AlphaGo at Ten: The Match That Announced the AI Era

In March 2016, in Seoul, South Korea, the world watched a machine do something the experts had said was a decade away. AlphaGo, the Go-playing program built by DeepMind, faced Lee Sedol, a 9-dan grandmaster and one of the strongest players in the game’s history, in a five-game match. AlphaGo won 4–1. Every game ended in resignation.

Go was supposed to be safe. Chess had fallen to IBM’s Deep Blue in 1997, but Go’s branching complexity, more possible positions than atoms in the observable universe, as every explainer noted, made brute force useless. AlphaGo didn’t brute-force. It used deep neural networks trained on human games and then on millions of games against itself, developing something its creators described as intuition. After the first victory, DeepMind CEO Demis Hassabis said simply: “We have landed on the moon.”

The $1 million prize went to charity, UNICEF and Go organizations, and the Korea Baduk Association awarded AlphaGo an honorary 9-dan rank. Lee Sedol, gracious in defeat, had given humanity its one bright moment in game four, the famous move that broke the machine’s composure. It remains the only game a human has ever won against AlphaGo in a formal match.

Why the match mattered

AlphaGo’s significance was never really about Go. It was the first demonstration, at public scale, that deep learning could master a domain defined by pattern, strategy, and judgment, not just calculation. The techniques proven in Seoul became the substrate of the decade that followed: the same research lineage leads, by way of transformers and scale, to the AI systems now rewriting software, science, and work. The match was the moment the field’s center of gravity moved from “if” to “when,” in front of a live audience.

Ten years on

The anniversary was marked this spring in the city where it happened: at the “Google for Korea 2026” event in Seoul on April 29, Hassabis and Lee Sedol reunited for a dialogue themed “10 Years of AlphaGo: A Vision of AI for Everyone.” A decade after a machine beat the best human at humanity’s hardest game, the conversation has moved from whether AI can match human judgment to how human judgment should govern AI. The prologue is over. The story it introduced is still being written, and this magazine intends to report it with the same seriousness the match deserved.