A Sony AI-developed autonomous table tennis robot named Ace has defeated high-level human players in regulated matches, as reported by Reuters. The robot is part of a category known as physical AI, where artificial intelligence is applied to machines operating in real-world environments. Designed for competitive sport, Ace combines high-speed perception systems with AI-driven control to execute shots under match conditions.
Competing under International Table Tennis Federation rules and officiated by licensed umpires, Ace won three out of five matches against elite players in April 2025 trials. It lost two matches against professional-level opponents. Subsequent matches in December 2025 and early 2026 included wins against professional players, according to Sony AI.
Peter Dürr, director at Sony AI Zurich and project lead, noted that table tennis remains a major open challenge for AI, unlike computer games where AI systems have surpassed human experts. Dürr explained that the sport involves high speed and variability of the ball, including complex spin and changing trajectories, requiring rapid sensing and coordinated movement.
Technical Capabilities and Training
Ace uses nine synchronised cameras and three vision systems to track ball movement and spin. It processes visual data fast enough to capture motion that would appear blurred to the human eye. The robotic platform uses eight joints to control the racket: three control positioning, two control orientation, and three manage shot force and speed. This configuration meets minimum mechanical requirements for competitive play.
Unlike many AI systems trained through human demonstration, Ace was trained entirely in simulation. This approach allowed it to develop its own strategies, resulting in play patterns that differ from human opponents. Dürr stated that the system learns to play not from watching humans but through self-training in simulated environments.
Professional player Mayuka Taira, who lost to Ace, said the robot was difficult to predict because it shows no visible cues during play. Rui Takenaka, an elite player who both won and lost against Ace, noted that the robot handled complex spins well but was more predictable on simpler serves. Taira added that the lack of emotional signals made it harder to anticipate its responses.
Humanoid Robot Race in Beijing
At the 2026 Beijing E-Town Humanoid Robot Half Marathon, humanoid robots competed over a 21-kilometre course. The event included more than 100 robots and approximately 12,000 human participants 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’s 57 minutes and 20 seconds recorded at the Lisbon Half Marathon in March.
Lightning collided with a barricade during the race but continued and finished first. Honor robots also placed second and third. Performance improved compared to the previous year, where the fastest robot finished in two hours, 40 minutes and 42 seconds. Organisers stated the event was intended to test humanoid robots in large-scale, real-world conditions.
According to Associated Press, another Honor robot completed the course in 48 minutes under remote control. However, race rules prioritised autonomous navigation, and Lightning was recognised as the official winner. Honor engineers said technologies developed for the robot, including structural reliability and liquid-cooling systems, could be applied in industrial scenarios.
Dürr noted that Ace demonstrates strong ability in reading ball spin and reacting quickly, while ongoing work focuses on improving adaptability during matches. The project team indicated that similar perception and control techniques could be applied to areas such as manufacturing and service robotics.
Looking ahead, Sony AI plans to continue refining Ace’s adaptability and real-time decision-making for broader applications. The humanoid robot sector is expected to see further testing in industrial and service environments, with future events likely to push autonomous navigation capabilities further.
