So I was at target and I just stopped one of the employees and asked them hey does this reflect how your day-to-day job is and they said yes directly.
They have this gun that’s made by I think zebra, and it basically tells them what to do all day every day for every minute and interacting with a human supervisor is by exception
I think Amazon is actually a better example here because warehouse workers have been living this way for I guess decades now with their per item countdown timers that they have to physically you know achieve or they get downgraded or whatever
Much of what we call "dystopian fiction" is actually just palette-swapped contemporary or historical fiction. People are appalled at the idea of {horrible thing} happening to them, not simply {horrible thing} happening in general.
> If you had an expert looking over your shoulder and telling you what to do, I expect most of you could do most of the work of an electrician … In fact, that's the bulk of how electricians learn their trade: through apprenticeship. People wearing glasses with built-in cameras, connected to today's strongest AIs could already do a lot with a bit of scaffolding.
Apprentices have their work directly checked by the person supervising them, who is also ultimately personally responsible for the results.
I am a programmer but also like to work with my hands—I’ve done some amount of "real" electrical, plumbing, carpentry, light construction, mechanical, and landscaping work. To me, this line of thinking overall falls into a category of "things computer programmers believe about other kinds of jobs," that I see fairly often on HN. It’s key to the whole argument here but kind of papered over with an unvalidated assumption it’d probably already work with today’s AI.
Anyone who has followed a DIY YouTube video more than a few times has encountered situations where having muscle memory and proprioceptive experience with actually manipulating the tools and objects you’re working with would significantly reduce the chance you mess it up. How tight is too tight? Am I going to shear this bolt head off? Is this saw blade getting dull and going to hurt me? Even the best guides and teachers don’t cover the combinatoric ways things can frequently go wrong–it’s accumulated knowledge in the heads and hands of people who don’t blog about it so it can be slurped into training data.
I think there’s an enormous chasm between "can AI provide reasonably correct guidance for an inexperienced person along a happy path to accomplish basic physical tasks" and "I no longer need to hire an electrician because AI guides me through every step."
The author uses a strange definition of teleoperation.
Following GPS directions is no different than assembling an Ikea furniture by following the instructions. So I guess Ikea has been teleoperating millions of humans for decades?
I think it has to do with the feedback loop. Ikea doesn't know what you're doing and can't adjust and adapt to provide you with new instructions if you take a wrong turn.
It's something that you'd pause and think about if you were trying to carefully craft a sensible argument. But the online attention economy now runs on AI op-eds. In this world of overabundance, the winning strategy is to produce outrageous takes and do it as quickly as possible.
I respect the author, but he succumbed to an... unnecessary temptation. There will be a dozen other "here's what AI means for humanity" posts on HN before the end of the week. For the human-written ones, I doubt that any of the authors will be proud of their analyses five years from now.
> The classic picture of AIs doing things in the world is robots, but I think a more realistic picture of the near future is computers telling people what to do.
As a purely qualitative description, this does not strike me as anything particularly new. Machines had been doing something like this long before the computer age. For example, as early as the late Middle Ages, mechanical wheel-driven clocks with striking mechanisms were used to call people to prayer.
Any process in which a person reacts to a machine can be characterised in this way. Every till that calculates how much one has to pay, every traffic light, every other machine-generated signal. When a warning light comes on in the car, we are prompted to do something: release the handbrake properly, fasten our seatbelt, visit a garage, ...
We are already so deeply embedded in such cycles that AI seems to be merely a new variation on them than a fundamental change.
The deep sense of unease regarding AI does not stem from a machine telling us what to do, but from the risk of losing control over the situation as a whole.
Interesting term that is in context of AI - "teleoperation". Sends a chill thinking about that world. But also the inevitability of a plausible future.
"So, teleoperation is probably very economically productive. Is it a good thing? I think mostly no, for several reasons." followed by several paragraphs discussing the negative consequences. This is celebrating?
Secondary reason: Oh, it will also make most people unable to have a living wage. We should probably do something about that, if it doesn't strain the economy too much.
The dehumanization of everyone who will still have a job? Nah, not a problem.
Though I think this dismisses the potential of people to organize, especially in jobs like those. This is why I'm skeptical that future will happen as easily as he is warning here.
Human extinction, mass starvation, and terrible jobs are all outcomes I think we should try hard to avoid. But I do think I have them in the right order!
(The post is also not about something I'm working on.)
One of the upshots of "pacing the frontier" is that we are all going to have to accept that compute will be a regulated, restricted resource. This includes both "how-much" restrictions (high-performance computers just won't be available to the public) and "what-kind" restrictions (software and hardware features will enforce policies on what software you may run, much like a game console or iPhone).
Back in the 2000s, one of my favorite programmers from the Before Times—Fabrice Bellard—came up with an algorithm for computing digits of pi that was so much better than the state of the art, he beat with a $3000 PC the record for speed of pi calculation, which had been established with a supercomputer.
Some bright spark—maybe they'll have Claude assist, maybe it'll trigger Claude's safety locks, I dunno—is going to derive a way to do transformer or other frontier AI inference in a really efficient manner on consumer hardware, making local models practical to run on the PCs people have today and nearly, if not actually, competitive with today's frontier models. "There is another theory which states that this has already happened"—perhaps several times already, and we just need a few more goes of the cycle. I find that for supposedly a crowd of the smartest dudes in the room, people in the Valley like to do things the stupid way around, by sheer fucking brute force: scale up massively, burn billions and trillions of VCbux on capex, aggressively conserve developer effort at the expense of everybody's CPU time and storage capacity—yours, your cloud provider's, your customers'... So I suspect, but at this time cannot prove, that there's a whole lot of headroom for optimization of AI inference, which has the potential to put some truly powerful models in the hands of ordinary people.
And that's very dangerous, because so much for pacing the frontier! If Joe Schmoe can build a mini-AI lab in his mancave, he can then gain access to a model that gives him all the big scary things Dario and Sam are worried about: instructions on how to build bombs or bioweapons, child porn generation, human sacrifice, cats living with dogs, mass hysteria! If you're concerned about an AI-mediated workplace panopticon, Joe Schmoe could, in principle, build one of those for his auto body shop with a local model. The only way to stop Joe from doing so is to stop Joe from acquiring the equipment it would take to run such a model.
Pacing the frontier is at least as much about pacing what you can compute as it is about what goes on in billion-dollar data centers. Any argument for limiting AI growth and development—and admittedly, there are some strong ones—is an argument against general purpose computing. So if we're really going to regulate AI, bend over for the laws that mandate OS/firmware/hardware-enforced restrictions on what binaries your computer may run. And so long open source. Oh well, those laws were coming anyway; that's what the age verification thing is all about. Society has determined that people can't be trusted with computers.
I don’t buy it. Your premise suggests new inference would make SoTA models performant on hardware that we already have in our hands. To prevent us from running local models, we would literally be forced to turn over existing hardware with such capabilities, prevented from building such machines without a suitable license, and so on. That sounds a lot like attempts at gun confiscation, and the result being that only criminals have weapons. This is not the way.
>Which is why I'm heartened to see a lot of support, including from the CEOs of Anthropic and OpenAI, for managing the pace at which these systems become increasingly capable.
This poor guy actually believes the PR, that the frontier model CEOs want to slow things down (and not just setup regulatory-enforced moats around their empires).
Sounds basically like the plot of the first Dystopia in Marshall Brain’s manna
https://web.archive.org/web/20260725174355/https://marshallb...
https://news.ycombinator.com/item?id=41743688
Wow, thanks for linking this. I must have read it once in the distant past but it seems far more relevant today.
I use this reference all the time because it’s literally and I do mean literally the state of a significant portion of retail workers
This is how target workers live
Can you elaborate? How do Target workers live?
So I was at target and I just stopped one of the employees and asked them hey does this reflect how your day-to-day job is and they said yes directly.
They have this gun that’s made by I think zebra, and it basically tells them what to do all day every day for every minute and interacting with a human supervisor is by exception
I think Amazon is actually a better example here because warehouse workers have been living this way for I guess decades now with their per item countdown timers that they have to physically you know achieve or they get downgraded or whatever
This is an academic resource that says similar: https://onlinelibrary.wiley.com/doi/10.1111/bjir.70051?utm_s...
Yup. Some people read dystopian fiction as an instruction manual.
Much of what we call "dystopian fiction" is actually just palette-swapped contemporary or historical fiction. People are appalled at the idea of {horrible thing} happening to them, not simply {horrible thing} happening in general.
> If you had an expert looking over your shoulder and telling you what to do, I expect most of you could do most of the work of an electrician … In fact, that's the bulk of how electricians learn their trade: through apprenticeship. People wearing glasses with built-in cameras, connected to today's strongest AIs could already do a lot with a bit of scaffolding.
Apprentices have their work directly checked by the person supervising them, who is also ultimately personally responsible for the results.
I am a programmer but also like to work with my hands—I’ve done some amount of "real" electrical, plumbing, carpentry, light construction, mechanical, and landscaping work. To me, this line of thinking overall falls into a category of "things computer programmers believe about other kinds of jobs," that I see fairly often on HN. It’s key to the whole argument here but kind of papered over with an unvalidated assumption it’d probably already work with today’s AI.
Anyone who has followed a DIY YouTube video more than a few times has encountered situations where having muscle memory and proprioceptive experience with actually manipulating the tools and objects you’re working with would significantly reduce the chance you mess it up. How tight is too tight? Am I going to shear this bolt head off? Is this saw blade getting dull and going to hurt me? Even the best guides and teachers don’t cover the combinatoric ways things can frequently go wrong–it’s accumulated knowledge in the heads and hands of people who don’t blog about it so it can be slurped into training data.
I think there’s an enormous chasm between "can AI provide reasonably correct guidance for an inexperienced person along a happy path to accomplish basic physical tasks" and "I no longer need to hire an electrician because AI guides me through every step."
The author uses a strange definition of teleoperation.
Following GPS directions is no different than assembling an Ikea furniture by following the instructions. So I guess Ikea has been teleoperating millions of humans for decades?
I think it has to do with the feedback loop. Ikea doesn't know what you're doing and can't adjust and adapt to provide you with new instructions if you take a wrong turn.
It's something that you'd pause and think about if you were trying to carefully craft a sensible argument. But the online attention economy now runs on AI op-eds. In this world of overabundance, the winning strategy is to produce outrageous takes and do it as quickly as possible.
I respect the author, but he succumbed to an... unnecessary temptation. There will be a dozen other "here's what AI means for humanity" posts on HN before the end of the week. For the human-written ones, I doubt that any of the authors will be proud of their analyses five years from now.
> The classic picture of AIs doing things in the world is robots, but I think a more realistic picture of the near future is computers telling people what to do.
As a purely qualitative description, this does not strike me as anything particularly new. Machines had been doing something like this long before the computer age. For example, as early as the late Middle Ages, mechanical wheel-driven clocks with striking mechanisms were used to call people to prayer.
Any process in which a person reacts to a machine can be characterised in this way. Every till that calculates how much one has to pay, every traffic light, every other machine-generated signal. When a warning light comes on in the car, we are prompted to do something: release the handbrake properly, fasten our seatbelt, visit a garage, ...
We are already so deeply embedded in such cycles that AI seems to be merely a new variation on them than a fundamental change.
The deep sense of unease regarding AI does not stem from a machine telling us what to do, but from the risk of losing control over the situation as a whole.
Sounds like we're just one step away from https://en.wikipedia.org/wiki/Manna_(short_story)
I think you might be thinking of https://en.wikipedia.org/wiki/Manna_(novel) ?
Weird to see Cary mentioned as the setting
Crap, you're right. The Brian Marshall one.
It's very prescient! Someone on LW linked me to it when I cross-posted there.
The wiki describing the US in this story reminds me of the opening scenes to Soylet Green.
or Mrs. Davis https://www.youtube.com/watch?v=PIOnrEujKl8
Avoiding this is part of why I think working on robotics is important. I don't want humans stuck being hands for an AI that doesn't have them.
I'm building this since 2022. It's not a new concept, but it is a new control plane and interesting things to solve.
Human + AI economies is the phrase I've been using. Maybe HAI Economies might be better.
A lot of people are doing my this already. There are also companies that build AI platforms for contractors to move them around to fill in gaps.
Interesting term that is in context of AI - "teleoperation". Sends a chill thinking about that world. But also the inevitability of a plausible future.
It's interesting how a more sophisticated term can either highlight (or obscure) the anxiety (or glibness) behind its tongue-in-cheek alternatives.
Isn't this just the "reverse centaur" again, but this time framed as something ostensibly desirable?
[author]
FYI, I don't think this is desirable. See the section of the post starting with "Is it a good thing? I think mostly no..."
> Human teleoperation naturally comes with on-the-job surveillance and micromanagement.
> This technology allows employees to be easily slotted in and out of different jobs with very little friction, which destroys their bargaining power.
> AI permanently blacklists employees from every minimum-wage job just for being late a few times
> This can easily lead to a race to the bottom where companies no longer have to providing fulfilling, dignified jobs.
If I were stuck in this hell, I would gladly assist any superintelligent AI taking a swipe at the humans who put me in it.
If there is one inexhaustible resource than the stream of ideas of capital owners how to make labor even more miserable.
...and of course he is a Rationalist/EA guy. Why am I not surprised?
The author of the article isn't advocating for teleoperated humans.
He's celebrating it as a reality that is already here.
"So, teleoperation is probably very economically productive. Is it a good thing? I think mostly no, for several reasons." followed by several paragraphs discussing the negative consequences. This is celebrating?
"Mostly" no. You can read a lot of warnings from rationalists that are oddly about the things they are actively working on.
I think the order of concerns is also telling.
Primary reason why it's bad: Increase risk of runaway superintellicence, XRisk, etc etc
Secondary reason: Oh, it will also make most people unable to have a living wage. We should probably do something about that, if it doesn't strain the economy too much.
The dehumanization of everyone who will still have a job? Nah, not a problem.
Though I think this dismisses the potential of people to organize, especially in jobs like those. This is why I'm skeptical that future will happen as easily as he is warning here.
[author]
Human extinction, mass starvation, and terrible jobs are all outcomes I think we should try hard to avoid. But I do think I have them in the right order!
(The post is also not about something I'm working on.)
https://erkansaka.net/2026/09/10/ai-doom-rhetoric-safety-hyp...
AI;DR
I don't see any celebrating in the article. The article makes it pretty clear that the author sees many problems coming from teleoperated humans.
Try reading the article before commenting
Aka Meatbag as a Service
I'm surprised the author is not panicking about the existence of customer support for decades.
This completely confused and misses the point of teleoperation vs following instructions/rules/protocols. Who set the GPS navigation destination?
Any sort of rule following is teleoperation by this definition then ⇒ the whole world is teleoperated.
Hospitals introduced intubation checklists to make sure protocols are followed and patients don't die. How about coding guidelines?
See also: "meat proxy", "reverse centaur"
One of the upshots of "pacing the frontier" is that we are all going to have to accept that compute will be a regulated, restricted resource. This includes both "how-much" restrictions (high-performance computers just won't be available to the public) and "what-kind" restrictions (software and hardware features will enforce policies on what software you may run, much like a game console or iPhone).
Back in the 2000s, one of my favorite programmers from the Before Times—Fabrice Bellard—came up with an algorithm for computing digits of pi that was so much better than the state of the art, he beat with a $3000 PC the record for speed of pi calculation, which had been established with a supercomputer.
Some bright spark—maybe they'll have Claude assist, maybe it'll trigger Claude's safety locks, I dunno—is going to derive a way to do transformer or other frontier AI inference in a really efficient manner on consumer hardware, making local models practical to run on the PCs people have today and nearly, if not actually, competitive with today's frontier models. "There is another theory which states that this has already happened"—perhaps several times already, and we just need a few more goes of the cycle. I find that for supposedly a crowd of the smartest dudes in the room, people in the Valley like to do things the stupid way around, by sheer fucking brute force: scale up massively, burn billions and trillions of VCbux on capex, aggressively conserve developer effort at the expense of everybody's CPU time and storage capacity—yours, your cloud provider's, your customers'... So I suspect, but at this time cannot prove, that there's a whole lot of headroom for optimization of AI inference, which has the potential to put some truly powerful models in the hands of ordinary people.
And that's very dangerous, because so much for pacing the frontier! If Joe Schmoe can build a mini-AI lab in his mancave, he can then gain access to a model that gives him all the big scary things Dario and Sam are worried about: instructions on how to build bombs or bioweapons, child porn generation, human sacrifice, cats living with dogs, mass hysteria! If you're concerned about an AI-mediated workplace panopticon, Joe Schmoe could, in principle, build one of those for his auto body shop with a local model. The only way to stop Joe from doing so is to stop Joe from acquiring the equipment it would take to run such a model.
Pacing the frontier is at least as much about pacing what you can compute as it is about what goes on in billion-dollar data centers. Any argument for limiting AI growth and development—and admittedly, there are some strong ones—is an argument against general purpose computing. So if we're really going to regulate AI, bend over for the laws that mandate OS/firmware/hardware-enforced restrictions on what binaries your computer may run. And so long open source. Oh well, those laws were coming anyway; that's what the age verification thing is all about. Society has determined that people can't be trusted with computers.
I don’t buy it. Your premise suggests new inference would make SoTA models performant on hardware that we already have in our hands. To prevent us from running local models, we would literally be forced to turn over existing hardware with such capabilities, prevented from building such machines without a suitable license, and so on. That sounds a lot like attempts at gun confiscation, and the result being that only criminals have weapons. This is not the way.
Interesting idea, but I couldn't get past this:
>Which is why I'm heartened to see a lot of support, including from the CEOs of Anthropic and OpenAI, for managing the pace at which these systems become increasingly capable.
This poor guy actually believes the PR, that the frontier model CEOs want to slow things down (and not just setup regulatory-enforced moats around their empires).
[flagged]