Just now, a debate erupted between China and the United States over robots

A "no acceleration, no remote control" robot video has surprisingly made Silicon Valley bigwigs restless

Recently, a robot video from a Chinese startup has caused a global sensation. Interestingly, this video not only showcases stunning technological prowess but also unexpectedly sparked a "debate over authenticity" across the Pacific.

A "debate" erupts between China and the U.S. over robots

Just recently, a video ignited discussions in the tech circles both domestically and internationally.

The star of the video is a humanoid robot from China. It can water plants, throw away trash, tidy up toys, and play frisbee with children, performing actions that are astonishingly smooth.

More importantly, the publisher—a Shenzhen startup named "MindOn Tech"—emphasized that the entire process was "no acceleration, no remote control," completely autonomously performed by the robot.

This newly established company has an impressive background, with its founder coming from Tencent. The hardware they used is the G1 humanoid robot from another Chinese company, Unitree.

"This is fake!" U.S. CEO questions

The video hit like a stone thrown into a calm lake, quickly stirring up waves. Its popularity soon attracted attention from across the ocean.

An American netizen directly tagged the founder and CEO of "American Unitree" Figure—Brett Adcock—on social media platform X, asking him, "Is this real?"

The CEO's response poured cold water on the heated discussion:

It looks like an open-loop playback RL controller (meaning it moves blindly without perception and runs on an open-loop playback).

Then at the end, when they grasp the object (indicating perception occurred in the loop), they removed the "no remote control" label from the bottom right of the video and performed manual remote control.

Brett Adcock seems to hold a generally skeptical attitude towards the achievements of Chinese robot companies. Previously, he publicly questioned another Chinese robot company—UBTECH—stating bluntly:

The mass production delivery video shown by UBTECH is "fake robots," likely computer-generated imagery (CGI).

With UBTECH's "precedent," when MindOn's video sparked heated discussions again, the external skepticism seemed to be reignited A debate surrounding "the authenticity of Chinese robot videos" has begun.

A Debate on "Truth and Falsehood"

In the face of skepticism, supporters have also presented evidence.

Some netizens quickly released a backup video, proving the authenticity of the video from different angles.

An American tech blogger named Mike Kalil even published a lengthy article detailing why he believes the video from Lingqi Wanwu "is not fake."

Kalil stated that the actions demonstrated by the robot in the video, such as pouring water, climbing onto a bed, and interacting with children, although impressive, have already been researched in academia (such as papers like ResMimic, HDMI, OmniRetarget, etc.). What MindOn has done is excellently integrate and engineer these cutting-edge research findings.

Kalil concluded that the technology of Lingqi Wanwu likely integrates several cutting-edge achievements in robotics research in recent years, such as using imitation learning and reinforcement learning to train robots millions of times in simulated environments and then transferring the learned capabilities to physical robots.

Behind the "Debate": The Emergence of a Route Dispute

The excitement of this cross-national "debate" is comparable to the robots themselves. But for investors, the more important question is: if this is true, what does it mean?

One analyst pointed out sharply:

If software systems like Lingqi Wanwu can achieve true general functionality on already mass-produced, relatively affordable hardware (like the Yushu G1), they could pose a serious threat to companies like Figure AI, 1X Technologies, and Tesla.

Currently, leading American players represented by Figure and Tesla mostly follow a "soft-hard integration" vertical integration route. They not only need to develop the "brain" (AI software) that drives the robots but also design and manufacture the "body" (hardware) of the robots themselves.

This is similar to Apple, which designs its own iOS system and also produces the iPhone. The advantage of this model is high integration and good experience, but R&D costs and product prices are also high.

In contrast, the video from Lingqi Wanwu showcases another possibility: the "Android model."

In this model, AI software (brain) and robot hardware (body) can be decoupled. Companies like Lingqi Wanwu focus on developing the strongest "brain," while hardware can be entrusted to companies like Yushu Technology, which have already achieved large-scale mass production at lower costs Once this model is successfully implemented, the competitive landscape of the humanoid robot industry will be completely rewritten.

The key to competition will shift from "who can create better robotic bodies" to "who can develop smarter robotic brains."

At the same time, the hardware integration strategy currently relied upon by American companies will be impacted. When inexpensive hardware platforms can also achieve advanced autonomous intelligence, the expensive customized hardware barriers will no longer be so solid.

The market may tilt towards a more open and flexible ecosystem, where software can run on any hardware. This is similar to the smartphone market; if the Android system had matured before the first generation iPhone was released, today's market landscape might be completely different.

From this perspective, the ongoing "debate" about the authenticity of videos is essentially a preemptive collision of two technological paths and two business models.

The anxiety of American CEOs may not be unfounded. What they see could be a highly disruptive "Android moment" quietly brewing in China.

In any case, the competition in humanoid robots has entered a heated stage. Chinese companies are unexpectedly joining this global competition in a comprehensive manner, from hardware manufacturing to software intelligence