Kevin Kelly once predicted in “Out of Control”: “The largest factory of the future may have only two employees: one human and one dog.”
The human’s job is to feed the dog, and the dog’s job is to prevent the human from touching the machines.”
This future is accelerating towards us.
The global AI robot industry is developing at an astonishing speed.
In the first half of 2025, the average revenue of the humanoid robot concept sector in the A-share market increased by 9.72% year-on-year, while the average revenue growth rate in the AI application field reached 18.7%, significantly higher than the overall level of the A-share market.
At the same time, the Chinese government issued the “Opinions on Deepening the Implementation of the ‘Artificial Intelligence+’ Action,” clearly supporting the development of intelligent robots. Beijing has even established an innovation center for humanoid robots and released the world’s first standard for the “Intelligent Classification of Humanoid Robots.”
01
Industrial robots are transitioning from “mechanical arms” to “intelligent hands.”
In 2023, traditional industrial robots accounted for only 60%, collaborative robots made up 30%, and adaptive robots with force control sensors, although only 10%, have become the fastest-growing “dark horse.”
The advantage of adaptive robots lies in their flexible production capabilities.
After being introduced in an automotive parts factory, the pass rate for polishing increased from 88% to 95%, and the annual maintenance cost per device was reduced by 30,000 yuan.
Henry Ford once said, “Efficiency is the cornerstone of industry.”
AI robots have elevated this concept to a new height.
Industry insiders point out that the cost reduction of industrial robots is not simply about “lower prices,” but rather a value reconstruction driven by adaptive technology that combines “increased precision + reduced costs.”
A netizen, “Manufacturing Master,” joked: “In the past, humans operated machines; now machines guide humans. Our factory’s experienced workers now have to consult robots for process parameters.”
02
In the medical field, AI robots are transitioning from “auxiliary tools” to “business partners.”
Estun’s 31-degree-of-freedom robot has captured 23% of the market share in the medical exoskeleton field, with revenue reaching 2.549 billion yuan in the first half of the year.
The North American market is leading the way, with its medical service robots already capable of high-value scenario applications—such as drug delivery robots in hospitals that can directly link to patient vital sign systems and automatically pause delivery when patient indicators are abnormal.
The principle of “first, do no harm” from the Hippocratic Oath is being programmed into the core algorithms of every medical robot.
Academician Wang Tianmiao from the Chinese Academy of Sciences pointed out that the medical care field is expected to be the first to achieve practical applications, including a series of subfields such as hospital applications and companion communication.
A netizen, “White Coat Angel,” commented: “Surgical robots performing operations, rehabilitation robots providing care; we nurses may need to transition to become robot coordinators in the future.”
03
Home service robots are moving from the laboratory to the living room.
Fourier’s GR-3 focuses on home services and plans to enter offline stores to provide services in the second half of the year.
However, challenges still exist—the single-unit cost of humanoid robots remains as high as 199,000 yuan, making it unaffordable for most families.
On the technical side, robots have a motion accuracy of less than 90% in complex environments, which still falls short of the requirements for home environments.
As mass production scales up and costs decrease, home service scenarios will gradually expand.
Bill Gates once predicted: “Robots will become standard in every household, just like personal computers.”
This prediction is coming true.
A netizen, “Smart Home Enthusiast,” complained: “My robot vacuum is already clumsy enough; I can’t imagine a humanoid robot bumping around in my house. But honestly, I would still buy one!”
Since the second half of 2024, many intelligent model startups have received hundreds of millions of yuan in financing:
Qianxun Intelligent raised 528 million yuan in Pre-A round financing;
Self-variable robots raised hundreds of millions in Pre-A+ round financing;
Star Sea Map raised 100 million USD in Series A financing;
Youlu Robot raised over 100 million yuan in angel round financing.
The large and small brain architecture has become the mainstream choice for the implementation of robot large model engineering.
This architecture uses pre-trained large models as the “thinking” system, with lightweight control small models completing the “reflex” from thought to action.
Professor Wang Tianmiao proposed that 2025 will usher in an industry landing driven by “intelligent boundary definition,” which will redefine models, computing power, and platform issues by defining boundaries in specific scenarios.
A netizen, “AI Dog,” humorously commented: “If robots have large and small brains, will they also, like us, have their brains saying it’s time to sleep while their small brains are still scrolling through short videos at midnight?”
Research and education have been listed as the first phase of application scenarios for robot implementation.
These scenarios do not require high generalization capabilities, and the ontology companies do not need to achieve technological breakthroughs at the software level but can quickly land a small number of delivery orders.
Einstein once warned: “Technology has surpassed our humanity, which may be the most critical issue of our time.”
A netizen, “Future of Education,” lamented: “My child is taking the college entrance exam this year and asked me what major has a future. I said robotics engineering, and he asked me: ‘Is it about building robots or preventing being replaced by robots?'”
According to Goldman Sachs, by 2035, the global humanoid robot market size is expected to reach 38 billion USD or even 154 billion USD.
However, in this transformation, the true winners are not just those companies that manufacture robots, but those who can dance in harmony with robots.
A netizen, “Friend of Machines,” profoundly stated: “Don’t worry that robots will replace all jobs; they just make us rethink what truly valuable human work is.”
The future has arrived; it just hasn’t been evenly distributed yet.