AlphaGo

Developed by Google DeepMind. Defeated world Go champion Lee Sedol in 2016 using deep neural networks trained with reinforcement learning. A significant milestone as Go is much more complex for brute-force methods than chess.