Sony AI: Ace, the Robot Challenging Table Tennis Elite

Sony AI recently made headlines with its table tennis robot, Ace, which successfully defeated high-level human players in official matches. This autonomous robot, developed by Sony AI, represents a significant advancement in the field of physical artificial intelligence, where machines are designed to operate in real-world environments. According to Reuters, Ace was designed to thrive in a competitive sports environment, requiring quick decision-making and precise motor control.

The Ace robot participated in matches governed by the rules of the International Table Tennis Federation, and these encounters were officiated by certified referees. During documented trials in April 2025, Ace won three out of five matches against elite players but also lost two matches against professional-level opponents. These performances demonstrate the robot's ability to compete with human players under real match conditions.

AI systems have often excelled in digital environments like chess and video games, where conditions are entirely simulated. However, as explained by Peter Dürr, director of Sony AI Zurich, physical and real-time sports like table tennis pose a major challenge. The system was developed to study how robots can react quickly and accurately in dynamic environments, a work detailed in a study published in the journal Nature.

Ace's architecture includes nine synchronized cameras and three vision systems that track the movement and spin of the ball. The system processes visual data at a speed sufficient to capture movements that are difficult for the human eye to discern. "It's fast enough to capture movements that would be blurry to the human eye," Dürr stated.

The robotic platform uses eight joints to control the racket: three for positioning, two for orientation, and three for managing the force and speed of the shots. This configuration was designed to meet the minimum mechanical requirements for competitive play. Unlike many AI systems trained through human demonstration, Ace was trained in simulation. This approach allowed it to develop its own strategies, resulting in gameplay patterns different from those of human opponents.

Mayuka Taira, a professional player who lost a match against the system, stated that the robot was difficult to predict as it showed no visible cues during play. Rui Takenaka, an elite player who both won and lost against Ace, mentioned that it handled complex spins well but was more predictable on simpler serves. Taira added that the absence of emotional signals from the system made it harder to anticipate its responses. "Since you can't read its reactions, it's impossible to feel which types of shots it dislikes or struggles with," she said.

Dürr asserted that the system demonstrates a strong ability to read the ball's spin and react quickly, while ongoing work focuses on improving adaptability during matches. The project team indicated that similar perception and control techniques could be applied to fields such as manufacturing and service robotics.

At the 2026 Beijing Humanoid Robot Semi-Marathon, humanoid robots competed on a 21-kilometer course in Beijing. The event brought together over 100 robots and approximately 12,000 human participants, who ran on separate tracks. A robot named Lightning, developed by Honor, completed the race in 50 minutes and 26 seconds. This time was faster than Olympic runner Jacob Kiplimo, who recorded 57 minutes and 20 seconds during the Lisbon half-marathon in March. Lightning hit a barricade during the race but continued and finished first. Honor robots also took second and third place in the competition. The performance improved compared to the previous year's event, where the fastest robot completed the course in two hours, 40 minutes, and 42 seconds. Organizers stated that the event aimed to test humanoid robots in real-world conditions on a large scale.

According to the Associated Press, another Honor robot completed the course in 48 minutes under remote control. However, the race rules favored autonomous navigation, and Lightning was recognized as the official winner. Honor engineers stated that the technologies developed for the robot, including structural reliability and liquid cooling systems, could be applied in industrial scenarios.