Hacker News Reader

Comment by jeremyberemy | Hacker News Reader

> Police use the technology to investigate crimes such as hit-and-runs and to locate missing persons. > No suspects have been identified.

Yeah, that tracks.

> Camera data is owned by the police department and permanently deleted after 30 days.

I assume this means that the police department deletes one local copy?

View on HN

> Camera data is owned by the police department and permanently deleted after 30 days

There's so much going on in that sentence! What a thing of pure PR beauty.

What does "owned" mean in this sentence? Probably fairly little.

What data does it even refer to? Just the actual video footage? Does the statement include derived data like license plate reads or ML models?

And surely if data is flagged as relevant to an investigation it is not deleted, so is there an implicit "*unless retained"?

You know, unless they get a national security letter, or change their TOS, or....

Or their app was configured to let other LE agencies access it without their knowledge in violation of state law (1). Only linked one case but afaik it happened in a number of CA cities.

(1) https://www.malwarebytes.com/blog/privacy/2026/02/flock-came...

https://www.ilsos.gov/news/2025/august-25-2025-giannoulias-a...

> Giannoulias’ Audit Finds License Plate Reader Company in Violation of Illinois State Law

> Flock Safety Shared Illinois Data with U.S. Customs and Border Protection Secretary, Orders Flock to Shut off Data Access

https://www.eff.org/deeplinks/2025/10/flock-safety-and-texas...

> Last May, 404 Media obtained data revealing the Johnson County Sheriff’s Office conducted a nationwide search of more than 83,000 Flock ALPR cameras, giving the reason in the search log: “had an abortion, search for female.” Both the Sheriff's Office and Flock Safety have attempted to downplay the search as akin to a search for a missing person, claiming deputies were only looking for the woman to “check on her welfare” and that officers found a large amount of blood at the scene – a claim now contradicted by the responding investigator’s affidavit. Flock Safety went so far as to assert that journalists and advocates covering the story intentionally misrepresented the facts, describing it as "misreporting" and "clickbait-driven."

> Earlier this month, there was purposefully misleading reporting that a Texas police officer with the Johnson County Sheriff’s Office used LPR “to target people seeking reproductive healthcare.” This organization is actively perpetuating narratives that have been proven false, even after the record has been corrected. According to the Sheriff in Johnson County himself, this claim is unequivocally false. That sheriff has since been arrested and indicted on felony counts in an unrelated sexual harassment and whistleblower retaliation case. He has also been charged with aggravated perjury for allegedly lying to a grand jury. EFF filed public records requests with Johnson County to obtain a more definitive account of events.

most likely the videos are deleted because it'd be a ginormous expense to keep it around; everything else, yeah, it's not being deleted.

It might be deleted at the PD, does not mean it wasn't uploaded to Flock DCs for metadata analysis and retained there.

My point is theres no way theybregain the video as anything other than cost of holding terabytes.

Of course they save the analysis and only thinkpeople care about the video. The journalists and police speak the same language of uncritical privacy violations.

yep it's pretty easy to scan a frame and store the info as "blue car 17:25:36 20260805 WXC7868" and toss the video info if all you want is civilian tracking.

Even better a single video frame because the OCR is demonstrably fallible.

What if "owning" here means the same as "owning" any movie online.

If questioned, I can already imagine someone defending that sort of agreement as a standard consumer, contractual terms and practices of the time.

Yes, the police own the videos they take of you? Like LLM companies own the models they digest from the texts they don't own?

I believe they're implying that Flock still has access to it even if the police technically retain legal ownership.

There's no such thing as legal ownership of data. There's IP rights, contracts and innumerable data regulation laws, but not actual ownership. "Ownership" here just provides a vague cover that sounds like something definite but is not.

They’re not taking the videos of people, their contractor is.

If the local police department were legally required to turn over all their computers and servers, it would only contain a fraction of what their contractor captured.

The contractor’s employees have also been proven to abuse access to the contractor’s video files across the nation. Something the local police department has little to no control over.

Police used Flock's license plate network to stalk women in at least 50 cases

https://www.msn.com/en-us/news/crime/police-used-flocks-lice...

Police officers are also statistically more likely to abuse their spouses. Don't let the women you care about get involved with cops.

Since people seem to be disagreeing with a fact:

https://digitalcommons.usf.edu/etd/1862/

https://academic.oup.com/policing/article-abstract/6/4/418/1...

https://www.sciencedirect.com/science/article/abs/pii/S13591...

> Don't let the women you care about get involved with cops.

I don't see any difference between what you just said and "13-85".

>and "13-85".

what does this specific lingo mean? normal searches and ai are both as confused as i am

I'm going to assume that commenter is referring to 'black crime rates' claiming that they make up 13% of the population while claiming that 85% of crime or some segment like violent crime is committed by them.

Being a cop is a choice. Being black is not.

But it's a fallacy to say all are wife beaters, like the parent said.

Recontextualized: *A person who chooses a career centered around the use of force is more likely to use force in their personal life*.

Ahdunno, where's the lie?

> where's the lie

Don't know, right there? I did martial arts and beat many people up, I even enjoyed it on many occasions. Doesn't mean I beat people in other contexts. It's called compartmentalizing.

For your compartmentalization efforts, did you get a bastardized American flag with a special thin colored line through it?

Cop culture nationwide is toxic as fuck. Can’t say the same about martial arts culture.

Parent didn't say "all are wife beaters" but that they are statistically more likely to be. The rephrasing in the sibling comment to mine is similarly accurate, even if you don't like the way it sounds.

60% of the spouses of police officers report being victims of domestic violence. The other 40% are too scared to answer honestly.

Seems like that might be indicative of a problem?

Not necessarily. One can accept that violent people do a violent job and commit violence in their home life.

Anyways, that study that everyone loves to cite is like 30 or 40 years old and unreplicated.

"not necessarly" and "more likely" are not at all at odds with each other.

People who use the threat of violence and actual violence to control people at work being more likely to do so at home to their own family seems self-evident to me.

Not making any moral judgements, if I used violence professionally I’d expect myself to be more violent in general.

Lesbians too [0]. And people who don’t make much money [1]. And people who aren’t Asian [2]. And athletes, and people who like sports [3].

Applying group statistics to judge individual people is almost never proper, and lies at the root of an awful lot of inhumanity. I hope and imagine the women in question are capable of thinking (and judging) for themselves.

[0] https://en.wikipedia.org/wiki/Domestic_violence_in_lesbian_r...

[1] https://www.cawc.org/news/socioeconomic-risk-factors-for-dom...

[2] https://bjs.ojp.gov/violent-victimization-race-or-hispanic-o...

[3] https://pmc.ncbi.nlm.nih.gov/articles/PMC10087409/#hsc14028-...

> Lesbians too [0]

> A 2010 CDC survey compared domestic violence victimization rates between lesbian, bisexual, and heterosexual women. The differences between most forms of reported domestic violence rates between lesbian women and heterosexual women were found to be non-significant due to insufficient statistical power.

> Applying group statistics to judge individual people is almost never proper

That's literally the job of statistics: make use of a lot of data about a population of instances to make a decision about an instance.

It's like saying, it's unfair to use safety data about Chevys in deciding whether to buy one.

We aren't judging individuals here, we're talking about trends and probabilies. I know a couple cops that are fine, and pretty sure aren't abusing their spouses, but that doesn't invalidate the statistics that they have more instances of abuse at a much higher rate in marriages/partnerships than the masses taken as a whole

exactly, and while correlation does not imply causation or something of the sorts lol.to add to the fire, the SOLE cop i know is an abusive alcoholic...

That's not too unexpected because the career often attract aggressive types of people

Flock data can be accessed by cops nationwide who have different (or no) restrictions on data retention or privacy compared to local police.

Flock data can be accessed by anyone. Literally. anyone. at anytime. Go see for yourselves.

https://measuredcollective.com/flocks-ai-surveillance-camera...

That's a bit misleading given that it's a subset. 404media cites "at least 60".

https://www.kcrg.com/2025/12/24/cedar-rapids-police-cameras-...

https://www.9news.com/article/news/local/colorado-news/dougl...

https://www.wabe.org/cobb-county-youtuber-documents-flock-ca...

https://www.wflx.com/2026/01/09/flock-safety-exposed-live-po...

https://www.thegazette.com/news/vendor-error-exposed-some-ce...

https://www.404media.co/flock-leaked-cops-license-plate-sear...

https://www.aclu.org/news/privacy-technology/tracking-alpr-c...

The list is pretty endless.

>The list is pretty endless.

Did you check your own sources? The first three articles you linked cites the same youtuber as in the gp. I'm not going to check any more.

I think you are overselling it a bit. 60-some cameras were misconfigured for a relatively short period of time.

This is false. They cannot be accessed by anyone.

> For an unknown period, at least 60 AI-powered surveillance cameras manufactured by Flock Safety were accessible to anyone on the internet.

Past tense.

Glad someone trusts these clowns.

Where I work, this wouldn't fly.

> Flock Safety acknowledged the issue in March, stating some California agencies “had their camera networks inadvertently accessible to out-of-state law enforcement agencies for a period of time,” according to a company statement. The company said it was “impossible to determine a specific cause” for the data breach.

Oh, that's lovely. So, if you don't know what the cause is, you can't really say you've mitigated it, can you? C'mere, I wanna honk your big red nose.

Flock has been known to lie and there isn't any reason to believe they are telling the truth. Why would they? There isn't some duty for them to tell the truth in an article.

Trying to parse what that sentence means is useless. It isnt designed to mean anything. Their various mouth pieces have claimed every manner of excuse.

Semantic games are for everyone else to play and lose. Corporations always win.

> deletes one local copy.

I think you’re being a little too generous here.

That whole paragraph is like a paid placement.

> Flock Safety cameras are automated license plate readers

Like Ring is a door chime.

> raise concerns about data being shared with third-party companies or federal agencies.

What, you don’t like “sharing”? It would be selfish not to “share”!

Truly the move from “send copies to” to “share” is one of the great linguistic dirty tricks the social-media-era firms have pulled.

Especially here, where what’s being “shared” is, in a meaningful sense, operational control and enforcement prerogative.

And don’t get me started on the minimizing formulation “third-party companies and federal agencies”…

No, Flock plays a semantic game where data on Flock's servers os "owned" by the client, so Flock is not responsible for the content or management of the data.

Isn't that what you'd expect from a camera service provider? I'd certainly want any videos taken by my own cameras to be owned by me.

https://en.wikipedia.org/wiki/Monkey_selfie_copyright_disput...

I don't see how that relates to any of this. If I record a video of a crowd in public, under US law I have copyright to that video. The crowd isn't creating the record, I am.

The issue of which group (Flock/police/city) has the real copyright versus an expensive license is interesting to contemplate, but that list doesn't include monkey-agency either.

I'm not familiar with North Dakota in particular, but at the federal level and in some states ALL works created by the government are exempt from copyright. The public paid for it, so the public owns it.

Eventually the copy in the officer's browser's cache expires...I figure that's what they mean.

> I assume this means that the police department deletes one local copy?

Ben Jordan got ahold of one of them and proved that they even had footage from the f'n factory they were made in! Flock is lying to cities and/or cities are lying to us

Reply on news.ycombinator.com