Sony's Ace Robot Masters Table Tennis, Signaling Leap in Physical AI

Daily Technology

23/04/2026

Sony has unveiled Project Ace, an autonomous robot demonstrating remarkable prowess in table tennis, capable of competing against professional players. This development marks a significant milestone in physical artificial intelligence, showcasing AI's ability to achieve expert-level performance in complex, real-world physical activities.

Key Takeaways

Sony's Project Ace robot can compete at a professional table tennis level.
The robot utilizes advanced sensory technology, reinforcement learning, and precision hardware.
Ace has demonstrated success against elite players, winning multiple matches.
This advancement signifies a new era for physical AI applications.

A New Challenger in Table Tennis

Sony's Project Ace represents a significant leap forward in the field of physical AI. The robot is designed to autonomously play table tennis, a sport renowned for its complexity, demanding split-second decisions, speed, and power. Peter Dürr, director of Sony AI and project lead for Ace, highlighted the game's intricate nature in a press release.

Technological Innovations Behind Ace

The robot's capabilities are powered by a sophisticated combination of advanced sensory technology, reinforcement learning, and high-precision hardware. Ace is equipped with nine active pixel-sensor cameras that enable it to accurately determine the ball's position in three-dimensional space. Additional cameras and systems are integrated to measure the ball's velocity and spin, crucial for anticipating its trajectory.

Beyond its visual perception, Ace features a control system based on model-free reinforcement learning. This allows the AI to adapt and make decisions dynamically without relying on a pre-programmed model, enabling it to learn and improve its gameplay in real-time. This intelligent control system is paired with high-speed robotic hardware, making Ace a remarkable fusion of art and engineering.

Performance and Future Implications

In rigorous testing, Ace competed against five elite and two professional-level table tennis players. The robot emerged victorious in three out of five matches against the elite players and demonstrated a strong offensive capability by scoring sixteen direct points during serves compared to the elite players' eight.

According to Peter Stone, chief scientist at Sony AI, Ace is more than just a proof of concept. It offers a glimpse into the future potential of physical AI. Stone stated that AI operating at an expert human level in such demanding conditions opens the door to a wide array of new real-world applications that were previously considered unattainable.

2026-04-14

Gemini Unveils Interactive Visualizations for Enhanced Chat Understanding

Google's Gemini AI now generates interactive visualizations in chat, enhancing user understanding of complex topics. Learn about this new feature, its capabilities, and availability.