Humanoid robots were still trading humans just a year ago, but this time they've taken a remarkable leap, shattering the human record in a half marathon.
At the 2026 Beijing Etown Half Marathon, a robot named Lightning crossed the finish line in just 15 minutes and 26 seconds, nearly seven minutes faster than the human world record and almost two hours ahead of last year's winning robot.
This year's race also featured a larger field of competitors, a sharp rise in autonomous navigation and a higher completion rate, even on a more demanding course.
So what's behind this rapid progress in humanoid robotics in China?
How will they evolve in the near future?
And what could that mean to our daily lives?
Hello and welcome to the panel discussion of World Today.
I'm Do Hongyu in Beijing.
To explore these questions and more.
We're joined by John Gunn, a professor with the University of International Business and Economics, Muhammad Ali Nasir, professor of economics at the University of Leeds, and Liu Shaoshan, Director of Embodied Intelligence with the Shenzhen Institute of Artificial Intelligence and Robotics for Society.
Thank you so much for joining us, dear panelists.
First of all, I'd like to get your very first impressions of this robotics marathon that just finished a few days ago.
So, Professor Gong, can we start with you?
What impressed you the most about the humanoid robots marathon?
I think what impressed me most is that it is able to run like humans in an entirely outdoor open space environment.
And there are a lot of things, not so much control in a, you know, as opposed to being a sort of a controlled indoor environment and can still run and runs like humans.
And also the you know the number of robots that have been participating in this from various companies.
That represents the overall level of advancement in humanoid robots technology in China.
I think that's probably the most impressive thing to me.
And then what about Dr. Liu?
Well, I think the existence of the Mirafon itself is very amazing.
Kudos to the local government who organized it.
And the second most impressive thing, I think, is that a newcomer got the first sixth place in this competition, although there are so many companies participating.
But a newcomer is able to revolutionize the whole competition.
That tells a lot about the barrier of the field.
Professor Nasir, what do you think of this humanoid robots marathon?
Yeah, it was fun to watch.
It was impressive, very impressive indeed.
But there are so many nice things happening in China and China is making so many advancements on various fronts.
So you say, wow, this is another one.
For me, there were several robots which are autonomous.
They didn't need any live guidance to run.
I think that's something which was very impressive and that shows the development on the AI front.
The speed of the robots.
If I'm not wrong, I think they have made a new world record of half marathon and the tough conditions in which they were operating.
I think as compared to the last year marathon in China, this time the conditions were a lot tougher.
So running over 20 kilometers in the conditions which are varying they are not controlled lab conditions.
I think that was very impressive.
So yeah, indeed, like other guests, I mean, I'm deeply impressed.
A lot of people have this question in mind.
We want robots to work for us.
We want them to serve us.
But what's the purpose of putting them through a half marathon?
Dr. Liu, what do you think are the key capabilities that they were tested?
Well, it's like why human beings have Olympic Games.
It's a test of the capability to interact with the physical environments.
I think those are the foundations for different tasks, whether it's a robot working in a factory or a robot working in your home.
It needs the basic capability to interact with the physical environments.
So I think marathon is a starting point.
Soon.
I would expect the whole marathon game would become a full Olympic game for robots in the near future.
Then, Professor Nasio, what do you think are the capabilities that were tested during this marathon for humanoid robots?
Yeah, I mean, these marathons, it is fun to watch them.
They bring several benefits, including the stress testing.
I mean, the first benefit I can think of is also the introducing a technology to the public and preparing them for the technological adoption of.
These robots will be becoming part of our society, and already in some way part of our society and our industry, but their usage will increase.
So in terms of introducing them to the public, showcasing them, that also has its value.
But, as Professor Liu said, there was element of stress testing, checking the hardware also checking the software and the AI capabilities, particularly for those robots which are autonomous, whether they can operate for a prolonged period of time autonomously.
That got checked and I think they did very well on that.
And then also to share experience with the colleagues and learn from their experience.
And those robots which might have malfunction at some problem for their manufacturer.
This was an opportunity to learn and improve and then come back again next time.
These were the benefit and these were the reason that they were there.
Indeed.
And then, Professor Gong, why is there a focus on human-like forms of robots?
Aren't robotic vehicles faster?
Aren't wheels faster?
Yeah, that's a very good question.
I think it also begs the question in general, why do we need robots of human-like forms in general?
That's actually a fascinating question.
I think the one important reason of a human like a robot is really the experience, you know, the user experience interacting with humans.
You know, there's this kind of a sort of emotional, sentimental feeling on the human part to act, interact with something that looks I like them.
Right.
I think that's that's pretty important.
The second reason, I think, is you know after all, the human like form imparts you know some level of a great level of complexity and sophistication.
You know, I think a very good example is that, you know, the function of a hand.
The hand is probably, as Elon Musk once said, the hand is probably the most sophisticated organ on the human body, and a robot replicating that function is immensely, immensely challenging.
I think having a humanoid robot running the marathon is just the mere beginning of all of this.
I think it represents Probably, you know, just the tip of the iceberg.
It's the beginning of a set of functions that we would expect from a human-like robot that is supposed to be functioned inside a home for various functions.
And running the marathon is probably the easiest one or is the beginning one.
So I think that's a reason why, you know, these companies participate in this race to demonstrate the capability.
I think this just represents the very beginning.
You know, the technology is improving and advancing rapidly.
And within a few more years we will have, you know, way more complicated robots, you know, interacting with humans in many capacities, in many environments.
Again, I want to emphasize that a big breakthrough, in my view, is to be operating in an uncontrolled open space environment.
That requires a great deal of situational awareness, a great deal of sensing capabilities, encountering things that are probably not expected.
These are all complexities, and if a robot can operate in that kind of a complex environment, I think it bodes very well for putting this thing into a much, much more complicated environment.
Again, this is the beginning and it looks like it's a big success.
And it's the fact that, you know, one robot runs faster than man is a feast in itself.
And I can't help thinking about defense applications, actually.
What about putting a robot on a battlefield that can run faster than a man, that can carry a lot more load than a man, that's not tired, doesn't know tiredness and can function in other aspects very much like a human?
So this is a great application.
Great.
And we're seeing the humanoid robots being more and more like humans.
And this is just the beginning.
Then, Professor Nasir, how are our daily lives connected with robots in the marathon?
How do the capabilities demonstrated by the robots translate to real world value and applications in our life?
Yes, indeed.
I mean, the marathon is, of course, fun and it is to introduce the robots to the world and what great works different companies are doing.
But in the real world setting, of course.
Once we have tested it in a condition like a marathon, which are tough conditions, then we could see a test that they would be able to use them in factories, in hotels, at home, for domestic use, for instance, sorting out your clothes or washing up your utensils after eating, which personally I don't like much.
Doing emergency responses, delivering mail or working as a firefighter in the retail space, hospitals.
There are many areas the robotics could be used and these capabilities which were demonstrated in these marathons, they show that yes, they are to some extent ready to work in these conditions, which are not controlled conditions as we may have in a lab.
So therefore, I think this marathon has many benefits to demonstrate these capabilities.
There are conditions in which the humans have been working and those sometimes, conditions are not very healthy in the case of, as you mentioned, some catastrophes like earthquakes, like floods, like fires.
I mean if there's a vast fire or forest fire in some of the countries, in Australia, in the US, they sometimes get huge fires and it is very dangerous to send human to fight these fires.
The use of robotics there, I think, could be very safe and very effective in very dangerous conditions.
So yes, this demonstration was indeed showing us that yes, there is a good potential of robot usage and there's a potential for further development.
And there's been a lot of development, particularly in China, on these fronts.
So I think this would be of social benefit not only for China or Chinese population, but for the rest of the world and for other countries.
I think these developments, as I see in the future, they are very valuable and we'll see their benefits in the coming years.
Great.
This year's robots performed much better than last year's, as we have just discussed.
The most obvious is about the speed, right?
The champion robot Lightning, or Flash, completed the 21 kilometer course in 50 minutes and 26 seconds.
That is not only quicker than the quickest man in the world, but also nearly two hours faster than last year's robotic winner.
So, Dr Liu, could you please tell us what technological advances have made this possible?
Well, i think the competition has not yet started.
Uh, if you look at last year, people think it's a show.
So just have a robot running, that's enough.
And this year we see some serious players.
For example, the newcomer got the first sixth place and if you examine its form factor, it's uniquely tailored for raising.
So I would say that's not just a humanoid robot, but a raising robot.
But if you think about it, a newcomer spends six months to a year's RD effort can get the first six places of this competition.
That means the competition has not yet started.
I'm not a little bit amazed by the fact that it's faster than a human.
It's definitely faster than a human.
It's a machine.
So we will see more competitions in the years to come as people take it more and more seriously.
Then Dr. Liu, what do you think has made the newcomer to win?
Maybe does he have a better body shape or better batteries, etc.?
They take the game seriously.
So they tailor-made the robot for the game instead of using a general form of robot.
If you look at the human being, right? a regular person or a normal person compared to using a boat.
The form factor and the shape of the bodies are very different.
So I think the newcomer tailor-made the shape for racing, whereas the other competition or the other companies they just use whatever they have to join the racing game.
We've also have seen some other changes.
Professor Gong have mentioned about the autonomous navigation right, which has expanded rampantly this year.
Last year, apart from the champion, all other teams relied on remote control.
But in this year's marathon, nearly 40% of the over 100 teams used autonomous navigation.
So, Professor Gong, what do you think have driven this shift?
Well, first of all, autonomous navigation.
Navigation is this kind of technology is progressing rapidly.
You know, my understanding is that it's progressing rapidly in the vehicle space.
I think there's a certain level of similarity between a running humanoid robot and a autonomously driving car.
There are certain, you know, technological similarities there, so there's a race in the vehicle space.
Practically every automaker is driving the technology development in that area, um and and my guess is that some of this is also being leveraged off into the human robot space humanoid robot space.
I think the underlying technologies has to do with the sensing technologies, using the technologies embedding the mapping stuff and other sort of AI-driven decision-making issues.
So I think a robot without that is probably unimaginable, especially a humanoid shaped robot.
So I think this is probably a necessary technology and every rubber maker is investing in that area.
So I'm not surprised to see that 40% of the teams used autonomous navigation technology.
I think this is probably the background.
I do want to make a point that once you have an ecosystem, I think clearly China is indeed cultivating an ecosystem.
From an individual company's perspective, developing a unique technology is relatively easier.
We have an expanding and enriching ecosystem with many companies contributing to technology.
You know the development of this line of technologies.
There's going to be talent spilled over.
There's going to be supply chain relationship spilled over.
So, you know, I think, you know, all of these factors I've just mentioned are driving this shift.
And China is advancing very rapidly.
So I'm not surprised to see that.
This year has been a great year for a lot of competitors.
We've seen a better completion rate, which jumped from 30% last year to 45% this year.
That's despite a more complicated track with more turnings, sharper turnings and slopes.
And even those who didn't complete the race, they have made a lot of progress.
For example, there's this very small, tiny robot which is about the height of a three or four year old kid.
It didn't finish even the first 100 meters last year, but this year it ran for 10 kilometers and it won a lot of love from the spectators also.
So, Dr. Liu, what has driven the overall improvement of humanoid robots in China over the past year?
Well, I think that's a very good question.
If the completion rate, regardless of the race is becoming harder, jumps from 30 to over 45, that says one thing.
That says all the individual technology is there.
Whoever does better optimization, and integration would have a better product.
But that implies a better question or better answer to the question that I think China has shifted from a demo phase to a product phase.
Because now people are getting serious about integrating individual technology into a product and the product can perform a real task.
The task for now is running a marathon, but we have achieved a very high completion rate.
So that's a very important shift in the past year.
And then, Professor Nasir, what do you think about what has driven the improvement of humanoid robots in China in just one year?
Yeah, I mean, part of it is, of course, learning from experience.
But of course, indeed, there's improvement, development in AI.
The pace of development is very fast there.
There's improvement in the in the structural of the robots and increase the structural stability and agility physical agility of the body and the shape.
So that has improved a lot.
And also there were improvement in the, I think the cooling system, which helped them running.
And then there was also this formula one style stocks The board staff, their batteries get changed and they get lubricated.
While they were fun to watch, but those improvements in their technology and tactics, I think they delivered better results this time.
And I think next time when we have the marathon, next time I expect that there will be even further improvements and there will be a lot of learning there from this time, and I expect there will be more contestants joining, which will increase, of course, the quantity and also the quality of the products and robots we will be seeing next time.
So I think there are several factors contributed to increase in the performance this time.
And maybe winning or losing isn't the purpose or the goal here.
Some argue that failing the race may be more valuable than winning it, because they might have more chances to expose their problems and to make improvements.
Professor Gong, what weaknesses have been exposed in this year's race?
Yeah, well, there are some reports.
I mean, I didn't witness the race on a spot, but I've read that there are a couple of things sort of reported quite interestingly.
One thing is that many participating robots have humans running beside it.
And some of them actually use the remote control to control the robots.
So my immediate impression is that it's not 100% entirely sort of autonomous navigation.
I mean there's some inference, some remote control happening on the side by a companion human running on the side.
That's one thing that i, i think uh desires a lot of improvement.
Um, the second thing is that i've seen reports saying that you know they change battery in the middle, even though they change it very, very quickly.
That means that you know the battery technology for a robot continuously running for about an hour, I guess, right.
A little bit short of an hour.
It's still quite problematic, I would imagine.
You know you can't add battery because it adds weight and then, you know, increases more consumption of power.
So it's going to be an optimized tradeoff.
And I think probably I don't know about the robot that's the championship robot.
Does it run all the way through without any change of battery or it also went through a change of battery?
We don't know.
But obviously, you know, many more robots that have participated. do change battery along the way.
So I think that's actually not so much a robot technology issue.
It's a battery issue.
And also in terms of weakness.
It also reports off quite a large percentage that actually didn't make through the finish line.
That means that the The technology is still not mature yet and it still waits for more improvement.
I guess.
These are the three things that I can think of.
Actually, the winner also changed its battery in the middle of the race.
I see.
And there were a lot of videos going viral online about the participating robots falling on the ground with their pieces running around on the track.
And a lot of staff just came with a stretcher to pick them up and to rescue them.
Then Dr. Liu, moving to you, where do you expect the biggest improvements by next year's race?
Well, every single aspect, right?
If you look at the upper limit of human and you look at the Olympic game, that's how fast we can run, say 95 seconds for 100 meters.
But if you examine machine, we don't know the upper limit yet.
So we still have to explore.
But that ties to a lot of different angles.
For example, battery, as we just discussed, how good your motor is.
I mean, if you compare that human or robot to a car, you still have a long way to go.
Imagine that's a Ferrari.
Assume that's the upper limit that the machine can do, then there's still a long way to go.
And then autonomy.
I think autonomy is another thing.
Some of the robots this year is autonomous.
Others are not.
Still remote control.
So we would like to see a game that says you cannot do remote control.
Let's switch to a fully autonomy mode.
So I think there are a lot more areas that we can improve upon.
Then, once that's accomplished, then we can turn the game from a marathon game into a full Olympic game with different categories.
Okay, thank you all for in-depth analysis.
Let's take a short break.
Coming back, we'll continue our discussion.
Hello, my name is Alessandro Golombievski Teixeira.
I'm a professor of public policy management at Tsinghua University in Beijing.
I am a great listener of The Wall Today.
In my opinion, The Wall Today is one of the best China radio programs.
In the world today, we can get the best news and analysis in what is happening now in the world.
So please come to join us.
Welcome back.
You're listening to the panel discussion of World Today with me, Dou Hongyu.
We've been unraveling the development of China's humanoid robots.
Let's continue our discussion.
And now the robots are being tested in much, much more scenarios.
There will be the second World Humanoid Robot Games in Beijing this August.
And the robots will compete in various sports apart from running, like gymnastics weightlifting, dancing.
And even the tug of war, I can't help picturing robots doing the tug of war.
So first, Professor Gong, compared with running, what other capabilities will these events test?
Well, I guess gymnastics, right?
To be more agile, to be more nimble, to be more athletic, I guess.
So it requires a lot of balancing skills and the actions are not very repetitive, unlike running.
We already have a taste of this during the CMG Spring Festival gala event.
They already demonstrated a lot of capabilities.
I guess, again, it's all about control.
You know, in that environment, we understand that there's a lot of control on the side.
Whereas this competition, I would imagine they would dispense some of these controlling capabilities see if the robots can truly do things on its own.
So I think that's certainly a much more high level of sophistication and capabilities.
These things are advancing very fast.
And you look at the number of companies participating in this marathon, immediately you can see how many companies will be participating in this intensely competitive environment.
I would expect great technological breakthroughs would happen.
These are more demanding capabilities will eventually develop.
And all of these things I think are quite useful for a true application inside a human family, inside a human house, and do wonderful things that we probably haven't imagined yet.
Maybe they will evolve into something that will work for us in our houses.
Then, Professor Nasir, Professor Gong just mentioned a few capabilities that were tested during gymnastics or dancing classes, like the flexibility, the balancing capabilities, and they may be beneficial to building on their capabilities to work for us in our houses.
Then how will these capabilities connect to other practical and real-world use cases?
Yes, I mean, indeed, the purpose the final goal is to make these robots useful for the society and when we check our test different capabilities and events like marathon where various capabilities are tested at the same time i think if you see like games where a robot is lifting weight That is basically demonstrating the weight lifting capabilities, whether it can lift 20, 30, 40 kilograms weight, and then whether that would be useful in the weight lifting, not only in industries, but at home.
And if they are doing the tasks which are more complex Or there are scenario-based events where they are to do, maybe sorting out clothes or washing some dishes or utensils.
Those 10 tasks then, would be, of course, translated into the real use, when they are used at home, they are used in factories or they're used in different social settings, like hotels are in the retail sector.
And one sector which I think is not that much talked about is maybe the agriculture sector.
I think that is an area where humans have been working for centuries and there's more scope of robots to be used.
And in the world humanoid robot games, the robots will be challenged in a lot of scenarios that you've just mentioned that are much more closer to us, like doing housework and firefighting and pharmaceuticals, like picking medicines for patients. and also retail.
These involve the capabilities to work with human beings.
When the robots dance or doing gymnastics, they don't have to interact with humans, right?
But, Dr Liu, how are these tasks like doing housework and working at pharmaceuticals?
Challenge the robots differently from the sports tasks.
Well, I think sports is about pushing the limit of an individual capability, whether it's running or lifting weights.
Then the real world task is about combining tasks, these different atomic operations or capabilities, into a scenario based skill.
So technology wise, there are two layers.
The first layer is about developing the fundamental capability.
That's the sports like competition.
Then we see a lot of progress in the past. a year or so.
Then, on top of that, we need to build a second layer such that different companies would integrate this individual capability into a sequence of tasks that the robot would perform.
And on top of that, you have to interact with users, human beings.
So right now we don't see a sports game for robots to compete on how well they can interact with the humans.
So I think that's the weakest link that we have to make up for.
But in terms of other functions, such as performing basic tasks, then we have seen a lot of progress in the past year.
So, Professor Gong, how close are we or how far are we to seeing humanoid robots working in our everyday environments, like homes or hospitals or tourism spots?
And what are needed to achieve the widespread real-life applications of such robots?
I think we are actually close, very close.
Um, it depends on you know how we define working in everyday environments in a home.
Um, if we're talking about, you know, doing choice at home um, you know, like cooking, like cleaning, that kind of things um, i think they are a little bit more complicated.
But if we're just talking about I mean, there's actually another entirely different set of applications inside a home that I'm seeing and actually I'm expecting and forecasting very rapid development and probably can enter into a home environment a lot sooner than people expect.
That is just like companionship.
Today, a lot of people raise dogs.
You ask, why do you raise dogs?
The dogs are being raised very much as a human friend, a human's best friend.
And they're serving the purpose of companionship dog.
You can't talk to a dog, but you can interact with a dog.
I think you know this kind of application is probably a lot closer.
There are a lot of singles out there who probably need some kind of companionship and I think to some extent it's already happening.
This kind of application, I think, can be relatively easier to be developed and uh, you know, with the ai large language model development, you know it fits very well with this kind of application and furthermore, i would imagine that the cost can come down very rapidly.
These things have been made in china with intense competition.
The cost can come down very, very rapidly and being a lot more affordable.
So I think you know to answer your question.
I think depends on how we define working.
What do we mean by working in a home, right?
And I think for a set of applications that are not very demanding on the side of sort of physical functions but more on the emotional and interactive capabilities, we are probably a lot closer than we think.
And I would imagine probably the first wave of robots getting to people's home will be of this kind.
And probably this is a very large market out there.
Yes, indeed.
And then Professor Nasir.
Given China's current progress in humanoid robotics, what expectations do you have for Chinese teams and the upcoming World Humanoid Robot Games?
I mean, yeah, with China, any expectation you have, however high it is, it gets beaten.
You know, they always amaze you.
So with this World Humanoid Robot Games we had last year, they were fun to watch.
So This time in August I think, maybe the second World Terminator Wars game.
I'm looking forward to them, to be honest.
I just got back from China last week after spending about a month and a half.
And now I'm looking forward to my second visit.
And I was thinking, if I go either in August or end of July, I might be able to watch some of these games.
So I'm looking forward to them.
And, as I heard the way they are planned, they will be for a few days and there'll be many different categories competitive events and scenario-based events.
Which I find very interesting.
There will be gymnastics, sports, dance, tug of war, various things.
And I think, yeah, that's something really something to look forward to, to watch.
And in terms of, as I said, expectations.
Coming back to your question, I mean, with all the investment made in this domain, in the AI, in the robotics, for the last few years and the rapid progress and the development made by China and the Chinese scientists, engineers and teams, I think that is very impressive.
So I am sure they won't disappoint me.
And then Professor Ghosn.
Building on that, where do you think does China currently stand in the global race for humanoid robot development?
Well, at the very front end.
I think we understand that there's a race between the United States and China in AI.
Know, humanoid robot is more than ai.
It's on top of ai, it's hardware development plus ai.
Um, and it's the hardware development side that china has a lot of experiences, i think have some advantages there.
Um, we have an ecosystem, we have very, you know, extensive manufacturing and uh Also, we have a constant advantage in terms of developing these kind of hardware.
And so we have a very large market.
And probably you know.
Quite interestingly, I want to point out this point is that the Chinese consumers are very much embracing, you know, these kind of new technologies.
They don't mind. you know, the prospect that the products are not so much mature initially.
I think Chinese consumers in general are more inclined to try new things.
So uh, you know, we have a very good environment which contributes to rapid market development uh, rapid market trials, uh.
So i think uh, you know definitely, to answer your question, we are developing, we are at the very forefront of this um and uh.
Um, we are not only at the very forefront of this, but also we have a large team of companies at the forefront of this.
So I think we're in a fairly good position right now.
Then Dr. Liu, as we know, AI chips are the soul of robots.
But have Chinese robot developers been affected by the restrictions on AI chip exports to China by the United States?
Well, used to be.
Although the in-body AI robots make a lot of headlines, but we have to notice that The processor business has made tremendous progress in the past few years in China as well.
The whole semiconductor supply chain.
If you examine the number of startups doing chips for robots in China, it's over 10.
In the United States, it's much less.
So we can project that in the next few years the chip will become very mature to support the whole ecosystem growth of robots.
So I don't think that's a constraining factor anymore.
And then Dr. Liu, here's a follow-up question for you.
How does the humanoid robot development in China differ from that in the United States, or are we moving in the same direction?
Well, we are more at the fence, if I'm completely honest with you, because we have a more mature supply chain, whether it's hardware components, whether it's battery, it's a transfer from the electric vehicle business.
We have a better electric vehicle business in China, and then we have a better supply chain for electric vehicles.
Thus we have a better supply chain for robots.
Uh, so we see a transformation like that.
Uh then, in software, we're catching up.
Uh today, i think, deep six just released their latest model.
Uh, some of these capabilities can be integrated into robotics.
And you just mentioned chips.
Then we are not doing bad in chip business.
But most importantly, we have more usage scenarios.
Chinese consumers are more welcoming towards different kinds of robotic applications, whether it's a drone delivering food for you or some other form factors.
So all in all, if you compare the two countries, I think China is leading right now.
China is leading.
And then, Professor Nasio, what do you think about the direction of development in China and the United States?
Are the humanoid robots developing in the same directions in the two countries, or are they taking different paths?
I think there is some level of divergence or different paths, you can say.
I mean, I would agree to the two participants.
There are two contestants, really.
I mean, if you say okay, which are the significant contestant or player in this race of AI, in robotics?
These are US and China.
And the growth, the rate at which China is growing, or the Chinese robotics there's a lot faster than the US.
So as the things are developing, I'll see the China will get well ahead of US in the coming days.
That's how I see it.
The Chinese employment of reports in industries and the different sectors is a lot higher than the US and it's higher than the rest of the world combined.
In terms of Chinese model of AI and robotics development and is different from the US, same as different industries in the China, as they differ in terms of their strategy and the culture from the US industries.
China is more open.
It focuses on innovation and then sharing this innovation and its fruits with the society, Chinese society and also with the rest of the world, where the US model is more focused on innovate, to then profit and to skim as much as you can.
So, where is the chase of China is the broad development. and some of it as earlier mentioned that the acceptance of the Chinese society of the new technology so the society accepts and loves to have new technology and their usage and that gives Chinese firms an advantage in a sense they will introduce something that would be welcomed by the public and also for the global south i mean if we look at in the case of let's say even evs electric vehicles you can see their usage is i mean very popular in the global south in many developing countries love to use the chinese evs in terms of their performance and also affordability.
So in China, this is the story of also scaling.
So you produce something and then you produce it on a larger scale, which reduces, of course, cost.
And then many people benefit from that in the China and outside the China.
Whereas in the case of the US, there is, of course, development and focus on the innovation and high-end innovation.
But I don't really know that whether US would be able to.
With that mentality or strategy, they would be able to scale it.
And if they can't scale it, then it would not maybe economically viable to use these products in the society, in the economy at a very large scale.
And they may then get beaten in this rate due to that strategy or that mentality they have.
And China will get ahead of it by providing products which are good, which are also cheap, So the Global South can also afford them and they have wider social benefits.
And Professor Gong.
Just now, Professor Nasir mentioned that China's advances in humanoid robotics will also benefit other countries.
I want to ask you what spillover effects could it have on the robotics development in other countries, and also for the countries that currently don't have the ability to develop its own technology.
Humanoid robotics, yeah well, I mean, there are plenty of examples out there, you know.
I want to point to the solar panel example.
Used to be very expensive, you know, with the Chinese companies relentlessly driving the cost down, Certainly we're very successful in that industry.
We've been dominant in the world market.
Now all of a sudden, many more consumers have access to this wonderful technology.
In a sense, they have light in the middle of the night.
For a lot of people on these remote islands in Indonesia, for example, a lot of people in remote places in Africa, for example, who have never would be able to have access to electricity.
Now they can buy these things.
So I think this is definitely a benefit for the rest of the world.
I think all of us can be assured that once humanoid robot industry is being developed and becoming rapidly growing In China, there will be benefits for the rest of the world.
There will be more affordable products accessible to many consumers around the world.
There will also be foreign direct investment coming from Chinese companies.
There will be technology spillovers, talent spillovers, and all of these things typically would happen when we have a very good, strong industry, a sunrise industry for the future.
So I think these are great benefits to the rest of the world.
And I think one very important distinction I need to point out is that here in the Chinese market, we still have a very strong government.
And the Chinese government usually makes a trade-off between the healthiness of the industry, the health of the industry, versus consumer welfare as well.
So we won't allow, for example, exorbitant um, you know, really obscene high profit kind of a pricing um for products on the chinese market and uh um i i gave you one very good example.
For example, you know the the medicines killing aids, for example, curing aids and other very difficult diseases.
You know some of these peers coming from pharmaceutical companies in the United States.
I mean they're being sold very high prices in third world countries.
But here in China, I think our government makes sure that these things are.
You know, we need the companies to be profitable.
But at the same time we also want to make sure that you know consumers are not going to be ripped off.
So I think from that perspective, I would argue that it's a very good thing for the international market, for people outside the country.
And then Dr Liu, looking beyond China's development plan for the next five years, has stressed the building of a solid foundation for artificial intelligence, which includes computing infrastructure, algorithm innovation and data resources.
So computing power, algorithm and data.
How will these new AI infrastructures propel the progress in humanoid robotics?
Well, I think that's a very good question.
As you can see, in these arrays, just a small portion of the robots have autonomy capability.
The AI infrastructure would support different kinds of autonomous machines, such as embodied robots.
But then now we haven't seen the real challenge yet.
For example, in this race, although it's a semi-structured environment, But it's nothing compared to the open field, such that if you want to deploy a robot in desert, compared to if you want to deploy a robot in the mountains, the intelligence capability is not there yet because the infrastructure is not ready yet.
Once we have a very mature infrastructure, including open source software such as DeepSeq computing clouds, data gathering capabilities,
Then we will form a closed loop to feed different models to different robots so that the robots will gain capabilities to work in different fields.
And then Professor Nasir, what's your view on this?
China is building new AI infrastructures.
How will that help with the progress in humanoid robots?
Yes, I mean as a part of a 15th five-year plan and, as we see, in the last few years China has focused on the, the technological self-reliance, and also there's been progress on the technology ladder and more focus on the high-end technology.
So there is robotics embodied AI.
This is a priority of the government.
There's investment On this side, their investment in integrating AI into the economy and increased use of robotics.
There is investment in advanced manufacturing and the goal is social modernization.
So we do see that now there is a focus of the government on the GDP growth, but not only the quantity but the quality of the growth.
And for that purpose, you need robotics.
For that purpose, you need development in AI.
There is increased expenditure I think more than 7 annual increase in the RD spending.
There's also increase in spending towards a low carbon economy and decreasing the carbon intensity of the chinese economy and again for that, the government is prioritizing the embodied intelligence and the ai and the robotics.
So, as the things are developing, i'll see more investment in these areas, which would then definitely deliver further in advancement and better products, whether they are in the AI or whether they are in robotics or in other areas.
And now there are a lot of concerns about the development of human-like robots because, as they are acting more and more like us human beings, sometimes it looks kind of scary for some people.
There are concerns about the safety and even consciousness about the robots.
Some worry that those human-like robots might one day get its own consciousness and hurt people, maybe.
So, Dr. Liu, how realistic are these fears?
I think safety fears is realistic, but consciousness, I think it's far, far away in the future.
With these robots integrating into society, I think we should worry about two things.
First, it's their reliability and safety.
What if someone hacked a robot from a remote location and Have the robot to do malicious operations.
I think that's the biggest concern.
The second concern is job market safety, meaning that the robots are here to replace jobs.
What do we deal with the societal changes?
I think those are the two main concerns, at least to me.
And then, Professor Nasir, what is needed to ensure the safe and responsible development of robots and also the safety of our jobs.
As far as the health and safety is concerned due to the malfunctioning of the robot, illegal or criminal practices?
I think In a country like China, where the environment is very secure and there's very good law and order situation.
But I think in terms of their safe usage.
I mean, of course, whenever we have the new technology, there's a debate two three, 400 years ago in England, when we had industrialization, then there was a movement that maybe this machine will take over the jobs.
There would be restructuring in the economies.
There would be different skill set that would be needed.
We would need jobs which would be maybe More humans.
Maybe we need the jobs where we need human experience, human touch, human involvement.
Those jobs would be probably required, more than those jobs, those tedious jobs which can be done by the machines.
Thank you all for your in-depth analysis.
They were Zhang Gong, a professor with the University of International Business and Economics, Muhammad Ali Nasir, professor of economics at the University of Leeds, and Liu Shaoshan, director of Embodied Intelligence with the Shenzhen Institute of Artificial Intelligence and Robotics for Society.
And that's all for this edition of World Today.
To listen to this episode again or to catch up on our previous episodes.
You can download our podcast by searching World Today or you can follow our X account and CGTN Radio.
Thank you so much for listening.
Bye for now.
