Microsoft really seems to be tripling and quadrupling down on total surveillance of the user's "own" system. If you haven't ditched MS yet, I'd consider it now.
Linux is becoming more and more viable for a gaming PC. For business uses a Linux desktop is usable but probably not ideal, but you also have macOS. I'd pick anything but Windows and MS stuff.
I think that's always been true. The ideas and reasoning process matter. So does the end product. If you produced it with an LLM and it sucks, it still sucks.
Consumer robotics strikes me as an engineering tar pit so deep it leads to hell. If full self driving is hard due to the long tail of unusual special cases, this is orders of magnitude worse.
Take FSD but multiply the number of actuators and degrees of freedom by at least 10, more like 100. Add a third dimension. Add direct physical interaction with complex objects. Add pets and children. Add toys on the floor. Add random furniture with non-standard dimensions. Add exposure to dust, dirt, water, grease, and who knows what else? Puke? Bleach? Dog pee?
Oh, and remove designated roads and standardized rules about how you're supposed to drive on those roads. There are no standards. Every home is arranged differently. People behave differently. Kids are nuts. The cat will climb on it. The dog may attack it. The pet rabbit will chew on any exposed cords.
We've all seen those Boston Dynamics robots. They're awesome but how durable would they be in those conditions? Would they last for years with day to day constant abuse in an environment like that?
From a pure engineering point of view (neglecting the human factor or cost) a home helper robot is almost definitely harder than building and operating a Mars base. We pretty much have all the core tech for that figured out: recycling atmosphere, splitting and making water, refining minerals, greenhouses, airlocks, and so on. As soon as we have Starship or another super heavy rocket that's reliable we could do it as long as someone was willing to write some huge checks.
And of course it's a totally untested market. We don't know how big it really is. Will people really be willing to pay thousands to tens of thousands for a home robot with significant limitations? Only about 25% of the market probably has the disposable income to afford these.
You'd have to go way up market first, but people up market can afford to just pay humans to do it.
> Will people really be willing to pay thousands to tens of thousands for a home robot with significant limitations?
The answer to that is no, probably for the foreseeable future. The robot demos we have no can't even fold laundry or put dishes away without being teleoperated. Both extremely basic tasks that any household robot would be required to do, along with other messy jobs that put it at risk as you said: taking out the trash, feeding the pets, cleaning up messes, preparing or cooking food, etc.
The price it would have to cost with current tech would be astronomically more than just hiring a human, and they would almost certainly come with an expensive subscription as well, whereas I can hire a human to come in and clean my home weekly for about $200/month.
Humans who aren't skilled require training regardless of how "unskilled" the task is.
Humans that are chronically unskilled also don't learn well, somewhat as a rule.
Humans that don't make much money have a high turnover rate from burnout. Additionally, those that can learn typically leave for greener pastures.
The bar isn't terribly high. Efficiency of scale in production will solve this eventually. I think the likely outcome is robots building themselves first.
Almost all developed economies are running into a fertility crisis right now, with labor shortages already appearing in the frontrunners of the trend, such as Germany.
Human work is going to cost more in the future, and immigration from countries such as Thailand or Vietnam is already slowing down. Even a mediocre robot will be sought after if it is the only choice you have.
I understand that. It's my personal opinion that one of the causes of low birth rates is that we continually choose to have robots solve our problems instead of choosing a human.
I think we could increase birth rates by making a taxation scheme in which the most marginally effective way to solve a problem is with a human, paid a wage which allows for that occupation to be a lifelong career.
They’ll be bought/leased, providing direct profit. Also, there’ll be maintenance revenue. I think they’re expected to cost around $30K.
In the case where they’re replacing a low-skill human worker, they’ll pay for themselves in 1-2 years…plus no sick days, no drug use, no theft, and they can work 24 hours a day, less any recharging time.
Once large swaths of the planet have been rendered uninhabitable from human activity, we'll require them to continue extracting profit from those areas. (this is a downer comment but also realistically the first thing that came to mind when trying to think of a use for them).
A teleoperated robot is little more than a human worker with extra steps. (And an expensive, clumsy human worker at that.) I can't imagine many situations where that would make sense instead of having a human do the work in person.
I could see teleoperated help catching on. Americans are weird about staff. When I visit my old-world family, it's seen as perfectly normal to have someone living in an attached apartment, handling the cooking the cleaning, etc. There are well-established etiquette rules, understood both by the staff and the family, which help navigate the rather complicated, radically unequal relationship between the two.
Americans by and large don't do that. We software developers have not that different of an income gap between us and minimum wage workers compared to my family overseas and their staff. Yet, it would be considered weird, extravagant even, for a $300-500k/yr developer to have dedicated help. We're far more comfortable with people we don't need to interact with directly, like housecleaners, landscapers, etc.
Teleoperated robots sidestep that discomfort, somewhat, by obscuring the the humanity of the staff. It's probably not a particularly ethical basis for a product, but when has that ever stopped us.
Maybe you can scale to have one operator operate ten or a hundred household robots at a time.
An autonomous robot that has 99% reliability, getting stuck once an hour, is useless to me. A semi-autonomous robot that gets stuck once an hour but can be rescued by the remote operator is tempting.
Expect security and privacy in the marketing for these things, too, but I don't think that's a real differentiator. Rich and middle class people alike are currently OK with letting barely-vetted strangers in their houses for cleaning the world over.
- Services like maids or cleaners are usually scheduled, maybe you have to wait and open the door etc. Maybe they can't make it that day because of snow storm etc.
- Services are normally limited to certain hours. With a remote operator, the robot could do laundry all night ran by someone in a different time zone.
- If needed could be operated in shifts.
- Other new use cases could arise, e.g. wellness check on elderly, help if fallen or locked out etc.
Low duty cycle. If one human can drive 20 robots, because most of them are sitting still most of The time, it starts to make sense. Vs a maid or butler that can obviously only really work one home at a time.
The person in a third world country is not a slave, they're doing the job for a few bucks a day because it's still better than other options available to them.
What is the difference between being a teleoperator in India for a californian family robot, and being a software dev for a company selling SaaS products to the US market?
Yeah but with a teleoperated worker you can have them work remote from a place with poor labor regulations and extremely low pay.
The future with this as a reality is a really dark place, where the uber wealthy live entirely disconnected from the working class except through telepresent machines half a planet away. That way the wealthy don't have to be inconvenienced by the humanity of the poors.
If a robot can do basic cleaning, laundry, and dishes, that's worth a lot to a lot of people. Dual-professional households have the money, and not having to do this housework could save some marriages.
I don't think it actually is worth a lot to people. I know dual-professional households who don't even use their dishwasher consistently, and multiple companies have gone bankrupt trying to bring automated laundry folding (which does exist in industry) to the consumer market.
Maid services are generally expected to handle "everything" for a pretty expansive definition of everything. They pick up scattered stuff and put in a sensible location, they arrange everything visible in an aesthetically pleasing way, they take out the trash, if there's some weird dirt that's hard to clean they creatively problem solve to find a way to get it off. I don't think there's a market for a service that can only handle basic cleaning.
(Will someone eventually invent a machine that can do all of that and more? Yes, probably, and they'll make billions when they do. But Tesla has offered no reason to believe this is on their horizon, and the focus on a humanoid form factor strongly suggests that they're optimizing for media appeal over practical capabilities.)
Maids are paid a VERY low wage in exchange for being able to take on an almost unlimited list of general tasks, from folding laundry to managing kids to mopping stairs. We are decades away from robots with that capability, and they are intended to replace people who are often not making even minimum wage? Please. Get real.
Robot vacuum with a mop, washing machine, tumble dryer and dishwasher reduce housework to like an hour per week, ie 30 min/person/week. This can be higher if you live in a big house, but if your marriage can’t tolerate 30 mins of house work a robot will not solve it.
> Dual-professional households have the money, and not having to do this housework could save some marriages.
Dual-professional households could hire a maid and pay for marriage counseling and still save money compared to a $20k robot plus whatever a subscription would run.
I can google "maid service seattle" and see dozens of entries. The first one in the yelp list is available to book and will clean a 1000 - 1500 sq ft, 2 bed, 2 bath house for well under $200. There's even a decent discount if you book is as a weekly or biweekly service.
That feels pretty affordable? I know it's a scale, but minimum wage here is $21/hr now.
I have enough time to take care of my own space, but for comparison Comcast internet is well over $120/month for crappy speeds. I think in comparison a little more than that for 1 deep cleaning a month is reasonable.
It's super cool but I see it as a much more flexible open ended take on the idea of procedurally generated worlds where hard-coded deterministic math and rendering parameters are replaced by prompt-able models.
The deadness you're talking about is there in procedural worlds too, and it stems from the fact that there's not actually much "there." Think of it as a kind of illusion or a magic trick with math. It replicates some of the macro structure of the world but the true information content is low.
Search YouTube for procedural landscape examples. Some of them are actually a lot more visually impressive than this, but without the interactivity. It's a popular topic in the demo scene too where people have made tiny demos (e.g. under 1k in size) that generate impressive scenes.
I expect to see generative AI techniques like this show up in games, though it might take a bit due to their high computational cost compared to traditional procedural generation.
Back in the 80's and 90's he was pretty well known for noting that if you owe a bank enough money, it is in the bank's interest not to let you fail.
My guess is that it also helps to owe lots of money to lots of banks at the same time. That way when one goes after you, the others will help you out or risk losing their money, too.
Starting way back in the early 2000s I was predicting all this and was consistently called nuts and paranoid.
In retrospect what has actually happened with mass surveillance has been far worse than what the most unhinged conspiracy nut on shortwave radio or some crazy end times Geocities web site was predicting back then. The predictions of the conspiracy nuts were conservative.
The big thing everyone got wrong was that we assumed people would care and put up resistance. We assumed people would choose technologies that protected their privacy and would get mad when highly invasive things were foisted on them. That never happened. Give people convenience and shiny and fun "content" like TikTok and YouTube and they'll consent to live in a total panopticon. They don't care.
We're also seeing that people will choose wealth and comfort over rights and freedom. This bargain is being made all over the world to varying degrees, and the trend is toward increasingly authoritarian societies that offer a comfortable lifestyle as long as you don't question it too much. A quote I read a while back described the emerging system like this: "it's Brave New World unless you question it, then it turns into 1984 real fast."
This is all a devil's bargain, but like the devil's bargain in fiction it's great at first. The devil really does deliver. It's all fun until you get dragged off to hell at the end.
I'm referring to mobile phones, software that constantly spies on you, location tracking, and mass data fusion without any regard for legal limitations or privacy.
Each individual data point seems normal or innocuous, but when you tie them all together and then leverage the tech panopticon you have an insane amount of detail on every person. There are no meaningful legal safeguards on how this data is used, especially when it's laundered through private contractors not subject to much oversight.
When you couple this with increasingly unlimited powers granted to law enforcement agencies, you get a situation where a system could decide you're a threat and some just comes and beats the shit out of you, takes your property, or shoots you, and you have little recourse.
The people cheering for this seem to think it'll never be used against them.
> I'm referring to mobile phones, software that constantly spies on you, location tracking, and mass data fusion without any regard for legal limitations or privacy
None of that is relevant to the article. It’s about HHS data being queried to give ICE probable addresses. What you’re doing is indistinguishable from whataboutism.
I don’t think that’s your intent. But we have an actual abuse of public data at hand here. Going on a tangent about dragnet surveillance is off topic and misleading.
> don't think the two topics are separable. This is a specific case of the general trend
They are and should be separable. DHS hoovering up government data is orthogonal to private data collection. They could become related. But they aren’t, and muddling a hypothetical problem with a clear, present and actual one is a good way to normalize the latter.
The should be separable, but they are not. Data collected privately absorbed by the government is a serious problem, even anonymized data can be de-anonymized if you can put more than one database next to each other. This allows for far more insights than any single database could give you and this is a real danger.
Keep in mind that DOGE made off with a huge stash of data, which combined with other data, such as voter registration data, twitter messages (public and private) and other such datastores could become an extremely efficient tool in messing with elections. The whole system is predicated on that being hard and so we trust the outcome of elections but with todays tools in the hands of the large US companies currently in cahoots with the Trump administration this is childs play.
> Data collected privately absorbed by the government is a serious problem
The data we’re talking about here are home addresses. HHS (or the IRS) having home addresses isn’t what most Americans would or should consider problematic.
This isn't about 'Americans' but about the negative set of HHS records compared to the records taken from for instance the IRS. Putting the one next to the other yields the names of individuals that were otherwise not standing out. ICE/Palantir/DHS should not have access to health records. The main reason for that is that people who are in the country may still require healthcare even if they have no other ties the US government. Of course, for some this is the desired outcome, they hope that those people will no longer avail themselves of healthcare at all with all of the predictable outcomes.
> ICE/Palantir/DHS should not have access to health records
Totally agree. Where I disagree is in saying the government shouldn’t have these records. Like, no. The government knowing where I live is not only fine but also sort of necessary. Just because it has some data doesn’t mean it can abuse it.
The government is not one entity and it is perfectly possible for one sub-entity to have certain data and for another not to be able to have that same data, in fact from being prohibited to have access (let alone use) that same data. Palantir is used as a way to gain access to data that should otherwise not be accessible and the fact that it isn't the health data itself is immaterial: it was collected in the process of providing healthcare and as such should be protected. That's the legal base, not to enforce immigration law. Unfortunately the USA does not have the equivalent of a GDPR (and even if it did it would have probably been killed by now).
> government is not one entity and it is perfectly possible for one sub-entity to have certain data and for another not to be able to have that same data
Sure. Again, we agree. What I’m pushing back on is the notion that it was inappropriate for any branch of the government to have these data, or that any of this has anything to do with private dragnets.
They’re addresses. This isn’t a possession problem, it’s one of access.
> the USA does not have the equivalent of a GDPR
You could have super GDPR that bans all private dragnets and HHS would still have home addresses. This is a Privacy Act and HIPAA problem.
Honestly so is the USA. Yes there is racism, more than I used to think, but it’s below average.
Not being racist is really a pretty modern idea that didn’t become popular anywhere until the last 50 years. For most of human history it was conventional wisdom that (whatever I am) is the obviously superior form of human. Statistically this is probably still what most living people think.
I mean of course whatever you are is the master race. Isn’t it obvious?
Linux is becoming more and more viable for a gaming PC. For business uses a Linux desktop is usable but probably not ideal, but you also have macOS. I'd pick anything but Windows and MS stuff.
reply