Overview
- The Nature paper published Wednesday documents April 2025 matches where Ace won three of five against elite players under ITTF rules with licensed umpires and lost both matches to Japanese professionals Minami Ando and Kakeru Sone.
- After the paper was submitted, Sony reports Ace beat professional opponents in December 2025 and again in March 2026, including a win over top‑25 player Miyuu Kihara, as the robot played faster and placed shots closer to the table edge.
- Ace tracks the ball with nine synchronized cameras and event-based sensors that read spin, then uses a reinforcement learning policy trained in simulation to drive an eight‑joint arm with about 20 millisecond reaction time.
- The system benefits from an external sensor array and a non‑humanoid arm, and pros learned to exploit patterns such as weaker returns to simple knuckle serves, which the research team says it is working to address.
- Researchers and commentators see uses in manufacturing, service and entertainment, yet they also flag dual‑use risks as physical AI gains speed and perception, marking a step beyond decades of ping‑pong robots that could only rally.