Hacker Newsnew | past | comments | ask | show | jobs | submit | HAL3000's commentslogin

> Neocortical networks, with thalamic and hippocampal system integrations, are sufficient to explain the entirety of human experience, in principle.

Where did you get that? That's not an established scientific theorem, it's a philosophical stance (strong physicalist functionalism) expressed as if it were empirical fact. We cannot simulate a full human brain at the correct level of detail, record every spike and synaptic change in a living human brain and we do not have a theory that predicts which neural organizations are conscious just from first principles of physics and network topology.

> We can induce emotions, sights, sounds, smells, memories, moods, pleasure, pain, and anything you can experience through targeted stimulation of neurons in the brain

That shows dependence of experience on brain activity but dependence is not the same thing as reduction or explanation. We know certain neural patterns correlate with pain, color vision, memories, etc. we can causally influence experience by interacting with the brain.

But why any of this electrical/chemical stuff is accompanied by subjective experience instead of just being a complex zombie machine? The ability to toggle experiences by toggling neurons shows connection and that's it, it doesn't explain anything.

> We've got a good enough handle on physics to know that it's not some weird quantum thing, it's not picking up radio signals from some other dimension, and it's not some sort of spirit or mystical phlogiston.

We do have a good handle on how non conscious physical systems behave (engines, circuits, planets, whatever) But we don't have any widely accepted physical theory that derives subjective experience from physical laws. We don't know which physical/computational structures (if any) are sufficient and necessary for consciousness.

You are assuming without any evidence that current physics + it's "all computation" already gives a complete ontology of mind. So what is the consciousness? define it with physics, show me equations, you can't.

> It's a computer, in the sense that anything that processes information is a computer. It's not much like silicon chips or the synthetic computers we build, as far as specific implementation details go.

We design transformer architectures, we set the training objectives, we can inspect every weight and activation of a LLM. Yet even with all that access, tens of thousands of ML PhDs,years of work and we still don't fully understand why these models generalize the way they do, why they develop certain internal representations and how exactly particular concepts are encoded and combined.

If we struggle to interpret a ~10^11 parameter transformer whose every bit we can log and replay, it's a REAL hubris to act like we've basically got a 10^14-10^15 synapse constantly rewiring, developmentally shaped biological network to the point of confidently saying "we know there's nothing more to mind than this, case closed lol".

Our ability to observe and manipulate the brain is currently far weaker than our ability to inspect artificial nets and even those are not truly understood at a deep mechanistic concept level explanatory sense.

> Your mind is the state of your brain as it processes information.

Ok but then you have a problem, if anything that processes information is a computer, and mind is "just computation" then which computations are conscious?

Is my laptop conscious when it runs a big simulation? Is a weather model conscious? Are all supercomputers conscious by default just because they flip bits at scale?

If you say yes, you've gone to an extreme pancomputationalism that most people (including most physicalists) find extremely implausible.

If you say no, then you owe a non hand wavy criterion, what's the principled difference, in purely physical/computational terms between a conscious system (human brain) and a non conscious but still massively computational system (weather simulation, supercomputer cluster)? That criterion is exactly the kind of thing we don't have yet.

So saying "it’s just computation" without specifying which computations and why they give rise to a first person point of view leaves the fundamental question unanswered.

And one more thing your gasoline analogy is misleading, combustion never presented a "hard problem of combustion" in the sense of a first person, irreducible qualitative aspect. People had wrong physical theories, but once chemistry was in place, everything was observable from the outside.

Consciousness is different, you can know all the physical facts about a brain state and still not obviously see why it should feel like anything at all from the inside.

That's why even hardcore physicalist philosophers talk about the "explanatory gap". Whether or not you think it's ultimately bridgeable, it's not honest to say the gap is already closed and the scientific explanation is "sufficient".


Well said.

We can stimulate a nerve and create the experience of pain, but stimulating the nerve does not create the memory of pain.

Nerves triggering sensations I can understand. But stimulating the same nerves, or creating the same electrical activations does not create the memory of the pain.


I don't know why you single out memories.

Certainly, if some mad scientist were to stimulate via an electrode some parts of your brain to make you experience pain, you will remember it. Also, it's not unreasonable to assume that it would be equally feasible to create fake memories by stimulating other parts.

If there is a hard problem, memory is not it.


It's some of the lower hanging fruit that is complex enough that we don't understand, but we don't have to question thoughts, or consciousness.

What I'm saying isn't that if you stimulate pain, you won't remember it.

False memories would be a challenge. How would you input a memory of going up a tree to rescue a cat. Where would you even begin?


Granted, memory is certainly more complex than basic feelings and can probably not be generated on demand by stimulating a few neurons in a single place. We are certainly far from being able to create memories by electric stimulations, but I see no reason to believe it's impossible. Therefore invoking "memory" (or any other complex though, really) does not refute "[the mind] is the sum of electrical and chemical network activity in the brain".

Anyway, I suspect in those discussions more time is spent disagreeing on the meaning of words than on the core concepts.


> It takes extraordinary skill to successfully juggle multiple ventures.

That's a myth. I've done that, and I know a lot of people who do that. Do you think Musk is writing sparse attention code for Grok? Does he even know how Grok's architecture works under the hood? Or that he designed the data centers? I mean, you delegate stuff. The only hard thing is getting the right people, but if you're a hyped up billionaire, it's easy mode because you can pay a lot, and people want to work for you. You just create an environment where they can achieve things.

There are times when the majority of your work is simply attending public meetings, podcasts, and doing interviews. People really overestimate what's involved in the work of a billionaire CEO. The people actually making things happen in space industry or AI work harder, longer, and solve more complex problems than any CEO and in some cases they need to work hard against the CEOs to actually make things happen.


> if you're a hyped up billionaire, it's easy mode because you can pay a lot, and people want to work for you

blue origin vs spacex says otherwise..


Blue origin just landed their new glen rocket. Not an easy feat.

Although I wish billionaires would fix homelessness, I think its good there’s more competition in the space launch industry.


Homeless is mostly a self inflicted issue that most don’t want to solve. So why should others try solve it?


All of the examples in videos are cherry picked. Go ask anyone working on humanoid robots today, almost everything you see here, if repeated 10 times, will enter failure mode because the happy path is so narrow. There should really be benchmarks where you invite robots from different companies, ask them beforehand about their capabilities, and then create an environment that is within those capabilities but was not used in the training data, and you will see the real failure rate. These things are not ready for anything besides tech demos currently. Most of the training is done in simulations that approximate physics, and the rest is done manually by humans using joysticks (almost everything they do with hands). Failure rates are staggering.


The last example they show (pick up package from pile, put it label-down on conveyor, repeat) seems to be the most realistic. They even have an uncut video of their previous model doing that for an hour on twitter [1].

I'm not sure that task needs a humanoid robot, but the ability to grab and manipulate all those packages and recover from failures is pretty good

1: https://x.com/adcock_brett/status/1931391783306678515


Economy of scale. This guy wanted a factory in China to do a custom run of a product where they didn't include the shell. It was cheaper to just buy 10,000 units and have a line to de-shell the items than it was to change the original line. That special purpose robot for the one task is going to get beat out by the general purpose robot thats being produced at 100x the volume.


> I'm not sure that task needs a humanoid robot

An industrial robot arm with air powered suction cups would do the trick... https://bostondynamics.com/products/stretch/ ...

... So the task they work best at is the task there is already cheaper better robots specialized for.


I feel like we're entering the era of general and inefficient solutions to problems.

Like LLMs being used to pick values out of JSON objects when jq would do the job 1000x more efficiently.

This is what this whole field feels like right now. Let's spend lots of time and energy to create a humanoid robot to do the things humans already decided humans were inefficient at and solved with specialised tools.

Like people saying "oh it can wash my dishes for me". Well, I haven't washed dishes in years, there's a thing called a dishwasher which does one thing and does it well.

"Oh it can do the vacuuming". We have robot vacuums which already do that.


There were plenty of digital circuit engineers back in the 90s that said microprocessors were general and inefficient solutions to problems.

And if you needed it programmable, well an FPGA was still almost as general and far more efficient than a microprocessor.

Guess what won.


As a hardware engineer I hear this a lot from software/electrical folks.

It's Moore's law that largely drove what you describe.

Moore's law only applies to semiconductors.

Gears, motors and copper wire are not going to get 10x faster/cheaper every 18 months or whatever.

10 years from now gears will cost more, they will cost what they cost now plus inflation.

I've literally heard super smart YC founders say they just assume some sort of "Moore's law for hardware" will magicallyake their idea workable next year.

Computing power gets, and will continue to get, cheaper every day. Hardware, gears, nuts, bolts, doesnt.


It is not the gears, motors and copper wire that are bottlenecking robots. It is the software and computing. We can already build a robot hand that is faster, stronger, more dexterous, etc. than a human hand. What we can't do right now is make the software to perceive the world around it and utilize the hand to interact with it at human levels. That is something that needs computing power and effective software. Those are things that get, and will continue to get, cheaper every day.


> It is not the gears, motors and copper wire that are bottlenecking robots.

It is those things that are bottlenecking the price of robots.

The price of something tends towards the marginal cost, and the marginal cost of software is close to $0. Robots cost a lot more than that (what's the price of this robot?).

Edit: In fact Figure 03 imply marginal costs matter:

  Mass manufacturing: Figure 03 was engineered from the ground-up for high-volume manufacturing


>We can already build a robot hand that is faster, stronger, more dexterous, etc. than a human hand

Can you attach it to a humanoid body that isn't wired?


Yes, but the two (software and hardware) scale very differently.

Once software is "done" (we all know software is never done) you can just copy it and distribute it. It is negligiblehow much it costs to do so.

Once hardware is done you have to manufacture each and every piece of hardware with the same care, detail and reliability as the first one. You can't just click copy.

Often times you have to completely redesign the product to go from low volume high cost manufacturing to high volume low cost. A hand made McLaren is very different than an F-150.

The two simply scale differently, by nature of their beasts.


China has shown that they don't scale all that differently. Yes the tooling is hard to build but after that you hit go and the factory makes the copies for you.

It's not quite startrek replicator but much closer to that than the US view of manufacturing where you have your union guy sitting in front of the machine to pull the lever.


This isn't true. They've showed that slave (or nearly) labor is cheap or even free.


This was somewhat true at once point but is a highly outdated view. Labor is no longer cheap in China relative to other nearby countries and there's a huge amount of automation with some factories that don't even turn on lights because they are effectively 100% automated.


Think about cars. Their manufacturers work really hard on efficient (cost and performance). And what people do with them is a very different story. It could see the same happening with robots.


I dont follow


The nuts and bolts won't change much but the software/compute controlling the thing likely will.


But the nuts and Bolts need to be paid for and manufactured for each robot. The software needs to be done once and then they can just click copy.


Human form robots are a case of Jake of all trades and master of none. Sure I have a dishwasher thats more efficient at doing the job than me but I still end up doing dishes because the cast iron frying pan can't going in there without ruining the polymerised layer of oils that have been baked into it and i would have to repeatedly oil and reheat it and stink up the house with smoke reseasoning it afterwards, and I have hand wash the thermos and travel mugs, and dishwasher arent good for the sharp knives and etcetera etc etc... sure the rumba can vacume very efficiently but it suck at gating around furniture leg or gaps to small for a 14'' diameter circle to fit through so I have to vacume all of the bits it can't get to. Sure the a robot lawn mower can do my yard very efficiently but it cant move the childrens toys out of the grass or open the gate to the front yard or close the gate to keep the dogs from running out the gate once its open. Specialized tools suck at edge cases. Human form robots if they ever works (big if) can do all of the edge cases and take advantage of all the tools made for humans I already have to do all of the those other jobs.


The real question, is whether the humanoid robot will ever be cheaper than hiring immigrant labour. Because that's a pretty damn low threshold


There isn't enough migrant to do all labor east asia will need as its population gets quickly older. Plus the societal aspiration of culture dissemination isn't there.


The problem is that human immigrant labor is, well, human. Cost isn't the only metric being considered.


have you ever googled a simple maths question? I often come back to that and realise we've been in this era for quite a while. Calculator would probably be 1000x more efficient!


Sure, but I have to launch the calculator, instead of just typing it into the ever present search bar of my persistent open browser.

If I could just type it into my shell, that would be nice. I’m sure there’s some command (or one could be trivially made) to evaluate an equation, but then you get to play game with shell expansions and quotes.

In emacs I have to convolute the equation into prefix.

All minor stuff but it adds up.


> I’m sure there’s some command (or one could be trivially made)

bc

https://www.gnu.org/software/bc/manual/html_mono/bc.html


From the demo videos on the site... it seems now that people have dishwashers they want a robot to load the dishwasher for them


That's not that weird, even if it is pretty pathetic. I don't understand it now but I used to dread "doing the dishes" when I was younger even though it was 95% just filling a dishwasher. Laziness drives technology an awful lot, at least from a product POV.


This unoptimizing has been going on since the start of time. Why are the values in json to start with? At somepoint the bad slow generalized version will overtake the specialized ways of doing things.


> LLMs being used to pick values out of JSON objects

Or the often-trumpeted "I had Claude do a killer refactor" that an IDE can do locally for zero power faster and 100% reliably


Isn't the entire point that LLMs are unreasonably better at even specific things that you would naively think a purpose build system would be good at.


Could you name an example aside rephrasing text?


Considering the entire thing is a text generator it's quite a statement to exclude that. Text is a huge world. Math, programming, stories, chatting, mail, search, website navigation, most things you want to do in the digital space involves text.


I think the use case here is smaller to medium size businesses that don't need a $150k suction robot arm 24/7, but do need 24/7 help with warehousing, packaging, restocking, taking inventory, sorting mailing, applying shipping labes, etc. With a single humanoid robot you can do all that for, at some point, possibly as low as $20k for a one-time robot purchase.


We're so far from that though. Even if we magically jump over the failure rates we're discussing here, the safety considerations seem to be far worse. These things are heavy and dangerous, c.f. Rodney Brooks' "never follow a robot up stairs" https://arstechnica.com/ai/2025/10/why-irobots-founder-wont-...


Why do you think a humanoid robot will be cheaper than a robot arm?


If humanoid robots can perform ok on a broader set of tasks then they could reach economies of scale that a robot arm might not.


To add to that, a good friend of mine is a welder and machinist (and still using Linux on the desktop years after I set him up). A robot 'helper' that just moves things around and maybe does basic machine work (cutting pipe and threading the ends, for example) would put his productivity through the roof. Same story with a guy who specializes in kitchen remodeling.

It's hard to find decent general purpose help these days and they would pay good money for a halfway useful helper.

Once it's able to weld... That's going to be a massive game changer, and I can see that coming 'round the corner right quickly.


Once it’s able to weld and climb then building skyscrapers will become a lot easier and cheaper as you don’t need safety equipment for them.


Yeah, just let them fall on top on someone/something below!


There are couple UR5 single arm cobots on eBay at $5.5k each right at this moment. The truth is that the value of humanoid is in it form, the novelty, the sense of accomplishment, not features.


If you found one for that price with the controller and pendent, please send me a link. I’ve looked a lot and have not seen any UR for remotely that cheap.


because they already are. an industrial arm from ABB is frequently over $100k. add in the cost to fit it with specialty equipment like vacuum suction for handling boxes, made by a small to medium size business, they'd probably charge another $50k. and if it breaks you need specialty mechanics and parts.

in a world with 500 million humanoid robots, parts are plentiful, theyre easier to work on due to not weighing 5000 pounds, and like the other person said, economies of scale


> one-time robot purchase.

With a hefty subscription to make it do anything useful.


I can already run the Qwen3 VL multimodal model for text, image processing, and speech recognition and generation on a well spec'd home workstation.

And the Unitree R1 already only costs $6k.

All the necessary pieces are aligning, very rapidly, and as James Burke has pointed out, that's when Connections happen.


the unitree r1 is effectively a useless toy. it's like positing about the future of robotics by looking at a sumo bot.

what it IS , however, is a remarkable achievement of commoditization; getting a toy like that with those kind of motors would have been prohibitively expensive anywhere else in the world; but much like the Chinese 20k EV, it's not really a reliable marker for the actual future; in fact bottomed out pricing is more-so an indicator of the phase of industrialization that country is in.


> the unitree r1 is effectively a useless toy

Only because it's not yet attached to a reasonable AI, which is my point. It's not going to do any heavy lifting, but it could easily do basic house chores like cleaning up, folding laundry, etc if it were. The actuators and body platform are there, and economies of scale already at work.

I guess some folks just can't or won't put 2 and 2 together to predict the near future.


Your reasonable AI cannot resolve the fact that its arm can only lift 2KG.

I am impressed by Unitree, but the problem that needs to be solved here is not just better software. Better hardware needs to come down in cost and weight to make the generalized robot argument more convincing.

There is still a long way to go for a humanoid to be a reasonable product, and that's not just a software issue.


> its arm can only lift 2KG

That covers more than 90% of the objects in my home, and most people's.

> the problem that needs to be solved here is not just better software. Better hardware needs to come down in cost and weight

I disagree. Software seems to be the main limitation to me at the moment. Bigger motors and batteries are readily available on the market already. Software is advancing rapidly, and seems to me will quickly be up to the task (i.e. within a few years), but at the moment is still the domain of research projects.

> There is still a long way to go for a humanoid to be a reasonable product

Whether or not you think it's a reasonable product, it's clearly already an available one which is already selling in volume. As with all things, future versions will be more capable.


Why are you comparing LLMs with robotics? What makes you think they are even remotely related problem sets?


With humans, we call that subscription a salary.


Of course it does not. We just add scanners in up down left and right of the conveyor. Never touching the package.


This was my solution as well. Why even have a robot? Give me a conveyor belt, some cameras and a moderately powerful SBC and I think I could probably manage a system that does one package a second with a fallback for humans to process what can't be processed by the machine.


The Boston robot is for packing boxes on and off the conveyor; like a transport car. And it helps having a few degrees of freedom to do that effectively.

If it's just checking or adding labels, it's silly to even use that.


That's the problem.

An obvious application, if this robot could do it, is retail store shelf restocking. That's a reasonably constrained pick and place task, some mobility is necessary, and the humanoid form is appropriate working in aisles and shelves spaced for humans. How close is that?

It's been tried before. In 2020.[1] And again in 2022.[2] That one runs on a track, is closer to an traditional industrial robot, and is used by 7-11 Japan.

Robots that just cruise around stores and inspect the shelves visually are in moderately wide use. They just compare the shelf images with the planogram; they don't handle the merchandise. So there are already systems to help plan the restocking task.

Technical University Delft says their group should be able to do this in five years.[3] (From when? No date on press release.)

[1] https://www.youtube.com/watch?v=cHgdW1HYLbM

[2] https://blogs.nvidia.com/blog/telexistence-convenience-store...

[3] https://www.tudelft.nl/en/stories/articles/shelf-stocking-ro...


The Telexistence demo isn't so bad, but I have no idea why we're trying to make human robots generally. The human shape sucks at a most things, and we already have people treating roombas and GPT like their boyfriends or pets...


What form factor would be better at going up and down stairs? Reaching to a high shelf? Getting between the refrigerator and counter to grab a key?


Because human work is designed for humans. If you want a drop-in replacement for human workers, humanoid robots are your best bet.


That doesn’t even remotely follow. Human work is designed for humans so if you want human work done you need a human to do it.

If you want to replace the human the best bet is to redesign the work so that it can be done with machine assistance, which is what we’ve been doing since the industrial revolution.

There’s a reason the motor car (which is the successful mass market personal transportation machine) doesn’t look anything like the horse that it replaced.


<HN voice>Technically, the motor car replaced the coach. A more accurate and enjoyable analogy would that car engines don’t look like horses.


hence horseless CARriage


I don't think that it's human work this would target, but instead work in shared human spaces.


We already have robots that work in shared human spaces, and our experience in that domain has shown that you need to put a lot of thought into how to do this safely and specifically how to prevent the robot from accidentally harming the humans. Ask anyone with a robotic cnc machine how they would feel about running the machine without its protective housing for example. I expect they will start to throw up just a little bit. Flexibility is exactly the opposite of what you need until we have a CV and controller combination that can really master its environment. I could forsee a lot of terrible accidents if you brought a humanoid robot into a domestic environment without a lot of care and preparation for example.


Sure but robots don’t join unions or ask for a pay raise or benefits.


Rodney Brooks (of iRobot fame) wrote an essay recently about why humanoids are likely decades and not years away from fulfilling their promise. It is quite long, but even a gpt summary will be quite valuable.

https://rodneybrooks.com/why-todays-humanoids-wont-learn-dex...

In short, he makes the case that unlike text and images, human dexterity is based on sensory inputs that we barely understand, that these robots don't have, and it will take a long time to get the right sensors in, get the right data recorded, and only then train them to the level of a human. He is very skeptical that they can learn from video-only data, which is what the companies are doing.


Most companies are using world model simulations for training nowadays, like Issac Gym or Mujoco, not just video data.


came here to see if anyone had read Rodney's recent essay - and to ask how does this announcement by Figure square with Rodney's essay.

The essay was long so I cant claim I read it in detail - one q in my mind is whether humanoids need to do dexterity the same way that humans do. yes they dont have skin and tiny receptors but maybe there is another way to develop dexterity?


You wrote what I wanted to write but I couldn't find such words.

Indeed, all the videos/examples are marketing pieces.

I would love to see a video like this "Logistics"[0] one, that shows this new iteration doing some household tasks. There is no way that it's not clunky and prone to all kinds of accidents and failures. Not that it's a bad thing - it would simply be nice to see.

Maybe they will do another video? Would love that.

[0]: https://www.youtube.com/watch?v=lkc2y0yb89U


This is always the case, though. The company is few years old. I'm no disciple of humanoids, but stuff has to start somewhere. Unfortunately hype > truth in order to get funding, so it produces incentives to cherry-pick like this.


Now the question is if this is GPT-2 and we’re a decade away from autonomous androids given some scaling and tweaks, or if autonomous androids is just an extremely hard problem.


https://www.figure.ai/company

"Building Figure won’t be an easy win; it will require decades of commitment and ingenuity."

"Our focus is on what we can achieve 5, 10, 20+ years from now, not the near-term wins."

At least it's not Musk's forever "next year".


> At least it's not Musk's forever "next year".

The problem with the principled approach to high-uncertainty projects is that if you slowly execute on a sequential multi-year plan, you will almost certainly find out in year 9 that multiple of the late-stage tasks are much harder than you thought.

You just don't know ahead of the time. Just look at how many corporations and research labs had decades-long strategies to build human-like AI that went nowhere. And then some guys came up with a novel architecture and all of sudden, you can ask your computer to write an essay about penguins.

Musk's approach is that if you have an infinite supply of fresh grads who really believe in you and are willing to work crazy hours, giving them a "next year" deadline is more likely to give you what you want than telling them "here's your slow-paced project you're gonna be working on for the next decade". And I guess he thinks to himself that some of them are going to burn out, but it's a sacrifice he's willing to make.


>Just look at how many corporations and research labs had decades-long strategies to build human-like AI that went nowhere.

They didn't go nowhere; they just didn't result in human-like AI. They gave us lots of breakthroughs, useful basic knowledge, and knowledge infrastructure that could be built off for related and unrelated projects. Plenty of shoot for the moon corporations didn't result in human-like AI either, but also probably did go nowhere, since they were focused on an all or nothing strategy. The ones that do succeed in a moonshot relied on those breakthroughs from decades-long research.

I'm not going to get into what Musk has been doing because I'm just not,


> Musk's approach is that if you have an infinite supply of fresh grads who really believe in you and are willing to work crazy hours, giving them a "next year" deadline is more likely to give you what you want than telling them "here's your slow-paced project you're gonna be working on for the next decade". And I guess he thinks to himself that some of them are going to burn out, but it's a sacrifice he's willing to make.

This feels incredibly generous. I'm pretty sure his approach is that he needs to keep the hype cycle going for as long as possible. I also believe it's partially his willingness to believe his own bullshit.


Most likely you are right. Best way to peddle a lie is to believe it.


I’d rather put my faith in his grand illusions than in your sanctimonious high priest pose.


who do you think spins "grand illusions" if not your "high priest" who dosn't even know you exist.


Musk really missed an opportunity to promise wrecking the govt “next year” - we all would’ve rolled our eyes a la “fully autonomous driving next year” and been eating our hats by now


For LLMs, the input is text, and the output is text. By the time of GPT-2, the internet contained enough training data to make training an interesting LLM feasible (as judged by its ability to output convincing text).

We are nowhere near the same for autonomous robots, and it's not even funny. To continue to use the internet as an analogy for LLMs, we are pre-DARPANET, pre-ASCII, pre-transistor. We don't even have the sensors that would make safe household humanoid robots possible. Any theater from robot companies about trying to train a neural net based on motion capture is laughably foolish. At the current rate of progress, we are more than decades away.


I would guess Amazon has a ridiculous amount of access to training data in its warehouses. Video, package sizes, weights, sorting.

I’m sure they could pretty easily spin up a site with 200 of these processing packages of most sizes (they have a limited number of standardized package sizes) nonstop. Remove ones that it gets right 99.99% of the time and keep training on the more difficult ones, the move to individual items.

Caveat: I have no idea what I’m talking about.


A more efficient way might be to train them in simulation. If you simulate a warehouse environment and use that to pre-train a million robots in parallel at 100x real time learning would go much faster. Then you can fine tune on reality for details missed by the simulation environment.


Does your estimate account for advancements in virtual simulation models that has simultaneously been happening? From people I speak to in the space (which I am very much not in) - they had mentioned these advancements have dramatically improved the rate of training and learning - though they also advised we're some ways off from showtime.


As Tesla could tell you with their failure to deliver self-driving cars, it doesn't matter if you have exabytes of training data if it's all the wrong kind of data and if your hardware platform is insufficiently capable.


Time will tell if that's true. We don't have the same corpus of data, that's true, but what we do have is the ability to make a digital twin, where the robot practices in a virtual world, what would happen. It can do 10,000 jumping jacks every hour, parallelized across a whole GPU supercomputer, and that data can be fed in as training data.


McD must be selling millions of burgers every day and cameras are cheap and omnipresent, so should not be difficult to get videos for single type of tasks.


There is no reason to employ humanoid robots in industrial environments when it will always be easier and cheaper to adapt the environment to a specialized non-humanoid robot than to adapt robots into humanoid shape. This is true for the same reason that no LLM is ever going to beat Stockfish at chess.


Robotics has a big training data problem. But your "we don't have the sensors" claim is absolutely laughable.

It was never about the sensors. It was always about AI.


No, it doesn't matter if you have a hypergenius superintelligence if it's locked in a body with no hardware support for useful proprioception. You will not go to space today.


A 'hypergenius superintelligence' could achieve most, if not all useful proprioception simply by looking at motor amperage draw, or if that's unavailable then total system amperage draw.

An arm moving against gravity has a higher draw, the arc itself creates characteristics, a motion or force against the arm or fingers generates a change in draw -- a superintellligence would need only an ammeter to master proprioception, because human researchers can do this in a lab and they're nowhere near the bar of 'hypergenius superintelligence'.


Lmao no. Every motor is a sensor. And the better my world model is, the less sensors I need to keep it up.


This is where I'm at. If you look at Boston Dynamics' first videos, they're 45 second clips of 4 legged robots walking in not even a straight line, just proving they could walk 5 feet over level ground without falling over. The top comment, from 4 years ago is "This was 11 years ago. Now these things are dancing." https://www.youtube.com/watch?v=3gi6Ohnp9x8

If you can make it look believable on camera for 15 seconds under controlled studio conditions... it's probable you can do it autonomously in 10-15 years. I don't think anyone is going to be casually buying these for their house by this time next year, but it certainly demonstrates what is realistically possible.

If they can provably make these things safe, it will have huge implications for in home care in advanced age, where instead of living in an assisted living home at $huge expense for 20+ years, you might be able to live on your own for most of that time.

I am cautiously optimistic.


The robot (BigDog) in that video shows numerous capabilities that Spot still can't do (climbing over terrain like that, being able to respond to a kick like that, the part on the ice, etc.). Even 16 years later.

This only highlights the fact that making a cool prototype do a few cool things on video is far, far easier than making a commercial product that can consistently do these things reliably. It often takes decades to move from the former to the latter. And Figure hasn't even shown us particularly impressive things from its prototypes yet.


It's an unfair comparison. Yes, they're both 4 legged 'dogs', but they use radically different design criteria -- design criteria that the BigDog was used to refine.

I'm not surprised that a Honda Civic can't navigate the Dakar Rally route..


I don't know if I caught your comment in my peripheral vision or what but GPT-2 is exactly where I conceptually placed this.

Neural networks for motion control is very clearly resulting in some incredible capability in a relatively short amount of time vs. the more traditional control hierarchies used in something like Boston Dynamics. Look at Unitree's G1

https://www.youtube.com/shorts/mP3Exb1YC8o

https://www.youtube.com/watch?v=bPSLMX_V38E

It's like an agile idiot, very physically capable but no purpose.

The next domain is going to be incorporating goals and intent and short/long term chains of causality into the model, and for that it seems we're presently missing quite a bit usable training data. That will clearly evolve over time, as will the fidelity of simulations that can be used to train the model and the learned experience of deployed robots.


Locomotion and manipulation are pretty different. The former we know how to do well -- this is what you see in unitree videos. Manipulation still not so much. This is not at all like GPT-2 because we still don't know what to scale (and even the data to scale is not there).


How does this square with the video where they showed it running continuously for an hour doing an actual Amazon package sorting job? https://www.youtube.com/watch?v=lkc2y0yb89U


> How does this square with the video where they showed it running continuously for an hour doing an actual Amazon package sorting job? https://www.youtube.com/watch?v=lkc2y0yb89U

The video shows several of glitches. From the comments:

  14:18 the Fall
  28:40 the Fall 2
  41:23 the Fall 3 
Also many of the packages on the left are there throughout the video.

But then I think lots of this can be solved in software and having seen how LLMs have advanced in the last few years, I'd not be surprised to see these robots useful in 5 years.


Three mistakes in an hour isn’t terrible, especially if that’s the last generation. As another commenter put it, this is the worst it’s ever going to be.


I only watched for 5 minutes, at double speed, but I saw a lot more than 3 mistakes fwiw


> this is the worst it’s ever going to be.

People keep parroting this line, but it's not a given, especially for such an ill-defined metric as "better". If I ask an LLM how its day was, there's no one right answer. (Users anthropomorphizing the LLM is a given these days, no matter how you may feel about that.)


Yeah but for things like packets dropped per hour there will probably be improvement.


> But then I think lots of this can be solved in software and having seen how LLMs have advanced in the last few years, I'd not be surprised to see these robots useful in 5 years.

Would asking the robot for a seahorse emoji leave you in a puddle of blood?


Thinking about it, I am sure it is only a matter of time until a self driving car or a robot will be used to kill a human. Or on a lower level for a DDoS attack - all cars/robots going to the white house.


> self driving car or a robot will be used to kill a human

They already have. We just don't hold the perpetrators accountable.


I think "has been used to kill" is a more serious accusation than "accidentally killed" (even if due to recklessness).

What case(s) are you thinking of?


Now there's an idea for Terminator 7.


I was confused by this because I assumed the robot fell over a few times - these are times that one of the piled up packages falls off or is a bit knocked off behind the robot (the second one seems to be knocked off by the elbow?).


Is it really sorting? All I see is the humanoid robot moving similarly shaped / sized packages from one conveyor belt to a platform to another conveyor belt. A little industrial automation design would be much more effective, cheaper, and faster compared to the task it is performing.


The actual sorting is typically automated with scanners reading the labels and shunting packages from one conveyor belt onto another, basically a physical sorting network.

Tasks left for human "sorters" to do are:

- put packages on conveyor belt so the scanner can read the label (as done by the robot in the video)

- deal with damaged or unreadable packages that can't be processed automatically

- when a package gets jammed and forces the conveyor belt to stop, remove the offending package before restarting

- receive packages at the other end and load them into vehicles

Generally the difficulty with all of these is dealing with variability and humans act as variability absorbers so the machines can operate smoothly.


Which is why robots that can also absorb variability in the same way humans do would be so valuable.


Plus it's missing the large stack of packages already in the corner that...seems like they will forever be stuck there.


I'm really confused by this video. What is it even supposed to be doing?

Is it supposed to be taking packages and placing them label face down?

I cannot understand how a robot doing this is cheaper than a second scanner so you can read the label face down or face up. I mean you could do that with a mirror.

But I'm not convinced it is even doing that. Several packages are already "label side down" and it just moves them along. Do those packages even have labels? Clearly the behavior learned is "label not on top", not "label side down". No way is that the intended behavior.

If the bar code is the issue, then why not switch to a QR code or some other format? There's not much information you need in shipping so the QR code can have lots of redundancy, making it readable from many different angles and even if significantly damaged.

The video description also says "approaching human-level dexterity and speed". No way. I'd wager I could do this task at least 10x its speed, if not 20x. And that I'd do it better! I mean I watched a few minutes at 2x speed and man is it slow. Sure, this thing might be able to run 24/7 without breaks, but if I'm running 10-20x faster then what's that matter? I could just come in a few hours a day and blow through its quota. I'd really like to see an actual human worker for comparison.

But if we did want something to do this very narrow task for 24/7, I'm pretty sure there are a hundred different cheaper ways to do it. If there aren't, then it is because there is some edge cases that are pretty important. And without knowing that then we can't actually properly evaluate this video. Besides, this video seems like a pretty simple ideal case. I'm not sure what an actual amazon sorting process looks like, but I suspect not like this.

Regardless, the results look pretty cool and I'm pretty impressed with Figure even if it is an over-simplified case.


> There should really be benchmarks where you invite robots from different companies

…and have a surprise dance-off.


It was interesting to see that they didn't show the robot folding laundry; rather, just it laying out the clothes.


There’s a scene about 2/3 through the first video where they show a brief clip of the robot folding and stacking a shirt. The quality and speed was roughly comparable to a 7-10 year old - slow and somewhat sloppy, but recognizably a folded shirt.


I think this is the frontier when it comes to "unstructured":

https://youtu.be/nmEy1_75qHk

They for sure did not anticipate that the user would backflip into their robot and knock it (and himself) out :D


I didn't know Ruby Rhod's great grandfather was already alive.


How did I know this was gonna be speed lol


No doubt. This video is just to hype investors. I'm sure they have more practical business plans than this.


I wonder if instead of making a robot to interact with the washing machine we should try to make a washing machine with an input and output hopper. Dump clothing in get clean folded clothing out.

You can control the happy path when the whole thing is your box.


+100!!! Please don't fall for the HYPE.

The current best neural networks only have around 60% success rates for small horizon tasks (think 10-20 seconds e.g. pick up apple). That is why there is so much cut-motions in this video. The future will be awesome but it will take time a lot of research still needs to happen (e.g. robust hands, tactile, how to even collect large scale data, RL).


> The future will be awesome

Perhaps this is a bit pedantic, but what about the probable eventual proliferation of useful humanoid robots will make the future awesome? What does an awesome future look like compared to today, to you?


It would be nice to be able to have the robots knock us up a house like the Alhambra place or whatever prefered architecture you have https://s7g10.scene7.com/is/image/barcelo/history-of-the-alh...


They could do all the back-breaking work. A domestic assistant would be great too.


Personal trainer. Driver. Valet. Private chef. Security guard. Doorman. Delivery driver. Rough carpentry. Baggage handler. Laundry handler. Janitor. Garbage collector. Pallet loader/unloader.

All with much improved privacy, reliability, order of magnitude lower cost, no risk of robbery/SA, etc. 24/7 operation even on holidays. Imagine service staff just sitting waiting for you to need them, always and everywhere.

Nevermind how much human lifespan will be freed from the tyranny of these mindless jobs.


> Go ask anyone working on humanoid robots today, almost everything you see here, if repeated 10 times, will enter failure mode

As someone who worked in the robotics industry, 90% of the demos and videos are cherry-picked, or even blatantly fake. That's why for any new robot in the market, my criteria is: Can I buy it? If it's affordable and the consumer can buy it and find it useful in day to day life, then this robot is useful and has potential; other than that, it's just an investor money grab PR hype.


This is so damning. How are you not afraid of retribution from big players in the AI space? You pretty much destroyed their company.


I'd watch that game show.


Oh wow, a robot that can play with my dog so I don't have to. That's exactly the kind of task I'd be relieved to automate.

The fabric wrap is idiotic. Insanely stupid. Let's have an expensive fabric-covered robot wash dishes covered in food. Genius. It's a good thing those "dirty dishes" were already perfectly clean. I doubt this machine could handle anything more. Put it in a real commercial kitchen and have it scrape oven pans and I'll be impressed.

I'm so glad I left robotics. I don't want to have anything to do with this very silly bubble.


It's a case of major FOMO. They would rather burn with the others who bet wrong than be the ones left behind.


You are purposely misinterpreting what he wrote. He said that it doesn’t matter how you die, it shouldn’t whitewash you. If you were radical and widely considered dangerous to the fabric of society, your death doesn’t magically absolve you of that or erase everything you said while alive.


> I'm saying the left has a bigger problem with violent rhetoric and actions

Here is some data that seems to say something different. It was posted as a response to Musk’s comment, "The Left is the party of murder."

https://x.com/SocDoneLeft/status/1965887912530293069

Btw It’s really crazy to read what a person who has 225M followers on X writes when he replies "Exactly" directly to claim that people who fund the Left, like Bill Gates, are murderers.


Looking at that source I’m skeptical of the validity of graph.

Anecdotally in recent years I generally see far more casual references to violence from left leaning people both online and in person. After the attempted assignation of Trump, my Facebook feed was full of left leaning friends saying “shame he missed!”. It was gross. Similar comments abounded on a Washington Post article about Kirk’s shooting. Or the guy who murdered the UnitedHealth CEO, etc.

On the linked graph take the case of that attempted assassination of Trump in Pennsylvania where the shooter is listed as “conservative/right leaning”.

However no motivation for that shooting has been found and the shooters politics were mixed. Seems he registered to vote as a republican but that’s not uncommon in a rural state as otherwise you don’t get to vote in primaries. He also donated to a democratic cause. His Wikipedia page lists his political beliefs as unknown.

Other cases I’ve looked into in my local Idaho area were listed as “right wing” or “white supremacist” but were a couple of members of a gang trying to free another who was imprisoned for dealing drugs.

Most of those drug gangs aren’t left or right leaning, just thugs.


40-50k for talent? No way. I have friends in Eastern Europe doing 120-130k USD working remotely for US companies.


> possible revolutionary products ahead.

Which products? EVs are a commodity. Self driving technology is better at Waymo, and in China, the latest Huawei version of self driving, installed in Avatar cars, is on par with Tesla’s and even better in some cases. What’s left? The Optimus robot? Unitree from China and Boston Dynamics (owned by Toyota), are ahead of Tesla. Not to mention the hundreds of startups in China working on the same thing, all using essentially the same transformer based architecture with only minor tweaks. There’s no moat this time. What Tesla still excels at is marketing and hype, but even that has its limits.


Basically everything you said here is completely untrue


That's quite the counterargument. Care to back it up with anything more than "you're wrong"? Right now this reads more like a dismissal than a rebuttal.


So much is wrong. First of all Boston Dynamics is owned by Hyundai not Toyota.

How many cars in the US is haiwei self driving in right now?

How many robots is Boston Dynamics selling?

You are arguing that vaporware is better than what Tesla has.


Langley


Sure, you are. You created this account 2 hours ago, all comments anti China, perfect english and you write about China as "their" country in one of the comments.


FWIW, I do the same thing when referring to the US if my being American is immaterial to the point or observation. It is a way of intentionally not privileging your opinion.


Rare to see "you're a bot" be invoked for anti-Chinese sentiment. Thanks for the tickle today!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: