AlphaGo

AlphaGo is an AI program developed by Google DeepMind specifically for the board game Go. It gained worldwide attention in 2016 when it defeated Lee Sedol, a world-class professional Go player ranked 9-dan — a landmark achievement in a domain long considered too complex for AI. AlphaGo was built by combining deep reinforcement learning with convolutional neural networks (CNN), leveraging both supervised learning from large databases of professional game records and reinforcement learning through self-play. AlphaGo evolved through several major versions: • AlphaGo: The original model trained on professional game records • AlphaGo Master: Achieved 60 consecutive wins in online play against top professionals • AlphaGo Zero: Reached superhuman performance using self-play alone, with no human game data whatsoever AlphaGo's emergence was a landmark moment demonstrating that AI could approach human-like intuition, creativity, and strategic thinking. It had a profound impact on subsequent AI research and public perception of AI's capabilities. Today, the technology behind AlphaGo has been extended to other fields including medicine, chemistry, and materials science — standing as one of the most iconic demonstrations that AI can surpass human limitations.