Why does the fire department need access to run facial recognition?
https://www.theguardian.com/world/2013/nov/20/us-uk-secret-d...
Arson investigation, identifying the people at the scene of a suspicious fire?
The police should be the ones investigating crimes, under extremely strict and limited guidelines (eg. 4th amendment) which in this case include not being allowed to use facial recognition software.
Part of the investigation is determining whether the event is actually a crime. I'd much rather have subject matter experts make the determination of arson vs. act-of-god rather than "every nail needs a hammer" police force.
Separating out duties to experts is more effective. Let the fire department investigate fires and then pass on the information for the police to secure the suspect/s and follow the justice system. Same with mental health emergency cases. More social workers and experts dealing with a variety of mental disorders will be better to work people in crisis since they are trained for that.
Actually neither happened. The article says they were not able to find any identifiable information online. They had to use drivers license instead.
> the fire marshal sent links to Clearview AI face search results, an archive of school play photos and another to an archive of high school formal photos. He said he couldn’t find associated social media but offered to get a driver’s license photo for the detective. “We have access to that,” he wrote.
I read the sequence as,
1. They started with a protest video
2. Clearview provided public images of the same person, but no name. It was certainly more identifiable (e.g., their high school).
3. Then somehow they get the driver's license photo. Do they use the original protest video, or the Clearview images? How does this search even work? Nobody knows. Lazy journalism.
As readers, we have no idea if the Clearview search was actually important, or a dead end.
>A minute later, the detective sent the fire marshal Ahmed’s name, date of birth and driver’s license number. Within five minutes, the fire marshal replied, “Bingo.”
I believe that is supposed to say that the fire marshal sent the detective the license information. The Fire Marshal was clearly able to find identifiable information online, in the form of multiple high school photos, but was unsuccessful getting a match to any social media accounts. So the facial recognition worked and found matches in Clearview AI’s database of scraped school photos, but not their database of scraped social media photos.
Then the Fire Marshal offered to get a driver’s license photo, and says he has access [presumably to the DMV database]. The fact the about a minute later, license information was passed, sounds like a search was run by the Fire Marshal, a match popped up, and he sent it to the detective. But it could be that the detective used high school photos (being higher quality and full front facing) to run a search against the DMV records (which the police have access to with “permission from supervisors”) but according to other articles about the NYPD in general, it doesn't seem like that are able to run facial recognition on DMV records.
Either way, I think the ID came directly from the information the Fire Marshal passed and the Judge said as much.
>The NYPD would not have identified Ahmed but for the FDNY’s Clearview AI search and accessing the DMV photo, the judge indicated in her ruling
If you believe the government would only use that data for just purposes then you probably wouldn't then believe that there is a 1A issue. But if you think the government would use it to identify persons at a protest and then take adverse actions against them on the basis of their presence alone (which to be clear, seems distinguished from the immediate instance) you would probably think there is a 1A issue.
SCOTUS ruled there are some instances where private use of a service is 1) effectively necessary for modern life and 2) leaks a huge amount of information about the person, then the government cannot utilize it without a warrant even if handed over or sold willingly by the third party.
I am suggesting that we likely need to expand Third Party Doctrine to things beyond cell tower data because 1) we don't have absolute control over how/where our images are used and associated with our names, and 2) the technology to later affiliate our always-on/always-visible identities (like faces, gaits, or fingerprints) with our names is getting better and better.
You're right that today this is not illegal, but I am pointing out that your argument for "what to do instead" is literally the precise argument for why it should be: it chills protected expression.
I don’t, I deleted my social media accounts a decade ago and wasn’t into posting my own photos prior to that anyway. But other people can post photos with me/including me and I can’t control that (and since I don’t use social media I don’t even know when they do that).
Right. Also make sure your friends don't. And your family. Good luck with that.
I am 100% sure of this because the government has been 100% consistent and 100% abusive about this, 100% of the time.
Even the Civil War was clearly orchestrated and the people were abused and not just rights, but the very core Constitution was essentially destroyed and nullified, and what we’ve had since is nothing more than an abusive invalidated social contract upheld my sheer force, delusion, and bribing. The delusion and bribery part being what keeps people from realizing that.
Curious why this is downvoted?
You attacked the idea of free speech for the other side in the same comment where you said this. I would assume based on reference to government infringements that you're referring to the first amendment as "free speech" if you hadn't specifically emphasized "idea"; conservatives have no real first amendment case, but they do get censored and suppressed by people with power. The idea of free speech is very much still in play when university admin cancels a guest speaker or a forum moderator only allows left-wing or non-political posts. What am I missing here?
Probably people reading the article title without reading the headline, not realizing that that it's not only literally about shouting in movie theaters.
But tbh most commenters/voters on this site are reflexively imperialist, which is not surprising for a forum run by (and for!) capitalists in the imperial core. That's doubtless a big factor as well.
It's the same question though; how many right will we be asked to give up in order to catch rock-throwing assailants.
I lean pretty heavily on privacy not being a right in the direct sense or the penumbral sense. That a company sells a product that uses facial identification is to me not super relevant. If they had shown the photo around the department, and to random people at the university, and someone said "I recognize that guy, he's so-and-so" then there would simply be no problem here. Similarly if someone was just browsing through old yearbook photos and recognized the face. The computer system allows this to happen at larger scale, but the same basic thing.
Walking around with a photo versus walking around with a hundred million photos and asking everyone simultaneously should not be considered about the basic same thing.
Like, this guy was identified off video of him throwing a rock at a protester that hit them in the face. By all accounts this is someone who is trying to violently suppress peoples rights. That he got off on police misconduct in the investigation is a loss to society, no matter how many waxing words try to twist him into being a "protester violated in his rights".
Right now, Al-Julani's forces are massacring Druze in Syria, Just lining civilians up and shooting them to death. That's what genocide is. THat's how easy it is to commit genocide. One bullet per life.
At what point do you listen to humanitarian organizations and the UN? Amazing that you think 24 people is a genocide but not 60,000+.
Edit: While I said "duty" I meant that I really hope the that the police investigate all allegations of hate crime assault properly.
The 2nd amendment and the notion that we have physical power over the government is going to be whittled away as facial recognition and omnipresent government spying via data brokers gives them all the info they need to spy on every citizen, all the time.
AI means they don't only collect all the data they want for when they need it ala NSA 2008, but they can have a robot army of analysts transcribe photos and phone calls instantly and analyze for sentiment, flag for review.
If we don't demand, as a society, that government stay out of the business of the people, and that the military stay out of the business of civil society (ICE/National Guard/Marines), we are in for true evil.
It's true in a sense though; if you give the government these tools they will use them to capture offenders and enforce the law. Abuse? I guess, because they are not allowed by law to use these tools. It comes down in my mind to the specifics of why they are disallowed. Is it false positives? Is it an equity issue? Those are the things that would inform my outrage further.
1. US foreign policy is uniparty. As terrible as this administration is, remember that quashing anti-war protests happened under Biden, too. Columbia, Hind Hall, etc were all under Biden. That being said, moving to deport or denaturalize pro-Palestinian protestors is new; and
2. The state will turn violent to quash anti-imperialist sentiment.
Let me give you some examples:
1. The MOVE bombing. In Philadelphia in 1985 there was a black liberation group called MOVE. After a day-long standoff with police, the police dropped a C4 explosive from a helicotper on the house. The resulting fire killed 11;
2. Kent State. In 1970, there was an anti-war protest at Kent State University in Ohio. The Ohio National Guard had been called in. The protestors were unarmed. The National Guard were at least 100 yards from the protestors. Yet at some point the protestors got scared and fired on the protestors, killing 4; and
3. At a pro-Palestinian protest at UCLA, the encampment was attacked by pro-Zionists. The police stood by and did nothing and the next day used that violence as an excuse to violently break up the protest.
Facial recognition, mass surveillance, social media checks at ports of entry, weaponized deportation, etc. The state simply will not tolerate anti-imperialist protests.
Only in regards to one foreign entity.
> Let me give you some examples:
2 of those 3 directly involve the US and US action. The outlier says a lot.
> The state simply will not tolerate anti-imperialist protests.
The current administration ran on an anti-imperialist platform. You can protest american, russian, french, chinese, british imperialism all you want. You can quote george washington's warning about empires and foreign wars all day long. What you can't protest is israel. Period.
You have the judge coming down on the side of privacy, which is good; but the circumstances of the particular case are troubling (allegations of someone throwing a rock at someone else).
I'd be happier gaining ground for privacy rights with cases about, e.g., blanket surveillance, using surveillance for political purposes, surveillance capitalism, etc. Then we figure out where the best lines are for when surveillance actually should or can be used.
(Edit: And ill-considered downvotes is why I'm not going to bother to try to have a meaningful discussion on HN.)
The system has tools like warrants for this. It appears to me as just sloppy policework.
More proper would be "NYPD Bypassed Facial Recognition Ban to ID Rock-Throwing Assailant"
Worth noting that the charges were dismissed, and that they added a hate crime enhancement based on why he threw the rock. The "hate crime" aspect does tie in to his protest activities, and like all hate crime enhancements is completely ridiculous, of course.
In the end, this is not a free speech issue except tangentially; it is a privacy issue.
You're preaching to the choir, everything is partisan now, including crime:
If the perp throwing rocks is in support of my political views then he's a victim of the fascist regime, but if he's from the opposite camp, he's a criminal fascist that's a threat to society.
That's how the media and people have trained themselves to see things. There's no bipartisanship, it's all propaganda for manipulation and engagement.
Which is why I tend to flag such topics as they're low quality rage bait that many people bite. I come here for technical talks not to read MSM garbage.
It read like a privacy issue. Then I read your comment, and was confused.
> More proper would be "NYPD Bypassed Facial Recognition Ban to ID Rock-Throwing Assailant"
This is inaccurate. The charges were dismissed. At best, it's an alleged rock-throwing assailant.
> In the end, this is not a free speech issue except tangentially; it is a privacy issue.
That's what the original headline suggested to me on first reading. Why did you think the headline was a free speech issue?
That being said, the threat of a government disobeying its own rules and policies is a deterrent to free speech.
At the time that they used the software, the charges had not been dismissed. They were looking for a person that in their eyes had committed a crime. This is right and proper; they don't have the right to hang the dude on the spot, they can arrest him and refer him for trial, which is what they did.
> That being said, the threat of a government disobeying its own rules and policies is a deterrent to free speech.
Why?
"This is right and proper"
I am not excusing their use of the technology, only that the state and specifically the police have a compelling interest to find people who commit crimes. There are lots of limitations on their powers to accomplish this end, but we do want police to investigate crimes.
The definition of "excusing": "attempt to lessen the blame attaching to (a fault or offense); seek to defend or justify."
I don’t like how enhancements are distributed like candy.
reverendsteveii•2h ago
Broke the law is the phrase we want here. They did an illegal thing. They didn't just scoot past a barrier, they violated people's rights.
gruez•2h ago
Claiming that an administrative policy against using facial recognition as a "right" seems like a stretch.
elashri•2h ago
This is such strange way to describe "right for privacy".
gruez•2h ago
johnisgood•2h ago
tetromino_•1h ago
actionfromafar•1h ago
FireBeyond•1h ago
The person hit by the rock is a victim of whomever threw it, be this person or another.
And this person is the victim of the police department's policy violation.
These things can coexist.
zimpenfish•1h ago
Allegedly. The article doesn't mention any evidence that he actually did.
> "Per the record before this court, there is no additional evidence connecting the defendant to the alleged incident — no surveillance video to and from his home, no independent identification by others in attendance."
No evidence.
> "This case is premised on the complainant's word that he was the target of criminal actions by another person, and that other person was the defendant."
Weak evidence (with potential bias.)
> "The NYPD digitally altered the defendant's DMV photograph [...] never sought the metadata which would clearly indicate how, when, and perhaps by whom the photo was doctored."
Manufactured evidence.
> "That statement alone renders these medical records discoverable as possible impeachment material, necessitating their disclosure [...] Yet the People [...] have articulated no efforts to obtain these records"
Withholding evidence from the defence.
All in all, utter bullshit from the prosecution.
kazinator•1h ago
elashri•1h ago
maeil•2h ago
Assuming you're a cop of course, otherwise we'll go to jail.
reverendsteveii•1h ago