Amazon’s Disturbing Plan to Add Face Surveillance to Your Front Door
aclu.org
aclu.org
After all, a good old security guard also does 'face recognition' of everyone passing by, but with a human there are obvious limits and a person to confront, charge or circumvent. With ubiquitous, tiny robots all over the place, society will get sharply stratified into the 'watched' and the 'watchers' and the latter will get access to nearly limitless data and power. That's not good.
Don't kid yourself, even the 'watchers' will be 'watched'. The people bringing all these systems together will not be your local yocal Barney Fifes. They will be professional and highly intelligent technical teams whose members have likely studied little other than security and intelligence gathering for their entire professional lives. These are not the type of teams that carelessly leave obvious security holes like that laying around.
Taken as a whole, the technologies coming out today will enable surveillance regimes the likes of which even the UK and China never previously would have dared to dream up.
thay's really important. Hitler, Stalin, the Stasi and many others would have been very pleased to have the systems we are building right now. Once a full country surveillance system gets in the hand of a ruthless dictator it will be very hard to get rid of that dictator.
People seem to not understand the asymmetry in difficulty between offence and defense. For something to be broken, there just needs to be one fuckup, or a few tiny fuckups that can be linked together. Smart people are not infallible. Everything is exploited eventually. The only foolproof thing to do is not to build the infrastructure in the first place.
But yes, I agree with your last point. There's no privacy anymore.
The more likely scenario is that this line of work would likely attract creepy antisocial perverts who like to watch, and love to exert authority -- essentially the people we would like to keep as far as possible from this kind of technology.
Yes, but the essential difference is that the good old security guard isn't photographing everyone passing by and putting those pictures into a database.
It wouldn’t be as concerning if it was only being stored in a local sqlite db.
I think facial recognition door systems are a fine idea if they can be implemented like a doorman: local storage only, carefully trained on local data. With current technology that would be impossible or extremely expensive, though.
I agree, with the addition that all actual processing is done locally, that would be fine. But that's not what this patent is talking about.
The difference between individual human watchers doing surveillance, face recognition, etc. and at-scale watchers doing the same thing is key.
The difference is no less than the old corner stores every few blocks vs Amazon's automated warehouses & shipping.
An ordinary drive with a bit of speeding here and there everyday goes from perhaps a ticket or two every decade, to 6 tickets every trip.
A meaningless wrong turn, I'm lost, let's look around and try to figure out where I am, turns from a non-event to a Suspicious Activity Report...
Just because the laws are word-for-word unchanged, the scaling of enforcement technology transforms the identical laws into a completely different society and experience (just as other scaling technology transforms our experience in a positive way).
This phenomena, as unintended consequences ("just trying to make the neighborhood safer...", but scaled 10^7), is perhaps as dangerous as Bostrom's AI warnings.
Can it be that the best way of avoiding that is avoiding speeding? It's interesting that you are lumping together objectively dangerous, even if prevalent in the US, public behaviour (deliberate speeding) with honest mistakes (making a wrong turn).
I regularly drive on highways with relatively dense traffic all going 10-20mph above the speed limit; in these conditions, it is objectively MORE dangerous to attempt to go the speed limit. I've seen the whole pack drive right by a state cop at 80+, and he'll do nothing because just to turn on his lightshow will likely cause a major wreck.
Outside of residential/school/congested areas with non-vehicular traffic & obstacles, it is NOT any numerical speed that causes issues, but the differential in speeds. Going too much faster -or too much slower- than surrounding traffic is what will cause accident situations (obviously, up to the limits of conditions, car preparation, driver skill & attention, etc.)
The Autobahns are regulated as an excellent example of this. In rural areas very high speeds are permitted, and lane discipline is rigorously enforced. You'll get busted way harder for driving slow in the left lane than for driving 300kph. And, in the more congested zones, the speed limits come down. And the safety records exceed those of the 65MPH US highways.
Speed does not kill, but rather driver ignorance does. A safe fast driver is better than an unsafe slow driver any day. And most US drivers are never taught even the basics of how to actually drive. I know otherwise very intelligent people who balked at taking their teens to driver safety & skills classes taught by SCCA and others (which objectively reduce accident rates) because they thought that teaching these skills (car control, balance, braking & turning at the limits, etc.) would give them wrong ideas...
The fix is simple: just make everyone observe speed limit with a bunch of speed cameras. In the meantime, fill the state coffers with fines. If people think the speed limit is too low, put pressure on politicians, make them run studies proving that it's safe to go 10-20mph more on that stretch and rise the speed limit if it is.
> it is NOT any numerical speed that causes issues
It absolutely is and you're just rationalizing. Kinetic energy rises quadratically with speed, and so does the risk of death and stopping distance.
> A safe fast driver is better than an unsafe slow driver any day.
That's a strowman. The point isn't that just you must go slower, it's that everyone must go slower and observe the actual speed limit. Speed limits aren't imposed just because, they are supposed to take into account tons of data that an average motorist can't possibly have in their mind. If people disregard and ignore them as "too low" instead of observing and fixing when needed, it just slowly corrodes civil society.
Not many people can afford their own good human security guard. Not everyone lives in areas where the police department is properly funded either.
This has to be the worst part of it, if you've ever been part of a suburban neighborhood that uses NextDoor as a community tool you'll know that Ring doorbell videos of "others" crank the pearl-clutching paranoia of bored homeowners up to 11. The comment sections will go from "Did you get a license plate?" to "Hurry, add his face to the database before he comes back to rob and murder us all!"
I was across town trying to get back to the highway, when I get confused in this residential neighborhood and end up on a dead end street. (Why so many dead ends?) As I get ready to turn back onto Pine, I notice this guy run out and hop in his car and start tailgating me. No way am I going home now, with this weirdo following me and probably wanting to steal my stuff. So I hit the gas, and duck down some side street hoping it’s not another dead end, but he stays on me. Luckily, I got a picture of him from the side mirror. If anyone sees this car CALL THE POLICE!
[ Thanked by 15 people ]
I jokingly wonder how many people I freaked out on nextdoor
But if I have video footage of someone approaching my door, not knocking and just looking aroud to then leave; and I'm given the option to track if that person shows up again in future video footage, you bet I would use that option.
This is not because I'm paranoic and I think right away that anyone showing up to my door will murder or rob me, but because said person already exhibited suspicious activity around my private property.
Why would I take for granted that this person was simply lost or innocently looking around? If that's the case, then my camera would never see them again. But if they show up a second time, I want to know that.
This kind of alerting is a bit like any of the automated alarms we have in IT (be it Nagios, CloudWatch, or whatever). They're necessary to tell us when things have gone wrong if they go off too frequently then they'll just get ignored.
In fact there was a parable about this as well, called "The boy who cried wolf".
Even if statistically the likelihood is low, I wouldn't say the fears of what this technology enables and could result in for someone simply being in the wrong neighborhood 'while black' are overly complicated or historically irrelevant/unfounded for certain areas of the US.
https://www.nytimes.com/2018/05/08/us/airbnb-black-women-pol...
Given this is an automatic sensor, people would just quickly just ignore it without even checking the footage - this even before any racial or social prejudices had chance to kick in.
We see this all the time with software monitoring solutions in IT what sysadmins ignore if they are too sensitive so my comment here isn’t without precedence either.
But essentially we are both arguing that an overly sensitive alarm is a bad thing - just from different perspectives. You make the point about discrimination and I make the point about the alerting being worthless in a sea of ignored alarms. Ultimately we’re both skeptical
You might be. I'm not. I never once said it's a bad thing, or described it as overly sensitive, to be fair. Just that I could see how certain communities might look at the technology and what it enables and have good reason to be incredulous about how certain neighborhood actors will use it.
That's about a nuanced as I can possibly get on the topic without trying to argue in favor or against these devices in full, or taking any absolute position.
Ring: Please provide a valid email address to continue.
Me: Turns to leave
Ring: Sorry we missed you! Please state your reason for visit and someone will reach out to you. Please provide a valid email address to continue.
Ring: Warning! This entry attempt have been recorded. An unauthorized entry attempt will be reported to the cloud and local law enforcment.
Where is the issue here? That we provide suspicious activity reports, or that our police SEEMINGLY can't reliably deal with those reports?
Feeding false positives to law enforcement causes them not to work what is important or to redirect resources to filtering reports.
The first thing that comes to my mind is that people who use these systems would be causing innocent others to be included in Big Data databases without their knowledge or consent. I think that it's not only immoral to do so, it is straight-up antisocial and harmful to community and society.
We have more than enough of that sort of thing already.
considering that it is easy to surf news sites and find video captured by home surveillance systems showing people taking packages, trying to get in, or having broke in, it is not unreasonable to install such a device.
the issue is that the data would be available to the police without the homeowners permission. people coming to the door have no expectation of privacy on property they do not own and if need be a simple sign on the property can declare as such.
>considering that it is easy to surf news sites and find video
This is confirmation bias in action. You need statistics on your residential area to make decisions around home security.
It is trivial to find videos of people committing crimes, but that does not imply anything about its likelihood. Meanwhile the state has an excellent record of abuse of power for ends that are unrelated to the reasons given for the power. See: the use of “terrorism” to justify militarizing domestic law enforcement, or parallel reconstruction.
The issue is that this idea that centralization of information will somehow avoid the pattern of abuse of centralized power in the hands of people with perverse incentives.
This video stream should be considered private property IMHO which requires a warrant like anything else.
It doesn't matter if it's an IP address scanning your network then hitting one system, then logging in and scooping up all your data versus a human staring at your house for a few days then walking up while you're not home and stealing your TV. You should know if someone is scoping out your perimeter in preparation for a breach, and post-breach you should know the history of the attack.
I know nothing about Next Door but from reading the comments it seems like the problem is having this information in untrained hands.
I'd go with 'undisciplined' hands versus 'untrained' hands, while admitting I'm unsure of the best way to articulate the difference in a manner that is empathetic towards the desires of residents who just want to feel like they're safe on their block but also addresses a certain pragmatic reality of what happens when 'good intentions' results in a patrol car pulling up next to you in your own neighborhood.
The reasonable expectation of your neighbors would be to assume they're normal people like you, not necessarily that they be trained in surveillance, criminal profiling or tracking persons of interest but instead disciplined in their vigilance as thoughtful and watchful community members who act when there is something demonstrably out of the norm that could affect the neighborhood.
Again, not sure how I can best articulate that thought, or even what steps a community has to take to avoid everyone "narcing" on everyone else for boring, mundane behavior (say, like for instance, retreading your steps through the neighborhood after you lost your phone or some other item the night before) while still remaining active enough to do the right thing when you see someone you know isn't your neighbor picking up packages off someone's doorstep.
This sort of thing is one of the reasons why I fear these systems. A world where this is common is a world where it's no longer safe to take a stroll through your own neighborhood.
I also wouldn't want to live in that world. There are many problems with it, but not being safe for "a stroll through your own neighborhood" is not a problem for that world.
I think that depends on what you consider "safe". Or perhaps I misunderstood what you were saying...
The minute you cross onto my property you're going to show up on multiple cameras that are pointed right at the boundary line. You rights to do whatever you want stop when you cross that line.
I dont think you could convince me that this makes you any less safe.
That's obviously fine for you, the property owner, camera operator, and a very reasonable person. But, imagine every house working that way as you walk around your neighborhood. Is everyone as reasonable as you?
I'm not trying to, because when it comes to my own security, what you think is or is not safe isn't relevant. Just as what I think isn't relevant to yours.
Im working for full 100% coverage of the entirety of my house, inside and out. The goal being able to control my smart house via gestures anywhere I am on the property via some machine learning and computer vision.
Also WYZE cams are 20 bucks a pop, and easily reflashed to do you own bidding. So it's not like this is some super expensive undertaking.
By the way, I have had my home pretty thoroughly automated for about 20 years now. I do recommend that (as long as you aren't using third party services -- that's a security risk all by itself). It's a lot of fun, not incredibly expensive, and pretty useful.
Maybe not but if the thieves are recognized automatically the police have a much better chance at recovery.
- put pervasive face recog databases on the big majority of the US population in hands of huge multinational companies (because America) so an opulent Joe The Programmer can feel a little extra safety over his stuff, because Those Scary Plebs can steal their Juicero juicer
- work on societal problems, reduce the inequality to reduce the risk of robbing and accept whatever risk is left because the societal cost of pervasive corp surveillance is just too high
The fact that sizable number of people feel that the former in any way preferable is pretty damning.
Most people would use that option if they had the opportunity; tracking potential troublemakers has obvious benefits. The problem is you are not being given the option "to track [a] person". You're being given the option to record video and apply an automated tool to identify "likely faces" and as way to compare those images against a local database using a definition of similarity that is poorly specified[1], extremely complex, and opaque/unexplained. Jumping right thinking of this as "tracking a person" ignores the many layers of metaphor shear between the desired goal and what the actual implementation does.
Metaphor shear is what you experience when you realize that interface doesn't match what is actually happening. The metaphor was lost somewhere in the hundreds of layers of abstraction connecting input to output. The map is not the territory. Efficient and powerful tools are very useful, but they also let you efficiently shoot yourself in the foot in powerful new ways.
Since metaphor shear is about the error between an expected ideal and actual measured reality, it compounds across layers of abstraction similar to how experimental error loses precision with each calculation. Any particular layer of abstraction is probably fine in isolation, but the interface metaphor loses significant digits as more and more layers are added.
If intellectual labor is going to be surrendered to a device, we need to be very clear and specific in what we are surrendering and what the device is expected to do.
(Video essayist Kyle Kallgren does a much better job of explaining[2] the problem of metaphor shear, using examples from sci-fi [Black Mirror, HHGttG, R.U.R.])
[1] poor in the context of trying to track someone. I'm sure the face matching algorithm is very well specified with respect to matching images of faces. The poorly specified part is when "matching images of faces" becomes "tracking someone" (what does "tracking" mean? How are probabilities about matching faces mapped to identifying "that person"?)
If you or some grandma install a cloud system that auto sends my info to multiple third parties I would think that is worse than creepy.
I could see grubhub delivery people end up at wrong adresses, a system detecting unusual arrival at 9pm, send alert, the system pulls other data about this person, combines it. This is shared with multiple groups of people. Now humans look through it and see her going to the sex toy store earlier in the day and the abortion clinic and the (insert non-popular religious place here) - and whatever.
All it takes is one person in multiple groups to get heated about this person's lifetsyle or 'suspicious activity' to put out a pull over and find some reason to get more info, write a ticket, fine, get them to rack up other issues.
In an extreme situation you end up one of the cog wheels that started a Sandra Bland situation because a suspicions wrong address was given via GPS to a delivery driver.
Even without wrongful deaths, having people look into your life like that and make jokes about you or whatever is pretty lame.
At least when this happens with Msoft's one drive, one could say that someone agreed to the terms of lack of privacy with the corp. Not that I would buy that as 100% true in 100% of most cases, but at least then it could be argued.
https://www.washingtonpost.com/local/social-issues/the-secre...
Quoted from the below link:
>"The cameras are equipped with artificial intelligence (AI). They can recognise people, who are among a crowd, by identifying faces. To ensure safety, the photographs of wanted people will be uploaded on the system. Whenever a person who is in the wanted list appears before the camera it will detect the person and send popup notifications to the three control rooms at Sabarimala," said IG Manoj Abraham, Cyberdome nodal officer.
http://www.newindianexpress.com/states/kerala/2018/oct/14/no...
While I dont know the specifics, police used this system to arrest 200 known protestors by preparing a database of 1500 people and had the system go through over 800 hours of footage.
If that is not enough, now they are planning to use drones with facial recognition as well.
http://www.newindianexpress.com/specials/2018/sep/17/kerala-...
I thought the plans to use face recognition in airports was scary, but they are taking it to an entirely different level altogether.
But what was happening on the ground was that the NYPD was identifying who potential leaders were by using face recognition technology and (we believed) cell phone tracking to identify people who consistently showed up to different meetings.
They would then use the same face identification technology to pick these people out of crowds, and arrest them before or when major events occurred, so they would be in holding during the event.
Eventually the individuals would either be charged for resisting arrest, or let go since they were arrested without reason. Which dovetails into the other NYPD/LAPD tactic during mass protests of , in their words: "arrest people now, end the protest now, pay the civil liberties fines 10 years later."
If this is true, and even if you (or someone) made it up but it is a practical possibility, then it has truly horrific consequences for social organization and the peaceful transition of power (democracy).
I've written about that before: https://news.ycombinator.com/item?id=15110645
This is bad.
If you take away more and more ways to (relatively) peacefully protest something (for example by arresting the leaders as you say). Then the frustration will just continue, and grow and grow until the only way is violence, say what Marx had in mind.
https://www.theverge.com/2017/3/22/15016984/nypd-video-surve...
> Similarly, FBI documents confirmed "a strategic partnership between the FBI, the Department of Homeland Security and the private sector" designed to produce intelligence on behalf of "the corporate security community." A PCJF spokesperson remarked that the documents show "federal agencies functioning as a de facto intelligence arm of Wall Street and Corporate America."
https://www.theguardian.com/environment/earth-insight/2013/j...
> then it has truly horrific consequences for social organization and the peaceful transition of power (democracy).
Snowden tried to tell everybody that. I recommend reading the actual documents in [2]; they paint a very clear picture about the tactics being used.
[1] https://en.wikipedia.org/wiki/COINTELPRO#Intended_effects
I think you're right, for the exact reason that you stipulated in your other post. I also think that our society is embracing technology in a way that makes this an increasingly difficult problem going forward. I don't want to put Kaczynski on a pedestal, as I significantly disagree with both his actions, and the conclusions he came to, but I think he and others were right to identify this as _the_ issue for western society going forward, when combined with wealth inequality.
If something so horrible (knowingly killing babies) can happen so habitually and be concealed for decades, then who the hell cares about freedom of speech or your civil rights?!? Seriously, wake the fk up!
> It’s not that Occupy dissolved, but you can only create a movement for direct democracy if you can get everybody out in some kind of public place. They have to be safe enough to go there. So if going to an Occupy march means risking getting beaten up with stick, or being thrown into prison, then people with children, old people, they’re just not going come. And then only the hardcore activists come. It’s that simple.
https://www.eurozine.com/anarchism-work-and-bureaucracy/
And then the society that lets that happen picks out the worst examples of misbehavior by hardcore activists, calls itself noble in comparison, and goes back to sleep.
The FBI was standing ready so assassinate all the leaders of Occupy "if deemed necessary". See my comments to other posters in this thread and the links I provided. Once the news leaked that snipers were in place even the hardcore Occupy members decided to fold their tent (literally). Most people can probably deal with a few beatings, some will even sue for brutality and win decent awards, and maybe get some police scum fired. But suing is not an option if you get killed, and most people (rightly) draw the line there and decided a protest is not worth it. The society that lets that happen is a gang of drooling imbeciles, zombified by decades of SSRI and anti-anxity therapy, psyops by media and government, and kept in constant fear by being on the brink of financial ruin 24x7. How can you expect people to care about political protests if most of them worry to death about the next day, every day?? https://news.ycombinator.com/item?id=18682331 https://www.cnbc.com/2017/08/24/most-americans-live-paycheck...
Summary of the second link above, if you don't read it - SEVENTY EIGHT percent of working Americans live paycheck-to-paycheck! You can only imagine the financial situation of "non-working" Americans. So, do you think in that state anybody would care the slightest bit about a political protest?? The 99% are nothing but slaves and the 1% own us. Any other discussion is empty talk...or a psyop.
The FBI refused to release documents or identify the group that had snipers tracking Occupy leaders in Houston but if you read the redacted documents it becomes quite clear it was a government agency. The redacted documents contain lines that basically said the sniper "group" would take out any Occupy leaders "if deemed necessary". That latter language is used to describe government agency activities, not some renegade alt-right group who decided to take out Occupy by force. If the FIB has knowledge of ANY criminal group ready to take out people on-demand "if deemed necessary" there would be quite a few arrests, even if no murders took place. The fact that FBI arrested nobody in that plot speaks volumes about who the snipers were. Either the FBI itself, the Houston SWAT team, or some 3-letter agency that officially does not exist.
Those same forces that were ready to take out Occupy leaders are the same forces that killed MLK. Most people, even African Americans, do not know that MLK's biggest threat to the established order was not his campaign for racial equality. It was his campaign for economic equality. As soon as he made the issue class instead of race, he was killed. MLK planned something called "Poor People's Campaign" and envisioned a million-people-strong demonstration in DC with the goal of non-violently interfering with government business and "occupying" the capital until the demands of the poor people are heard. Does it sound eerily similar to the Occupy group? Where do you think the Occupy leaders got the idea from? It was a re-incarnation of the MLK campaign from the start. And just like back in MLK's days, the government stood ready to murder the inconvenient leaders. Thankfully, it did not get to that point but it is quite obvious that individual rights and the law in general in the Western capitalist "democracies" are long gone.
https://www.huffingtonpost.com/obery-m-hendricks-jr-phd/why-... "...Yet in December, 1999, after hearing the testimony of over 70 witnesses, including the owner of a restaurant close to the murder scene who admitted his complicity in the plot to kill King, an interracial jury in Shelby, Tennessee unanimously concluded that King’s murder was the result of a conspiracy of unnamed “GOVERNMENTAL AGENCIES.”
It is way overdue that we ALL wake up and realize the truth - we ALL live in an international fascist dictatorship that knows no boundaries and obeys no laws. The laws are just for us, to keep us from disturbing the status quo. And no matter how hard organizations like ACLU et. al fight for the rights of the common person, it is a lost cause from the start as these secret entities do NOT follow laws. They cannot be reformed. President Kennedy said the same thing about the CIA (that it cannot be reformed) and he was killed a month later. I think it is rather obvious where this whole process is going.
Unfortunately, history repeats itself, even for the people who learn from it.
FR is like having a human lookout who is really good with faces.
Beyond that, it is all humans using this capability for good or bad purposes.
Nevertheless, is as scary as you think...
With all that said, what did you think would happen when people openly started allowing listening/video devices on the inside/outside of their homes? Of course people (law or others) are eventually going to want a way to utilize that data if it's useful to them. The same way that I'm sure Facebook/Twitter/etc. work with law enforcement with their data.
Amazon filed this patent now, instead of earlier, purely for the creation of "outrage articles"? Can you walk me through the logic here?
The news is successful in scaring us, but what can we do against this type of thing? What can I as an individual do? I can install ad-blockers, etc on my PC, but how can I prevent the technology described in the article from impacting my life or that of my children? I don't want them to grow up in a 1984 or Brave New World scenario. What can I do today to stop from, at the very least, my country from going down a path where this is considered normal?
* Push your politicians to support GDPR-style legislation
* Donate to your politicians so they actually listen
* Run for office
* Try some weird tech solution like dazzle clothing (this won't work most likely)
* Take inspiration from Exit, Voice, and Loyalty, and emigrate to a place with governance that more closely reflects your own values. A place covered by the GDPR would be a good start. _Presumably_ if your country loses enough talented people it will work to retain them. But, that's a pretty long-term approach and I have doubts about it's true efficacy (My homeland is hardly building great bike infrastructure since I left, and that was one of the biggest motivations)
* (Also pretty extreme and of dubious efficacy long-term) live off-grid somewhere remote and not walk by people's houses.
Probably a lost cause, though. I find it infuriating that my voice is being listened to by Alexa, etc. when I'm in other people's homes, but they don't seem to care. Maybe civilization will collapse and render this a moot point.
I don't know? I think idea number 2 from your list would likely be extremely effective if you had a large enough amount of money so that you could give large amounts to a large number of politicians.
But, yeah, not many people out there with that kind of resource. For the majority of us, resisting these sorts of technologies and tracking is a lost cause.
> Push your politicians to support GDPR-style legislation
Luckily I live in the EU, only 'luckily' might be an overstatement with regards to privacy if you look at the article 13 issue. Apart from that, at least to what I perceive, privacy carries at least some meaning here still.
> Donate to your politicians so they actually listen
I do not have enough money to make politicians listen I think.
> Run for office
This seems like one of the only conceivable routes to take that could have some real impact. I don't want to run for some political office, but at some point I might have to.
Correct me if I am wrong, but I think Nietzsche thought that we shouldn't give power that those who want it. Right now there are a lot of people in power who want that political power and use that to advance themselves in life, rather than helping society. Not only in politics however; Amazon's developers who work on these intrusive technologies are just as guilty. I would not work on this type of technology because I can't justify this to myself as a person, but apparently many people leave the person they are at the door when they go to work, developers and politicians alike.
> ...emigrate...
I could consider this at some stage, but that would be to protect myself and my family against imaginable changes to political situations with regards to privacy, not as much as a way to force government to change their views.
> ... live off-grid
This sounds like a great way to draw attention to yourself. Besides, I don't want to live in the woods, I want to live in society without my privacy being violated constantly by advertisers, other companies and government. Because everybody has a right to their thoughts staying their own.
Yeah, it just means you won't have to walk by the doorbells. I wasn't really serious about it. And between cheap drones, IR cameras, etc. if somebody wants to find you, they will.
At this point I think your only real hope is finding a place that aligns with your values, moving there, and working to keep it good. Effecting real change is 99% luck and places heading towards dystopia have a lot of wealthy forces arrayed against you.
Could this bring about a police state? Yes absolutely. But the technology is possible to build, so someone will build it, or at least patent it. There's no preventing the technology from existing.
We don't prevent a police state from happening by freaking out about the technology that enables it. We prevent it by voting for responsible leaders. So let's just assume that the technology for government oppression is going to exist and make sure we structure our government in such a way that it doesn't oppress us.
"Amazon PeopleTech’s mission is to re-invent ways for our employees to connect with one another. You will work closely with our team of engineers in designing a strategy for new projects like our new Facial Recognition technology for Amazon’s Time and Attendance systems"
I love connecting with people!
My point is, I don't like it, and this is actually where we're heading with this crap. The same tracking we currently have online, but in the real world.
1. No question this raises concerns about the system pinging authorities as someone who has been flagged in the system is detected.
2. This will sell like hotcakes with people who are on NextDoor. Everyday there is an alert about a burglary, stolen packages and people snooping around their properties. These people are very likely to approve sending any images/video of interest to police. (Right now it's very mechanical, scroll through video, capture and send --people will put out requests for footage very often).
I think the only thing we can do is call for strict regulation on its use. It's gonna get deployed and it's gonna get used.
I could imagine selling a service that multiplexes video cameras and only sends Amazon devices live video when there is really motion and not just when they say there is motion. Meaning, put the security control back on the home owner.
I have always felt it would be very useful to be able to identify people in videos who may have committed a crime locally or elsewhere in the city, to learn of their movements, where they might live, what other crimes they committed, etc. This would help police prioritize some non-violent crime effectively, get a high return on the time they spend investigating property crimes (actually catching criminals), which would then in turn serve as a deterrent that is outright missing right now. I would gladly contribute my video to that cause, as would almost everyone in my neighborhood.
A word on the commenters here going on about 'pearl-clutching': that's just the latest in pejoratives that are nothing but empty stereotypes. Most people on NextDoor in my city are reasonable, but are regularly unfairly characterized by the far-left progressive cabal in this manner. Caring about the safety of yourself, your loved ones, and your property is not a bad thing. We work and live in part for those very things. If you disagree, then try outlawing door locks. It is unbelievable to me that the rights of criminals are more important to commenters here than the rights of those who want to protect their own. And worse yet, the conspiracy-theory notion that we are about to be taken over by a tyrannical government because of doorbell recordings. If this was a topic they disagreed with, those same commenters would readily call this a "slippery slope fallacy".
A further aside: The ACLU has been wrong a lot on this topic - for example when they called AWS's Rekognition service a "facial recognition service" instead of what it is, a generic object recognition service. They also did not seem to understand how false positives/negatives are managed and why it is fine to use an imperfect object recognition service as a first-pass filter. And they are wrong again here - people are allowed to film on public property and share it with whomever they want. They're allowed to perform whatever analysis they want and share it with whomever they want, period.
Often a company is motivated to increase their IP portfolio. In turn they motivate their engineers by offering lucrative bonuses for each patent filed. At my last company we played this game and it was roughly $3k per patent. In general, out of the huge number of patents that get filed in a big company, very few---if any---have any hope of ever seeing the light of day as an actual invention.
While I agree that the technology itself is somewhat alarming and I personally disagree with the idea of anyone building it, let's maybe save the outrage until they announce that they are actually going to build this thing and deploy it in the real world?
Perhaps enough outrage now would prevent it from being deployed, or would get them to engage in at least a tiny amount of mitigation of the worst aspects of it.
By that time someone will make "the train had sailed" comment. I.e. claim that nothing can be done, because consumers had spoken and the technology is already being deployed en mass.
https://motherboard.vice.com/en_us/article/evkgpw/smart-door...
[1] https://www.reddit.com/r/Damnthatsinteresting/comments/9ys1e...