Google’s Nest will provide data to police without a warrant
petapixel.com
petapixel.com
In my experience, these capabilities are only used to protect the politically connected or for oppression.
I’d be less skeptical if the program allowed any homeowner to pull any nest footage that included their property, and also allowed individuals to pull any footage that included them.
Of course, that will never happen.
The next day, they found someone offering to sell the same bikes online (in the same geographical area even). They gave this info to the police, basically "hey here is someone trying to sell known stolen property" and the police told them to try and set up a meeting with the sellers themselves! Called back a week later, still had not even assigned a detective.
$20k is not pocket change, so it does make me wonder exactly what kinds of property crimes they do investigate, if any.
Why would they need to justify anything? They have no one to answer to. Look at groups like the Uvalde police who are internationally known fuckups at this point and they’re still just throwing other people under the bus left and right in response. The police only do their job when they feel like it and that’s usually never unless it’s writing down that they went to guard a construction site for the overtime pay.
I think it's a few things contributing to it. First is not having enough man power. And also a lot of stuff isn't worth the expense of investigating. A stolen laptop, phone, or bike just isn't worth the cost of detectives hunting it down. $20k of police work to get a $1000 laptop doesn't make much sense.
And it’s not like those cases even go to trial. I don’t have the exact data on hand, but I’d be seriously surprised if >90% of small-time drug crimes ended up with a plea deal that most of the time doesn’t even involve pleading guilty to a felony.
Police regularly spend countless man-hours investigating people for selling $20 worth of marijuana. Those are not open and shut cases, they have to spend time building cases, use confidential informants or undercover officers to gather damning evidence that won't be thrown out in court, etc.
Unfortunately this situation is by no means unique.
> That has to be unsettling.
And now you know why there has been a growing movement predicated on the idea that the vast sums of money we spend on police may not be well-spent.
The police get a lot more leeway to do bad things because some percent of society just defaults to giving them the benefit of the doubt.
A town of 300 people I grew up next to, always had their one cop ticketing every car that didn't slow down from the posted 55 mph (but honestly, it was always faster) to 30 mph, or accelerated over 30 before they passed the city limits sign.
The city government loved that guy, because he brought in thousands of dollars every week.
When the NYPD threw a fit over de Blasio, they famously refused to arrest anyone unless "it was absolutely necessary". Why? The police union wanted to hurt NYC's[0] budget.[0] It's the same reason why civil forfeiture is so prevelant.
[0] https://www.theatlantic.com/national/archive/2014/12/the-ben...
But yes, in the general case you're right that law enforcement is a revenue generator.
Corruption is rampant in the "developed" world but we just use different words for it.
Data ownership issues like this in the US will never be pushed by existing political interests, hopefully we'll see more EFF-ish PACs in the future that take a page from the fascists and provide the written laws and bills they would like enacted to state and municipal governments so local representatives can edit the title block and get back to harassing their interns.
Springfield, Massachusetts if you're curious.
I don't know how exactly it's in the US, but in EU security cam footage is considered to be personal information and as result of that GDPR applies. So whoever controls the security camera is the data controller and is required to e.g. provide access to people who were recorded. In this case it would likely be the home owner and unless Google was acting as data processor for the home owner they wouldn't be allowed to process that data.
I don't think you have gdpr obligations if your private cameras are just capturing your own grounds and home.
But it DOES apply if your private cameras capture footage of public places.
Interestingly, it looks like it must apply to all cameras in and around AirBNBs.
2.3 Household exemption
As far I understand it, if the cameras are pointed towards your home the exemption applies. If there is even a partial view of the street it would not apply (like the case is with many doorbell cameras). If there was a fence and the camera did not see above the fence it would likely apply though if there is a gate on the view it might not.
"Google will allow law enforcement to access data from its Nest products — or theoretically any other data you store with Google — without a warrant."
We choose to ignore risks all the time. We also choose to panic about stuff all the time. You're in control of what goes into what pile.
Police don't need a warrant to get into your gmail or icloud or whatever else you host with 3rd parties
Prior ignorance of said policy (which is standard for most users), for one.
Perhaps instead of pinning all of the blame on users, we could have the companies producing labyrinthian ToS contracts written by top-grade lawyers and full of legalese (that no layperson should be expected to understand) shoulder at least some of the blame?
This doesn't even touch on the fact that many topics (as related to data aggregation and privacy) are highly technical and require at least a few years of post-secondary to even begin wrapping your head around (e.g. de-anonymization via large sparse datasets is not something I can reasonably teach my 85-year old parent, nor to my child, both of which use Google services in some capacity).
But, yes... Let's blame it on Average Joe, who just wants to watch their dog for a few minutes while at work and saw an ad on TV about a convenient way to do so. Shame on them for not being both a lawyer and a CS graduate.
I wholly agree.
But we're quickly approaching (and in some cases, past) the point where proper due diligence requires a 4-year post-secondary education in a related CS field, if not more.
We're talking about products that take multiple domain experts several years of collaboration to create. How is it reasonable to expect my mechanic, accountant, etc. to do their due diligence on how that product processes their data, especially when it's processed in a black-box created by several other domain experts, and their only source of information is purposefully opaque terms written by lawyers?
I don't think that's the case here or indeed very commonly. You don't necessarily have to understand implementation details if some core tenet of popular ethics is being violated. One key feature of the domain -- namely that you don't own "your data" and so you don't get to decide what happens with it -- is pretty clearly in violation of principles that the vast majority of Westerners would at least profess to hold. Beyond the motivating principle that third parties should be required to receive explicit whitelist access to use privately-owned data, "implementation details" refers mostly to policy and enforcement, not really technologies.
“It is difficult to get a man to understand something, when his salary depends on his not understanding it.” — Upton Sinclair
This, in effect, nullifies all but the most important components of a Tos.
Privacy to tech like this is very hypothetical till it happens, and it'll rarely happen. If it's not in our faces we won't vote against it.
I think this is the entirely wrong framing. My other comment covers some of it, but specifically in regards to your comment: it's a lack of education, not the embodiment of short-term thinking.
And really, we can't expect every person that uses Google (or whatever other large tech company) to thoroughly understand all of the bits and pieces of technology that could be used to fuck them. Or how things that we've been told are anonymous/private become non-anonymous/non-private when combined with other sparse data. These are complex topics that even many technologists don't understand (or are outside of their field of expertise).
These companies hire top lawyers to write complex ToS, use as many dark-patterns as legally possible, do illegal things until they get caught doing so, evolve their terms frequently, etc. Yet somehow they've convinced everyone to blame the layperson. It's remarkable, really.
What would be really swell is if we could, you know, not have companies spend millions of dollars on how-to-fuck-your-user initiatives.
Ie if we expect corporations to not fuck you over, who is there to enforce that? Who has the power to keep them in check? Okay, maybe Government should hold that role - but who then keeps the government in check? Who ensures that the spying or privacy from the Government is kept in check? etc
Ultimately the buck always stops at the individual. And we have to be hyper aware of long term implications, because money, greed and power has deep, deep pockets (as you also mentioned) and the fight will be never ending.
We, as a community, have de-propritized education, health care, public safety, privacy, etc. Sure, powerful forces have been pushing for that exact thing, but we can't expect them to "just be nice" or w/e.
I'm very pro "Big Government". However my ideas behind big government will not work without individual responsibility. Until then citizens are purposefully and willfully giving their power away with every tiny step. The blame is on us, and our current state is inevitable. My 2c.
The main point I wanted to get across is that it's baffling that companies aren't blamed in these conversations. It's always the user who is blamed ("well you read the ToS didn't you!"). And that's dumb, because the vast majority of users aren't lawyers and don't have CS degrees -- both of which are becoming increasingly required to provide informed consent to a ToS. (edit: in every other contract I sign, a lack of informed consent is grounds to void the contract, exception being tech-company ToS contracts)
If you still want to blame my 85-year old parent for not understanding what Google is doing with his data, go for it, I guess. Just seems stupid to do so, because he barely can open up a web browser but is somehow expected to understand the complexities of data aggregation and what impact it will have on him. And as time marches on, it's equally ridiculous to suggest that he just never use a computer to avoid the issue.
>And we have to be hyper aware of long term implications,
Without post-secondary education in niche fields, this is becoming impossible. Especially across multiple services with changing terms, in countries with changing laws, in a world where technology evolution outpaces curriculum changes.
I agree, but again i go back to, "but how else can it work"?
Of course i don't expect everyone to be knowledgeable on all low level systems. However, to the point of your 85 year old grandma, she is a tiny demographic in a much larger, much more reasonably informed demographic who also completely ignore the implications.
Name a demographic that isn't wildly ignorant of things that are reasonable to know?
But again, i repeatedly fallback to "But who else can do this?". This is why i'm pro Government, but not until people start pushing for responsibility on this front. It may not be reasonable for your grandma to be responsible for Google Data stuff, but she _(and the rest of us)_ have sat around for dozens of years watching authority figures have little to no accountability or oversight.
The issue isn't about Google. The issue is about us, and our inability to build a government and authority system that is in-line with our views. We hand our power over with no thought or oversight and then we're shocked when it all comes back against us. This has nothing to do with Google or CS, imo.
My argument is that the "reasonably informed demographic" is incredibly small. I can only say the same thing so many times, though, so I'm not sure how to explain it in a different way.
To restate my example, even very smart CS graduates may not realize that anonymized data joined with other anonymized data can result in de-anonymized data, because the linking and de-anonymization of sparse datasets is a niche subfield that has only recently begun being explored.
Many people may think they are reasonably informed (they look into the ToS, see that data is anonymized, and decide that they are okay with that) without knowing that the data may later by de-anonymized through advanced statistical analysis they've never been exposed to in all their schooling. So while they thought they were informed, they weren't. This repeats across several domains.
>But again, i repeatedly fallback to "But who else can do this?".
Why is that when a problem is identified, people demand a solution be provided at the same time? I don't have a solution, sorry. But that shouldn't preclude me from identifying a problem.
I honestly did not expect saying basically "Let's put some of the blame on Google, because they're the ones with the dark patterns and lawyers and experts, rather than solely blaming the layperson" would be met with much pushback.
I think we're in agreement here. To be clear, i'm mostly talking about intent, an attempt to stay informed and a willingness to act - to push for centralized leadership who is informed.
Ie as i said before, your grandma is not expected to know this. She is expected to fight for a government that will be, and that will also be able to be held accountable.
We have neither the oversight on government current, nor the willingness to act. Your grandma built the same world we are building today. One of inaction and obfuscation.
If society cannot be informed and active on what is essential to build that world (whatever that may be), then we are doomed. Currently, the population at large is not. At least, not from what i can see in action.
Who defines "reasonable"?
When you get delayed on a flight due to a maintenance issue, are you equipped to determine if that delay was reasonable? Most likely not, although many mechanically inclined people may be in a position to make that call. Those same people may not be in a position to arbiter the reasonableness of Google's ToS (side-stepping the whole obfuscation of details that was previously covered).
When society gets reasonably complex, we out-source those decisions. In the example of the aircraft, we have a regulatory body who makes the rules about what is reasonable. It wasn't always like that, of course, but the need grew out of the growing complexity and risk profile. So to your question and an earlier point, there may be room for regulatory bodies as an alternative for "how else can it work?".
If you cannot determine factors by which outsourcing is successful or not, by which it is accountable or not, democracy fails, and you can no longer outsource it.
This isn't about being informed on obscure topics. As i said this has nothing to do with Google. It's about our willingness to fight for a government that can handle this, and fight to control said government.
This is absolutely by design and part of a larger pattern of propaganda that keeps Americans scared of the government and in love with the idea of becoming billionaire CEOs themselves because it's "moral". That holy "free market" has rewarded those rich people for being some damn smart and efficient--they deserve it, not the damn communist free loader leftists who hate America.
-Ron Swanson
/s, obviously
That's an odd take, I honestly don't find anything about this article, or the broader topic of privacy and overreach by companies and law enforcement, amusing in any way.
Looks like the newspeak got you.
This includes ALL your data. Gmail, Google, Android.
So unless you're opting for iOs (provided they're not doing the same as Google here) and not using Gmail or Google you're still falling under "Surveilled by the gov via tech company who serves their interests and not yours, even if you pay money for their services".
GrapheneOS isn't for everyone. Neither is iOS, or Android with Google, or the Pinephone/Librem 5. And that is fine. It's good to have options.
It's a great strategy as Google makes 80%+ of it's revenue from Ads, and their ad model does not work well with a privacy first mindset.
Apple designs with privacy in mind. Google designs with invasiveness.
0: https://www.barrons.com/articles/apple-revenue-ads-516478777...
That Ads devision can operate in a privacy first mindset while still seeing huge growth YoY. Where-as if Google operated in a don't be evil/privacy first mindset, they'd loose substantial revenue and likely no longer be profitable.
I find that hard to believe when so many of their devices' functionality depends on you sending data over to them. Unless you go out of your way to make sure you're blocking all your devices from phoning home or sending any data over to Apple, then any supposed privacy benefit becomes a lie.
Either you're the only owner of your data or your data is not, by definition, private to you.
I'm afraid your conviction in Apple is the product of a well-crafted fantasy by their marketing department, instead of based on some deep rooted philosophical belief regarding the rights and privacy of their users.
Leith, D. J. (2021). Mobile Handset Privacy: Measuring The Data iOS and Android Send to Apple And Google. URL: https://www.scss.tcd.ie/doug.leith/apple_google.pdf
i.e Right to repair, Side Load Apps, no draconian app store policy, etc etc etc
Embrace both privacy and user freedom. that would be great
Is the phone listening anyway? Maybe, but that violates a privacy expectation, and there may be recourse if someone discovers it's doing that.
This is all publicly available info, and perhaps there’s no reason why you should trust me any more than you trust Amazon as a company, but as one privacy-conscious engineer to another, I promise that your ambient conversations are not being stored or sent to Amazon and that any data you delete in the app (either by specifying an auto-delete period or manually deleting it) is actually, really, truly deleted.
A process running locally on your Alexa device listens for the “wake word”.[0] This audio is only processed locally within a constantly-overwritten memory buffer, it is neither stored nor transmitted. Only once the wake word is detected does Alexa begin to transmit an utterance to the cloud for processing. I’ve worked with the device stack and it really isn’t transmitting your ambient conversations.
Within the Alexa app[1], you can see and hear all of your stored data and can delete any of it. You can also control the duration after which it is auto-deleted. From working with ASR datasets, I can confirm that deleted audio (and the associated text transcript) is really deleted, not just hidden from your view.
I never owned an Alexa or other smart home device before I worked on it, and I’m not sure I’d buy another company’s device where I lack the same ability to “trust but verify”, so I’m not sure how much weight my word should carry. But I can give my word that Alexa is not transmitting your ambient conversations or just setting “deleted=true” in a database when you tell it to delete your data.
[0] see page 4 of https://d1.awsstatic.com/product-marketing/A4B/White%20Paper...
The government isn't particularly interested in ending this problem either, I suspect this is due to a combination of industry capture and intelligence agency interest in these products.
Oh, but there is. It's subservient to the manufacturer, not to you.
I'm still annoyed that no country decided to classify "selling" things where the manufacturer keeps complete control and denies you access as fraud. But just because no legal system decided it's a crime, it doesn't mean it's right.
You think the problems are a mistake or otherwise something to be "fixed" the products are working exactly as both the government and the manufacturers want them to, and it has nothing to do with "intel agencies"
Having your entire life "cloud connected" and them complaining about privacy, is like opening a window then complaining that the house is drafty.
I love home automation, not a single component of my home automation is cloud connected, if more people would accept, learn and support non-cloud systems, services and protocols everyone would be better off
They say there is nothing wrong with the technology, but then go on to list things that are in fact wrong with the not-technology.
The problem with the technology is it allows the company, political power, or police to access the data with out user permission, that is a technological problem, belief that just passing the correct laws to resolve this technology problem defy's recorded history and logic
[citation needed] because basically every Google (and Nest) product have used pretty decent protocols since Snowden leaks.
Google and Nest use transport encryption to ensure only the device and google can see the data, but once it stored google can see everything
Zero knowledge data Encryption would require that ONLY the owner of the device has access to the data, not google
and example of this is say BitWarden Password Manager, even though the data is stored on BitWarden Servers, No one at bitwarden as access to any users password vault, The user holds the key, and only the user can view the passwords.
• better if you could please provide links to how-to starter guides on how to do this
• I share your concerns, but one obvious huge drawback to a home system that's 100% off-cloud is security camera footage: how do you prevent a violent home invasion or burglary where the first thing they do is at gunpoint force you to show them where your server is so they can disconnect/destroy it? Not having a solution could easily be a $50,000 mistake or worse. Perhaps you just live in an area where violent home invasions are not a regular occurrence. It's not just about porch pirates stealing packages, or thermostats. - Home Assistant
- Encrypted NVRI would love to see a source of this being a common tactic of thieves, home invasion in general are extremely rare as most thieves are cowards and want to attack soft targets i.e unoccupied homes. Then of the home invasions I am aware of, I know of ZERO where the "first thing they did" was force the homeowner to show them the location of the storage server.
Even if that is a "common occurrence" which I doubt, what stops a home invader from cutting off the internet before the attack? and many of these cloud connected cams are wifi, there are several very easy attack vectors to knock them offline. I think your strawman is weak and easily defeated in a number of ways
>>> better if you could please provide links to how-to starter guides on how to do this
Some of the Technology I use, or sources I visit to look for new things
https://www.home-assistant.io/
Basic statistics: [1]: The US averages ~1 million burglaries/yr, 3,300/day, or one every 25.7 seconds. (Most burglars are unarmed; most properties are unoccupied when robbed; most burglaries don't involve violence; most burglaries only last <10min; only 12% of home invasions are planned in advance; 88% of burglars may be robbing to support their drug habit; 85% are not professional burglars)).
See [2] for an up-to-date dataset of individual incidents involving recent deaths or injuries in violent home invasions, listing no. of casulaties and deaths.
See [3],[4] for aggregate US statistics from 2017-2020, citing FBI/DOJ.
Some complicating factors in crime-tracking statistics [5]:"Elusive statistics":
> "The term "home invasion" isn’t standardized across jurisdictions, or universally defined; or necessarily used in crime-tracking data. Some jurisdictions use it and some don’t, but there is no universal definition or dataset. What would be called a home invasion is often reclassified as the eventual crime committed by the suspect — a robbery, a physical attack or so forth."
According to [4]
> 7. Only 7% of home burglaries involve violence. > A State of the USA report found that 7% of these (on average) resort to violence on a yearly basis; though, only a few (12%) burglars had used a firearm. The most common type of violence present in burglaries is a simple assault, with 36% of cases resulting in only minor injuries and psychological trauma.
I see little point in arguing about how "rare" or "common" this is; personally I know people this happened to. You could argue that when we subtract cases where the occupants shot the burglars, and exclude that some of the fatalities are homes of drug dealers or people who are known to carry large amounts of cash, that the overall number of injuries law-abiding-occupant home invadees who were threatened, injured or killed is "comparatively low".
For example, [6] reporting from ABC7 on a recent string of violent home invasions in South San Jose, CA; this sort of thing is admittedly rare:
- *"5/31/2022 home invasion in the 600 block of N Capitol Ave: 5-6 suspects bound an elderly couple with belts, brandished a handgun, assaulted the victims and stole valuables.*
- *The suspects, using a stolen vehicle, [then] committed a second home invasion in the 1000 block of Summerdale Dr, where they held a male victim and his 15-month-old child at gunpoint... threatened to shoot the child and kidnapped the victim, forcing him to go to his bank to withdraw money. The suspects also stole various valuables from the home."*
Nothing (other than speed, preparedness and competence; home invasion crews do not in general act like Ghost Recon). But that's missing the point; the truncated footage will still show when and how the home invasion happened, the number of criminals, their general description, may also show vehicles or weapons.
None of us are interested in debating strawmen; please just tell us your suggested best-practices in home security for self-hosted/non-cloud solutions.
[1]: https://www.bankrate.com/insurance/homeowners-insurance/hous... "Bankrate.com: Burglary statistics 2022"
[2]: https://www.gunviolencearchive.org/query/2109008e-ddd0-4318-... Gun Violence Archive: Home Invasions [i.e. the subset of home invasions that involved a gun]
[3]: https://comfyliving.net/home-invasion-statistics/ "21 Important Home Invasion Statistics to Consider in 2022" (cites statistics from recent years)
[4]: https://policyadvice.net/insurance/insights/burglary-statist... 27 Alarming Burglary Statistics for 2022
[5]: https://www.politifact.com/factchecks/2017/oct/25/michael-mo...
[6]: https://abc7news.com/south-san-jose-home-invasions-robberies...
You can still have off site backups and recordings on an innumerable count of hosting or storage solutions, encrypted with your own keys.
With these cameras and recognition algorithms, you don't even need people to go to the doors. Just pull the feeds.
TikTok could be a spyware (lol) that requires your SSN and people would STILL download it and defend it just because it brings them mindless 10 second videos. I remember reading 1984 as a kid and thought it was so far fetched, that nobody would willingly let society get to that point... but it just only made more sense as I got older... people really just don't care...
https://stevenmcollins.com/auto-license-plaste-to-have-rfidt...
https://www.nbc12.com/story/20500107/virginia-lawmakers-want...
NYC was trying to do congestion pricing, in which case it would be pervasive.
It is still possible to avoid, but difficult.
Sometimes I think people covet privacy for its own sake, and don't think about the practicalities. The whole point of living in a collective society is that we give up some freedoms for the sake of overall increased prosperity, that's always been the tradeoff, and this is just one of those tradeoffs.
https://louisville.edu/law/library/special-collections/the-l...
I'm not denying the right exists, I'm saying we give up our other rights all the time for the benefit of society, why is privacy any different?
Further, it's a spectrum. You're not putting a camera up for police to peruse at their leisure, it's only in specific situations.
I couldn't find evidence that mass surveillance is good a society.
“Convenience, less human hours wasted on boring stuff” is fine as an individual consumer mindset, but does not form sufficient criteria for evaluating complex social systems.
Kinda-good, 1, so and so can just go check the camera that points at the central plaza fountain that anyone can access, and sees that his spouse has arrived and is waiting for him as agreed.
or Bad, 2, cameras all over that everyone has no idea who controls, watches, and/or is recording
In the vein of practicalities, both Google and the justice system (USA for me) are monstrously large bureaucracies known to make difficult-to-redress errors. Google's capricious account banning, police raiding incorrect addresses, eg. The decision to share with them more information than the law requires is one I'd prefer to make myself.
Or just plain out refusing to fix errors where they would be relatively easy to fix; compare Scalia's "it's fine to fry a provably innocent person as long as the procedures are followed" argument.
A numerate person would know how rare these things you're afraid of are, and not let those fears drive how they live. I (hopefully) follow that path, and I recommend you check it out!
I read what you wrote in the same way I suspect you would read someone who is afraid of space because meteors have killed people (as a rough example).
It just doesn't seem like the rates at which the things you're worried about are happening in a volume that would actually matter to a society.
This is a step beyond that. Warrants are granted at the discretion of a judge, the bar is high, the scope is narrow and you (theoretically) have recourse if it’s abused. Here, the discretion is Google’s, the bar is nonexistent, the scope is unlimited and you have zero recourse if you think you’ve been wronged.
This wouldn’t be an issue if people trusted Google or the police. But they don’t, and it’s pretty easy to imagine ways in which this could be abused to harm people.
Let’s say you live in Texas and get abortion pills in the mail. If the police have a warrant to search your house for something unrelated, they (theoretically) can’t see the pills and decide to charge you with an unlawful abortion (unless they were “in plain view”, etc). But if Google gives police access to footage of your house extrajudicially, police can use anything they see as evidence against you. And make no mistake — things like that will happen as a result of this policy.
Google is smart enough to know that "snitching" on its users is bad for business.
Think "track a burglar as he moves through a neighborhood" not "snoop (illegally) on the contents of people's mail".
Just because you currently seem to have a "failure of imagination" does not mean that law enforcement, corrupt/facist/dystopian government officials or even unscrupulous employees within the tech sector itself will not absure their observational powers now or in the future.
Or maybe you just haven't been paying attention to the news for the last "n" decades.
We all deal with "sky is falling" person on our development teams, and it's pointless doomsaying then as it's pointless doomsaying now.
When the bad things happens, we can react. Until then, let's try to stay reasonable.
It’s not illegal, though. Google is (presumably) fully legally in the clear to just hand over footage to police. That means that, if Google decides to hand over your footage to the police, anything on tape can be used as evidence against you. And Nest offers indoor security cameras, so your entire house could be fair game.
> Google is smart enough to know that "snitching" on its users is bad for business.
Is it bad for business? That’s not clear. Your whole argument is that this is fine so long as you think the likelihood of abuse is low. My guess is that it actually won’t hurt Google’s business at all, even as we start to discover police misusing this.
“It only happens to a handful of people! And anyway, they were [doing drugs/stealing/etc], so they deserved it. It’ll never happen to me!”
My argument is not that it's fine as long as the likelihood for abuse is low, my argument is that it's fine as long as there hasn't been any actual abuse. When something does happen, we can respond to it.
Until then, it's not reasonable to go through a bunch of worst-case scenarios.
All I’ll say is that I’d prefer to stop the thing that will very clearly harm someone before that actually happens, not pretend we haven’t seen a million times before what happens when you give police more power.
It sounds arrogant, to me, to be so sure you know what the outcome will be of a given situation, considering all of the necessary ingredients for that outcome have existed for some time and what you claim is "inevitable" has still not happened.
"Convenience and if I go missing mysteriously Google will hand info over to the people trying to find me? Win-win."
Because every argument I've seen along those lines fall apart after drilling down a few layers.
Perpetual and pervasive surveillance completely changes the human expectation of public spaces.
For most of human history, even the most public of spaces was effectively mostly private most of the time. A few decades back you could be in, say, Times Square between thousands of people and yet your actions and presence was completely private. Because unless you did something particularly attention-grabbing, nobody would notice or much less remember. Even if you did do something that people took notice, few of them would really remember you individually a day later (let alone years later). If the cops wanted to interview people about your presence they'd have to find those few that noticed and remembered you, nearly impossible. Thus even public spaces were for most practical purposes private space, and that is the expectation human minds has.
You might say even decades back they could assign people to watch you. And they did. But only for select people so there had to be some cause and selectivity (sometimes unjust, but still). It was literally impossible to watch everyone everywhere all the time.
That's what changes now. It is possible to track everyone everywhere all the time. And not just at the moment, but recorded for posterity for future ML queries.
So this is just an article pointing out something in the TOS.
National security requests are common for any big company[0-2] since they'd rather play ball today, under their own terms, than resist and trigger new legislation that forces them to hand over information in any warrantless circumstance.
0: https://www.apple.com/legal/transparency/us.html
1: https://transparencyreport.google.com/user-data/us-national-...
2: https://transparency.fb.com/data/government-data-requests/
Even in the worst case (law passes, court upholds, and people re-elect the politicians) at least we have a definition somewhere on what is and what isn't ok. Today where it's ambiguous and "Company X may or may not" leaves a person to wonder.
Exactly what it takes to be reasonably convinced that I would be saving someone's life is of course not a simple question, but in at least some situations it seems like it would be worth the tradeoff.
I'm not going to go into any detail for people's privacy, but it took over an hour to figure out who a regular member of a community was in real life and then get emergency services in another country to them. Got independent confirmation they got dragged to a hospital in time. They weren't exactly grateful afterwards, but they stayed in the community and didn't repeat the attempt.
Why won’t Google Nest and Amazon Ring do this?
I have 6 Nest cameras throughout my home ( not the doorbell product)
It’s one of those perverse things that they are some of the few companies that actually don’t sell user data, and yet they’re the ones that are most frequently accused of it.
I don’t think it helps us in keeping big tech companies to account when we spread false information about them.
You said it yourself... How would you analyze the data if it's encrypted?
https://support.apple.com/en-gb/guide/icloud/mme054c72692/1....
https://support.apple.com/guide/icloud/set-up-homekit-secure...
Yet it is still end to end encrypted because the processing happens locally on device (on the hub) instead of on a cloud server.
I have the ability to detect people, animals, vehicles, and packages and set recording settings based on that. In addition to if I am home or not based on the existing detection round that built into HomeKit.
This narrative around Apple's products missing major features is no longer true and has not been for a while.
But it seems to work decent enough. However the face recognition works off of the faces you have categorized in photos. So actually recognizing specific people will only be as good as that database is. I assume in the backend it is largely the same tech so you can get an idea of how well it will work based on that.
I can't compare it to other offerings since I have never used them, but it does what I need it to do.
I have had issues with false positives, but I also am using a not great quality camera with it so it is limited with that. I plan on getting a better camera but just have not had a need for it yet.
Also, can Android users use it?
Technically yeah you need to leave the Apple TV running (or a HomePod) but it just stays in standby mode anyways. But I don't know 100% how that works power wise.
When I look in my Home app at the hubs, I see all of my HomePods and Apple TV's there. The living room Apple TV is marked as "Connected" and the others "Standby". Assuming that means the Apple TV is the one processing things, it does not appear that the Apple TV is doing a ton. The light isn't on and it isn't generating really any heat.
To my knowledge Android can't use HomeKit but maybe there are third party ways to use HomeKit with an Apple hub. But I would be surprised if those third party ways could access the secure video. But I honestly don't know.
I don't see how that's possible. See the instructions at:
https://support.apple.com/guide/icloud/set-up-homekit-secure...
Nothing in the process requires a token/authentication/blessing be done from a pre-existing hub machine that would have to control the encryption. Apple can remotely grant your device HomeKit cloud video based on nothing more than your ability to attest ownership of your new device.
And if they can do that, they can certainly recover the video internally for anyone who wants it. They just choose not to, and you believe them (where you clearly don't for Amazon).
Also, they can charge police for these data.
I won’t buy their products because of their decision to monetize over privacy. What I find interesting is that their products aren’t cheaper than other privacy preserving products from Wyze and Apple so this extra revenue is pure profit or gravy. This isn’t some essential need of their business model, they just don’t respect user privacy enough to avoid this extra revenue.
UX might be a bigger issue, as it is harder to view this data in any device and still keep keys unaccessible by Google. But still not too hard. People use Whatsapp web in daily basis.
Main problem is the ML use for video feed to trigger some stuff etc.
> In response to a request for information if we believe disclosure is in accordance with, or required by, any applicable law or legal process, including lawful requests by public authorities to meet national security or law enforcement requirements;
IANAL but that sure sounds to me like they will provide data in cases where they aren't legally required to (as in a warrant) as long as they believe it's "in accordance with" the law. Which could mean almost anything other than a clearly illegal request.
https://www.wyze.com/policies/privacy-policy#c
I like Wyze and think they are acting in good faith, but I wouldn't assume they would say no to a request from government/police. They may offer local recording and RTSP, but AFAIK they are still cloud connected unless you otherwise block them or use alternate firmware and hence not private.
Ring says it offers end-to-end encryption [1].
[1] https://blog.ring.com/products-innovation/ring-announces-end...
Unfortunately, either Ring or Nest is required in my neighborhood given our property crime issues.
What do you do with that footage? I have a full resolution video of a dude stealing $300 worth of crabbing equipment from my porch. What’s next? Police is not going to do anything about it even though there’s probably technically a way to track him down.
https://support.ring.com/hc/en-gb/articles/360054941511-Unde...
Many of these features are key selling points, so most consumers aren't going to be interested in turning end-to-end encryption on.
It also doesn't work on battery-powered devices, presumably because they offload as much work as possible to the server to save power.
It's a sensible design choice given the current constraints of the hardware.
Really almost all the top comments on this topic are people overreacting to the headline without realizing that the circumstances involved are emergencies and not just routine LE requests. Clickbait has ruined everything.
Every supposed attempt to protect you is really just an attempt to justify an inherently unethical business model because it is profitable.
They'll try to notify users? After their data has been shared with police.
Well that's reassuring.
If Google had a magic system for getting information to a kidnapping victim reliably, they wouldn't need to share user's private data with law enforcement, they could just share the information on where to reach the kidnap victim.
Not at all - it's merely somebody's _claim_ that there's some danger.
FWIW, anyone still doing business with Google probably has a relatively high trust of Google, so that comparison is probably closer to equivalent than many might imagine for Google's users.
What this will predominantly be used for is abusive partners, who also happen to be law enforcement officers (police) to terrorise their victims, or, more generally, abuse of powers by LE.
Here in Australia the police already have 30% more domestic violence charges than the average, and have been caught way too many times abusing their powers as LE.
Also, a cop doing this would leave a trail of evidence of their irrefutable guilt.
You're being hyperbolic.
Fine, every routine LE request is now an emergency. Because one thing US LEOs are known for is their reserve in only requesting just the absolute minimum to do an investigation.
"A robot may not injure a human being or, through inaction, allow a human being to come to harm."
“A provider like Google may disclose information to law enforcement without a subpoena or a warrant ‘if the provider, in good faith, believes that an emergency involving danger of death or serious physical injury to any person requires disclosure without delay of communications relating to the emergency,'”
s/provider like Google/robot/
s/may disclose.*if the provider/may act to protect any person if the robot/
s/requires disclosure without delay of communications/requires action/
“A robot may act to protect any person if the robot believes that an emergency involving danger of death or serious physical injury to any person requires action without delay of communications relating to the emergency,'”
What's the penalty for police lying to Google?
But it seems like Apple is the only cloud based offering that cannot actually respond to these requests even if they get one?
I just checked my HomeKit Secure Video camera and it seems like it isn't just the default setting, but the only setting. I see no way to tell it to not encrypt the data when storing it in iCloud (not sure why they would give that option at all TBH)
This is in line with the reasons cops can do things that otherwise normally require a warrant. This is generally 'a good thing.' The issue, for me, is in the abuse and interpretation of these exemptions.
But would they actually do it?
Recall at the time I mentioned to friends… I bet police get direct access and within 5 years, here we are lol
Many utility companies still do https://themoneyninja.com/free-google-nest-thermostat/
I only have nest protect (stove heating) with occupancy sensor/pathlight and I really dont think it will record motion events nor submit them later (while connected). Could be wrong.
Facebook also paid me several hundred dollars because they were auto tagging faces for Illinois users.
Check out the Illinois BIPA, it's the reason tech companies are trying to get federal privacy legislation that preempts state law.
The friendliest city I've ever lived, amazing public transit and where I met my first real girlfriend.
We have facial recognition stations too.
:(
Edit: This same issue was also a problem in the 1990s with private CCTV! If a police officer or detective tells a business owner that a crime has been committed and there might be evidence on the tape, the owner doesn't have to show the police the tape without a warrant. But they usually did, because it looks suspicious if they don't.
Edison Carter: Security Systems has its tendrils into every element of our society - the government, our homes, the police, the courts - I'm not gonna spike this story just because it deals with dollar amounts beyond your comprehension! It's too important!
Murray: ...cerebral...
Theora Jones: Murray, we're trying to play this takeover as a threat to our average viewer. Nobody knows who's doing it. I mean, we all deal with SS every day - what if some really dangerous people got control of it?
Murray: Who do you think controls it now?
Folks - while it's definitely disturbing from a legal perspective and needs to be addressed 'proportionality' is also something we have to consider.
If this only happened '7 times' in a year, I'm doubtful this is arbitrary handing over of data.
'Real Life' is complicated, and sometimes processes don't line up as we would like. Neither Google nor the Police are corrupted institutions, rather, they can in some instances do things we don't want them to or that are actually bad. Meaning, I suggest in all likelihood there was likely some kind of material reason fore these '7 disclosures'. They used the term 'emergency' and I buy it.
What we need to do is close the loophole and adapt the judicial process so as to be able to accomodate such situations.
People are yelling about 'Orwell' - well that Dysoptia takes more than '7 times a year' in a nation of 350M to have surveillance abused on them. Again, needs to be sorted out, but we have to contextualise this.
I moved off Gmail to them about 9 months ago and it's be good.
The worst part is that Dreamhost's spam filtering is abysmal compared to Google's. But my desktop is always on and I just keep Thunderbird running on it. TB's spam filtering is pretty good after training.
I figure some of the dedicated email hosting services do a better job with spam.
These are exigent circumstances, which allow police to enter your home without a warrant. Extending that access to include video cameras seems reasonable.
This is a "In Case of Emergency Break Glass" type of situation.
If a police officer comes to your house without a warrant are they allowed to break in? If not they shouldn’t be given access to my camera feeds without my permission.
The move by google here just re-affirmed that I made the right choice to not buy their products.
Absolutely. If they hear gunshots or screams of “help! Help!” they will break the door down. I’m sure there are other triggers, too.
https://support.ring.com/hc/en-us/article_attachments/360053...
looks like someone just mashed the keyboard after a long day of work.
That being said, that power is obviously easily misused without checks and limits.
You can see it. No theorizing. Click the big red "Submit Emergency Request" button.
The problem is there isn’t any broken glass here. No judge is stamping this, and I see no duty to warn the customer about what device and data was accessed, and there doesn’t seem to be a way to opt out ahead of time. This removes a lot of checks in the name of expediency.
You can get a lawyer, go to court, and demand an accounting.
You don't have that control here. Your footage can just walk off, no clue. No accounting, no reporting, no post-hoc punishment for mis-use.
> You don't have that control here. Your footage can just walk off, no clue. No accounting, no reporting, no post-hoc punishment for mis-use.
The supreme court recently ruled that a woman beaten by police didn't have standing to sue. Good luck with breaking qualified immunity in a court of law in the coming years.
The only difference being that when you pull the alarm for lols there are consequences -- for you. Members of law enforcement on the other hand do that on the regular and face zero consequences.
That said, Google doing this might ease the pressure to enable this across the board through legislation. So maybe a positive in the end; I wouldn't really expect any privacy with Nest anyways.
The "process" to which you are "due" when it comes to these things is if the police try to use this information to prosecute you, they have to explain it in court, whereas if they obtain it through a search warrant, they have to explain themselves in advance. It's all "due process".
I sympathize with this dilemma. It's a real-life version of the Trolley Problem[1].
Or imagine an idle threat is made off-hand in a private gmail, so pre-emptively disclose that threat to law enforcement?
In the sci-fi Minority Report scenario, Google could use AI language models on all gmail to make a judgment call on "emergent danger of serious physical injury". If I plan to go bunjee jumping with an unreliable provider, Google scans my gmail and tells law enforcement?!?
I know a person who was banned from twitter and was nearly fired because he dead-named somebody. The explanation was that it was literal violence against the person he dead-named, because so many trans people self harm if they don't feel that their identity is affirmed. The person who was banned says it was a force of habit because the person had only very recently went public with their new identity, but it did not stop the ban hammer. He almost lost his job as well because the tweet was posted in a public slack channel. HR left it up the person who was dead-named to make the decision, and they decided not to fire him.
I could absolutely see a scenario like the above happening in gmail or some other Google product, and the decision-maker deciding that it was a threat to a person's life and should be disclosed. To some people that makes perfect sense. To others it does not. The point is simply that when we let humans make subjective decisions like that, we get all the downsides of human judgment.
You say they hand all the information to the police to prevent package stealing? Iam pretty sure thats not the reason at all, maybe a very very little part of it.
I'm done with all these devices, yet I know that my phone will still have this ability even if I've disabled the hotword functionality.
Luckily for us, there are still the single board computers with cameras attached to them, which can do the same job without the could.
At this point, what doesn't Google know or is unable to accurately infer about any given person on the internet?
https://epic.org/documents/google-purchase-tracking/
I'm sure they have other programs I haven't heard of.
Adding this for the people who keep downvoting this:
https://www.mintpressnews.com/national-security-search-engin...
"Google – one of the largest and most influential organizations in the modern world – is filled with ex-CIA agents. Studying employment websites and databases, MintPress has ascertained that the Silicon Valley giant has recently hired dozens of professionals from the Central Intelligence Agency in recent years."
We are imprisoned by our addiction to convenience. I am in the middle of dealing with this now, switching my services where I can: dumb phone, linux laptop, swisscows, openstreetmap and an old Garmin GPS, and just being OK with uncertainty.
I want this. Which model did you get and does it have up-to-date maps?
> Apple’s default setting for their doorbells is end-to-end encryption which means the company is unable to share user video at all.
I had a Nest at my previous house - decided to try the Wemo Homekit doorbell this time and I'm very pleased with it. I also use an Apple TV to drive all the content on my TV and the Homekit video notifications are amazing - instantaneous since the Apple TV is my local homekit hub. Way faster and far more reliable than the Nest doorbell ever was.
You’re done, and you’re welcome.
If you’re really lazy at least use something like Apple HomeKit with supported cameras.
You get both security AND privacy.
We may share your personal information in the instances described below
Third parties as required to (i) satisfy any applicable law, regulation, subpoena/court order, legal process or other government request, (ii) enforce our Terms of Use Agreement, including the investigation of potential violations thereof, (iii) investigate and defend ourselves against any third party claims or allegations, (iv) protect against harm to the rights, property or safety of Eufy, its users or the public as required or permitted by law and (v) detect, prevent or otherwise address criminal (including fraud or stalking), security or technical issues.
If we should ever file for bankruptcy or engage in a business transition such as a merger with another company, or if we purchase, sell, or reorganize all or part of our business or assets, we may disclose your information, including personally identifiable information, to prospective or actual purchasers in connection with one of these transactions.
We may also share information with others in an aggregated and anonymous form that does not reasonably identify you directly as an individual.
That itself is the most bizarre behaviour.
It's very rough around the edges but if you are up to contribute it's a good base.
https://github.com/tcz/choicecam-server https://github.com/tcz/choicecam-client
The client is a React Native project, should run on Android or iOS.
Most consumers don't want to (or can't) deal with managing encryption keys, and aren't going to be keen on the idea that if you lose the key, you lose the data permanently.
[1] https://www.reuters.com/article/us-apple-fbi-icloud-exclusiv...
Most people in tech still use @gmail addresses, do you think regular people bother to have their own private cloud for data storage ?
Granted, they could be using Proton or some other alternative. But you still have the same inherent problem that someone else hosts/owns your email data.
Of course that doesn't help when you've had a Gmail address for 15 years and can't get rid of it
The risk surface for self-hosting is higher, on multiple axes, than cloud-hosting. Hosting your own cloud is the "I don't trust auto mechanics so I'm going to become an auto mechanic" approach and most people have neither the time nor the talent.
Not sure how anyone who understands how Internet works could claim Google is in any way trusted, or trustworthy. Or has any reputation it could lose by a mere data leak.
Contrast that with, well, gestures widely to most of the Internet. While Google's security model is "We secure your data," so the trust interface between the end-user and Google is significant, Google for their part does an execllent job holding tightly to that data. If anything, the biggest risk working with Google is losing access to your own data because they stop believing you are you, not some third party getting your data out of Google's clutches.
Hackers go after value targets which are companies or organizations holding a lot of valuable data. My little home server with social media, XMPP, Home Assistant and Nextcloud for pictures and such is not something they can do anything with. Good enough security is built into most self-hosting platforms.
I predict a correlation between small self-hosting projects and more entries on shodan.io. Your small home server is probably secure enough... Probably. But you're the sort of person that posts on a site called "hacker news..." How much should we trust the average soul to do the bare minimum to not get owned? Do we imagine they're checking in regularly on https://www.cvedetails.com/vulnerability-list/vendor_id-1723... ?
I didn't want to bother dealing with the advanced calculator but the AWS EC2 instance to handle this bespoke application would be $130k/yr with one year upfront and 12 month reservation:
https://calculator.aws/#/estimate?id=b3de412ba6a7748e9acd782...
This doesn't include bandwidth (which AWS has an insane markup on) or nearly enough storage.
Needless to say you can lease the hardware from $VENDOR and host it for years with an all you can eat 10gig port. Oh and the cost is absolutely fixed - no worries about getting that shocking AWS bill we all know of.
With a lease and hosting it's still an operating expense and with Section 179 it's actually preferable from a tax standpoint.
With datacenter remote hands and vendor support you're still not dealing with hardware - you don't have to even step foot in the hosting facility. Ever. If a hardware component fails the vendor dispatches someone and it gets handled (same day). That said hardware failures are (in my experience) exceedingly rare until you get to some significant scale where it's a numbers game.
Because ec2 is OS up anyway from a dev ops standpoint it's pretty much identical.
Let's say you keep it for three years and upgrade. The old hardware is still worth something so you can repurpose, sell, trade, whatever.
Rough math but let's call this approach 1/4 the cost (but likely much less).
At AWS markup you can hire a dedicated full time person to manage just one server and still come out ahead. Want five of them around the world? Now you're saving a TON of money vs different AWS regions.
Even with this when I tell people we deployed our own hardware they look at me like I'm crazy. We have an entire generation in tech from jr devs to the C suite to investors that are terrified of hardware and will pay anything to avoid it.
Because of this they don't even know what hardware, bandwidth, etc actually costs. It's assumed the cost is whatever $BIGCLOUD offers it at. I know this is an extreme case but it probably happens more than most would think.
It’s worthless because it’s end-to-end encrypted. They can hand over the data but no one can view it.
There are safe ways of using the cloud.
Recently, a certain head of state's son had 100s of GB of iCloud backups thrown onto a torrent, and within a day rogue manchildren living in their parents' basements cracked most, if not all of it open.
With the backup password in hand, all one needs is this README.md file [0] to be off to the races.
[0] https://github.com/avibrazil/iOSbackup/blob/master/README.md
That's how encryption works...
Please stop posting this same topic in your comments.
If you don’t control the keys, to me, that’s not end-to-end encrypted.
Source: https://support.apple.com/guide/security/data-security-sec49...
"The authentication is based on Ed25519 public keys that are exchanged between the devices when a user is added to a home. After a new user is added to a home, all further communication is authenticated and encrypted using Station-to-Station protocol and per-session keys"
"The user who initially created the home in HomeKit or another user with editing permissions can add new users. The owner’s device configures the accessories with the public key of the new user so that the accessory can authenticate and accept commands from the new user. When a user with editing permissions adds a new user, the process is delegated to a home hub to complete the operation. "
https://support.apple.com/guide/security/data-security-sec49...
Also, since you brought it up, appears Ed25519 vulnerability has been reported:
https://www.google.com/search?q=Ed25519+exploit
Also, are these HomeKit “keys” in iCloud backups unencrypted? Meaning that the HomeKit data is encrypted, but the keys are not; to be clear, not saying they are, asking if they are.
down with surveillance capitalism!!