrey Hinton, known as the 'father of deep learning,' and Demis Hassabis, the 'father of AlphaGo.' Among them, one Korean is mentioned: Lee Sedol, the 9-dan Go master who faced off against the AI Go program AlphaGo.

In March 2016, at Gwanghwamun in Seoul, a historic Go match took place between a human representative (Lee Sedol) and AI (AlphaGo). It was a challenge from AI, armed with massive data learning, to the realm of Go, which had long been considered the sanctuary of human intuition and creativity. The result was a decisive victory for AlphaGo, with four wins and one loss. It marked a turning point in human civilization. Shocked by the collapse of human intuition in the face of AI's overwhelming computational power, Lee Sedol declared his retirement, saying, "If I were born again, I would want to become someone who creates AI." He is now studying AI while serving as a distinguished professor at UNIST and as a board game author specializing in AI.

"The era of humans competing against AI is over. Our roles are different now; humans must become those who set the direction and take final responsibility." After his retirement, Lee Sedol shared this key insight. At a keynote lecture titled "Human Competitiveness in the AI Era," hosted by the Korea Artificial Intelligence Association on December 5, he said, "When learning Go, there are forbidden moves that you are taught never to play. But AI places stones there without hesitation." He confessed, "I realized that it is not AI but rather humans who are trapped by fixed ideas." This suggests that AI can be a powerful partner in human learning and growth.

We are entering an era in which those who master AI are securing top positions in every field. The current world number one in Go, Shin Jinseo, is a prime example. Nicknamed "Shin-Gongjineung" (Shin Jinseo + Artificial Intelligence) for his computer-like precision, Shin Jinseo has become an "unbeatable champion" since studying Go with AI. AI has become an indispensable partner in training national Go teams, with 70 to 80 percent of training now conducted through AI. Although the Go community initially had concerns about AlphaGo's arrival, it has since found a way to coexist and thrive alongside AI.

Demis Hassabis, the creator of AlphaGo, went on to develop AlphaFold, an AI for protein structure prediction, which earned him the 2024 Nobel Prize in Chemistry. Neural network technology that began with AI Go programs has now solved major challenges in life sciences through collaboration with humans.

Since the launch of ChatGPT in November 2022, major AI technology companies that previously released new models every six months to a year have begun upgrading their models every week to month in the second half of this year. The AI technology race is intensifying. The Korean government also announced that it will distribute graphics processing units (GPUs) secured from Nvidia to industry, academia, research institutions, and national AI projects starting in February next year, signaling its late entry into the "global AI top three" competition.

AI is now spreading uncontrollably. What would take over a year in the human world is changing in less than a day in the AI world. There is growing anxiety that the speed of inequality in society will accelerate, making it harder to catch up if you fall behind in the AI race. The year 2026 marks the 10th anniversary of the so-called "AlphaGo vs. Lee Sedol moment." Once again, a golden opportunity is upon us. How much do we truly know about AI, and how prepared are we? We must reflect on why we failed to fully capitalize on the "Lee Sedol effect" ten years ago. The lesson from Lee Sedol is the realization that we must go beyond competing with AI and instead enhance our abilities through collaboration.

Cho Youngcheol, Content Editing Team 1 Leader