- Who Is Lee Sedol?
- What Happened in the AlphaGo vs. Lee Sedol Match?
- What Were Move 37 and Move 78?
- Why Did Lee Sedol Retire?
- What Do Todd's Frameworks Say About Lee Sedol?
- What Business Lessons Come From Lee Sedol vs AlphaGo?
- Frequently Asked Questions
- Sources
- About the Stagnation Assassin
Four games to one. And the one is the game people still talk about.
In March 2016, Google DeepMind’s AlphaGo beat the South Korean Go champion Lee Sedol 4-1 in Seoul. Lee’s single win came in Game 4, built on his Move 78, which DeepMind’s own data said had about a 1 in 10,000 chance of being played, according to Google. The short answer to what Lee Sedol teaches leaders: when a machine changes the rules of your game, the human advantage is the unexpected move, made by someone willing to abandon the orthodox playbook.
This piece covers who Lee Sedol is, what happened in the match, why Move 37 and Move 78 mattered, why he retired, and what business leaders should take from the most famous human-versus-AI contest of the decade.
Who Is Lee Sedol?
Lee Sedol, also written Lee Se-dol, is a South Korean professional Go player, born in 1983, who won 18 international titles and was widely considered the greatest player of his decade, as Google DeepMind describes him. Go is an ancient board game with far more possible positions than chess, which is why it was long seen as a game computers would not master for many years.
That reputation made him the natural human champion for DeepMind’s challenge. Beating him wasn’t just a technical milestone. It was a public test of whether a machine could beat the best human at the game most associated with intuition.
What Happened in the AlphaGo vs. Lee Sedol Match?
The five-game Google DeepMind Challenge Match was played in Seoul in March 2016. AlphaGo won Games 1, 2 and 3, which decided the match, then Lee won Game 4 and AlphaGo took Game 5. The final score was 4-1. DeepMind says the match was watched by over 200 million people worldwide, and that AlphaGo was awarded an honorary 9-dan professional ranking, the first time a computer Go player received the highest certification.
Google said it would contribute the $1 million prize money to organisations that support science, technology, engineering and maths education, as its post-match blog explains. The British Go Association published a game-by-game report.
The result surprised professionals, and it surprised DeepMind too. Its own team wrote that AlphaGo winning four of five was a surprise “including ours.”
Free from Todd
I write The Unaccredited MBA for people who’d rather learn from real turnarounds than from a classroom. It’s free by email, and lessons like this one are what it’s built on.
What Were Move 37 and Move 78?
In Game 2, AlphaGo played Move 37, a shoulder hit on the fifth line that human professionals would almost never choose. DeepMind estimated a human would play it about 1 time in 10,000. It looked like a mistake to many commentators. It turned out to be the move that won the game, and DeepMind describes it as upending centuries of traditional wisdom.
Two games later, Lee answered. His Move 78 in Game 4, a wedge in the centre of the board, was just as improbable by DeepMind’s data, again about 1 in 10,000. Commentators nicknamed it “God’s Touch.” AlphaGo responded poorly in the moves that followed, and Lee won. DeepMind’s own Game 4 commentary walks through it.
Lee himself has been modest about it. In later interviews he has described his winning move as something closer to a trick than a perfect move, one that worked because the program misjudged it. That makes it a more useful lesson, not a weaker one: he found the place where the system’s assumptions broke.
Why Did Lee Sedol Retire?
Lee retired from professional Go in November 2019. Speaking to Yonhap News Agency, he said: “Even if I become the number one, there is an entity that cannot be defeated.” The BBC and others reported the same.
That is a striking statement from the only human to win an official game against that version of AlphaGo. It shows how the meaning of being the best changed for him once machines entered the game. The top human ranking still existed. It just no longer meant the strongest player on the board.
Business leaders are heading into similar territory. In more and more tasks, the question isn’t whether your best person beats the software. It’s how you redesign the work so that people and software together beat a competitor using either one alone.
What Do Todd’s Frameworks Say About Lee Sedol?
Move 37 is a near-perfect illustration of what my book Stagnation Assassin calls an orthodoxy: a belief an industry treats as permanent truth when it is really just a temporary equilibrium. Centuries of Go wisdom said a fifth-line shoulder hit that early was wrong. A system with no attachment to that wisdom tested it and won. As I wrote in the book: “Most organizations can’t smash orthodoxies because they can’t see them since these beliefs operate below conscious awareness.”
The Orthodoxy-Smashing Framework gives operators methods to find those beliefs on purpose before a competitor or a machine does, including the Outsider Exercise, the History Audit and the Why Chain. If you want to find the one costing you most, start with which orthodoxy is costing you.
Lee’s Move 78 also connects to the pattern reader, one of the four roles in my Four-Position Framework. He recognised that AlphaGo had its own blind spots and played into one. Every leadership team needs someone looking for where the dominant system, whether that’s a market leader or an AI tool, is predictable. I wrote about how to identify pattern readers in your company.
What Business Lessons Come From Lee Sedol vs AlphaGo?
- Treat “everyone knows” as a hypothesis. Move 37 broke centuries of received wisdom. Your industry has rules like that too.
- Look for the system’s blind spot. Lee won by finding where AlphaGo’s evaluation broke. Every dominant player, human or machine, has one.
- Don’t measure yourself only against the machine. The strategic question is how people and AI together beat competitors, not whether a person beats the tool.
- Keep a pattern reader on the team. Someone has to spot where the game is changing before it shows up in results.
- Learn from the loss. Four losses taught the Go world more than any win could. Study what the new entrant is doing differently.
AI also changes the productivity equation. In the Karelin Method, productivity is Activity × Efficiency × Focus. Tools like AlphaGo raise the efficiency ceiling, but they only pay off when leaders aim them at the few activities that matter most.
Frequently Asked Questions
Did Lee Sedol beat AlphaGo?
Lee Sedol won one game, Game 4, in the March 2016 match, but AlphaGo won the five-game match 4-1.
What was Move 78?
Move 78 was Lee Sedol’s wedge in Game 4 against AlphaGo. DeepMind estimated it had about a 1 in 10,000 chance of being played, and it led to his only win of the match.
What was AlphaGo’s Move 37?
Move 37 was AlphaGo’s unusual fifth-line shoulder hit in Game 2. DeepMind estimated a human would play it about 1 in 10,000 times, and it helped AlphaGo win the game.
Why did Lee Sedol retire from Go?
He retired in November 2019, telling Yonhap that even if he became number one, “there is an entity that cannot be defeated.”
What does AlphaGo teach business leaders?
That long-held industry rules can be wrong, and that AI changes the strategic question to how people and machines together beat competitors. Todd Hagopian’s Orthodoxy-Smashing Framework helps teams find those rules first.
Free from Todd
Before you go: if any of this sounded like your company, score your Stagnation Genome. The free worksheet shows you which of the five genes is active, so you treat the cause, not the symptom.
Sources
- Google DeepMind: AlphaGo
- Google: What we learned in Seoul with AlphaGo
- DeepMind: Game 4 commentary (PDF)
- British Go Association: Lee Sedol v AlphaGo match report
- Yonhap: Go master Lee says he quits unable to win over AI Go players
- BBC News: Go master quits because AI 'cannot be defeated'
About the Stagnation Assassin
Todd Hagopian is VP of Product Strategy and Marketing, the founder of Stagnation Assassins, and a business transformation author and speaker. Known as The Stagnation Assassin™, he has led transformations at Berkshire Hathaway, Illinois Tool Works, and Whirlpool that generated $3B+ shareholder value across five turnarounds. He is the author of Stagnation Assassin: The Anti-Consultant Manifesto and The Unfair Advantage: Weaponizing the Hypomanic Toolbox. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.
Your move this week: List the three ‘Move 37s’ in your industry: assumptions an AI system, a new entrant or an outsider would ignore. Run one through a Why Chain this week, asking why it’s true until you hit an answer nobody can defend. If you want this thinking in front of your team, see my speaking page, or start with my books.