I want to believe its true, but this seems like quite a leap forward, at that price point. Even without AI, would we know if the videos were just a person in a suit?
The leap is not really there yet and it's cheap because you are the product. The robot will be a massive headache, will work poorly for most tasks, frequently break and require maintenance. In exchange for $500/mo and providing those test hours in a novel environment and the data that goes with it, you get to have a robot in your house that occasionally does something right. The bet being made here is that they can turn that data hose into a useful robot before this poor customer experience tanks their brand.
Presumably, this is a way to collect diverse training data for the robot to be trained on. Wash and fold as a service is valuable (to some people), and presumable the “extra steps” are offset with the in-home aspect of this.
Meanwhile, the ethical considerations are huge. Laborers are literally training their replacement, and probably at questionable wages. They’re also explicitly inviting someone into your home remotely, and that person can see and interact with your house. Feels like a privacy and safety risk. Additionally, it seems likely that this would be a literal Trojan horse to allow international labor to work within the US without dealing with actual immigration. Oh and just for good measure, it’s taking the jobs traditionally held by some of society’s least privileged and most desperate workers.
Anyways, if it actually works, I want one.
Edit: I feel compelled to note that apparently they’re hiring in Palo Alto for these roles, today.
I'm wondering about teleoperation. Many housekeeping activities require an incredible level of attention to detail, precision, and real-time awareness. For example, consider manual dishwashing, small sewing jobs, knife operation, or repotting houseplants.
Even if latency were not an issue, the operator would need to excel at all these tasks.
By the way, we've had robotic surgery [1] for years. These machines are very expensive, and it takes months, if not years, to learn to operate them flawlessly.
I agree with you. This is brilliant. Tele-operation of the humanoid vs. waiting for AI is the key here. Then off-shore the tele-operation when you've smoothed out the edges.
I'll personally wait to own best hardware (Unitree) and purchase my own 3P tele-operation service contract.
Oof. The roomba guy said that the form factor of robots inform customer expectations. I keep thinking about that and wincing when I see these humanoid robots. Even if there's impressive engineering that goes into them, people are going to expect they can do human things. When they can't, they're going to be disappointed.
I expect my robot vacuum to vacuum the floor, because it's a little wheeled disc on the floor. It's not going to be able to cook for me. But this thing? Yea, it should cook for me.
Form factor isn't what is important. What is important is the jobs it gets done. If it can do useful jobs we will accept the form factor. Roomba doesn't look like any other vacuum I've never owned, but it gets the floors clean so we learn that form factor means clean floors.
Humanoid only seems useful if it can do stairs - something many form factors fail at. Though I'd expect a centaur form factor could do stairs better and probably is cheaper.
Could it do stairs if it was wheeled? Not every house has them, but most do (where I live basements are universal in single family houses, but other regions basements are rare)
So, like a large scale scam to get your "downpayment" ?
How would remote human operators scale, especially for the $20k "ownership" model? I presume the actual hardware probably costs them at least $10k to make, so after about 400 hours of "remote operator use", it's all loss on the company?
I suspect they have a limit on use, or a pay-to-use-remotely thing they neglected to announce.
The part of the video showing the robot putting glasses in the upper cabinet. It is something normal for humans, but it felt scary watching it being done by that robot. Maybe it was how it was handling the glass, maybe another kind of uncanny valley, or how I think present software should handle that task today. But I don't think it is ready yet to match our expectations.
The issue is might be the latency with teleop, by the time the operator realizes the glass is about to tip over, it's probably already fallen on the floor. So the robot can really only do one object at a time, and has to move about awkwardly. I do like all the ideas though, I hope they can get it to a polished state.
Although these particular units are designed for home use, commercial applications are not far off, perhaps in the order of months.
Small and medium-sized businesses will start thinking that it's much better to lease a unit for $500/mo. than $2,000/mo. in payroll for one human. Then they own the unit after 3 years. We're going to need some form of UBI soon.
80 comments
[ 3.4 ms ] story [ 83.9 ms ] threadMind blowing.
> Use NEO as a mobile bluetooth speaker anywhere in your home.
https://news.ycombinator.com/item?id=43132260 ("Neo Gamma (Home Humanoid) (1x.tech)"—48 comments)
TBD on if it ships on time, how good it is, etc, but fuck, this is pretty cool.
What else are we getting AT BEST beside taking out the trash and gimmicks?
Is this a humanoid robot that's controlled by someone in a call center remotely doing your laundry?
Putting aside ethical reservations about how much they are probably paying per task, that feels like wash and fold with extra steps.
Presumably, this is a way to collect diverse training data for the robot to be trained on. Wash and fold as a service is valuable (to some people), and presumable the “extra steps” are offset with the in-home aspect of this.
Meanwhile, the ethical considerations are huge. Laborers are literally training their replacement, and probably at questionable wages. They’re also explicitly inviting someone into your home remotely, and that person can see and interact with your house. Feels like a privacy and safety risk. Additionally, it seems likely that this would be a literal Trojan horse to allow international labor to work within the US without dealing with actual immigration. Oh and just for good measure, it’s taking the jobs traditionally held by some of society’s least privileged and most desperate workers.
Anyways, if it actually works, I want one.
Edit: I feel compelled to note that apparently they’re hiring in Palo Alto for these roles, today.
That one doesn't have to do, hence the appeal.
Imagine being a kid and waking up to this sitting in your room, silently watching you sleep.
Imagine how terrified your dog is going to be of this thing, shuffling around or getting stuck with its foot on the edge of a rug.
Imagine finding it going through your underwear drawer when you come home from work early.
How'd they somehow revive Gene Roddenberry to come and pose with Neo?
* Water Plants
* Turn off lights
* Get the door
* clean up trash
* Load/Empty dishwasher
* Tidy House
* Laundry
* Bartend Party
* Feed Pets
* Play music as the most over engineered Bluetooth speaker
I'll be curious if they move those positions to a lower cost-of-living area as they scale up.
1: https://1x.recruitee.com/o/robot-operator
By the way, we've had robotic surgery [1] for years. These machines are very expensive, and it takes months, if not years, to learn to operate them flawlessly.
[1]: https://en.wikipedia.org/wiki/Robotic_surgery
I'll personally wait to own best hardware (Unitree) and purchase my own 3P tele-operation service contract.
I expect my robot vacuum to vacuum the floor, because it's a little wheeled disc on the floor. It's not going to be able to cook for me. But this thing? Yea, it should cook for me.
Humanoid only seems useful if it can do stairs - something many form factors fail at. Though I'd expect a centaur form factor could do stairs better and probably is cheaper.
How would remote human operators scale, especially for the $20k "ownership" model? I presume the actual hardware probably costs them at least $10k to make, so after about 400 hours of "remote operator use", it's all loss on the company?
I suspect they have a limit on use, or a pay-to-use-remotely thing they neglected to announce.
It's pretty clear that they're still working on the AI training so 'human in the loop' is not part of their long term business model.
Small and medium-sized businesses will start thinking that it's much better to lease a unit for $500/mo. than $2,000/mo. in payroll for one human. Then they own the unit after 3 years. We're going to need some form of UBI soon.