Chinese robot runs 100M sprint quicker than Usain Bolt's world record
theguardian.com
theguardian.com
Watching robot after robot being taken away on a stretcher was pretty funny.
We've had purpose-built machines that outperform every human at specific tasks for over a century now. Just because a subset that's popular these days (the "humanoid" and dog-like variants) invoke some psychological and cultural (we've been primed by media for over 150 years at this point) fears, doesn't make them more capable than they seem.
Sure, the thing can run real fast, but looking at it, that's pretty much all it actually can do. In contrast to a boring old industrial robot, though, being able to win a half-marathon or a 100m dash against any hypothetical human isn't all that useful, no?
Emptying a washing machine located in the cellar/basement, taking the clothes to a room on the ground floor, folding and sorting them before putting them in the closet still is quite the challenge.
Wirelessly hooking such robot up to a data centre running heavy machine learning models isn't a proper solution either, because in contrast to digital agents, latency matters quite a bit in physical interactions.
Until then human is the best at navigating and moving across various terrain.
Also, further counterpoint, if we reject industrial produced food, who spent $139 billion at McDonald’s in 2025?
perhaps in the US, sure, you guys are weird
But in Europe, McDonalds is a drop in the bucket compared to all the rest of the restaurants. People DO prefer real stuff
As it stands I am not hopeful. I believe the frog will be well and truly boiled long before we reach the rejection en masse stage and it will be too late to change course
What does that even mean?
BTW I consumed Soylent regularly for several years while living in the US and was never "disgusted" by it. Would love to get my hands on some now as it's been so long. You should watch your arguments.
For many humans "watch your arguments" carries the same tone as "pay attention to those boxes you're standing on".
A reasonable assumption here would be the GP questioning the assumed universal truth of:
the same way [ all people ] are disgusted when eating a "powder shake" kind of meal compared to a whole, earthy, flavorful, home-made mealHowever had, I watched it, and I have some problems here.
First: these robots have barely any head. I think for a humanoid robot, you need to have a proper head with some weight in relation to the rest, otherwise it is just SIMULATING being humanoid. Then the feet; this is also a problem. Specifically the feet are not like human feet at all, they are more like stubs or elongated stubs with push-back being made-super easy. Usain Bolt does not have such short jokes of feet, so this comparison is unfair. Same with the hip area - I am sorry, but the hip structure of these robots is ABSOLUTELY not at all like a human. I am not even calling these humanoids.
Don't get me wrong, it is good that science and technology progresses and such robots may do things that robots solely with a wheel alone, may be unable to do, such as running, climbing, you name it. But I simply can not call them "humanoid". They are, IMO, simply not humanoid and not even humanoid-like.
For "proper" science fiction, have a look at the first Blade Runner movie, but also the first terminator. Initially Cameron thought Arnold was unfit because he was massive and the infiltrating cyborg should melt into the environment easily (which was better in Terminator 2, T1000 was kind of more "logical", but I hated the background storyline in Terminator 2, it made no sense to me, no clue why people preferred that over the original terminator movie). I am aware that these running robots don't aim to be a look-alike replacement, but they simply also should not be called humanoid robots right now. They simply are not humanoid. Thus one should not call them that way.
It's true that the zoomers got the shit end of the stick, economically and culturally. It's also true that they are literally paid to be polite and still refuse to do so.
Just because you're going through the apocalypse doesn't mean you gotta take it out on everybody else. That's like rule #1.
Yep. A lot happened since then!
- 9/11
- 2008 recession
- 20+ years of warfare
- Student Debt Crisis
- Housing Crisis
- COVID
- Now Venezuela, Iran War, Cuba on the on-deck spot
- Tariffs massively increasing costs of goods
You're talking about 25-36 years ago. I remember all of it.
Gen Z has a much higher debt burden per degree, they have a much higher home price-to-income ratio, much higher baseline living costs, and they are socially and culturally vastly different from the people in charge in this country right now. Now they're dealing with AI pushing down both wages and the number of positions, especially entry level ones.
It's no fucking wonder!
The headline by itself doesn't really mean much, anyone can build a purpose-built robot to beat specific human limitations. The question is can it have that kind of performance without completely foregoing most other tasks and abilities.
My bigger problem is that this is labelled as "humanoid robot". IMO that is more like some machine gadget that looks very remotely like some non-descript model of a humanoid, but not anything at all like a humanoid robot. I am a bit spoiled by Blade Runner here - these robots right now come nowhere close to anything like that right now, even if they run faster than Usain Bolt (but they run differently, so they are cheating; look at the angles, the tiny feet, the cheating in the hip area, the barely existing head - so many things are CHEATING here. I can not call this humanoid robot).
You might have different proportions due to different weight distribution or use cases (maybe the legs are made less flexible if the workspace has fewer obstables, for example.
Maybe "bipedal" is a better descriptor, since wheels are a proven way to go faster than Usain Bolt, but walking-style strides are new.
You should ask those questions in the same spirit re a 1 vs a 5 vs a 20 year old. May lead to some insight on development-level-based capability.
Anyone could build a robot that could run on two legs at that speed for 100m? I think someone would have done it before. Even the other robots in the race were slower.
Your bipedal one might be able to walk or run or jump up there though.
> I don't see high jumping
They did high jump in this very same contest.
Killer robots which can't be outrun and which can give chase through urban landscapes
> What a time to be alive
Indeed.
First hiking 10 miles into enemy territory. Then laying down in low-power mode, for 120 days. Finally waiting for "important enemy person" to pass by, and then using a weapon with super-human sensory skills + weaponry. Note - Does not need to be humanoid.
We need to start Waging Peace, as our ability to kill each other is outstripping our willingness to live peacefully together.
Usain Bolt is incredibly fast for a human, but cheetahs are twice as fast still (and, presumably, can climb stairs too).
I do appreciate the engineering achievement on its own, but there seems to be an implied "and will replace humans" vibe here that doesn't make a lot of sense to me.
Totally. In a little while, you might even be able to marry a robot and even raise robot kids.
What a time to be alive indeed!
I highly doubt that that it will happen. But just mocking the celebration of paving a path towards the death of all things that makes being a human worth all the pain..
I'd say that's the problem. There are people who think humans are uniquely special/purposed, and AI is systematically destroying that notion. They need to wake up, or they may find themselves jumping off a bridge or something.
Correlation =/= causation.
I don't know. We know that the physical changes happening in brain cannot be the cause of the sensations that we experience.
Because dreams. If the physical changes in brain is the cause of the sensation of vision, then how can we see things in dream without the exact physical changes happening in the brain that could have caused it when awake?
What this tells me is that there is something (consciousness?) that just "measures" the brain activity, and constructs a sensation from the measurements, and thus sense it. This thing is also capable of constructing the sensation without the measurements, which makes dreams possible..
How does it work? Not a clue. But the point is that the physical changes in the brain cannot be solely responsible for our sensations.
Running fast or jumping high are what makes being a human worth all the pain?
People can like GPS while disliking nuclear weapons without any contradiction.
I dislike the vast waste of resources happening during the current AI bubble, and don’t really care about generative AI.
The ones I see deployed in construction do stuff like drilling holes in hard to reach places, or laying bricks. None of them look like people, though. Much like the typical household dishwasher hasn't been named Alice or Doris for the past 50+ years, but rather Bosch or Samsung and doesn't use actual hands.
But yeah, humanoid robots do look cool and many movies and books told us to be fearful of them.
Such robot would easily smash the current marathon record as well and the only thing I'd be impressed by, would be the battery capacity tbh.
If you were ever trapped on the top floor of a burning building, should we just send a car up there?
Nobody is surprised by advancement like this, but there is still a lot of engineering that went into it, not talking about how it’s now almost consumer grade tech that you can buy.
Riddle me this then: what role does a robot that has been built to run real fast play in that scenario?
Sure, a good deal of engineering went into this, but the hard part - solving bipedal locomotion - had been done long before that. Apart from showing off it's not really advancing the field and so there's not too much left to be impressed by (honestly, the battery and actuator efficiency that allowed the thing to do a half marathon was MUCH more impressive, at least to me). It seems to me to be more analogous to car tuning.
It will run real fast up the stairs, I guess? Is it a good enough riddle?
Running up the stairs is the easy part after all. The difficult part in such scenario is lack of orientation due to smoke, obstacles like locked doors or fallen debris, unstable walls and ceilings, locating people and getting them out safely. All of those things require a good amount of autonomy, on-the-spot physical problem solving and judgement.
To this effect, the recent DARPA Triage Challenge, which has its final taking place in November this year, seems like a much more thorough and practical demonstration of current capabilities. Something to look forward to if you're interested in the current state-of-the-art w.r.t. real-world applications of robotics.
Everyone has the same basic objective and measurement norm, so we can see how performance improves year over year, and how different manufacturers handle a specific known goal, even if it's not super-representative of real world use cases.
Billion-dollar supercomputers don't spend most of their time running LINPACK, and some probably perform better on specific tasks, but it doesn't make the Top 500 list any less of a thing.
> Lightning the humanoid robot clocks 9.32sec in Beijing
Bipedal, hips, knees, feet. Looks humanoid to me.
If cloudflare puts this in a grid of images and asks me to pick the humanoids, I’m clicking it
> A machine is stronger and faster than a human? Shocking. Next we'll learn that cranes can lift heavy loads faster than a human with ropes and pullys.
It’s funny how absurdly dismissive and condescending tech people can be about difficult problems in other people’s fields that they know absolutely nothing about. Numerous developers, informed by maybe a YouTube video, Reddit thread, or equally misguided explainer post on another developer’s blog, have tried to ‘correct’ me about topics in a field in which I hold a degree and have plenty of professional experience. Other tech people might be impressed because they have matching blind spots in their mental models of the world, but people who actually understand, rather than baselessly assuming they understand, see someone being smug about their own ignorance.
"Machines can do stuff humans cannot do" isn't exactly groundbreaking in 2026.
The only interesting part of this story is that this particular machine form factor looks pretty much like a human
It's still not exactly shocking news that it can outrun a human. Of course it can. We're well past John Henry vs the Machine at this point in history
That's the whole goddamn point of this. Fast stable autonomous bipedal locomotion was out of reach of human technology for decades. Since this year it isn't. This was impossible year ago, despite all of the achievements of humanity, cumulative, till the year 2025, included. Now it isn't.
"That's the whole point" that we don't need humans anymore?
Great. I'm a human. You want me to celebrate? No thanks
As for the fear of human labour not being needed at scale due to me as most others needing an income, I'm optimistic there for two reasons. First, large unfed masses tend to start revolutions, something both governments and the not-working are interested in avoiding. Second, large amounts of young idle people has tended to result in stupid destructive behaviour at scale. So I'm confident there will be created opportunities for paid work for mostly everyone, and the conditions won't be too poor as there would be no profit motive to push people too far. Rather, the people having both the resources, time, and energy to consume, would still be what props up the system (until we can agree on trying something better now that we have excellent tools for guaranteeing transparency and optimising allocation at scale in real time).
This has nothing to do with existential happiness or contentment
It is about existential fear that if I am no longer needed then there's nothing useful I can provide to earn an income. At best I wind up living a destitute existence or homeless, at worst I wind up being executed by murderbots owned by rich people to reduce the risk that I might take action against them
> As for the fear of human labour not being needed at scale due to me as most others needing an income, I'm optimistic there for two reasons. First, large unfed masses tend to start revolutions, something both governments and the not-working are interested in avoiding.
In the past governments and rich people still needed working people's labor to keep their lifestyles. In a fully automated world that's no longer the case. I am very confident that if the ruling class of society could slaughter all of us and still maintain their lifestyles forever they would in a heartbeat. Foolishly, we are delivering that possibility to them in the form of automation.
What the rich want is not to be content. They want to get richer.
And it's terribly hard to extract money from other rich people, because they want to get richer, not poorer too.
The easiest way to get richer (and pretty much only feasible way) is to take away the money from the poor. And the best way to do that is to take those suckers' money by giving them products and services that are worth a little bit less than the money you take from them.
But if the poor have no money the whole system that served the rich so well breaks. Some of them already see that. There are millionaires advocating higher taxation of millionaires. Because they see what's up. Spontaneously the money flows only one way, from the poor to the rich. And there is less and less money to extract, that could make you wealthier than your peers. But if you tax all rich equally, which is a fair game as it doesn't disadvantage you. And then transfer that money back to the poor, it's a new pool you can compete for with other rich and get wealthier than them if you succeed.
Have you seen in the game of the monopoly when the one person gets winning hard, it's natural instinct to borrow money to other players, without collateral, against the rules of the game or even gifting it? Because for the rich the pleasure is in taking not in having. You are sure it all ends up in your pocket anyways, because you see yourself as the best at the game.
This flow of money is in play for decades and you can plainly see its effects in economic charts. It can't stay like that regardless of how much more automation we'll have. The strong will need to create a mechanism for funneling resources back to consumers, before they run dry. Employment barely works for that purpose. Without customers there's no business, and without business, getting richer is very, very hard.
You might think of yourself as a worker and believe that's your utility. But in reality your utility lies in the fact that you are first and foremost a consumer. And your ability to consume is a vital interest of the rich, even if not all of them realize that yet.
Machines using fossil fuels have been able to do more than men or beast for centuries now.
So a purpose-built (look at the thing - running is all it can actually do) two legged fast running robot doesn't seem all that impressive today. Cool? Sure! Useful? Not so much. Impressive? Whatever... it's just showing off at this point, much like Boston Dynamic's Atlas series that went nowhere (in contrast to this machine: this one did half marathon distance:)
The US is behaving like an industry incumbent, like Blackberry when the iPhone came out. That the US could fall behind on technology seems to have not crossed many minds. It will be very hard to overcome that first mover advantage.
More like "If this forum created by a guy from Silicon Valley..."
Or, to be more accurate, the distance at that speed.
Because, of course the machine which does not tire and is deliberately designed to be durable and exceed natural human tolerances are better than humans. Cranes can lift far more than the strongest man.
But I understand that historically, bipedal machines have had balancing issues. The number of slight adjustments needed are significant. And doing it while at that speed and also not tumbling over is an achievement.
Humans on enhancements vs robots?
I'm so tired of my messages taking 2 weeks to deliver! 3000km shouldn't take that long! Personally I'm really looking forward to a world where my message can take maybe even a week to deliver to someone 2000km - 3000km away. What a time to be alive.
Yes it is true that we have had machines on rubber wheels that can move very fast on open roadways. But these new machines are also capable of walking up and down stairs, navigating obstacles, using doorways, and generally operating in the same complex spaces that humans do.
Now apparently they can run really fast too.