I think the person requesting to access the data was doing the right thing and I agree with the judge’s ruling.
The fact that they’re gonna shut it down, implies the scale of indiscriminate nature of data capture and the volume of data being captured.
These cameras are popping up all over the nation and if people realize how much data is being captured and where that data is going (or who it’s being sold to) and how it’s being used by government and private entities they would be appalled.
There’s been exposés about these cameras, everything from AI misidentification of “stolen” (not) vehicles and erroneous arrests and police encounters, to analysis of shopping patterns being sold back to private entities for better ad targeting. It’s wild.
The laws need to be updated. CCTV in public used to be fine because no one was actually watching it unless there was an incident. Now it’s possible to have AI watch every camera and correlate everything everywhere we need new privacy laws to reflect this capability.
I don't mind a local AI on an airgapped security camera network monitoring a camera and issuing an alert to a security guard. The issues are internet connectivity, data retention/mining/sale, and non-local processing (ie handing stuff off to a third party that does who knows what and probably doesn't take security seriously).
Just as two trivial examples, even though neither affects me personally:
The estimated number of heroin users in the UK exceeds the total prison population. The number of class-A drug users in the UK is estimated to be so high that if they actually followed the minimum sentencing guidelines for possession, it would cause a catastrophic economic disaster both from all the people no longer working and also all of the people who suddenly had to build new prisons to hold them. I'm not interested in drugs (and I don't live in the UK, but I assume the UK isn't abnormal in this regard).
Another example is road traffic law. Even just speeding offences, I think you probably catch everyone who actually drives in the UK, often enough that after a month the only people left allowed to drive would be people like me who don't even own a car.
The entire legal system has to be radically changed with far less punishments for almost everything if you have perfect, or even 30% of the way to perfect, surveillance.
> Another example is road traffic law. Even just speeding offences, I think you probably catch everyone who actually drives in the UK, often enough that after a month the only people left allowed to drive would be people like me who don't even own a car.
You don't have to strip the driving licenses. You should impose a fine and not an extremely painful one for starters.
And then probably within less than a year the whole population will drive properly.
I think I'm in favor of indiscriminately fining everyone speeding at every camera, but I realize there is no privacy-preserving way to do it today thus I will be against it.
(I'm a driver and car lover who is never speeding)
That's covered by "system has to be radically changed". UK driving licences give you room for 12 "penalty points" worth of mistakes before you risk being banned from the road, of which speeding costs you at least 3: https://www.gov.uk/penalty-points-endorsements/endorsement-c...
4 strikes today with current speed enforcement is fine; 4 strikes in a hypothetical where 100% of infractions are caught, I aver will catch (almost) everyone who actually drives within a month.
Where I live almost every main road now has average speed cameras, leaving only the small residential streets without, and they are generally short enough that few people are speeding anyway. In general I approve, especially in residential areas, although the surrounding area has gone through a process of downgrading speed limits for no obvious reason, such that it seems their only intent is to annoy drivers so that they want to stop driving. Almost all the country roads in the county I live adjacent to were downgraded from national speed limit to 50mph about 10 years ago. It didn't seem to be anything to do with safety, just seemingly out of spite. I heard rumours that they were also trialing drones to spot offenders (and presumably with sufficiently good cameras to read the plate from up high). Recently, many residential places have dropped from 30mph to 20mph pretty uniformly, again seeming nothing to do with safety as it's entire suburbs (but not all suburbs in the city), even on long straight roads with excellent visibility that would have been safe with limits above the original 30. Doing 20 on these seems completely unnecessary and it's hard to see it as anything other than a revenue generation tactic.
If notice of the infractions is a envelope delivered days later and sitting on on end table for a week, typical drivers are going to rack up 4 strikes before seeing the first one!
Perfect enforcement has to come with immediate feedback
I dunno. We got car insurance once that had a "put this spy device in your car for a couple weeks for a lower rate" deal and I felt like I was driving a lot less-safely when I was constantly worried about looking like I was driving safely.
Like, to pick an example that's specifically speed-limit related, if more people really tried hard to do 25MPH (the marked speed limit on many of them) or under for the entire length of an interstate off-ramp, I think we'd be spending more money on brake pads and there'd be a lot more cars getting rear-ended. Sticking to that speed the entire length is silly and not very safe, and things work just fine without people doing that. Tons of other edge cases like that where you're technically breaking some law or another for a little while, but things work way better if you do. Plus practically every one of these laws has some kind of judgement-call clause that applies to modify it, and I don't want the people making those judgement calls to know that if they do what seems right to them in the moment, there's a very high likelihood they'll be hassled for it.
I've never seen an interstate off-ramp in the US with a 25MPH speed limit (white sign). I've seen 25MPH advisories (yellow sign), but those aren't a legal limit. Advisory signs are the maximum safe speed for the worst possible conditions (road covered in ice).
Interstate, no, and actual speed limit, no, but the one I always think of is this, on US 101, near Steamboat Island in Olympia where you come off a 60mph two lane state highway to an off-ramp that is 430' long and has an advisory 10mph speed for a very unforgiving 90/180 degree turn (160' diameter circle) - with the ability to rear-end people joining the highway if you miss the first part of the turn or the ability to t-bone people if you miss the second. ChatGPT describes the braking required as being at the high end of moderate, almost hard braking, for a sustained 9 seconds to slow sufficiently to make the turn, assuming you start braking the moment you are in the exit lane (and there's no advanced warning, just a single "10mph" yellow sign as you enter that lane): https://imgur.com/a/uns86kp
If speeding is not something that happens constantly, then a radar could detect the instances of speeding, and only turn on a camera when a speeding car is nearby. This would keep the majority of passing cars from being recorded, and would record the fewer cars the fewer drivers would be speeding.
How long until the AI estimates how fast you were going based on the time you were tagged at 2 cameras? The system says travel between these cameras should take 3 minutes. You made it in 2:45. No review, just a ticket in the mail.
That's how it actually works in some places (https://www.carwow.co.uk/blog/average-speed-cameras-how-do-t...). You don't need AI because we already know that distance = speed * time. We can calculate to a high degree of certainty - with high school math - that you had to have 1. been speeding or 2. bent space time if you cover the distance between them in too short of time.
Several months later, the state changes the speed limit on the street without changing the ticket check. Some time after, traffic authorities add a bypass road or a fast lane to deal with the new traffic problems and forget to update the timings. Then it's reported that a rural municipality set timing thresholds just below the speed limit to cover budgetary shortfalls.
Eventually the camera manufacturers offer "easy setup" systems that dynamically adjust based on actual journey times. This works fine until the first office holiday, when the routine congestion is gone.
If their clocks are actually synchronized and can show that in court then unless you've got a warp drive and can bend space time it can be pretty definitive that one was speeding between those two points. Especially if you're on something like a toll road which utilizes transponders for billing. I'm almost surprised this isn't already done in those situations.
You don't even need AI for this. Its pretty basic.
It is similar to how air enforced speed limits are done. They just paint two lines on the highway. A plane overhead times your car between the lines.
The point he’s making that people violate the letter of the law in many, many small ways, and to prosecute people for all of them would be a crippling burden on both individuals and the economy.
Yeah it's a bit of wishful thinking from my side, I confess.
> exactly why we end up with surveillance states
Hate that. This is why I wrote that it looks like there is no privacy-preserving way to trap ALL speeding drivers, although I have been corrected on that specific part.
What I wanted to say is that I would love for road infractions to be fined all the time so that people would know that there is no free ride.
I'm going to argue the other side: in Chinese cities like Chongqing they've seen a drastic reduction in crime after blanketing the city with cameras and monitoring technology.
Whole categories of crime disappeared. Women and elderly feel safe to walk the streets at night. No one locks their bike anymore in Chongqing.
I care about privacy, but I think we should be smart enough to work out a way to get some of those benefits without going full 1984. For example by having surveillance that can only be queried by an AI with very strong guard rails.
Admittedly, I live in a country with very strong democratic institutions, and I trust we would take action the moment something gets abused or surveillance overreaches. I would probably feel differently living elsewhere.
> I trust we would take action the moment something gets abused or surveillance overreaches.
The thing about turnkey authoritarian solutions is that once something happens it's likely too late to take action. However there are often alternative solutions that physically constrain the system such that substantial abuse is impossible without time consuming and expensive physical modifications. The traditional speed cameras in the UK for example.
Cameras, AI integrated at the edge, software that can't be updated remotely, the full stack publicly audited, that only output video data when a suspected violent crime is flagged. Something like that might work. I'm not optimistic such a solution would see much support though.
You'd also probably want a policy put in place in advance to quickly pull them down if certain criteria are met. But again, I'm not optimistic about the prospects.
Well I agree, and my hopes aren't very high of this actually happening. Our politicians tend to be clueless with anything tech related, their opinions calibrated by what they saw in Hollywood movies, where anything tech related always turns into "black mirror". (By contrast, allegedly over half the Chinese politburo has an engineering degree of some kind).
But we could start small, with just one neighborhood, a pilot project where the kinks get worked out and slowly scaled up. Getting permission for a small scale pilot shouldn't be impossible.
Everybody made that exact same "slippery slope to Nazi Germany" argument when euthanasia was legalized here. That was decades ago. There have been several attempts to broaden or narrow the scope of those laws and the democratic institutions did just what they were designed to do, making changes judiciously.
If you are worried about the slippery slope, then you are really worried that democracy does not work as intended. (And depending on where you live that may be a very reasonable worry). By the way, Nazi Germany was not really a surveillance state, perhaps you are thinking of East Germany?
> If you are worried about the slippery slope, then you are really worried that democracy does not work as intended.
Not really. That's a well established failure mode. People's perceptions can gradually shift as they become accustomed to the new way of doing things.
Personally I'd be less concerned about a slippery slope and more concerned about abrupt changes in policy. All infrastructure should be designed with the worst case scenario in mind. It's naive to assume that things will never get worse suddenly or that we will have plenty of warning or even a meaningful opportunity to react.
There's no way to deploy a system like the one you're describing without being abused for authoritarian overreach. It's simply a matter of time, and once it is deployed for authoritarian overreach, the only way back will be paid for in blood.
> Even just speeding offences, I think you probably catch everyone who actually drives in the UK,
So we have a fucktonne of speed cameras allover the place: https://www.speedcameramap.co.uk/ (you need to zoom in there are so fucking many)
But we have less redlight cameras than the US. we also have hatching cameras (yellow hatched boxes mean no stopping, usually at junctions) we also have bus lane cameras, where if you drive in a buslane you get a fine.
For the Speed cameras, they are normally put there based on evidence of road deaths linked to speeding. I dont like speed cameras, but they do serve a purpose.
When you get a speeding ticket, if its your first offence, you can take a speed awareness course, and you won't get points on your license. otherwise its three points and a £100 fine. The points age out after 3 years. the maximum you can normally get is 12 points on your license.
Its only in extreme cases do you get a ban, or license revoked.
The reason why people are still able to drive are numerous:
1) its been a gradual evolution.
2) we have fairly robust training for drivers (theory, comprehensive real world test)
3) Evidence based placement. Its not like they just shove these things where poor people live (or in the US where the city has zoned living for people with more melanin than others). If there are higher than average road crashes, the road is re-made to make it safer, speed limits dropped, traffic calming put in place, then speed cameras.
4) You are expected to follow the traffic rules
5) the traffic rules are actually pretty sensible.
> we also have hatching cameras (yellow hatched boxes mean no stopping, usually at junctions)
Weirdly I've never encountered these in the US (only red light cameras) and do we ever need them. I'm generally opposed to government associated cameras due to concerns about turnkey authoritarianism but if we have to have cameras at intersections they could at least curb the awful self centered behavior.
Most states ban speed cameras and many ban red light cameras as 4th amendment violations. You cannot face your accuser when your accuser is a robot. As a result, speed camera tickets have always used some legal sleight-of-hand to nail you into confessing, and this has become unpalatable.
If we raised speed limits (almost) across the board to the actual safe limits of modern cars, I think a lot more people would be ok with speed cameras. There would still be a constitutional problem, however.
Data from various states raising their limits over the last few decades is that every 5MPH increase in state speed limits brings with it an 8.5% increase in traffic fatalities on freeways. With the 70MPH speed limit common for freeways through most of the US, we're already up 25% on road fatalities over the 55MPH that was chosen for gas mileage during the oil crisis.
For cities, pedestrians struck at 25MPH are already at a 10% chance of death, which reaches 50% at 40MPH.
Nobody goes 70 mph on freeways. They go 80 mph on that road because it's the speed of traffic. If you declared a "speed limit reset" and raised the speed limit to 80 mph, people won't be going 90 mph. They will be going the speed limit.
In the UK 80-90 was quite normal 25 years ago, and off peak you'd find the outside lane of the M40 doing over 90 a fair amount
It's not today as there's far more traffic. It's rare to have the opportunity outside of a few areas (south of Bristol on the M5, north of Kendal on the M6 etc). When I first learned to drive I'd do M62 to M5 in well under an hour, today it's about 20 minutes longer.
> You cannot face your accuser when your accuser is a robot
I don't really get this take though. If one contests the ticket, have it go to a manual review where someone looks at the tape and confirms the calibration history of the equipment, and boom now that person can be the official "accuser".
If its the argument that they might not have been the one actually driving the vehicle, just make the laws relate to the registered owner of the vehicle is ultimately responsible for the safe operation of the vehicle. If the owner wants to try and prove it wasn't them, they can deflect that in court and prove it was actually their friend or whoever.
If we had a CCTV of a murder happening we wouldn't just go "well that's just a robot not a real person making the recording, guess we need to toss that video!" I don't understand how we take that kind of position when it comes to moving violations.
Also FWIW the 4th Amendment has nothing about facing your accuser. Its the sixth amendment that talks about "...to be confronted with the witnesses against him".
> So we have a fucktonne of speed cameras allover the place: https://www.speedcameramap.co.uk/ (you need to zoom in there are so fucking many)
Doesn't seem that many compared to what I was describing. At the scale of a country, "a lot" != "a high %".
Your point 3 is the biggest divergence between them.
Point 5 is only kinda true, the failure mode is weird: there's good reasons why the speed limit isn't enforced until you're significantly over it, but that in turn means it has to be set lower than physics and reaction times dictate, which in turn means people push back against them. 20 zones knowing people will do 25, chosen because if they were 30 zones people would do 35 and 35 is too fast, that kind of thing.
People who know that, let themselves go a bit over the limit; but a bit over means they get caught some of the time because of the same small occasional variations that are the reason why the speed limit isn't enforced at x+1 mph in the first place.
I was a bit unclear. I agree, I don't want the government using AI to identify all violations of the law. That sounds like a very straightforward dystopia.
What I don't mind is private companies using AI analysis to support their security guards. I object to any sharing of the data with third parties though. It should be illegal for the data to leave their internal network and it should be illegal to retain it for more than a few days.
I don't care if grocery store loss prevention has eyes on every aisle. My concern is data warehousing and subsequent misuse.
> The entire legal system has to be radically changed with far less punishments for almost everything if you have perfect, or even 30% of the way to perfect, surveillance.
Prosecutorial discretion means they can just collect evidence and choose not to charge you unless they want to leverage you for something. This already happens, but universal surveillance means it can literally happen to anybody, because everybody breaks the law in some way due to how many laws we have.
Discretion is the real problem I think. It seems extreme, but maybe discretion should be eliminated: if you commit a crime you will be charged. This will at first result in way too much prosecution, which will lead to protests and hopefully repealing laws and we'd end up in a better place where the law is understandable and predictable by mortals.
Speeding is a special case, because it's unclear what the lawmakers, road designers, and police intend. When the speed limit is 65 mph, do they actually intend for everyone to go no faster? I don't think so. I think the lawmakers, if driving in traffic, want people to go a bit faster. Same with the police. And I think the road designers design the roads knowing most people will speed.
I want to follow the law. But when it comes to speeding, it's hard for me to follow the letter of the law, because all the parties involved in creating and enforcing the law don't want me to follow the letter of the law. So I instead follow the intent of the law, and speed up to 9mph. When Google Maps pops up a "police ahead" warning, I don't slow down at all, because I'm following the intent of the law, and that's what police around where I live enforce. If I'm driving in other areas of the country, I'm less certain what police want, so I'll be more likely to follow the letter of the law.
If there was automated strict enforcement of speeding, then it would be clear to me that the letter of the law is the intent, so I would gladly obey the letter of the law. There would certainly need to be a transition period with clear warnings that in the future, the letter of the law will be enforced, instead of the current status of something looser.
Some states follow Assumed Maximum Posted Speed (in certain places) and others are Absolute Maximum Posted Speed. It is not absolute in an Assumed Maximum Posted Speed state that driving faster than the posted speed is against the law and deserving of a fine, merely it is prima facie evidence that you were driving dangerously but can be challenged and overturned. For example, in Minnesota, outside of municipalities on highways (there may a few more qualifiers like posted speed is 55 mph or higher and might need to be a divided highway, I don't remember 100%) an officer can pull you over and issue you a ticket for merely driving faster than the posted speed limit. You can even admit you were driving faster (I don't recommend this). You can still challenge the ticket in court. If you can convince a judge that your speed was safe, the judge can let you off. If the weather is dry, temperature moderate, visibility great, no other people or vehicles around you, you were able to safely slow down, and (prima facie evidence) that you posed no risk as no one was injured by you driving faster. In Wisconsin though they are an Absolute Maximum Posted Speed state so if you are found to have been driving faster than the posted speed limit, that's enough to ensure you can be fined.
Or are you asking (2) how we wound up in this situation as a society?
(1) I think what I think for several reasons. Basically everyone speeds. Probabilistically that includes he very lawmakers writing the laws, the police, and the road designers. I've also read some articles talking about road design, and in it it's mentioned that the designers factor in that most people will speed if the road conditions are amenable. I've also seen police cars driving around without their lights on, passing people at higher than the speed limit, and when unable to pass, the appear annoyed to me.
(2) I think this situation arose in sort of a "normalization of deviance" manner. Police didn't want to be too strict, or didn't want to bother fighting tickets for people speeding only a little, so only gave tickets for people speeding a lot. Then over time many people realized that, and started speeding a little. More are and more people started speed just to fit in with the surrounding traffic, until eventually everyone was speeding. Peer pressure. I've heard driving the same speed as the surrounding traffic is generally safer than driving significantly slower (or faster). Once everyone is speeding, that includes lawmakers, road designers, and police. And they factor that in when they write laws, design roads, and enforce laws.
There were also fairly contiguous moral panics in the late 1960s (teenage boomers crashing muscle cars) 1970s (fuel use) and early 1980s (slightly older boomers driving drunk) that lead to the regulations being written far in excess of what there's popular support to enforce which is a huge contributor to why the enforcers and judiciary are essentially responsible for dialing it back to something that doesn't make the system look stupid.
In Washington state, the State Patrol has gone on the record multiple times and with the media saying their enforcement is based on "Under 10 over", i.e. they won't pull you over for doing 69 or less on the Interstate in a 60 zone, or 79 or less in a 70 (assuming speed is the only issue).
"it's unclear what the lawmakers, road designers, and police intend"
In many cases, there's a gap between the original intention and the current need.
Many speed limits and policies were established in an era of fewer cars, but also much less capable cars with fewer safety features - many speed limits were established before the adoption of ABS, stability control and airbags, and more recent innovations in lane-keeping and adaptive cruise control.
Modern cars may be capable of travelling at greater speeds with greater safety, but there's a more recent recognition of the increase in emissions pollution from increased speed. Speed-limits typically remain grandfathered in at their original value (which may have been set 30, 40, 50 years ago), regardless of the change in context.
Then there are some pecularities such as the UK default of 60mph for a single-track road, but if you were to try that in many rural locations (think Cornwall, Scotland, Wales) you would likely find yourself upside-down in a ditch.
This post highlights the absurdity of some of the limits!
The UK NPCC (National Police Chief's Council) have a published policy where enforcement effectively starts at 10% +2mph over the speed-limit (whilst allowing officers to use individual discretion if they feel there are aggravating factors).
Counterpoint: with mobile devices, and increasingly, control and information features of automobiles themselves, distracted driving is increasingly a concern.
There's also the point that driving capabilities vary wildly by individual, and often decline drastically with age. Recent case-in-point, an elderly driver in San Francisco who killed a family of four (a mother, father and two daughters, waiting at a bus stop, not in the roadway at all), let off with a minimal sentence, raising much public furore:
> Speed-limits typically remain grandfathered in at their original value
That depends on where you are. In Texas, state highway speed limits are determined though a traffic study[1]. The monitor traffic for a while, then set the limit to the 85th percentile.
People can use this to get out of speeding tickets. If you find that it's been a long time since a speed study was done on the road you were on, the judge might throw the ticket out.
There are some hard limits though. For example, the maximum speed limit that can be set on a road is 85 mph.
> UK default of 60mph for a single-track road, but if you were to try that in many rural locations (think Cornwall, Scotland, Wales) you would likely find yourself upside-down in a ditch.
The way the national limit is framed is more limit than road speed. It's interesting how we think of the limits as drivers: we get frustrated when other people go slower than the limit, we don't treat it as a limit, we treat it as the speed you should be traveling at.
I live fairly rural in New Zealand (UK expat) and even though you necessarily get a lot of speed variation on the roads around me, due to being winding, having farm traffic, sometimes narrow, you still get idiots who have to be going at the exact limit (or over) and tailgate 1m behind any vehicle in their way. Including trucks who can't really see them when they do that. I enjoy driving fast on those roads but I still don't understand the impatience.
I have a friend that was just fired from a job, driving 18-wheelers, because he was being monitored by an AI, and the AI malfunctioned, yelling at him for hours to put on his seat belt (it was on). He put a piece of tape over the speaker, and was fired for that.
One of the best, and most experienced big rig truckers in the area. They lost an invaluable employee, and he got another job in minutes (truckers are still a valuable commodity).
One of the things about computers, is that they can’t cut you (or themselves) slack.
My understanding is this varies wildly by jurisdiction.
But it doesn't matter to my point: assume it's just about income, put the cameras everywhere, how fast does it make driving unaffordable?
Same for anything with a fine, e.g. would littering fines be enough to replace taxation, but in a weird way that also puts the messiest subset of the population into life-long bankruptcy to pay those fines?
(Of course, current AI is nowhere near good enough for this, if it was those Amazon "AI-powered" "cashier-free" shops wouldn't have secretly been done by humans watching CCTV).
Determining where the cameras are placed and what to alert on are also important and unresolved issues.
Simply getting alerts from a camera can cause people to believe that the area is a high-crime area, when it's merely a consequence of having a camera there.
Poor people are more like to be in public areas than rich pedophiles who can buy an island or ranch so they and their friends can enjoy wonderful secrets out of the eye of any Flock camera.
If the camera alerts on AI facial recognition for wanted criminals, and facial recognition causes disproportionally higher false alerts for people of south Asian heritage than of Anglo-Norman heritage, then systemic racism is built into the system, which we should all mind.
I'm not talking about monitoring public spaces or searching for criminals. I don't want either of those things and I'm generally opposed to the government operating cameras. I just don't mind private businesses using them to support their existing security guards so long as they don't mishandle or abuse the data.
I'd even be in favor of entirely banning the use of facial recognition technology in conjunction with security cameras. Have them alert on concrete suspicious activity.
I took your list ("The issues are internet connectivity, data retention/mining/sale, and non-local processing") as being incomplete. The examples I gave were to give examples of additional issues. There are equivalents for my examples to private businesses, even putting recognition systems to the side.
I personally have noticed that "alert" and "suspicious" tends to mean "something unusual", and not "something illegal". Increasing alerts results in forced normality.
On the flip side, if the information was there and not used, then the security guards are blamed for not connecting the dots, so investigating alerts becomes a CYA task.
As an example, security guards have harassed people on public sidewalks who are legally taking pictures of the building they are guarding. They are incentivized to investigate the alert, face no consequences (so long as that harassment doesn't itself break the law) for a false alert, and risk losing their job if the photographs are used for nefarious purposes. Adding air-gapped AI may help the security guards, while increasing the amount of harassment.
Yes, I have had a security guard stand over me while I delete a photograph I took of a building while in a public park. I think I was not legally required to follow request. I wasn't going to risk escalating the confrontation over a picture of a neat-looking gargoyle. No, I don't want AI enabling more of that harassment.
I just watched Enemy of the State (late 1990s sequel to The Conversation [1974]) — one of the major plot elements is having to physically acquire the footage/tape (from isolated witnesses/cameras); whereas today, everything feeds into one central company [0].
[0] whom then repackages streams and sells to anybody — mostly law-enforcement — with no 4th Amendment protections ('cause it's a private company brah!).
No. Everyone should be restricted from buying (or better: collecting) it, otherwise you just created the business model for evil corp that does the job and collects your tax dollar to do the same thing.
Easy. If you ban the government to buy that data, but evil corp is allowed to buy it, you now created a business model where evil corp is paid by your government for some compiled version of said data as a service with hefty additional fees on top.
And if you think that is unrealistic you may need to do some research into who pays the bills of Palantir et al.
So if the goal of the ban is to prevent government overreach, you need to ban collection for everybody, except maybe well documenter scientific applications and specific private areas.
Yep, ban collection and pruchase of such data for everyone. Exceptions usually mean private companies hop in to offer the "service".
I think the current insane development are surveilance capitalists, trying to rush their panopticon to solidify their power. Guess that means no reasoable privacy law for the US, even under hypothetical president newsom.
If the government is only restricted from buying the data, then they'll just have someone else buy it. Palantir is not the government. So they can buy the real time feed, analyze it in real time, and give the real time results of that analysis to the government without issue.
Restricting the government from buying that data does nothing. If you want to stop the government taking advantage of the data, then you would have to outlaw the collection of the data altogether. So that the initial collection of the data by anyone, is illegal.
Personally, I don't think that's gonna happen. There's way too many people making way too much money telling the government who hangs out with who, who cheats with who, and so on and so forth.
It's a free charity service, and the hundred million dollars per year the taxpayers pay the same person for some other excuse has nothing to do with it, honest
Oh how the public opinion has been moved already. Rewriting your argument to echo the sentiment from a generation ago:
> The laws need to be updated. Having police officers monitor public streets was fine because they wouldn't actually recall anything unless there was an incident. Now it's possible to go back and review specific footage and identify everyone on those camera's -- we need new privacy laws to reflect this capability.
This is data collected with public funds — our money — for public purposes.
Not only should it be available to any US resident by request, it should be public, as in in an online library, and any US resident without a criminal record should be able to get continuous access, not only a batch of records (yes, keep out anyone with a restraining order or any other crime).
It is our tax dollars, any of us should be able to do research on the data. Including watching the watchers. Where do the government employees go and when? Where do the Flock employees go, and when?
Or, if that kind of instantly-available stalking of anyone is too much of a risk, shut it down. Hard. All of it.
The real-world dynamics of the system is either 1) everybody's motions in public are public, or 2) it is a tool of a totalitarian state. There is no other option, and option 2 is intolerable in a free society.
If only; they temporarily shut off their cameras, while other jurisdictions look to change their laws removing this data from the public record. Neither of these moves are close to "wins"
>The fact that they’re gonna shut it down, implies the scale of indiscriminate nature of data capture and the volume of data being captured.
Or it implies that the .gov, it's agents and those associated with it are not squeaky clean and that any aggrieved party being able to request footage would be bad for the .gov.
Yeah, it appears they have a lot of things backwards, for example:
> “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
The whole point is that they should have been collecting data on perpetrators of crimes only in the first place, not a massive dragnet.
It feels so in bad faith as there’s a claim ICE could use the footage if it’s public domain. They already can and that’s one of the big arguments against it you clowns.
Yeah, that's even more important, isn't it. It begs the question, framing the conversation as "Who should have access to this invasive, privacy-destroying footage" rather than "WTF? Stop this."
Public traffic cams have existed for a while. I don't think anyone would complain as much if the ALPR feed was public. Politicians and government officials want to use surveillance as a means of control not to benefit the public and that's what galls people.
> the scale of indiscriminate nature of data capture and the volume of data being captured
It took a lot of naivete, to put it gently, and head-in-sand attitude to believe otherwise. Flock had everything in place to collect a treasure trove of data but they would decide not to do it? Out of principle? Or even if we take the very charitable interpretation that they don't do it today, but also that they'll never cave in to the pressure to do it in the future?
> Everett Mayor Cassie Franklin said the city disagrees with the ruling and is concerned about who could obtain the footage. “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
These people are fooling themselves if they think that keeping the cameras but not allowing the public to see the data will stop domestic abuse or stalkers. We've already seen these cameras used to stalk people and it wasn't random members of the public doing it, it was police officers. As long as this data is being collected it will be abused. If not by the public, then by police, or by Flock employees, or by hackers. The only way to protect people is to not gather the data at all. Anyone who keeps these cameras doesn't actually care about the public's safety.
The mayor is making a great argument for not blanketing your city in a surveillance dragnet.
Policymakers were warned about precisely these dangers ahead of time. They went ahead anyway, and now they want to play blameless and are trying to shift the blame on anyone but themselves.
The defense of the photos not being government business until accessed seems shaky. That the physical camera installations were purposeful intentions to conduct government business in those areas is a reasonable line; this doesn't set precedent for Google's information becoming public records because the police might do a google search, to use an extreme example.
The proposed legislative amendment that would exclude Flock footage from public records (which would make this judgment moot) makes sense in the light of red light cameras already being excluded by the same legislators. However, I'd like to see a more incisive law covering both that would compel a reasonable amount of public insight into the footage.
The defense of the photos not being government business until accessed seems shaky.
It's reminiscent of the NSA's argument that data "collection" occurs only when a search is performed on existing "gathered" data. File under "Stuff that's only legal when the government does it."
My prediction: “It’s not a “search” when an AI looks at the stored database and does sentiment analysis, because an algorithm doesn’t violate your privacy. It’s only a “search” after it’s flagged and an actual human with human opinions sees your private chats criticizing the supreme leader.”
Scott Alexander has a decent article (or rather a guest blog post) at https://www.astralcodexten.com/p/all-lawful-use-much-more-th... that brings up the subject in the context of Anthropic and OpenAI's dealings with the Department of War (sic), and how their contracts with the DoW might be interpreted with regard to mass surveillance of Americans.
Worth checking out. I'm not personally knowledgeable enough to vouch for the veracity, though.
Please contact your representative in Olympia and tell them to kill the bill that would exempt this footage from public record requests. It passed the Senate on Feb. 6.
I might be good with legal guarantees, meaning jail time for those involved, that the only place images on these devices went was local to the municipality collecting them and that they were only accessed for very well defined reasons by very specific people.
The core issues are that aggregation and exfiltration of this data means that privacy is dead and the AI world allows analysis for almost no cost. We need an idea in our laws that puts back the limited scope that technology has removed. If the police have to expend one person's worth of time to listen to a wiretap then it really isn't possible to get out of control. We need that level of cost associated with ALPR and all surveillance so that the abuse of these systems doesn't get out of control. Make it appropriately hard and it won't be a problem.
Then the feds come in with a national security law and bypass all those state/local protections and slurp it all up into an AI-powered Palantir database, the very existence of which is classified. Suddenly you’re the victim of parallel construction and don’t even know it.
The database CANNOT EXIST SAFELY. Why don’t people who “might be okay with this IF…” understand that?
Collect the data and it WILL be misused, eventually, with 100% certainty. Has nobody read Snowden’s book? They even have a name for intel agents casually spying on their partners/crushes.
The law does not apply to everyone equally. The intelligence agencies get to break any laws they want without consequences, by longstanding tradition (remember 007’s “license to kill” or the CIA’s famous heart attack gun?). There are NO legal safeguards that can prevent abuse, no matter how you word them, because there will always be some animals who are “more equal than others” to whom they do not apply (“national security carve-out”, “LEO exception”, etc).
Sadly, those to whom they do not apply are now coordinating with the new wannabe SturmAbteilung in what are called “fusion centers”.
Or Martin Niemoller: a good Protestant German pastor who viewed the anti-theist attitudes of the Socialists and Communists as more threatening than Nazis. And then the Nazis put him in a concentration camp.
There is another looming threat of modern day surveilance: previously hidden correlations.
The data you found benign sharing in the past might allow unpleasant conclusions in the future and might not even come from you personally. Think about what toys you bought for your kids, or in what college milieu your worldview developed.
And making it easier, better, faster, and more common isn't going to be a good thing. Reminding us that this already happens in smaller scale isn't really adding anything helpful considering the entire goal is to prevent it happening more in the future.
> Cameras that automatically capture images of vehicle license plates are being turned off by police in jurisdictions across Washington state, in part after a court ruled the public has a right to access data generated by the technology.
For reasons I have been unable to articulate, I have believed that if license plate readers are a thing, someone should open-source one that the general public could share.
When a comment comes back that someone malicious will use the tech to stalk someone I really have no answer for that. That is indefensibly a bad thing.
Of course the tech exists nonetheless (not open source as far as I know) and it could be argued it is even being used maliciously—or at least without the kind of judicial oversight we assumed would be in place. But one actor behaving badly does not justify the use of the tech by everyone.
I read about the citizens of Minneapolis using decidedly low-tech means to track ICE vehicles prowling their community and it suggested a scenario where an open license-plate-tracking solution could "balance the scales" a bit (that is if you believe the scales to be imbalanced).
I imagined not a license-plate reading dragnet but rather software where you had first to enter in very specific strings of ASCII characters and the software would only announce when there was a specific string match from the camera.
To that end I vibe-coded an app for iOS in about 15 minutes using iOS's Vision framework and the built-in phone camera. Anyone could do the same.
Nonetheless I only tested it with "HELLO" and "WORLD" using scraps of paper in my kitchen and never tried it outside as the craziness in Minneapolis seemed to have quieted down.
There had been a fairly decent open APLR library called OpenALPR. It used to be updated fairly regularly, but then was "acquired" and now sits abandoned for almost a decade without any updates, while the company that acquired it (REKOR) has commercialized it and extended it as a closed source API and platform.
ALPR is fairly trivial now days, and a modern CPU can process thousands of plates per second (dozens of plates per frame, hundreds of frames per second coming multiple cameras pointing at different lanes of traffic).
The "product" around all of these is linking the plates traveling together, tracking routes through various cameras, identifying common travel patterns, and more disturbing travel patterns that are outside norm for a given plate or route (i.e. do they normally drive these 3 cameras M-F, but this Th, they went a different route and stopped somewhere for a while).
All that used to be done by the police/detectives/investigators on their own. Now the AI is automating this, and that is truly terrifying, especially for how often misreads occur.
Ive wanted to know this for a long time. Why isn't there some package I can download to a raspberry pi with a camera, and contribute to a public website that shows things like, oh I don't know real time positions of government vehicles...
I always think that the best way to get stuff regulated is make those in power feel the risks.
So if people started using something like flock to embarrass politicians, business leaders, or newspaper leader writers then suddenly privacy might become a big issue.
don’t worry, normal people will never ever see this data to use it against the powerful.
our public data can only be seen by billionaires and cops, not us.
it can be used against us, but never the other way around. the faster we realize this, the faster we can move out of our “divisive” phase and get back to making billionaires dreams come true.
Indeed - but as compute costs come down, it doesn't have to be that way - ultimately there are 8 billion eyes out there.
There are also public data examples - for example the public data on charter flights or ship locations had people like Elon Musk bleating about privacy.
My issues is with false arrests due to people blindly following AI. People pulled out of their cars are gunpoint due to the AI misreading a single letter on the plate, the lady who was arrested for stealing amazon packages just because flock saw her enter and exit a neighborhood and she had to get doorbell camera to prove her innocence.
Also i think it should be public record about every single point of data flock collects and retention period. I am not saying they have to release the data itself, but I want to know what my tax dollars are collecting, how long they are keeping it, who can access it and who they are selling, er i mean sharing it with.
> For now, Everett’s Flock camera network remains offline, as the debate over transparency, privacy and public safety continues in the Legislature. The bill in Olympia that would put guidelines on Flock's data has passed in the Senate.
> “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
No concern over the dozens (or hundreds?) of cases of police or government employees themselves doing exactly what they’re afraid of here. Strange.
> No concern over [...] government employees themselves
Three paragraphs later someone else is paraphrased as including immigration enforcement agents among the problematic users, and in the current political environment, federal law enforcement being made more effective might be the real problem for state and local government.
While I agree with the risks of DA/stalkers getting that data, this data is not known for being well protected against LoveInt. Quite the opposite it is usually sold on grey markets.
They know, but if they acknowledge it it would make some people mad. Anyone who works with or it’s associated with ALPR knows why you have to have audit logging and access controls if you give law enforcement access to it.
I am less worried about Flock ALPR (which are aimed in the direction of traffic flow to read rear number plates) as I am about the THOUSANDS of facial recognition cameras installed in the last year in all four directions at nearly every intersection in southern Nevada and many many cities in southern California (LA notably excepted). These are mounted above the stoplights and aimed against traffic at stoplights to read faces.
I mention these locales specifically only because I have directly observed them. I would be surprised if this isn’t also happening in many other US metro areas, given how eagerly DHS/TSA/CBP/ICE are mass collecting facial geometries at every available opportunity.
Flock cameras are advertised as ALPR but have facial/vehicle/etc recognition systems running on them as well. You can see one panning up to track other people near the end of this Benn Jordan video on the cameras.
The statute in Washington that allows these to be used to enforce speeds and signals also prohibited photographing the driver using another or the same camera.
shutting down the cameras rather than releasing the data tells you everything about what was being collected. the system was designed around the assumption that nobody outside law enforcement would ever see the footage.
"The masses/general populace are the enemy" - once you understand that this is the fundamental belief at the root of the elites behaviour, everything will make sense. Flock cameras and AI surveillance is designed to reign in 'the enemy'.
A mix of public (city councils) and private (think HOAs that then donated access/equipment to the city) contracted with Flock in the past few years.
The questions of exactly who, when, and why, are very muddy especially with the HOAs who operate rather privately.
The government's power over its citizens should mainly derive from its legitimacy in the eyes of citizens, not from its technological and military superiority over the people.
This kind of ubiquitous surveillance is so dangerous to democracy and the ability of the citizens to do anything without the government knowing.
I think the person requesting to access the data was doing the right thing and I agree with the judge’s ruling.
The fact that they’re gonna shut it down, implies the scale of indiscriminate nature of data capture and the volume of data being captured.
These cameras are popping up all over the nation and if people realize how much data is being captured and where that data is going (or who it’s being sold to) and how it’s being used by government and private entities they would be appalled.
There’s been exposés about these cameras, everything from AI misidentification of “stolen” (not) vehicles and erroneous arrests and police encounters, to analysis of shopping patterns being sold back to private entities for better ad targeting. It’s wild.
The laws need to be updated. CCTV in public used to be fine because no one was actually watching it unless there was an incident. Now it’s possible to have AI watch every camera and correlate everything everywhere we need new privacy laws to reflect this capability.
I don't mind a local AI on an airgapped security camera network monitoring a camera and issuing an alert to a security guard. The issues are internet connectivity, data retention/mining/sale, and non-local processing (ie handing stuff off to a third party that does who knows what and probably doesn't take security seriously).
Even with that, I do mind.
Just as two trivial examples, even though neither affects me personally:
The estimated number of heroin users in the UK exceeds the total prison population. The number of class-A drug users in the UK is estimated to be so high that if they actually followed the minimum sentencing guidelines for possession, it would cause a catastrophic economic disaster both from all the people no longer working and also all of the people who suddenly had to build new prisons to hold them. I'm not interested in drugs (and I don't live in the UK, but I assume the UK isn't abnormal in this regard).
Another example is road traffic law. Even just speeding offences, I think you probably catch everyone who actually drives in the UK, often enough that after a month the only people left allowed to drive would be people like me who don't even own a car.
The entire legal system has to be radically changed with far less punishments for almost everything if you have perfect, or even 30% of the way to perfect, surveillance.
> Another example is road traffic law. Even just speeding offences, I think you probably catch everyone who actually drives in the UK, often enough that after a month the only people left allowed to drive would be people like me who don't even own a car.
You don't have to strip the driving licenses. You should impose a fine and not an extremely painful one for starters.
And then probably within less than a year the whole population will drive properly.
I think I'm in favor of indiscriminately fining everyone speeding at every camera, but I realize there is no privacy-preserving way to do it today thus I will be against it.
(I'm a driver and car lover who is never speeding)
That's covered by "system has to be radically changed". UK driving licences give you room for 12 "penalty points" worth of mistakes before you risk being banned from the road, of which speeding costs you at least 3: https://www.gov.uk/penalty-points-endorsements/endorsement-c...
Indeed. 4 strikes and you are out seems fine to me (as a UK driver). You can also opt to take a speed awareness course in leu of the points, I believe
4 strikes today with current speed enforcement is fine; 4 strikes in a hypothetical where 100% of infractions are caught, I aver will catch (almost) everyone who actually drives within a month.
Where I live almost every main road now has average speed cameras, leaving only the small residential streets without, and they are generally short enough that few people are speeding anyway. In general I approve, especially in residential areas, although the surrounding area has gone through a process of downgrading speed limits for no obvious reason, such that it seems their only intent is to annoy drivers so that they want to stop driving. Almost all the country roads in the county I live adjacent to were downgraded from national speed limit to 50mph about 10 years ago. It didn't seem to be anything to do with safety, just seemingly out of spite. I heard rumours that they were also trialing drones to spot offenders (and presumably with sufficiently good cameras to read the plate from up high). Recently, many residential places have dropped from 30mph to 20mph pretty uniformly, again seeming nothing to do with safety as it's entire suburbs (but not all suburbs in the city), even on long straight roads with excellent visibility that would have been safe with limits above the original 30. Doing 20 on these seems completely unnecessary and it's hard to see it as anything other than a revenue generation tactic.
If notice of the infractions is a envelope delivered days later and sitting on on end table for a week, typical drivers are going to rack up 4 strikes before seeing the first one!
Perfect enforcement has to come with immediate feedback
I remember a case from decades ago when someone went from clean license to a ban within 3 miles
I think what he is trying to say though is that people will start to drive slower, especially if they need to drive to function/get to their job.
I dunno. We got car insurance once that had a "put this spy device in your car for a couple weeks for a lower rate" deal and I felt like I was driving a lot less-safely when I was constantly worried about looking like I was driving safely.
Like, to pick an example that's specifically speed-limit related, if more people really tried hard to do 25MPH (the marked speed limit on many of them) or under for the entire length of an interstate off-ramp, I think we'd be spending more money on brake pads and there'd be a lot more cars getting rear-ended. Sticking to that speed the entire length is silly and not very safe, and things work just fine without people doing that. Tons of other edge cases like that where you're technically breaking some law or another for a little while, but things work way better if you do. Plus practically every one of these laws has some kind of judgement-call clause that applies to modify it, and I don't want the people making those judgement calls to know that if they do what seems right to them in the moment, there's a very high likelihood they'll be hassled for it.
I've never seen an interstate off-ramp in the US with a 25MPH speed limit (white sign). I've seen 25MPH advisories (yellow sign), but those aren't a legal limit. Advisory signs are the maximum safe speed for the worst possible conditions (road covered in ice).
I otherwise agree.
Interstate, no, and actual speed limit, no, but the one I always think of is this, on US 101, near Steamboat Island in Olympia where you come off a 60mph two lane state highway to an off-ramp that is 430' long and has an advisory 10mph speed for a very unforgiving 90/180 degree turn (160' diameter circle) - with the ability to rear-end people joining the highway if you miss the first part of the turn or the ability to t-bone people if you miss the second. ChatGPT describes the braking required as being at the high end of moderate, almost hard braking, for a sustained 9 seconds to slow sufficiently to make the turn, assuming you start braking the moment you are in the exit lane (and there's no advanced warning, just a single "10mph" yellow sign as you enter that lane): https://imgur.com/a/uns86kp
I have to kind of agree with you on all points here.
If speeding is not something that happens constantly, then a radar could detect the instances of speeding, and only turn on a camera when a speeding car is nearby. This would keep the majority of passing cars from being recorded, and would record the fewer cars the fewer drivers would be speeding.
This is exactly how non-AI assisted speed cameras [1] have worked for almost four decades. You don’t even need video for it.
[1]: https://en.wikipedia.org/wiki/Gatso
If speeding is not something that happens constantly, then a radar could detect the instances of speeding
And if it is, which it generally is, it means the speed limits are not set appropriately. But that point always seems to be overlooked.
How long until the AI estimates how fast you were going based on the time you were tagged at 2 cameras? The system says travel between these cameras should take 3 minutes. You made it in 2:45. No review, just a ticket in the mail.
That's how it actually works in some places (https://www.carwow.co.uk/blog/average-speed-cameras-how-do-t...). You don't need AI because we already know that distance = speed * time. We can calculate to a high degree of certainty - with high school math - that you had to have 1. been speeding or 2. bent space time if you cover the distance between them in too short of time.
Several months later, the state changes the speed limit on the street without changing the ticket check. Some time after, traffic authorities add a bypass road or a fast lane to deal with the new traffic problems and forget to update the timings. Then it's reported that a rural municipality set timing thresholds just below the speed limit to cover budgetary shortfalls.
Eventually the camera manufacturers offer "easy setup" systems that dynamically adjust based on actual journey times. This works fine until the first office holiday, when the routine congestion is gone.
If their clocks are actually synchronized and can show that in court then unless you've got a warp drive and can bend space time it can be pretty definitive that one was speeding between those two points. Especially if you're on something like a toll road which utilizes transponders for billing. I'm almost surprised this isn't already done in those situations.
You don't even need AI for this. Its pretty basic.
It is similar to how air enforced speed limits are done. They just paint two lines on the highway. A plane overhead times your car between the lines.
The point he’s making that people violate the letter of the law in many, many small ways, and to prosecute people for all of them would be a crippling burden on both individuals and the economy.
One obvious response is that if 100% enforcement of laws is causing social problems, maybe you just have bad laws and need to change them.
[flagged]
Yeah it's a bit of wishful thinking from my side, I confess.
> exactly why we end up with surveillance states
Hate that. This is why I wrote that it looks like there is no privacy-preserving way to trap ALL speeding drivers, although I have been corrected on that specific part.
What I wanted to say is that I would love for road infractions to be fined all the time so that people would know that there is no free ride.
I'm going to argue the other side: in Chinese cities like Chongqing they've seen a drastic reduction in crime after blanketing the city with cameras and monitoring technology.
Whole categories of crime disappeared. Women and elderly feel safe to walk the streets at night. No one locks their bike anymore in Chongqing.
I care about privacy, but I think we should be smart enough to work out a way to get some of those benefits without going full 1984. For example by having surveillance that can only be queried by an AI with very strong guard rails.
Admittedly, I live in a country with very strong democratic institutions, and I trust we would take action the moment something gets abused or surveillance overreaches. I would probably feel differently living elsewhere.
> I trust we would take action the moment something gets abused or surveillance overreaches.
The thing about turnkey authoritarian solutions is that once something happens it's likely too late to take action. However there are often alternative solutions that physically constrain the system such that substantial abuse is impossible without time consuming and expensive physical modifications. The traditional speed cameras in the UK for example.
Cameras, AI integrated at the edge, software that can't be updated remotely, the full stack publicly audited, that only output video data when a suspected violent crime is flagged. Something like that might work. I'm not optimistic such a solution would see much support though.
You'd also probably want a policy put in place in advance to quickly pull them down if certain criteria are met. But again, I'm not optimistic about the prospects.
Well I agree, and my hopes aren't very high of this actually happening. Our politicians tend to be clueless with anything tech related, their opinions calibrated by what they saw in Hollywood movies, where anything tech related always turns into "black mirror". (By contrast, allegedly over half the Chinese politburo has an engineering degree of some kind).
But we could start small, with just one neighborhood, a pilot project where the kinks get worked out and slowly scaled up. Getting permission for a small scale pilot shouldn't be impossible.
So you want to slippery slope your way into Nazi Germany?
Everybody made that exact same "slippery slope to Nazi Germany" argument when euthanasia was legalized here. That was decades ago. There have been several attempts to broaden or narrow the scope of those laws and the democratic institutions did just what they were designed to do, making changes judiciously.
If you are worried about the slippery slope, then you are really worried that democracy does not work as intended. (And depending on where you live that may be a very reasonable worry). By the way, Nazi Germany was not really a surveillance state, perhaps you are thinking of East Germany?
> If you are worried about the slippery slope, then you are really worried that democracy does not work as intended.
Not really. That's a well established failure mode. People's perceptions can gradually shift as they become accustomed to the new way of doing things.
Personally I'd be less concerned about a slippery slope and more concerned about abrupt changes in policy. All infrastructure should be designed with the worst case scenario in mind. It's naive to assume that things will never get worse suddenly or that we will have plenty of warning or even a meaningful opportunity to react.
Are you saying democracy is working as intended, but you don't like the outcome?
There's no way to deploy a system like the one you're describing without being abused for authoritarian overreach. It's simply a matter of time, and once it is deployed for authoritarian overreach, the only way back will be paid for in blood.
> Even just speeding offences, I think you probably catch everyone who actually drives in the UK,
So we have a fucktonne of speed cameras allover the place: https://www.speedcameramap.co.uk/ (you need to zoom in there are so fucking many)
But we have less redlight cameras than the US. we also have hatching cameras (yellow hatched boxes mean no stopping, usually at junctions) we also have bus lane cameras, where if you drive in a buslane you get a fine.
For the Speed cameras, they are normally put there based on evidence of road deaths linked to speeding. I dont like speed cameras, but they do serve a purpose.
When you get a speeding ticket, if its your first offence, you can take a speed awareness course, and you won't get points on your license. otherwise its three points and a £100 fine. The points age out after 3 years. the maximum you can normally get is 12 points on your license.
Its only in extreme cases do you get a ban, or license revoked.
The reason why people are still able to drive are numerous:
1) its been a gradual evolution.
2) we have fairly robust training for drivers (theory, comprehensive real world test)
3) Evidence based placement. Its not like they just shove these things where poor people live (or in the US where the city has zoned living for people with more melanin than others). If there are higher than average road crashes, the road is re-made to make it safer, speed limits dropped, traffic calming put in place, then speed cameras.
4) You are expected to follow the traffic rules
5) the traffic rules are actually pretty sensible.
> we also have hatching cameras (yellow hatched boxes mean no stopping, usually at junctions)
Weirdly I've never encountered these in the US (only red light cameras) and do we ever need them. I'm generally opposed to government associated cameras due to concerns about turnkey authoritarianism but if we have to have cameras at intersections they could at least curb the awful self centered behavior.
Most states ban speed cameras and many ban red light cameras as 4th amendment violations. You cannot face your accuser when your accuser is a robot. As a result, speed camera tickets have always used some legal sleight-of-hand to nail you into confessing, and this has become unpalatable.
If we raised speed limits (almost) across the board to the actual safe limits of modern cars, I think a lot more people would be ok with speed cameras. There would still be a constitutional problem, however.
What are "the actual safe limits of modern cars"?
Data from various states raising their limits over the last few decades is that every 5MPH increase in state speed limits brings with it an 8.5% increase in traffic fatalities on freeways. With the 70MPH speed limit common for freeways through most of the US, we're already up 25% on road fatalities over the 55MPH that was chosen for gas mileage during the oil crisis.
For cities, pedestrians struck at 25MPH are already at a 10% chance of death, which reaches 50% at 40MPH.
Nobody goes 70 mph on freeways. They go 80 mph on that road because it's the speed of traffic. If you declared a "speed limit reset" and raised the speed limit to 80 mph, people won't be going 90 mph. They will be going the speed limit.
So "the actual safe limits of modern cars" just means "the speed everyone is currently driving"?
In the UK 80-90 was quite normal 25 years ago, and off peak you'd find the outside lane of the M40 doing over 90 a fair amount
It's not today as there's far more traffic. It's rare to have the opportunity outside of a few areas (south of Bristol on the M5, north of Kendal on the M6 etc). When I first learned to drive I'd do M62 to M5 in well under an hour, today it's about 20 minutes longer.
> You cannot face your accuser when your accuser is a robot
I don't really get this take though. If one contests the ticket, have it go to a manual review where someone looks at the tape and confirms the calibration history of the equipment, and boom now that person can be the official "accuser".
If its the argument that they might not have been the one actually driving the vehicle, just make the laws relate to the registered owner of the vehicle is ultimately responsible for the safe operation of the vehicle. If the owner wants to try and prove it wasn't them, they can deflect that in court and prove it was actually their friend or whoever.
If we had a CCTV of a murder happening we wouldn't just go "well that's just a robot not a real person making the recording, guess we need to toss that video!" I don't understand how we take that kind of position when it comes to moving violations.
Also FWIW the 4th Amendment has nothing about facing your accuser. Its the sixth amendment that talks about "...to be confronted with the witnesses against him".
Other states deal with it by making it a civil infraction, not a moving violation, bypassing the 4th amendment issue.
> So we have a fucktonne of speed cameras allover the place: https://www.speedcameramap.co.uk/ (you need to zoom in there are so fucking many)
Doesn't seem that many compared to what I was describing. At the scale of a country, "a lot" != "a high %".
Your point 3 is the biggest divergence between them.
Point 5 is only kinda true, the failure mode is weird: there's good reasons why the speed limit isn't enforced until you're significantly over it, but that in turn means it has to be set lower than physics and reaction times dictate, which in turn means people push back against them. 20 zones knowing people will do 25, chosen because if they were 30 zones people would do 35 and 35 is too fast, that kind of thing.
People who know that, let themselves go a bit over the limit; but a bit over means they get caught some of the time because of the same small occasional variations that are the reason why the speed limit isn't enforced at x+1 mph in the first place.
I was a bit unclear. I agree, I don't want the government using AI to identify all violations of the law. That sounds like a very straightforward dystopia.
What I don't mind is private companies using AI analysis to support their security guards. I object to any sharing of the data with third parties though. It should be illegal for the data to leave their internal network and it should be illegal to retain it for more than a few days.
I don't care if grocery store loss prevention has eyes on every aisle. My concern is data warehousing and subsequent misuse.
> The entire legal system has to be radically changed with far less punishments for almost everything if you have perfect, or even 30% of the way to perfect, surveillance.
Prosecutorial discretion means they can just collect evidence and choose not to charge you unless they want to leverage you for something. This already happens, but universal surveillance means it can literally happen to anybody, because everybody breaks the law in some way due to how many laws we have.
Discretion is the real problem I think. It seems extreme, but maybe discretion should be eliminated: if you commit a crime you will be charged. This will at first result in way too much prosecution, which will lead to protests and hopefully repealing laws and we'd end up in a better place where the law is understandable and predictable by mortals.
Speeding is a special case, because it's unclear what the lawmakers, road designers, and police intend. When the speed limit is 65 mph, do they actually intend for everyone to go no faster? I don't think so. I think the lawmakers, if driving in traffic, want people to go a bit faster. Same with the police. And I think the road designers design the roads knowing most people will speed.
I want to follow the law. But when it comes to speeding, it's hard for me to follow the letter of the law, because all the parties involved in creating and enforcing the law don't want me to follow the letter of the law. So I instead follow the intent of the law, and speed up to 9mph. When Google Maps pops up a "police ahead" warning, I don't slow down at all, because I'm following the intent of the law, and that's what police around where I live enforce. If I'm driving in other areas of the country, I'm less certain what police want, so I'll be more likely to follow the letter of the law.
If there was automated strict enforcement of speeding, then it would be clear to me that the letter of the law is the intent, so I would gladly obey the letter of the law. There would certainly need to be a transition period with clear warnings that in the future, the letter of the law will be enforced, instead of the current status of something looser.
> When the speed limit is 65 mph, do they actually intend for everyone to go no faster? I don't think so.
Where is this supposed ambiguity coming from?
Some states follow Assumed Maximum Posted Speed (in certain places) and others are Absolute Maximum Posted Speed. It is not absolute in an Assumed Maximum Posted Speed state that driving faster than the posted speed is against the law and deserving of a fine, merely it is prima facie evidence that you were driving dangerously but can be challenged and overturned. For example, in Minnesota, outside of municipalities on highways (there may a few more qualifiers like posted speed is 55 mph or higher and might need to be a divided highway, I don't remember 100%) an officer can pull you over and issue you a ticket for merely driving faster than the posted speed limit. You can even admit you were driving faster (I don't recommend this). You can still challenge the ticket in court. If you can convince a judge that your speed was safe, the judge can let you off. If the weather is dry, temperature moderate, visibility great, no other people or vehicles around you, you were able to safely slow down, and (prima facie evidence) that you posed no risk as no one was injured by you driving faster. In Wisconsin though they are an Absolute Maximum Posted Speed state so if you are found to have been driving faster than the posted speed limit, that's enough to ensure you can be fined.
Are you asking (1) why I think what I think?
Or are you asking (2) how we wound up in this situation as a society?
(1) I think what I think for several reasons. Basically everyone speeds. Probabilistically that includes he very lawmakers writing the laws, the police, and the road designers. I've also read some articles talking about road design, and in it it's mentioned that the designers factor in that most people will speed if the road conditions are amenable. I've also seen police cars driving around without their lights on, passing people at higher than the speed limit, and when unable to pass, the appear annoyed to me.
(2) I think this situation arose in sort of a "normalization of deviance" manner. Police didn't want to be too strict, or didn't want to bother fighting tickets for people speeding only a little, so only gave tickets for people speeding a lot. Then over time many people realized that, and started speeding a little. More are and more people started speed just to fit in with the surrounding traffic, until eventually everyone was speeding. Peer pressure. I've heard driving the same speed as the surrounding traffic is generally safer than driving significantly slower (or faster). Once everyone is speeding, that includes lawmakers, road designers, and police. And they factor that in when they write laws, design roads, and enforce laws.
There were also fairly contiguous moral panics in the late 1960s (teenage boomers crashing muscle cars) 1970s (fuel use) and early 1980s (slightly older boomers driving drunk) that lead to the regulations being written far in excess of what there's popular support to enforce which is a huge contributor to why the enforcers and judiciary are essentially responsible for dialing it back to something that doesn't make the system look stupid.
In Washington state, the State Patrol has gone on the record multiple times and with the media saying their enforcement is based on "Under 10 over", i.e. they won't pull you over for doing 69 or less on the Interstate in a 60 zone, or 79 or less in a 70 (assuming speed is the only issue).
Many speed limits and policies were established in an era of fewer cars, but also much less capable cars with fewer safety features - many speed limits were established before the adoption of ABS, stability control and airbags, and more recent innovations in lane-keeping and adaptive cruise control.
Modern cars may be capable of travelling at greater speeds with greater safety, but there's a more recent recognition of the increase in emissions pollution from increased speed. Speed-limits typically remain grandfathered in at their original value (which may have been set 30, 40, 50 years ago), regardless of the change in context.
Then there are some pecularities such as the UK default of 60mph for a single-track road, but if you were to try that in many rural locations (think Cornwall, Scotland, Wales) you would likely find yourself upside-down in a ditch.
This post highlights the absurdity of some of the limits!
https://www.reddit.com/r/CasualUK/comments/1dng5z9/genuinely...
The UK NPCC (National Police Chief's Council) have a published policy where enforcement effectively starts at 10% +2mph over the speed-limit (whilst allowing officers to use individual discretion if they feel there are aggravating factors).
https://library.college.police.uk/docs/NPCC/Speed-enforcemen... [PDF]
Counterpoint: with mobile devices, and increasingly, control and information features of automobiles themselves, distracted driving is increasingly a concern.
There's also the point that driving capabilities vary wildly by individual, and often decline drastically with age. Recent case-in-point, an elderly driver in San Francisco who killed a family of four (a mother, father and two daughters, waiting at a bus stop, not in the roadway at all), let off with a minimal sentence, raising much public furore:
<https://www.sfchronicle.com/crime/article/sf-west-portal-cra...>
<https://www.sfchronicle.com/crime/article/san-francisco-west...>
The driver was speeding (70 mph in a residential area), and possibly driving the wrong way on the street.
> Speed-limits typically remain grandfathered in at their original value
That depends on where you are. In Texas, state highway speed limits are determined though a traffic study[1]. The monitor traffic for a while, then set the limit to the 85th percentile.
People can use this to get out of speeding tickets. If you find that it's been a long time since a speed study was done on the road you were on, the judge might throw the ticket out.
There are some hard limits though. For example, the maximum speed limit that can be set on a road is 85 mph.
[1]: https://www.txdot.gov/safety/driving-laws/speed-limits/speed...
In other words, 15% of people are ALWAYS speeding, by definition.
So with perfect enforcement (punish everyone over the threshold), how do you enable an 85th percentile rule?
> 15% of people are ALWAYS speeding, by definition
Not necessarily. If the data is heavily clustered around a single value, the percentile collapses to that number.
> UK default of 60mph for a single-track road, but if you were to try that in many rural locations (think Cornwall, Scotland, Wales) you would likely find yourself upside-down in a ditch.
The way the national limit is framed is more limit than road speed. It's interesting how we think of the limits as drivers: we get frustrated when other people go slower than the limit, we don't treat it as a limit, we treat it as the speed you should be traveling at.
I live fairly rural in New Zealand (UK expat) and even though you necessarily get a lot of speed variation on the roads around me, due to being winding, having farm traffic, sometimes narrow, you still get idiots who have to be going at the exact limit (or over) and tailgate 1m behind any vehicle in their way. Including trucks who can't really see them when they do that. I enjoy driving fast on those roads but I still don't understand the impatience.
The purpose of a system is what it does. In Australia, they want you to go 50. In the USA, they want a reason to fine you.
> In the USA, they want a reason to fine you.
In my area of the US, I don't think so. I'll go 9mph over the speed limit when I drive by a police car. I've never been pulled over.
In some other areas of the US, you're right, which is why I'm less likely to speed in other areas.
I have a friend that was just fired from a job, driving 18-wheelers, because he was being monitored by an AI, and the AI malfunctioned, yelling at him for hours to put on his seat belt (it was on). He put a piece of tape over the speaker, and was fired for that.
One of the best, and most experienced big rig truckers in the area. They lost an invaluable employee, and he got another job in minutes (truckers are still a valuable commodity).
One of the things about computers, is that they can’t cut you (or themselves) slack.
Speed cameras are a source of income, they are not for enforcement or safety.
My understanding is this varies wildly by jurisdiction.
But it doesn't matter to my point: assume it's just about income, put the cameras everywhere, how fast does it make driving unaffordable?
Same for anything with a fine, e.g. would littering fines be enough to replace taxation, but in a weird way that also puts the messiest subset of the population into life-long bankruptcy to pay those fines?
(Of course, current AI is nowhere near good enough for this, if it was those Amazon "AI-powered" "cashier-free" shops wouldn't have secretly been done by humans watching CCTV).
Two things can be true at once
Determining where the cameras are placed and what to alert on are also important and unresolved issues.
Simply getting alerts from a camera can cause people to believe that the area is a high-crime area, when it's merely a consequence of having a camera there.
Poor people are more like to be in public areas than rich pedophiles who can buy an island or ranch so they and their friends can enjoy wonderful secrets out of the eye of any Flock camera.
If the camera alerts on AI facial recognition for wanted criminals, and facial recognition causes disproportionally higher false alerts for people of south Asian heritage than of Anglo-Norman heritage, then systemic racism is built into the system, which we should all mind.
I'm not talking about monitoring public spaces or searching for criminals. I don't want either of those things and I'm generally opposed to the government operating cameras. I just don't mind private businesses using them to support their existing security guards so long as they don't mishandle or abuse the data.
I'd even be in favor of entirely banning the use of facial recognition technology in conjunction with security cameras. Have them alert on concrete suspicious activity.
I took your list ("The issues are internet connectivity, data retention/mining/sale, and non-local processing") as being incomplete. The examples I gave were to give examples of additional issues. There are equivalents for my examples to private businesses, even putting recognition systems to the side.
I personally have noticed that "alert" and "suspicious" tends to mean "something unusual", and not "something illegal". Increasing alerts results in forced normality.
On the flip side, if the information was there and not used, then the security guards are blamed for not connecting the dots, so investigating alerts becomes a CYA task.
As an example, security guards have harassed people on public sidewalks who are legally taking pictures of the building they are guarding. They are incentivized to investigate the alert, face no consequences (so long as that harassment doesn't itself break the law) for a false alert, and risk losing their job if the photographs are used for nefarious purposes. Adding air-gapped AI may help the security guards, while increasing the amount of harassment.
Yes, I have had a security guard stand over me while I delete a photograph I took of a building while in a public park. I think I was not legally required to follow request. I wasn't going to risk escalating the confrontation over a picture of a neat-looking gargoyle. No, I don't want AI enabling more of that harassment.
I just watched Enemy of the State (late 1990s sequel to The Conversation [1974]) — one of the major plot elements is having to physically acquire the footage/tape (from isolated witnesses/cameras); whereas today, everything feeds into one central company [0].
[0] whom then repackages streams and sells to anybody — mostly law-enforcement — with no 4th Amendment protections ('cause it's a private company brah!).
And the government needs to be restricted from buying data it wouldn’t be permitted to collect itself.
No. Everyone should be restricted from buying (or better: collecting) it, otherwise you just created the business model for evil corp that does the job and collects your tax dollar to do the same thing.
How can an evil corp that isn't the government tax you?
Easy. If you ban the government to buy that data, but evil corp is allowed to buy it, you now created a business model where evil corp is paid by your government for some compiled version of said data as a service with hefty additional fees on top.
And if you think that is unrealistic you may need to do some research into who pays the bills of Palantir et al.
So if the goal of the ban is to prevent government overreach, you need to ban collection for everybody, except maybe well documenter scientific applications and specific private areas.
Yep, ban collection and pruchase of such data for everyone. Exceptions usually mean private companies hop in to offer the "service".
I think the current insane development are surveilance capitalists, trying to rush their panopticon to solidify their power. Guess that means no reasoable privacy law for the US, even under hypothetical president newsom.
Meh.
If the government is only restricted from buying the data, then they'll just have someone else buy it. Palantir is not the government. So they can buy the real time feed, analyze it in real time, and give the real time results of that analysis to the government without issue.
Restricting the government from buying that data does nothing. If you want to stop the government taking advantage of the data, then you would have to outlaw the collection of the data altogether. So that the initial collection of the data by anyone, is illegal.
Personally, I don't think that's gonna happen. There's way too many people making way too much money telling the government who hangs out with who, who cheats with who, and so on and so forth.
>give the real time results of that analysis to the government without issue.
Presumably they’re not doing this for free.
It's a free charity service, and the hundred million dollars per year the taxpayers pay the same person for some other excuse has nothing to do with it, honest
When passive monitoring turns into analytics -> public watching turns into stalking.
it’s ok for me to observe a person in public park - it should be ok to watch camera for an activity
But if it’s not ok for me to stalk someone I think it should be illegal for a network of cameras to watch my movements too!
Slapping AI onto stalking is still stalking.
It was not fine then -- what we have now is simply even worse. We do not need to make concessions to our oligarchs: none of this is OK.
Oh how the public opinion has been moved already. Rewriting your argument to echo the sentiment from a generation ago:
> The laws need to be updated. Having police officers monitor public streets was fine because they wouldn't actually recall anything unless there was an incident. Now it's possible to go back and review specific footage and identify everyone on those camera's -- we need new privacy laws to reflect this capability.
The judge is right, and should go even further.
This is data collected with public funds — our money — for public purposes.
Not only should it be available to any US resident by request, it should be public, as in in an online library, and any US resident without a criminal record should be able to get continuous access, not only a batch of records (yes, keep out anyone with a restraining order or any other crime).
It is our tax dollars, any of us should be able to do research on the data. Including watching the watchers. Where do the government employees go and when? Where do the Flock employees go, and when?
Or, if that kind of instantly-available stalking of anyone is too much of a risk, shut it down. Hard. All of it.
The real-world dynamics of the system is either 1) everybody's motions in public are public, or 2) it is a tool of a totalitarian state. There is no other option, and option 2 is intolerable in a free society.
>> The fact that they’re gonna shut it down
If only; they temporarily shut off their cameras, while other jurisdictions look to change their laws removing this data from the public record. Neither of these moves are close to "wins"
>The fact that they’re gonna shut it down, implies the scale of indiscriminate nature of data capture and the volume of data being captured.
Or it implies that the .gov, it's agents and those associated with it are not squeaky clean and that any aggrieved party being able to request footage would be bad for the .gov.
Yeah, it appears they have a lot of things backwards, for example:
> “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
The whole point is that they should have been collecting data on perpetrators of crimes only in the first place, not a massive dragnet.
It feels so in bad faith as there’s a claim ICE could use the footage if it’s public domain. They already can and that’s one of the big arguments against it you clowns.
beyond that it also implies "but the bad guys could then trawl all this data; only us, the good guys should be able to do that!"
Yeah, that's even more important, isn't it. It begs the question, framing the conversation as "Who should have access to this invasive, privacy-destroying footage" rather than "WTF? Stop this."
Public traffic cams have existed for a while. I don't think anyone would complain as much if the ALPR feed was public. Politicians and government officials want to use surveillance as a means of control not to benefit the public and that's what galls people.
> the scale of indiscriminate nature of data capture and the volume of data being captured
It took a lot of naivete, to put it gently, and head-in-sand attitude to believe otherwise. Flock had everything in place to collect a treasure trove of data but they would decide not to do it? Out of principle? Or even if we take the very charitable interpretation that they don't do it today, but also that they'll never cave in to the pressure to do it in the future?
> Everett Mayor Cassie Franklin said the city disagrees with the ruling and is concerned about who could obtain the footage. “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
These people are fooling themselves if they think that keeping the cameras but not allowing the public to see the data will stop domestic abuse or stalkers. We've already seen these cameras used to stalk people and it wasn't random members of the public doing it, it was police officers. As long as this data is being collected it will be abused. If not by the public, then by police, or by Flock employees, or by hackers. The only way to protect people is to not gather the data at all. Anyone who keeps these cameras doesn't actually care about the public's safety.
Two such instances of police using Flock to track current or former romantic partners:
* https://www.kansas.com/news/politics-government/article29105...
* https://www.fox6now.com/news/milwaukee-police-officer-charge...
The mayor is making a great argument for not blanketing your city in a surveillance dragnet.
Policymakers were warned about precisely these dangers ahead of time. They went ahead anyway, and now they want to play blameless and are trying to shift the blame on anyone but themselves.
Cops and DV. And crash coverups.
this is not an honest argument, it's just a variation of the "think of the children" strategy.
[dead]
This is a good article about some of the legal particulars. https://www.heraldnet.com/2026/02/24/snohomish-county-judge-...
The defense of the photos not being government business until accessed seems shaky. That the physical camera installations were purposeful intentions to conduct government business in those areas is a reasonable line; this doesn't set precedent for Google's information becoming public records because the police might do a google search, to use an extreme example.
The proposed legislative amendment that would exclude Flock footage from public records (which would make this judgment moot) makes sense in the light of red light cameras already being excluded by the same legislators. However, I'd like to see a more incisive law covering both that would compel a reasonable amount of public insight into the footage.
The defense of the photos not being government business until accessed seems shaky.
It's reminiscent of the NSA's argument that data "collection" occurs only when a search is performed on existing "gathered" data. File under "Stuff that's only legal when the government does it."
My prediction: “It’s not a “search” when an AI looks at the stored database and does sentiment analysis, because an algorithm doesn’t violate your privacy. It’s only a “search” after it’s flagged and an actual human with human opinions sees your private chats criticizing the supreme leader.”
What should I be reading?
Scott Alexander has a decent article (or rather a guest blog post) at https://www.astralcodexten.com/p/all-lawful-use-much-more-th... that brings up the subject in the context of Anthropic and OpenAI's dealings with the Department of War (sic), and how their contracts with the DoW might be interpreted with regard to mass surveillance of Americans.
Worth checking out. I'm not personally knowledgeable enough to vouch for the veracity, though.
An older article was reposted here yesterday:
https://news.ycombinator.com/item?id=47210572
Please contact your representative in Olympia and tell them to kill the bill that would exempt this footage from public record requests. It passed the Senate on Feb. 6.
This is the bill: https://app.leg.wa.gov/billsummary/?BillNumber=6002&Chamber=...
Find your reps: https://leg.wa.gov/legislators/
Washington State Legislative Hotline: 1-800-562-6000
I might be good with legal guarantees, meaning jail time for those involved, that the only place images on these devices went was local to the municipality collecting them and that they were only accessed for very well defined reasons by very specific people.
The core issues are that aggregation and exfiltration of this data means that privacy is dead and the AI world allows analysis for almost no cost. We need an idea in our laws that puts back the limited scope that technology has removed. If the police have to expend one person's worth of time to listen to a wiretap then it really isn't possible to get out of control. We need that level of cost associated with ALPR and all surveillance so that the abuse of these systems doesn't get out of control. Make it appropriately hard and it won't be a problem.
Then the feds come in with a national security law and bypass all those state/local protections and slurp it all up into an AI-powered Palantir database, the very existence of which is classified. Suddenly you’re the victim of parallel construction and don’t even know it.
The database CANNOT EXIST SAFELY. Why don’t people who “might be okay with this IF…” understand that?
Collect the data and it WILL be misused, eventually, with 100% certainty. Has nobody read Snowden’s book? They even have a name for intel agents casually spying on their partners/crushes.
The law does not apply to everyone equally. The intelligence agencies get to break any laws they want without consequences, by longstanding tradition (remember 007’s “license to kill” or the CIA’s famous heart attack gun?). There are NO legal safeguards that can prevent abuse, no matter how you word them, because there will always be some animals who are “more equal than others” to whom they do not apply (“national security carve-out”, “LEO exception”, etc).
Sadly, those to whom they do not apply are now coordinating with the new wannabe SturmAbteilung in what are called “fusion centers”.
> The database CANNOT EXIST SAFELY. Why don’t people who “might be okay with this IF…” understand that?
You have to point out the Jews in Amsterdam. They had nothing to hide--until they did.
https://en.wikipedia.org/wiki/1943_Amsterdam_civil_registry_...
Or Martin Niemoller: a good Protestant German pastor who viewed the anti-theist attitudes of the Socialists and Communists as more threatening than Nazis. And then the Nazis put him in a concentration camp.
https://en.wikipedia.org/wiki/Martin_Niem%C3%B6ller https://en.wikipedia.org/wiki/First_They_Came
There is another looming threat of modern day surveilance: previously hidden correlations.
The data you found benign sharing in the past might allow unpleasant conclusions in the future and might not even come from you personally. Think about what toys you bought for your kids, or in what college milieu your worldview developed.
"Using metadata to find Paul Revere", a what-if of the American revolution.
https://kieranhealy.org/blog/archives/2013/06/09/using-metad...
This is a myth that continues to be propagated. Using previously unobservable correlations is a bedrock concept in Federal investigations in the US
https://en.wikipedia.org/wiki/Parallel_construction
And making it easier, better, faster, and more common isn't going to be a good thing. Reminding us that this already happens in smaller scale isn't really adding anything helpful considering the entire goal is to prevent it happening more in the future.
URL is 404'ing. Another article..
> Cameras that automatically capture images of vehicle license plates are being turned off by police in jurisdictions across Washington state, in part after a court ruled the public has a right to access data generated by the technology.
https://www.geekwire.com/2025/washington-state-cities-turn-o...
Awesome. I think I'll put in an open records request for the cameras down the street in my little Wisconsin town. See what happens
Funny I was thinking of doing that in my little Wisconsin town too. Howdy sorta neighborish HN user.
Wonder if we should coordinate doing it simultaneously in like 10,000 cities and towns?
And my axe! Let me know if you do. (Also WI)
*Flock (YC S17)
https://www.ycombinator.com/companies/flock-safety
For reasons I have been unable to articulate, I have believed that if license plate readers are a thing, someone should open-source one that the general public could share.
When a comment comes back that someone malicious will use the tech to stalk someone I really have no answer for that. That is indefensibly a bad thing.
Of course the tech exists nonetheless (not open source as far as I know) and it could be argued it is even being used maliciously—or at least without the kind of judicial oversight we assumed would be in place. But one actor behaving badly does not justify the use of the tech by everyone.
I read about the citizens of Minneapolis using decidedly low-tech means to track ICE vehicles prowling their community and it suggested a scenario where an open license-plate-tracking solution could "balance the scales" a bit (that is if you believe the scales to be imbalanced).
I imagined not a license-plate reading dragnet but rather software where you had first to enter in very specific strings of ASCII characters and the software would only announce when there was a specific string match from the camera.
To that end I vibe-coded an app for iOS in about 15 minutes using iOS's Vision framework and the built-in phone camera. Anyone could do the same.
Nonetheless I only tested it with "HELLO" and "WORLD" using scraps of paper in my kitchen and never tried it outside as the craziness in Minneapolis seemed to have quieted down.
I moved on to other projects.
There had been a fairly decent open APLR library called OpenALPR. It used to be updated fairly regularly, but then was "acquired" and now sits abandoned for almost a decade without any updates, while the company that acquired it (REKOR) has commercialized it and extended it as a closed source API and platform.
ALPR is fairly trivial now days, and a modern CPU can process thousands of plates per second (dozens of plates per frame, hundreds of frames per second coming multiple cameras pointing at different lanes of traffic).
The "product" around all of these is linking the plates traveling together, tracking routes through various cameras, identifying common travel patterns, and more disturbing travel patterns that are outside norm for a given plate or route (i.e. do they normally drive these 3 cameras M-F, but this Th, they went a different route and stopped somewhere for a while).
All that used to be done by the police/detectives/investigators on their own. Now the AI is automating this, and that is truly terrifying, especially for how often misreads occur.
Ive wanted to know this for a long time. Why isn't there some package I can download to a raspberry pi with a camera, and contribute to a public website that shows things like, oh I don't know real time positions of government vehicles...
I always think that the best way to get stuff regulated is make those in power feel the risks.
So if people started using something like flock to embarrass politicians, business leaders, or newspaper leader writers then suddenly privacy might become a big issue.
don’t worry, normal people will never ever see this data to use it against the powerful.
our public data can only be seen by billionaires and cops, not us.
it can be used against us, but never the other way around. the faster we realize this, the faster we can move out of our “divisive” phase and get back to making billionaires dreams come true.
Indeed - but as compute costs come down, it doesn't have to be that way - ultimately there are 8 billion eyes out there.
There are also public data examples - for example the public data on charter flights or ship locations had people like Elon Musk bleating about privacy.
My issues is with false arrests due to people blindly following AI. People pulled out of their cars are gunpoint due to the AI misreading a single letter on the plate, the lady who was arrested for stealing amazon packages just because flock saw her enter and exit a neighborhood and she had to get doorbell camera to prove her innocence.
Also i think it should be public record about every single point of data flock collects and retention period. I am not saying they have to release the data itself, but I want to know what my tax dollars are collecting, how long they are keeping it, who can access it and who they are selling, er i mean sharing it with.
This appears to be an informative link;
https://www.everettpost.com/local-news/everett-temporarily-s...
The link is broken. Here is a working one. https://www.king5.com/article/news/community/facing-race/was...
This just redirects me to the king5 YouTube channel
Does them removing it simply because it’s public record imply that they were up to no good?
They're not removing cameras.
> For now, Everett’s Flock camera network remains offline, as the debate over transparency, privacy and public safety continues in the Legislature. The bill in Olympia that would put guidelines on Flock's data has passed in the Senate.
Well if they had nothing to hide... /s
> “We were very disappointed,” Franklin said. “That means perpetrators of crime, people who are maybe engaged in domestic abuse or stalkers, they can request footage and that could cause a lot of harm.”
No concern over the dozens (or hundreds?) of cases of police or government employees themselves doing exactly what they’re afraid of here. Strange.
> No concern over [...] government employees themselves
Three paragraphs later someone else is paraphrased as including immigration enforcement agents among the problematic users, and in the current political environment, federal law enforcement being made more effective might be the real problem for state and local government.
State and Local LE are also worrying. It’s common enough that they use these systems to stalk exes and such.
While I agree with the risks of DA/stalkers getting that data, this data is not known for being well protected against LoveInt. Quite the opposite it is usually sold on grey markets.
They know, but if they acknowledge it it would make some people mad. Anyone who works with or it’s associated with ALPR knows why you have to have audit logging and access controls if you give law enforcement access to it.
Or for what can already be purchased from a data broker on the open market.
Somewhat related discussion on Redmond Washington & Flock cameras: https://news.ycombinator.com/item?id=45879101
I am less worried about Flock ALPR (which are aimed in the direction of traffic flow to read rear number plates) as I am about the THOUSANDS of facial recognition cameras installed in the last year in all four directions at nearly every intersection in southern Nevada and many many cities in southern California (LA notably excepted). These are mounted above the stoplights and aimed against traffic at stoplights to read faces.
I mention these locales specifically only because I have directly observed them. I would be surprised if this isn’t also happening in many other US metro areas, given how eagerly DHS/TSA/CBP/ICE are mass collecting facial geometries at every available opportunity.
Flock cameras are advertised as ALPR but have facial/vehicle/etc recognition systems running on them as well. You can see one panning up to track other people near the end of this Benn Jordan video on the cameras.
https://youtu.be/vU1-uiUlHTo
> These [cameras] are mounted above the stoplights and aimed against traffic at stoplights to read faces.
Those cameras are for traffic detection, so the signal can turn green when a car comes. They aren't reading faces.
The statute in Washington that allows these to be used to enforce speeds and signals also prohibited photographing the driver using another or the same camera.
> I would be surprised if this isn’t also happening...
"I wouldn't be surprised if this is also happening..."
Only the second sounds correct to me.
I believe both to be correct but “would be surprised if this isn’t happening” to be a stronger statement than “wouldn’t be surprised if it were”.
shutting down the cameras rather than releasing the data tells you everything about what was being collected. the system was designed around the assumption that nobody outside law enforcement would ever see the footage.
LLM comment.
The fact that they shut it down to avoid it becoming public record proves that light is still the best disinfectant against vermin.
The above link 404's for me, but https://www.wltx.com/article/news/nation-world/281-53d8693e-... works.
"The masses/general populace are the enemy" - once you understand that this is the fundamental belief at the root of the elites behaviour, everything will make sense. Flock cameras and AI surveillance is designed to reign in 'the enemy'.
So whenever these cameras could actually be used by the public (potentially for good), the government is allowed to just shut them down?
Anyone can tell, why were those cameras installed in first place? Some company just said "lol for the fun" or what? Who paid for them?
A mix of public (city councils) and private (think HOAs that then donated access/equipment to the city) contracted with Flock in the past few years. The questions of exactly who, when, and why, are very muddy especially with the HOAs who operate rather privately.
Thanks. I don't live in US and this subject is quite unknown to me
Relevant federal payola:
- https://www.opensecrets.org/federal-lobbying/clients/summary...
- https://www.opensecrets.org/federal-lobbying/clients/lobbyis...
According to the article, the Flock cameras are still in place but are "offline".
Why does that not convince me?
Are there cameras pointed at the offline Flock cameras? I sure hope so because it would be a shame if they disappeared...
The government's power over its citizens should mainly derive from its legitimacy in the eyes of citizens, not from its technological and military superiority over the people.
This kind of ubiquitous surveillance is so dangerous to democracy and the ability of the citizens to do anything without the government knowing.
Great now let’s follow suit in all 50 states.
Red states have zero crime, so they don't need them in the first place. /s
This isn’t a red/blue issue.
Flock is no more populate on the right than it is on the Left.
[dead]