The Chengdu performance provokes the question of whether live entertainment has reached a new technological threshold. In Chengdu, Wang Leehom’s “Best Place Tour” married high-intensity musical performance to advanced robotics, with six Unitree G1 humanoid robots arrayed at center stage alongside human dancers. The robots, costumed in billowy trousers and sequined overshirts, demonstrated movements of noteworthy precision coupled to the rhythm of the artist’s composition “Open Fire.”
A salient moment occurred when all six robots executed synchronized Webster flips, a gymnastics move that requires fine-grained rotational control and robust landing mechanics. The following sequence shows the improved dynamic-balance system on the G1, whose responses are implemented at millisecond time scales. This speed allows for rapid adaptation under changing loads and impacts while maintaining postural stability after high-intensity aerial maneuvers. Collision with performers and stage elements was avoided by the onboard environmental-recognition systems operating within the dense and highly illuminated stage environment, further emphasizing the complexity of their motion-planning algorithms.
These humanoids were developed with the ability to outperform conventional levels of human flexibility by Unitree Robotics, a Hangzhou-based firm responsible for the G1. The platform integrates dexterous hands, high-torque joint actuators, and AI-driven motion-control software, and is capable of blending pre-programmed choreography with real-time adjustments. Industry observers have described the Chengdu performance as an advance over prior public demonstrations such as at the Spring Festival Gala, where the robots performed simple turns with handkerchiefs expanding toward more intricate, high-impact dancing routines that approach human performers in their timing and form.
The choreography reflected progress in AI-assisted motion synchronization. Rather than repetitive, looped gestures, the G1 units exhibited diversified sequences of arm waves, leg kicks, spins, and jumps nearly all of them precisely aligned to the musical tempo and the live movements of Wang Leehom. Achieving such integration would essentially require not only accurate kinematic modeling but also predictive motion algorithms capable of anticipating beats and transitions for smooth interaction between a human and robotic dancer. Audience responses were immediate and strongly affirmative. The Chengdu audience gave a standing ovation, and onlookers recorded the scene with mobile devices. Videos went viral across the internet from Weibo to Instagram to X, with many drawing comparisons to Tesla’s Optimus robot. Elon Musk weighed in briefly on the footage, commenting, “Impressive.”
Within hours, the post had tens of millions of views and sparked debate over what the future of performance art might look like. To some observers, the show proved that “music and technology can blend seamlessly together,” while to others, it demonstrated the failure of robots to capture something essential to live dance: the expression of the human face. Offstage, Unitree sees its dancing robots becoming part of everyday life. The company has recently teased a feature called “Keep the Music Going, Keep the Dance Flowing,” in which G1s would keep rhythm with music at home. It is still not available to consumers, but the tease provides one possible vision of the future in which AI choreography will be regular fare in home entertainment, further erasing the line between professional stagecraft and personal leisure. The Chengdu concert was more than just a viral feat of engineering; it was a public stress test of humanoid robotics in a challenging, high-visibility environment.
What the G1 was able to do-acrobatics under live lighting, sound, and audiences-suggests significant strides in joint flexibility, balance algorithms, and environmental sensing. For the technologically savvy entertainment fan, it shows the degree to which AI and robotics have continued to advance in their integration with human-led creative expression and how close such collaborations are to becoming standard features in the global live-performance landscape.
