Ten years after a milestone victory, AI now dominates Go training. Players are figuring out what that means for the game.

Technology Review, published by MIT, is a leading source of news, analysis, and commentary on emerging technologies, scientific advancements, and their societal impacts. Covering topics such as artificial intelligence, biotechnology, renewable energy, and more, Technology Review provides  reporting and  articles on the intersection of technology and society. Readers can stay informed about the latest breakthroughs and trends shaping the future of technology.

MIT Technology Review

Ten years after AlphaGo defeated Lee Sedol, AI has fundamentally transformed professional Go. Players now train primarily with KataGo, an open-source successor that outpaces AlphaGo, mimicking its moves to stay competitive. This has homogenized playing styles, erased centuries-old heuristics, and shifted the game's creative center from opening moves to the middle game. Top players like world-ranked Shin Jin-seo match AI suggestions over 37% of the time, yet still struggle to extract the underlying principles from AI's opaque reasoning. On the positive side, AI has democratized training, helping female players close the competitive gap and reach new milestones. Lee Sedol, who retired after his 2016 defeat, reflects that Go has shifted from art to rote replication, while current players seek to preserve human storytelling within the game's AI-dominated framework.

AI is rewiring how the world’s best Go players think