We don't need TSA strip searching grandmothers flying to visit their kids. We don't need Flock cameras recording the movement of every citizen at all times. We don't need gun detection AI at schools.
But, it's easy to make people scared and scared people will support unneccessary things that make them feel safe against their fears no matter how unlikely or irrational, so here we are.
Aka the fact that there has been a kid with a gun at a school somewhere, sometime, doesn't justify deploying aggressive technology to all schools, for all time.
Especially when the school system doesn't have sufficient policy wherewithal to prevent armed law enforcement responses to known false positives.
>The AI surveillance system’s error prompted the false alarm and set off a chain of human failures
>armed police were dispatched anyway because school and police officials failed to have sufficient protocols in place to prevent cancelled alerts
This is a mistaken setup, by humans. The human override (cancelling the false alert) hasn't worked.
But if AI is sufficiently tempting, then it is the problem. People have instincts and will operate on them, and if something exists to trigger those instincts, then it is as much the problem as the people who made the decision.
Especially since many of these AI solutions are known for lying about their own effectiveness in marketing and practice (e.g. Flock) and impeding transparent oversight.
Feels like the main use case of AI is to offload responsibility. It's been very visible in corporate situations that people are extremely risk-averse because a mistake could cost them their career. This culture has penetrated other parts of the society. Nobody wants to manually recall the alarm because if they do and something happens, they'll be fucked.
Similarly to why Spain security checks on high-speed trains. Nobody wants to be the politician that removes them and then is blamed for an attack.
The most ironic part of this story is that the people on site actually weren't afraid and canceled the alarm anyway. But the dumb system still called the cops. You try to do the human thing, and the code just trips you up.
Or, chillingly more and more accurate, the humans in the loop don't actually understand how the system functions, so they don't even know they need to cancel the alert on the police side too.
The cancel button is broken, no doubt about it. But honestly a model that spends 90% of its time talking about chips is clearly half-baked and shouldn't have been released to the public yet
So, apparently Omnialert sells enabling computer vision on existing cameras and backs up mistakes with human review. They at least claim no facial recognition, just guns. Reviews on reddit say the system (and similar ones) has many false positives, and failed to detect real guns like rifles.
Garbage in, garbage out. The system should not have been deployed because of it's incompetence if nothing else.
It sounds like AI was the problem. Omnialert's ai gave it a 90% probability. The company also had multiple failures before it got out but the alert is designed to be sent quickly.
False positives are not necessary a problem. A false positive is just a wrong suspicion.
In this particular case, the automated alert was just the beginning of the chain of events. The chain could have been stopped well after, and it looks like that they even tried to stop is, but were unsuccessful. That's the problem.
Not sure what the race of the Students has to do with the story, or is the implication that AI would not have identified the bag of Doritos as a weapon on white Students?
The AI must have been trained on all the police reports and news articles where police mistook common everyday items for deadly firearms when black people hold them such as cordless phones, cellphones, wallets, canes, hair brushes, gas pump nozzles, bicycle parts, sun glasses, pill bottles, ipods, Wii remotes, toy trucks, cologne bottles, sandwiches, bibles, candy bars, shower rods, flashlights, broomsticks, underwear, beer bottles, wrenches, and vape pens. AI really is just a reflection of ourselves.
Tests for AI racial desrimination are not mandated, yet, but are an inevitability.
We could work up a joke about driving while bieng AI, ok too egdy and wry to be very funny, but still.
They most likely don’t feel that way. Check out a few headlines from weirder parallel universes where identity-highlighting culture somehow evolved different:
“Ginger high school students sue after AI mistakes…” - with an inherited body variation.
“Goth high school students sue after AI mistakes…” - with a sociocultural group.
Most likely those would raise some eyebrows as to “ok, but why a highlight, what importance does it serve here?”
There’s likely no brunette or hippie fragility involved there. So, it’s possible to imagine there’s possibly some different idea rather than a feeling of threat here. Don’t so easily dismiss it to a convenient fragility excuse here. It’s surely not that simple.
And if anything, explaining the importance (if you know it and don’t just conform) would do a lot more bridge-building than accusing of fear. For all it’s worth, the article does a really poor job at this, mentioning a single Black student with a bag of chips, and not covering how that expanded onto their peers.
> “Goth high school students sue after AI mistakes…” - with a sociocultural group.
That's pretty much our universe.
After the Columbine massacre in 1999 there were a lot of rumors and headlines about how goth culture was a major contributing factor to blame, with Marilyn Manson as a scapegoat.
A Goth kid wearing a trenchcoat was eyed suspiciously as being the next killer in hiding.
There was a widespread satanic panic at the time, including training some police to recognize "satanic ritual abuse". Had there been AI the time you know that some of the training data would have included these so-called "indicators", leading to false notifications resulting in a AI mistakes.
Yes, both examples are not entirely random groups.
Problem is in the article itself being sloppy. Unlike your comment, it highlights a detail in the title then totally fails to provide any more details that connect to it. There’s that one story and then a statement that some communities are affected more, and that’s about it. And this explanation gap doesn’t feel right: the most important detail of the story (for why else would it get into the title) ultimately gets a couple relevant sentences in the whole text - that just cannot be right for some.
And then people ask “why Black” and other people accuse of racism… and so goes the misunderstanding.
Just one of many possible explanations, of course.
Your criticism seems directed towards the HN submitter, as the actual title is:
"Lawyers’ Committee Files Lawsuit on Behalf of Black Students Detained and Searched Due to AI Alert Falsely Identifying a Bag of Chips as a Firearm"
while the alternatives you mentioned, like “Ginger high school students sue after AI mistakes…” are variants of the HN title "Black high school students sue after AI mistakes Doritos bag for gun".
Since you think the headline is supposed to contain the most important details, can you explain why "high school" is one of the most important details, even more important than Black, as your objection has only been to that last detail.
Then repeat with why we need to know it was specifically "Doritos" chips.
You wrote "totally fails to provide any more details that connect to it" but it links to https://www.lawyerscommittee.org/ai-civil-rights-act/ with more details which connect to it, where you can see their viewpoint (quoting the one-pager):
```As AI has increasingly been unleashed on the public, millions of Americans are impacted by the outputs of complex algorithms that few of us understand. Algorithms are used to make decisions about all aspects of our lives, determining who gets bail, who can rent a house, and where we can go to school.
Those decisions haven’t been made equitably or fairly. Over and over again, promises of innovation have resulted in products that lead to discrimination, disproportionately harming Black communities and other communities of color. Because these algorithmic technologies are built using data that reflects generations of redlining and segregation, they often replicate and reinforce generations of discrimination. Because the teams that develop and deploy these products often don’t resemble the people that are affected by them, blind spots and biased assumptions are baked in. Because inadequate testing doesn’t reflect the diversity of the real world, our lives and experiences are reduced to the narrow parameters of lab conditions which work against us.
The irony is deep. The tools of the future are locking us into the mistakes of the past.```
Image training data can only ever hold surface details, and will thus always results in discrimination. Factors that will impact the decision process, among others, are gender, age, race, physical fitness, clothing, hair color and style, accessories, gate, and more. Algorithms based on image training data is guarantied to make unfair and biased decisions.
Change gender, age, and many of the other aspects in the correct direction and the AI in this case would likely not have flagged the bag as a gun. There is no way to fix AI so they make fair and unbiased decision in that regard, and so the tool is unfit for that purpose.
It's worse than that. Over 99% of the alerts did not require a police response, the police knew they received a bunch of false alerts, and in fact this alert had already been cancelled by designated representatives of the school and police department.
The police still (according to the complaint) came in with guns drawn, and carried out an illegal detention and search of bystander students who did not appear in the
Alert.
And you should question why you feel the need to imply others were threatened when they didn't mention that word or give any indication that they were.
"Students of demographic historically and currently disproportionately harmed by police violence sue school for using shoddy AI to call the cops on them saying they have a gun."
Being wrong has worse consequences for black students than white students so you can maybe imagine why they might be extra mad at such a system being implemented even if the false positive rate was equal.
> The false alert eventually led to armed local police swarming the teenagers on campus just steps away from their school’s entrance. They were ordered to their knees, handcuffed, and searched, despite doing nothing wrong.
Good thing there's no chance of a cop "fearing for their life" and someone ending up dead in a situation like this.
Or "Student of demographic currently disproportionately responsable of homicides sue school for using shoddy AI to call the cops on them saying they have a fun"
Look I see what you're trying to get at here and the data somewhat supports you but it mostly doesn't. The part where the data backs you up is that black students are over represented relative to student population in shootings that take place on school grounds. Where it doesn't is that white students are also over represented and a majority of school shooters and an even greater majority of mass/indiscriminate shooters.
Following your logic if you're going to tune your algorithm to be more wary based on risk of harm it would be pointed squarely at white boys, then black boys, then hispanic boys, then asian boys.
> Students who choose to bring unapproved snacks to class shouldn't face detention
If there are clear rules that forbid it, why not, for example on repeat offences?
Having said that, it doesn’t seem that was the case here. FTA:
“No child in Baltimore County should face this kind of danger while waiting outside his own school for a ride home from his mother,” said Freida Winfield, who was on her way to pick up her son Keith and the students when they were all ordered to their knees, handcuffed, and searched.
Because it is a.) idiotic b.) adults doing it are commiting much worst offence then the original one.
Power abuse is worst then unaproved snack. Ability to make rules is not carte blanche for mistreatment, even if you feel good and powerful when mistreating people.
I can't believe what I'm reading. Are you seriously suggesting there is an amount of Doritos that warrants use of force on school children?
I'm vexed. What amount of Doritos do you think would be appropriate before calling the misnamed American death squad to come and shoot some black kids?
It would entertain me to no end suggesting any child possessing multiple shipping containers filled with Doritos should be investigated. But that might be considered silly.
```37. On Monday, October 20, 2025, Plaintiffs attended football practice after school. Practice ended at approximately 6:30 PM. After practice, their coach passed out individual bags of Doritos to the team, as he occasionally did after practices. Taki took several.
38. Plaintiffs then went outside the school to wait for Keith’s mom to pick them up, just as they did many days after practice.
39. Plaintiffs convened near the picnic tables near the back entrance of the school. As they waited for their ride, they were talking and eating the Doritos their coach had given to them.```
You never know when you will need an extra of energy to run from a mass shooter, because politicians refused to do anything about it, literally, in centuries.
----------
A better way to solve the problem would be weapons that call the police and sleep when entering some protected areas, or enter in self-blocked mode if manipulated to remove this feature.
Your fridge can call South Korea all day, but nobody can built such type of weapons? for Pete's sake.
The police would love to have the series number of the weapon that is being fired inside a school before to reach the scene. Much faster response to help the family also. (Often mass shooters shoot their family first. If you reach them sooner, many of this victims could survive instead to bleed to death).
The gun makers would salivate about a gun that go to sleep automatically when entering into a school area.
First because the invaluable publicity of being the safest family brand in the market (only maker that really had the fucking galls to do something about it). The parents would love to have the option to block a gun from being used in a configurable range of areas of their choice, like for example their own bedrooms.
Second because the maker would avoid events that make they look bad or be sued. Their insurance company would love that also.
But most of all, because that would increase the profits. A self-blocked gun must be bought again, or fixed by the maker for a price. And families would gladly want to pay this price. Still much better than seeing your confused boy rotting in jail after a temporary depression, an adolescent broken heart or a drug/booze mistake.
Old brands will never change until forced to change, and politicians will not move a finger, so make a "Tesla for guns" and take all the market for yourself by assault. If you can't do it for the children, at least do it for capitalism.
Any electronic lock with a chip can usually be bypassed in ten minutes with a soldering iron, or just tossed in a faraday bag so it can't ping anything
>weapons that call police and self-block if they enter into some protected areas, or that self-block if manipulated to remove this feature.
And, in a life and death situation, you'd be dead because, unlike the one that the bad guy is carrying, your gun can't connect to the network to authorize a discharge.
Or when it does, it's too late; latency was too high.
The myth of the child hero is that, just a myth. Weapons fail 200 times a year, each year, to stop children to shoot other children or their teachers in USA. School shootings rarely end with a "childhero" or a "janithero" killing the shooter before they can kill tens of people. At this moment is too late yet. Self blocking doors, is a realistic solution. Your children being put in the situation to kill somebuddy or being killed just because they went to school is not. Is unfair to ask them to manage this psychologically. Many times they will end shooting collateral innocent victims by mistake and left scarred for life, or will be mistaken by the aggressor by police, and killed.
So your fix will never happen, because exactly zero adolescents should be allowed to bring a hidden gun at their school and because is very unfair to ask them to fix the problem, so you don't need to lose a dollar. Doors that detect guns are placed yet for some reason. The idea of allowing only the "good children" to carry guns inside, just because, is 120% nuts. Adolescents are hormonal imbalanced by definition.
Why to spend money on this doors if you have guns available in the teacher's room? Every mass shooter would go directly there first.
Interactions with police and BAME people have a significantly increased risk of accidental death, in many economies.
DOXXING, false theft reports, accidental "its a gun" reports, even on their own property.
Treating this as "no big deal they just got cuffed for a bit" is missing the lede: this was a bullet dodged (literally) and in many instances would be about murder, not cuffing (it is murder, even when police do it, if there is in fact no reasonable apprehended belief in risk to life)
> Interactions with police and BAME people have a significantly increased risk of accidental death, in many economies.
The use of "BAME" makes me think you're in Britain, but at least in America (where this story happened), there is evidence[0] to show that there is no statistically significant differences based on race in civilian deaths at the hands of police, when controlled for situational and behavioral context.
no statistically significant differences based on race in civilian deaths at the hands of police
rather it concluded:
On the most extreme use of force– officer-involved shootings – we find no racial differences in either the raw data or when contextual factors are taken into account.
George Floyd, for example, was a death at the hands of the police, but not a shooting. That detail aside, it did accept that:
On non-lethal uses of force, blacks and Hispanics are more than fifty percent more likely to experience some form of force in interactions with police.
Other studies with differing data sets and methods have found greater risks for non whites in respect of being shot and killed.
( Not all shootings are deaths, some are winged or discharged, not all studies look at the same data pools, etc ).
I don't know what BAME is; I'm assuming non-white.
What you say does happen, but I would say the much bigger problem in black communities is black-on-black violence. For some reason that's a taboo topic, but we've got to talk about it if we want to make things better.
It's a dated term from the United Kingdom (now disavowed in .gov.uk publications) that has fewer African Americans per capita and many many more "Asian" variation (generally meaning people who have arrived from Pakistan).
Not defending any terms here, just explaining - the unhinged aspect comes from a clash of UK and US terminology.
Absolutely, that was intended when I phrased it that way originally, also to highlight the fact that the subject is two different terminologies from two different countries with two different (albeit intertwined to a degree) histories.
Call it it an example of Australian absurd understatement.
and, FWiW, the UK had a small influx of African Americans post WWII who were there for the war and stayed (apparently) for the general acceptance ("better than Mississippi") .. which led to the Rolling Stones and other acceptably white Brits introducing Black American music to the white Americans of the USofA (or so the story goes).
How would you feel if the cops showed up to your work and marched you out of the the office in cuffs, or past your child’s school, or your church, or your potential romantic interest? And then you learned they did it because of wildly irresponsible use of AI tools and that they clearly knew it was because you carried a bag of Doritos, but did it anyway because of professional incompetence? What if you were startled and moved furtively when they yelled at you and got shot because of the same error?
Police carry guns and are trusted with a great deal of responsibility. There are clearly defined precedents, police department policies, and standards and they failed to uphold them.
It is because of AI tools. They integrated unreliable tools and put in place unreliable mitigation for that unreliability, which led to the outcome every even mildly interested observer and human rights group warned would happen.
The need for for gun detection AI at US schools is already massively concerning
Obvious counterpoint - there is no such need.
We don't need TSA strip searching grandmothers flying to visit their kids. We don't need Flock cameras recording the movement of every citizen at all times. We don't need gun detection AI at schools.
But, it's easy to make people scared and scared people will support unneccessary things that make them feel safe against their fears no matter how unlikely or irrational, so here we are.
Aka the fact that there has been a kid with a gun at a school somewhere, sometime, doesn't justify deploying aggressive technology to all schools, for all time.
Especially when the school system doesn't have sufficient policy wherewithal to prevent armed law enforcement responses to known false positives.
These kids could have been shot.
AI is not the problem here.
>The AI surveillance system’s error prompted the false alarm and set off a chain of human failures >armed police were dispatched anyway because school and police officials failed to have sufficient protocols in place to prevent cancelled alerts
This is a mistaken setup, by humans. The human override (cancelling the false alert) hasn't worked.
But if AI is sufficiently tempting, then it is the problem. People have instincts and will operate on them, and if something exists to trigger those instincts, then it is as much the problem as the people who made the decision.
Especially since many of these AI solutions are known for lying about their own effectiveness in marketing and practice (e.g. Flock) and impeding transparent oversight.
Feels like the main use case of AI is to offload responsibility. It's been very visible in corporate situations that people are extremely risk-averse because a mistake could cost them their career. This culture has penetrated other parts of the society. Nobody wants to manually recall the alarm because if they do and something happens, they'll be fucked.
Similarly to why Spain security checks on high-speed trains. Nobody wants to be the politician that removes them and then is blamed for an attack.
>Feels like the main use case of AI is to offload responsibility
Yes, and that's a wild thing to do, anal_reactor.
>Nobody wants to manually recall the alarm because if they do and something happens, they'll be fucked
The weird thing about this particular case is that they recalled the alarm, and the cops still showed up. Someone definitely set that system up wrong.
>Nobody wants to be the politician that removes them
I also agree that this is why many bad systems stay in place. Another oft-cited example is Japan's weird porn censorship law.
The most ironic part of this story is that the people on site actually weren't afraid and canceled the alarm anyway. But the dumb system still called the cops. You try to do the human thing, and the code just trips you up.
Or, chillingly more and more accurate, the humans in the loop don't actually understand how the system functions, so they don't even know they need to cancel the alert on the police side too.
Using AI is one of the problems. It seems to me it has the same level of reliability as rolling a dice.
Sounds like AI isn’t the only problem here.
Oh yes. In fact, it's the least of the problems. Any detection system can have a false positive. The important thing is how people handle its output.
It might be the least of the problems, but it's still a pretty big fucking problem.
The cancel button is broken, no doubt about it. But honestly a model that spends 90% of its time talking about chips is clearly half-baked and shouldn't have been released to the public yet
So, apparently Omnialert sells enabling computer vision on existing cameras and backs up mistakes with human review. They at least claim no facial recognition, just guns. Reviews on reddit say the system (and similar ones) has many false positives, and failed to detect real guns like rifles.
Garbage in, garbage out. The system should not have been deployed because of it's incompetence if nothing else.
Doesn’t sound like AI was the problem. Multiple human errors and process errors led to the armed response.
It sounds like AI was the problem. Omnialert's ai gave it a 90% probability. The company also had multiple failures before it got out but the alert is designed to be sent quickly.
False positives are not necessary a problem. A false positive is just a wrong suspicion.
In this particular case, the automated alert was just the beginning of the chain of events. The chain could have been stopped well after, and it looks like that they even tried to stop is, but were unsuccessful. That's the problem.
The first link in the chain is a problem. We put people in prison for making false accusations even when the victim is shown to be innocent.
This is like saying the Soviet busted-ass ICBM early detection system wasn't the problem because humans figured out it was giving false alarms.
Some discussion on the October 2025 incident at the time:
https://news.ycombinator.com/item?id=45684934
[dead]
Unsure if you got a problem or not? Call the cops and now you know you have atleast 1 problem.
Fuck this dystopian shit.
ATF has entered the chat
[flagged]
Not sure what the race of the Students has to do with the story, or is the implication that AI would not have identified the bag of Doritos as a weapon on white Students?
> is the implication that AI would not have identified the bag of Doritos as a weapon on white Students?
Yes
[flagged]
You're uninformed if that's your take.
[flagged]
The reason you're calling good is decades of biased data as input.
How? The topic of not making your AI model racist has been around for a long time.
[dead]
The AI must have been trained on all the police reports and news articles where police mistook common everyday items for deadly firearms when black people hold them such as cordless phones, cellphones, wallets, canes, hair brushes, gas pump nozzles, bicycle parts, sun glasses, pill bottles, ipods, Wii remotes, toy trucks, cologne bottles, sandwiches, bibles, candy bars, shower rods, flashlights, broomsticks, underwear, beer bottles, wrenches, and vape pens. AI really is just a reflection of ourselves.
[dead]
Tests for AI racial desrimination are not mandated, yet, but are an inevitability. We could work up a joke about driving while bieng AI, ok too egdy and wry to be very funny, but still.
You should question why your immediate reaction is to feel threatened that race was mentioned in the story
They most likely don’t feel that way. Check out a few headlines from weirder parallel universes where identity-highlighting culture somehow evolved different:
“Ginger high school students sue after AI mistakes…” - with an inherited body variation.
“Goth high school students sue after AI mistakes…” - with a sociocultural group.
Most likely those would raise some eyebrows as to “ok, but why a highlight, what importance does it serve here?”
There’s likely no brunette or hippie fragility involved there. So, it’s possible to imagine there’s possibly some different idea rather than a feeling of threat here. Don’t so easily dismiss it to a convenient fragility excuse here. It’s surely not that simple.
And if anything, explaining the importance (if you know it and don’t just conform) would do a lot more bridge-building than accusing of fear. For all it’s worth, the article does a really poor job at this, mentioning a single Black student with a bag of chips, and not covering how that expanded onto their peers.
> “Goth high school students sue after AI mistakes…” - with a sociocultural group.
That's pretty much our universe.
After the Columbine massacre in 1999 there were a lot of rumors and headlines about how goth culture was a major contributing factor to blame, with Marilyn Manson as a scapegoat.
A Goth kid wearing a trenchcoat was eyed suspiciously as being the next killer in hiding.
There was a widespread satanic panic at the time, including training some police to recognize "satanic ritual abuse". Had there been AI the time you know that some of the training data would have included these so-called "indicators", leading to false notifications resulting in a AI mistakes.
Yes, both examples are not entirely random groups.
Problem is in the article itself being sloppy. Unlike your comment, it highlights a detail in the title then totally fails to provide any more details that connect to it. There’s that one story and then a statement that some communities are affected more, and that’s about it. And this explanation gap doesn’t feel right: the most important detail of the story (for why else would it get into the title) ultimately gets a couple relevant sentences in the whole text - that just cannot be right for some.
And then people ask “why Black” and other people accuse of racism… and so goes the misunderstanding.
Just one of many possible explanations, of course.
Your criticism seems directed towards the HN submitter, as the actual title is:
"Lawyers’ Committee Files Lawsuit on Behalf of Black Students Detained and Searched Due to AI Alert Falsely Identifying a Bag of Chips as a Firearm"
while the alternatives you mentioned, like “Ginger high school students sue after AI mistakes…” are variants of the HN title "Black high school students sue after AI mistakes Doritos bag for gun".
Since you think the headline is supposed to contain the most important details, can you explain why "high school" is one of the most important details, even more important than Black, as your objection has only been to that last detail.
Then repeat with why we need to know it was specifically "Doritos" chips.
You wrote "totally fails to provide any more details that connect to it" but it links to https://www.lawyerscommittee.org/ai-civil-rights-act/ with more details which connect to it, where you can see their viewpoint (quoting the one-pager):
```As AI has increasingly been unleashed on the public, millions of Americans are impacted by the outputs of complex algorithms that few of us understand. Algorithms are used to make decisions about all aspects of our lives, determining who gets bail, who can rent a house, and where we can go to school.
Those decisions haven’t been made equitably or fairly. Over and over again, promises of innovation have resulted in products that lead to discrimination, disproportionately harming Black communities and other communities of color. Because these algorithmic technologies are built using data that reflects generations of redlining and segregation, they often replicate and reinforce generations of discrimination. Because the teams that develop and deploy these products often don’t resemble the people that are affected by them, blind spots and biased assumptions are baked in. Because inadequate testing doesn’t reflect the diversity of the real world, our lives and experiences are reduced to the narrow parameters of lab conditions which work against us.
The irony is deep. The tools of the future are locking us into the mistakes of the past.```
Image training data can only ever hold surface details, and will thus always results in discrimination. Factors that will impact the decision process, among others, are gender, age, race, physical fitness, clothing, hair color and style, accessories, gate, and more. Algorithms based on image training data is guarantied to make unfair and biased decisions.
Change gender, age, and many of the other aspects in the correct direction and the AI in this case would likely not have flagged the bag as a gun. There is no way to fix AI so they make fair and unbiased decision in that regard, and so the tool is unfit for that purpose.
It's worse than that. Over 99% of the alerts did not require a police response, the police knew they received a bunch of false alerts, and in fact this alert had already been cancelled by designated representatives of the school and police department.
The police still (according to the complaint) came in with guns drawn, and carried out an illegal detention and search of bystander students who did not appear in the Alert.
Where do they mention feeling threatened?
And you should question why you feel the need to imply others were threatened when they didn't mention that word or give any indication that they were.
"Students of demographic historically and currently disproportionately harmed by police violence sue school for using shoddy AI to call the cops on them saying they have a gun."
Being wrong has worse consequences for black students than white students so you can maybe imagine why they might be extra mad at such a system being implemented even if the false positive rate was equal.
> The false alert eventually led to armed local police swarming the teenagers on campus just steps away from their school’s entrance. They were ordered to their knees, handcuffed, and searched, despite doing nothing wrong.
Good thing there's no chance of a cop "fearing for their life" and someone ending up dead in a situation like this.
Or "Student of demographic currently disproportionately responsable of homicides sue school for using shoddy AI to call the cops on them saying they have a fun"
Look I see what you're trying to get at here and the data somewhat supports you but it mostly doesn't. The part where the data backs you up is that black students are over represented relative to student population in shootings that take place on school grounds. Where it doesn't is that white students are also over represented and a majority of school shooters and an even greater majority of mass/indiscriminate shooters.
Following your logic if you're going to tune your algorithm to be more wary based on risk of harm it would be pointed squarely at white boys, then black boys, then hispanic boys, then asian boys.
Correct, because no self respecting white student would be eating Doritos; they prefer Takis
Students shouldn't be bringing bags of Doritos to class either.
Students who choose to bring unapproved snacks to class shouldn't face detention and search by the police.
> Students who choose to bring unapproved snacks to class shouldn't face detention
If there are clear rules that forbid it, why not, for example on repeat offences?
Having said that, it doesn’t seem that was the case here. FTA: “No child in Baltimore County should face this kind of danger while waiting outside his own school for a ride home from his mother,” said Freida Winfield, who was on her way to pick up her son Keith and the students when they were all ordered to their knees, handcuffed, and searched.
Because it is a.) idiotic b.) adults doing it are commiting much worst offence then the original one.
Power abuse is worst then unaproved snack. Ability to make rules is not carte blanche for mistreatment, even if you feel good and powerful when mistreating people.
I can't believe what I'm reading. Are you seriously suggesting there is an amount of Doritos that warrants use of force on school children? I'm vexed. What amount of Doritos do you think would be appropriate before calling the misnamed American death squad to come and shoot some black kids?
Oops. I misinterpreted ‘detention’, reading it as school detention.
It would entertain me to no end suggesting any child possessing multiple shipping containers filled with Doritos should be investigated. But that might be considered silly.
Now that you mention it, I do think the silly example must be teased out here.
For example, where did a child get the money for all those Doritos? Plus, presumably, the cost of the containers and shipping?
Where do they store the shipping containers? Where does one even buy that many Doritos?
And what are they going to do with all those chips?? This is more than just selling sneakers online, this is a whole operation.
This deserves at least a local news segment.
lmao
They weren't in class when this happened.
Do you often make up excuses as a form of post-hoc justification?
They didn't bring the Doritos and it wasn't during class, nor in a school building, nor during normal school hours.
From the linked-to complaint at https://www.lawyerscommittee.org/wp-content/uploads/2026/11/...
```37. On Monday, October 20, 2025, Plaintiffs attended football practice after school. Practice ended at approximately 6:30 PM. After practice, their coach passed out individual bags of Doritos to the team, as he occasionally did after practices. Taki took several.
38. Plaintiffs then went outside the school to wait for Keith’s mom to pick them up, just as they did many days after practice.
39. Plaintiffs convened near the picnic tables near the back entrance of the school. As they waited for their ride, they were talking and eating the Doritos their coach had given to them.```
You never know when you will need an extra of energy to run from a mass shooter, because politicians refused to do anything about it, literally, in centuries.
----------
A better way to solve the problem would be weapons that call the police and sleep when entering some protected areas, or enter in self-blocked mode if manipulated to remove this feature.
Your fridge can call South Korea all day, but nobody can built such type of weapons? for Pete's sake.
The police would love to have the series number of the weapon that is being fired inside a school before to reach the scene. Much faster response to help the family also. (Often mass shooters shoot their family first. If you reach them sooner, many of this victims could survive instead to bleed to death).
The gun makers would salivate about a gun that go to sleep automatically when entering into a school area.
First because the invaluable publicity of being the safest family brand in the market (only maker that really had the fucking galls to do something about it). The parents would love to have the option to block a gun from being used in a configurable range of areas of their choice, like for example their own bedrooms.
Second because the maker would avoid events that make they look bad or be sued. Their insurance company would love that also.
But most of all, because that would increase the profits. A self-blocked gun must be bought again, or fixed by the maker for a price. And families would gladly want to pay this price. Still much better than seeing your confused boy rotting in jail after a temporary depression, an adolescent broken heart or a drug/booze mistake.
Old brands will never change until forced to change, and politicians will not move a finger, so make a "Tesla for guns" and take all the market for yourself by assault. If you can't do it for the children, at least do it for capitalism.
Any electronic lock with a chip can usually be bypassed in ten minutes with a soldering iron, or just tossed in a faraday bag so it can't ping anything
>weapons that call police and self-block if they enter into some protected areas, or that self-block if manipulated to remove this feature.
And, in a life and death situation, you'd be dead because, unlike the one that the bad guy is carrying, your gun can't connect to the network to authorize a discharge.
Or when it does, it's too late; latency was too high.
The myth of the child hero is that, just a myth. Weapons fail 200 times a year, each year, to stop children to shoot other children or their teachers in USA. School shootings rarely end with a "childhero" or a "janithero" killing the shooter before they can kill tens of people. At this moment is too late yet. Self blocking doors, is a realistic solution. Your children being put in the situation to kill somebuddy or being killed just because they went to school is not. Is unfair to ask them to manage this psychologically. Many times they will end shooting collateral innocent victims by mistake and left scarred for life, or will be mistaken by the aggressor by police, and killed.
So your fix will never happen, because exactly zero adolescents should be allowed to bring a hidden gun at their school and because is very unfair to ask them to fix the problem, so you don't need to lose a dollar. Doors that detect guns are placed yet for some reason. The idea of allowing only the "good children" to carry guns inside, just because, is 120% nuts. Adolescents are hormonal imbalanced by definition.
Why to spend money on this doors if you have guns available in the teacher's room? Every mass shooter would go directly there first.
Myth?!
If teachers carried, these shootings would not go very far, if happen at all.
[flagged]
Interactions with police and BAME people have a significantly increased risk of accidental death, in many economies.
DOXXING, false theft reports, accidental "its a gun" reports, even on their own property.
Treating this as "no big deal they just got cuffed for a bit" is missing the lede: this was a bullet dodged (literally) and in many instances would be about murder, not cuffing (it is murder, even when police do it, if there is in fact no reasonable apprehended belief in risk to life)
> Interactions with police and BAME people have a significantly increased risk of accidental death, in many economies.
The use of "BAME" makes me think you're in Britain, but at least in America (where this story happened), there is evidence[0] to show that there is no statistically significant differences based on race in civilian deaths at the hands of police, when controlled for situational and behavioral context.
[0]: https://fryer.scholars.harvard.edu/publications/empirical-an...
That study (and there are many) did not conclude
rather it concluded: George Floyd, for example, was a death at the hands of the police, but not a shooting. That detail aside, it did accept that: Other studies with differing data sets and methods have found greater risks for non whites in respect of being shot and killed.( Not all shootings are deaths, some are winged or discharged, not all studies look at the same data pools, etc ).
I don't know what BAME is; I'm assuming non-white.
What you say does happen, but I would say the much bigger problem in black communities is black-on-black violence. For some reason that's a taboo topic, but we've got to talk about it if we want to make things better.
>BAME people
What?
Black, Asian, and minority ethnic
Why is Asian included? What does “ethnic” mean? Why did you move on from BIPOC? This is absolutely unhinged.
It's a dated term from the United Kingdom (now disavowed in .gov.uk publications) that has fewer African Americans per capita and many many more "Asian" variation (generally meaning people who have arrived from Pakistan).
Not defending any terms here, just explaining - the unhinged aspect comes from a clash of UK and US terminology.
> the United Kingdom ... has fewer African Americans per capita
Quite a surprise.
They maybe pulled above their weight in other descendants of the Atlantic slave trade - https://www.youtube.com/watch?v=hpypYcMe16I
I think GP is poking fun at the fact that the UK would indeed have fewer African Americans per capita than America.
Absolutely, that was intended when I phrased it that way originally, also to highlight the fact that the subject is two different terminologies from two different countries with two different (albeit intertwined to a degree) histories.
Call it it an example of Australian absurd understatement.
I've seen people refer to African people living in Africa who grew up in Africa as African Americans.
Is that better or worse than referring to his Muskiness the same?
BTW, FYI- https://news.ycombinator.com/item?id=50529748 (I'm still pressed for time, but it's a partial answer)
and, FWiW, the UK had a small influx of African Americans post WWII who were there for the war and stayed (apparently) for the general acceptance ("better than Mississippi") .. which led to the Rolling Stones and other acceptably white Brits introducing Black American music to the white Americans of the USofA (or so the story goes).
Armed confrontations with law enforcement comes with a non-zero chance of being fatal for anyone, but especially young black men. It's a huge deal.
> Some people were cuffed and searched
Some children were cuffed and searched.
This is a joke, right?
The Iranian school was also flagged by an error prone system, human oversight did not matter, some people were bombed as a consequence.
It is in no way a justified detention and search.
How would you feel if the cops showed up to your work and marched you out of the the office in cuffs, or past your child’s school, or your church, or your potential romantic interest? And then you learned they did it because of wildly irresponsible use of AI tools and that they clearly knew it was because you carried a bag of Doritos, but did it anyway because of professional incompetence? What if you were startled and moved furtively when they yelled at you and got shot because of the same error?
Police carry guns and are trusted with a great deal of responsibility. There are clearly defined precedents, police department policies, and standards and they failed to uphold them.
No this failure isn't because of AI tools. It's because Bob was supposed to tell Dick it was a false alert and to call it off but he forgot.
It is because of AI tools. They integrated unreliable tools and put in place unreliable mitigation for that unreliability, which led to the outcome every even mildly interested observer and human rights group warned would happen.
[flagged]