In a landmark achievement for physical AI, Sony AI announced on April 22, 2026, that its Project Ace robot has become the first autonomous system to consistently defeat elite human professional table tennis players. While AI has long conquered virtual games like Chess and Go, the physical realm presented a “reality gap” that was thought to be years away from closing.
Faster Than Human Reaction
Table tennis is one of the most demanding tests for robotics, requiring rapid decision-making and precise execution at the edge of human reaction time. Project Ace utilizes a high-speed perception system consisting of nine active pixel sensors. Unlike previous robots, Ace doesn’t just “react”; it uses reinforcement learning to predict the spin and trajectory of the ball before it even crosses the net. This breakthrough proves that autonomous systems can now perceive, plan, and act in dynamic, adversarial physical environments—laying the groundwork for robots that can safely assist in high-pressure medical or emergency response scenarios.
