In March 2016, the Four Seasons Hotel in Seoul became the epicentre of an existential reckoning. Across a wooden Go board, one of the greatest human minds of his generation faced an opponent that possessed neither heartbeat nor hesitation. Lee Sedol, an eighteen-time world champion and a titan of the ancient game of Go, sat opposite Aja Huang, a lead programmer acting as the physical proxy for DeepMind’s AlphaGo. Over the course of five games, watched by hundreds of millions globally, the parameters of machine intelligence were irrevocably redefined.
The AlphaGo match was not merely a technological demonstration; it was a profound cultural moment. For decades, computer scientists had conquered classical games through brute-force computation. IBM’s Deep Blue dispatched Garry Kasparov in chess in 1997, evaluating 200 million positions per second to overwhelm human tactical foresight. But Go, a game originating in China over 4,000 years ago, resisted such approaches. It is a game of spatial influence, abstract strategy, and pattern recognition, boasting more possible board positions than there are atoms in the observable universe. Experts widely predicted that an artificial intelligence capable of defeating a top-tier professional Go player was still a decade away. DeepMind accelerated that timeline drastically, and in doing so, forced a global conversation about the nature of creativity itself - a theme increasingly dominant across contemporary culture.
The Weight of Four Millennia
To understand the sheer magnitude of Lee Sedol’s challenge, one must understand the reverence afforded to Go in East Asia. It is not merely a board game; it is a philosophical pursuit, an art form that reveals the moral and intellectual character of its players. Lee Sedol, known for his aggressive, brilliantly unorthodox style and fierce independence, was the quintessential Go artist. Having turned professional at the extraordinary age of 12, he had dominated the international circuit for a decade. He approached the 2016 match with a quiet, almost dismissive confidence, publicly predicting a 5-0 or 4-1 victory in his favour.
It was a sentiment shared by the global Go community. Prior AI systems had struggled to reach amateur levels, primarily because Go cannot be solved by looking ahead at every possible outcome; the branching factor is simply too vast. Human professionals rely on "shape" and "intuition" - a subconscious feeling for where a stone belongs. DeepMind solved this by moving away from brute force, utilising deep neural networks combined with a Monte Carlo tree search. AlphaGo possessed a "policy network" to select the next move and a "value network" to predict the winner of the game from the current board state. It didn't just calculate; it evaluated.
Game One systematically dismantled the human assumption of superiority. AlphaGo played with a relentless, suffocating efficiency, forcing Lee into a resignation. The defeat sent shockwaves through South Korea, plunging the nation into what the local press dubbed the "AlphaGo shock." But it was Game Two that fundamentally shifted our understanding of machine capability.
Move 37: The Alien Intelligence
If Game One proved AlphaGo could calculate a victory, Game Two proved it could create art. On its 37th turn, the AI placed a black stone on the fifth line from the edge of the board - a move so unconventional that human commentators initially assumed Aja Huang had misclicked the board. In the millennia-long history of Go theory, early plays on the fifth line are universally discouraged because they sacrifice immediate, secure territory for vague, long-term influence.
Fan Hui, the European Go champion who had played and lost to an earlier version of AlphaGo months prior, famously described the move during the broadcast: "It’s not a human move. I’ve never seen a human play this move. So beautiful."
Move 37 was not an error; it was a profound strategic calculation that paid dividends dozens of turns later, securing a victory for the machine. It demonstrated a terrifyingly elegant form of alien intelligence. The AI was not merely mimicking human strategy; it had transcended it. By playing millions of games against itself, AlphaGo had innovated in a space humans had studied rigorously for thousands of years, revealing blind spots in our collective intuition. This marked a paradigm shift in how we perceive machine learning - no longer just a tool for processing data, but a mechanism capable of generating original thought, a concept now driving the vanguard of digital design and algorithmic art.
When AlphaGo secured the series by winning Game Three, the atmosphere in the press room was funereal. Demis Hassabis, CEO and co-founder of DeepMind, articulated the gravity of the moment, telling reporters, “To be honest, we are a bit stunned and speechless.” Hassabis, however, remained acutely aware of the human element facing the algorithm, adding, “AlphaGo can compute tens of thousands of positions per second but what's incredible is Lee Sedol can compete with that just with his mind and his ingenuity. He stretched AlphaGo to its limits.”
Move 78: The Hand of God
The five-game series had been decisively lost, but the defining moment of the match - and perhaps the most poetic moment in the history of human-computer interaction - arrived in Game Four. Exhausted, facing an opponent that had systematically dismantled his lifelong expertise, Lee Sedol found himself in a desperate, highly complex mid-board fight.
Then came Move 78.
After nearly thirty minutes of deep thought, Lee wedged a white stone between two of AlphaGo’s black stones. It was a brilliantly disruptive, highly counter-intuitive play that defied traditional Go wisdom. DeepMind's control room monitored the AI's internal evaluations, which suddenly plummeted. The machine, which had played with cold, godlike perfection for three and a half games, failed to anticipate the move. AlphaGo's win probability dropped, and the program began to output nonsensical, amateurish moves, effectively entering a state of computational confusion. Lee Sedol had found the singular weakness in the algorithm’s sprawling neural network.
Lee forced a resignation in Game Four. The resulting celebration was a raw outpouring of human relief and admiration; journalists cheered, and Lee entered the post-match press conference to a standing ovation. Demis Hassabis immediately acknowledged the brilliance of the human champion, posting on social media: “Lee Sedol wins game 4!!! Congratulations! He was too good for us today and pressured AlphaGo into a mistake that it couldn't recover from. Lee Sedol is an incredible player and he was too strong for AlphaGo. We are also very happy because this is why we came here - to test AlphaGo to its limits.”
Move 78 was quickly dubbed the "divine move" or the "hand of God" by Go commentators. It was a testament to human resilience - a reminder that even in the face of overwhelming computational superiority, human intuition can still summon moments of transcendent brilliance. Hassabis has repeatedly reflected on Move 78 in subsequent years, describing it as a rare instance of "pure human genius" that effectively broke the AI's logic.
An Entity That Cannot Be Defeated
Lee Sedol lost the fifth and final game, concluding the series with a 4-1 defeat. While he was celebrated as a hero globally - the lone human to draw blood against the machine - the match left an indelible mark on his psyche and the wider Go community. Following the Seoul match, AI systems in Go continued to evolve at a blistering pace. DeepMind released AlphaGo Zero, a subsequent iteration that learned the game entirely by playing against itself, bypassing human data completely. It achieved superhuman performance in mere days, defeating the version of AlphaGo that beat Lee Sedol by a staggering 100 games to zero.
The gap between human and machine became an unbridgeable chasm. Professional Go players began studying AI games intensely, treating the algorithms as infallible oracles. The nature of the profession shifted; the ultimate pursuit was no longer human mastery, but an exercise in machine interpretation.
In November 2019, Lee Sedol shocked the world by announcing his retirement from professional Go competition at the age of 36. His reasoning was stark, devoid of the usual athletic clichés about spending time with family or physical decline. He articulated an existential resignation that resonates far beyond the confines of a board game, touching upon the anxieties of automation in every creative field.
"With the debut of AI in Go games, I've realized that I'm not at the top even if I become the number one through frantic efforts," Lee stated upon his retirement. "Even if I become the number one, there is an entity that cannot be defeated."
The Legacy of a Singular Victory
Lee Sedol’s retirement marked the end of an era, not just for the game of Go, but for the human relationship with computation. To this day, he remains the only human to have ever won a formal, tournament-style game against AlphaGo. That singular victory, secured by the sheer, uncalculatable brilliance of Move 78, stands as a monument to human ingenuity in the twilight of our intellectual supremacy in deterministic games.
The 2016 match in Seoul was a mirror reflecting our own brilliance and our inherent limitations. Lee Sedol did not just play against a machine; he played against the accumulated weight of human technological progress. His defeat forced us to confront a future where machines will routinely surpass us in fields once thought to be the exclusive domain of human cognition.
Yet, the enduring narrative of the AlphaGo match is not solely one of human obsolescence. It is about the elevation of the discipline itself. AlphaGo showed humanity entirely new ways to play, breaking millennia of dogmatic thinking and revitalising the strategic depth of the game. Lee Sedol, in his final years as a professional, played some of the most inspired Go of his career, pushed to new heights by the very entity that would eventually precipitate his retirement.
When humanity yielded to code on that Seoul game board, we lost our undisputed dominion over a beautiful, complex game. But we gained a profound realization: the machines we build are capable of devastating, alien creativity, and confronting them can still draw out the very best of what it means to be human.
