Back to Newsroom

I spent a day at a robot “carnival” in Shanghai. Here’s what I saw.

By Modelverse Editorial·August 25, 2026·2 min read
I spent a day at a robot “carnival” in Shanghai. Here’s what I saw.

Shanghai’s robot “carnival,” hosted by an R&D center on the city’s outskirts, drew families to a public holiday showcase featuring more than 100 companies involved in research, development, manufacturing, and marketing of robotic systems. The event highlighted a variety of platforms, from miniature quadruped pets to larger humanoid prototypes, with activities ranging from interactive play to skill demonstrations.

China’s dominance in humanoid production was evident: nearly 90 % of the roughly 13,000 two‑armed, two‑legged robots shipped worldwide in the previous year were manufactured domestically. Attendees watched a mini robotic dog attempt stunts, children guided a quadrupedal robot up stairs, and remote‑controlled bots engaged in a water‑bead duel. Nearby, a DexForce arm brewed coffee while another pair of robotic arms folded a basket of T‑shirts, illustrating both entertainment and functional capabilities.

Despite the festive tone, the exhibition acknowledged persistent obstacles for humanoids—safety concerns tied to their weight and balance, high costs, limited battery life, and ongoing debate over whether a human形状 is the optimal robot morphology. These factors have slowed broader adoption elsewhere, making public exposure events a strategic step toward familiarity and acceptance.

Why this matters

The carnival’s emphasis on direct interaction serves as an evidence‑grounded inference that increasing public familiarity can mitigate skepticism and accelerate demand for embodied AI systems, even while technical hurdles such as power efficiency and actuation robustness remain unresolved. By aligning showcase activities with national five‑year plan goals, China is leveraging mass‑market events to feed feedback loops into R&D pipelines, potentially shortening the iteration cycle for humanoid platforms that must meet both performance and safety thresholds before scaling beyond novelty. This approach contrasts with regions where humanoid development proceeds primarily in isolated labs, suggesting that sociotechnical exposure may be a decisive factor in shaping the trajectory of commercial deployment.

Share this article

Found this insightful? Share it with your community on Reddit, X, or copy the link.

ai-newsbriefmit-technology-review

Footnotes & Primary References

Related content

Situational Awareness, star AI hedge fund that nearly imploded, now being probed by the SEC

The AI hedge fund went from "the talk of Wall Street" to "subject of federal subpoenas" faster than you can say "diversify your portfolio."

Read article

Wire It, Run It, Deploy It: AI Workflows in Gradio

Wire It, Run It, Deploy It: AI Workflows in Gradio

Read article

Disrupting a new covert influence campaign from Russia

OpenAI banned Russia-origin accounts using AI to promote a fake Israel-based think tank and a “sovereignty” index praising Russia and criticizing the West.

Read article