Chicago PD Believes It Can See The Future, Starts Warning Citizens About Crimes They Might Commit
from the buttle/tuttle dept
We’ve talked a lot over the years about the attempts to get out “ahead of crime” by using computer programs and algorithms to try and predict who might commit a crime. Predictive computing can then either target specific areas or specific people that might be in need of some extra law enforcement attention. Except as we’ve noted repeatedly, these programs are only as valuable as the data they use. Garbage in, garbage out, but in this case you’ve got a human being on the other end of the equation whose life can be dramatically impacted by law enforcement holding what they believe is “proof” that you’ll soon be up to no good.
With that in mind there’s growing concerns about efforts in Chicago to use predictive analytical systems to generate a “heat list” — or a list of 400 or so individuals most likely to be involved in violent crime. The Chicago efforts are based on a Yale sociologist’s studies and use an algorithm created by an engineer at the Illinois Institute of Technology. People who find themselves on the list get personal visits from law enforcement warning them that they better be nice. The result is a collision between law enforcement that believes in the righteousness of these efforts and those who worry that they could, as an EFF rep states, create “an environment where police can show up at anyone’s door at any time for any reason.”
Law enforcement and the code creators, as you’d expect, argue that it’s only the bad guys that need to worry about a system like this:
“A press liaison for the NIJ explains in an email: “These are persons who the model has determined are those most likely to be involved in a shooting or homicide, with probabilities that are hundreds of times that of an ordinary citizen.” Commander Steven Caluris, who also works on the CPD’s predictive policing program, put it a different way. “If you end up on that list, there’s a reason you’re there.”
Unless law enforcement makes a mistake, your data is wrong (which it often will be), or we decide to expand the program significantly, right? Another concern bubbling up in Chicago is that the programs are effectively using racial profiling to target already-troubled areas where crime naturally would be greater due to poverty, without anybody bothering to perform a deeper analysis of why those areas might be having problems (aka targeting symptoms, not disease):
“…how are we deciding who gets on the list and who decides who gets on the list?” (EFF staff attorney Hanni) Fakhoury asks…”Are people ending up on this list simply because they live in a crappy part of town and know people who have been troublemakers? We are living in a time when information is easily shareable and easily accessible,” Fakhoury says. “So, let’s say we know that someone is connected to another person who was arrested. Or, let’s say we know that someone’s been arrested in the past. Is it fair to take advantage of that information? Are we just perpetuating the problem?” He continues: “How many people of color are on this heat list? Is the list all black kids? Is this list all kids from Chicago’s South Side? If so, are we just closing ourselves off to this small subset of people?”
Chicago PD denies that there’s any “racial, neighborhood, or other such information” being used in their heat list calculations, but a FOIA request to actually confirm that was denied, under the pretense that releasing such information could “endanger the life or physical safety of law enforcement personnel or any other person.” So yeah, there’s great transparency at work here as well.
Predictive computing is excellent for a good many things, from improving traffic congestion to designing sewer networks, but calculating the future movements of highly complicated and emotional human beings is a bridge too far. It’s not particularly difficult to imagine a future where law enforcement (not always known for nuanced thinking or honest crime stat record keeping) starts using their belief in the infallibility of mathematics as the underpinnings for bad behavior, with the horrible experiences of the falsely accused dismissed as anecdotal experiences (“well shucks, most of the time the system is right, so its existence is justified”). It might just be time for a re-watch of Terry Gilliam’s Brazil with an eye on reminding ourselves what a simple clerical error can do to the Archibald Buttles of the world.
Filed Under: chicago, data, police, sociology, statistics
Comments on “Chicago PD Believes It Can See The Future, Starts Warning Citizens About Crimes They Might Commit”
I got a GREAT idea!
Let’s give people who love liberty and freedom MORE reason to hate and resent us! – Police
Re: I got a GREAT idea!
Let’s give the ACLU a reason to sue us on behalf of an innocent citizen for slander! – Police
Re: I got a GREAT idea!
Another concern bubbling up in Chicago is that the programs are effectively using racial profiling to target already-troubled areas where crime naturally would be greater due to poverty, without anybody bothering to perform a deeper analysis of why those areas might be having problems (aka targeting symptoms, not disease)
Actually, it’s the laziness idea. Go for the low hanging fruit instead of doing proper investigative work. Pretty much the default for law enforcement these days.
...
So… We combine no-knock warrants with a pre-crime list, add in swat teams and take away Constitutional restraint while multiplying a lot of military force…
Those computers do not compute well for LEOs getting away with murder.
Re: ...
I keep thinking “Minority report”.
Re: Re: ...
We all saw a real world example of the infallability of math when derivatives crashed the finance system didn’t we.
And of course, EVERY police organization will assure us that the fall under the “TOO BIG to fail” umbrella. If we disband them or even just cut back on their military weaponry the “Lives will be lost !!!”.
Watch for some idiot to propose being on the list automatically labels you a felon so you can no longer own a gun.
Re: Re: Re: ...
us that the fall under the “TOO BIG to fail” umbrella
You are confusing Tom Cruise with Resident Evil.
Re: Re: Re: ...
Bullshit.
No “math” was at fault in that crash. The fact is, the math was applied to financial instruments that contained wildly over-valued components (and some fraudulently mis-valued). Now, was there some kind of math error involved there? No.
It was fraud. Too many 1%ers making too much money off the overvaluation of homes, and when the only way to maintain the impetus of the bubble was to find new buyers…well, the government, under the guise of “helping the poor” was perfectly willing to front your tax dollars to MAKE new buyers out of people who really couldn’t afford homes.
Those ridiculous mortgages, lumped together, were the “rotten apples” (or ‘tranches’, I believe they were called) of the derivative market.
The math itself, and the fallibility thereof, had nothing to do with it.
Re: Re: Re: ...
First, remove the police unions, then make all complaint investigations a third party office completely separated from the police. Then we’ll have a chance of taking things back.
Re: Re: ...
Pre-crime division
Seems to me that if the public were to make “a list” based on actions and past homicides most law enforcement institutions and government officials would rank pretty high on the “heat list” ,This is a very slippery slope they’ve decided to descend and need to be careful or they may just end up being on the wrong side of it. I hope they make this a law pass it let us be the judges and jury of our elected officials ,the heat list will soon be a flaming pit of horror.
it would be much easier for the people to watch each party or each agency ,so bring it.
Re: Re:
What on earth makes you think anyone holding a government job won’t automatically be exempt? I can already hear the rationale: “people in certain positions will necessarily produce large numbers of false indications, so in the interests of accuracy they will have to be omitted from the database.”
All the Troubles of the World
“These are persons who the model has determined are those most likely to be involved in a shooting or homicide,…”
If that includes being shot by police in their own homes in the middle of the night, those computers might just be onto something.
Does this remind anyone else of minority report ?or do I watch to much sci-fi.
Re: Minority Report
No, that was based on psychic feelings of a few individuals. Much more accurate, I suspect, than a bunch of data entered by whatever minimum-wage clerk they have at the keyboard this week.
Re: Re:
Kind of a cross between “Minority Report” and “Person of Interest”.
Re: Re:
Before it’s crime, it’s pre-crime!
ITM No Agenda producers! 😉
Re: Response to: Anonymous Coward on Feb 26th, 2014 @ 6:07am
First thing that came to my mind!
Re: Re:
I think we’ll soon need to drop the “fiction” part of science fiction.
This whole profiling algorithm system, sounds like a huge waste of time and money. Even if this alleged prediction system were miraculously correct with it’s predictions (highly unlikely), it still doesn’t predict the exact time and place a supposed future crime would occur.
The only thing this profiling system is good at, is adding people who have yet to commit a crime to a list. I hope this list isn’t available for employers, otherwise this system will simply compound crime, by making it hard for people to find jobs. Which pushes them towards crime!
Obvious question, considering it's Illinois
Just wondering, given the very high correlation between this position and felony conviction, is the Governor of Illinois at the top of the Pre-Crime department’s heat list?
Minority Report Redux
At this point, it doesn’t take too much insight to understand what’s going on. The system’s been in place for decades already. Now that the light has been cast on it, they are threating us with it. They don’t even need to tell us how they’re doing it because we all know, thanks to Snowden.
Thanks to grants to small cities and towns, they have military force available to carry out their threats. Deep Politics is coming to the surface, or at least is beginning to show its fa?ade. Keep in mind: this is only the tip of ice burg. They are scared. Their mass has become too big and is out of balance. Be cautious around scared creatures; you don’t know what they will do when frightened.
You need to extricate yourself from the system: stop feeding it. Stop using credit. In fact, avoid using federal reserve notes. This is the blood. The future is neighbor to neighbor.
Since when did the police start believing in crystal balls? Also do they not know about self fulfilling prophecies; which hassling someone could make come true?
Idiots with Statistics
Few things are worse than idiots using statistics. Statistical illiteracy is bad but half-literacy is worse.
Social Workers
If they really wanted to stop the crimes before they happened, perhaps they should send out social workers to find out what is wrong and try to help the person before they commit the crime.
Re: Social Workers
They might go for that if we could privatize the social workers. Is there some way corporate america could profit off the social workers? if so, you might be on to something.
Re: Social Workers
Ummm, that is exactly what they said they are going to do.
Idea stolen from Futurama
Is Matt Groening and David X. Cohen getting royalties for their obviously stolen idea?
http://www.comedycentral.com/episodes/2hgv6c/futurama-law—oracle-season-6-ep-616 (1:50 is where the Future Crimes Division is introduced)
Re: Idea stolen from Futurama
http://www.imdb.com/title/tt0181689/
Re: Re: Idea stolen from Futurama
FYI, you can’t ‘steal’ an idea
Re: Re: Re: Idea stolen from Futurama
If police can predict future crimes then surely ideas can be stolen.
Re: Re: Re:2 Idea stolen from Futurama
Yes, but only ideas from the future.
This will occur only when the Future Crimes Division learn how to predict future ideas and who would have really thought of them, then they can know whom to arrest when that person comes up with an idea they couldn’t have possibly come up with on their own in a million years.
Ideas stolen from the past will still only be investigated by the Past Crimes Division. For what purpose, no one knows.
So, Chicago has cleared all current crime backlog and now has time and resources to throw at non existent crime? Maybe they should be sharing the secret of their success.
If we don’t fight against this NOW, they WILL want to recreate the Minority Report future. Because that’s how they think, and that’s what you get when you allow the government and authorities to put “safety” above EVERYTHING ELSE.
“Knock, knock – it’s the police. We’ve been alerted by our software system that you have an 80 percent chance to punch someone this week. We’re here to arrest you.”
This stuff WILL happen, if we LET IT happen. But it’s going to be a trend, not something that happens all of the sudden. That’s why it’s important to kill it before it becomes too strong of a trend, and before the lobbying for such technologies becomes too strong in Congress, whether from law enforcement themselves or from companies making money off it.
Re: Re:
This shows that Chicago police are using “magical thinking” when it comes to computers.
If the “prediction” hits then it is proof of its greatness.
If it misses then “something got left out” so they need more surveilance.
Rachets just like Copyright.
It beats working
Chicago’s queue of unsolved homicides, rapes, assaults, and other violent crimes is enormous. But working through those is tedious and unglamorous, plus it takes intelligence and diligence.
But this — hey now, THIS is sexy and fun. Much better than pulling out an old file and trying to find a fresh insight in it or attempting to chase down a lead. Never mind all the actual real live crime victims and all the actual real live criminals on the streets, no no no, they’re not important. Let’s worry about theoretical possibilities, about hypothetical situations, about could-be maybe futures that may never come to pass.
slow clap Congratulations, Chicago PD. Well done.
Watching too much TV
I think they’ve been watching too much “Person of Interest”.
They usual trope comes out:
“Law enforcement and the code creators, as you’d expect, argue that it’s only the bad guys that need to worry about a system like this:”
And I provide the usual response:
So what constitutes a crime? Say, at sometime in Dearborn MI, walking around with your head uncovered if you are a woman becomes a punishable offence. You can no doubt think of other behaviours that are not criminal now, but could be made criminal in the future. And given that there is a new federal regulation about every 3 hours (I really don’t know on this–could be every 3 min) crime becomes a creeping thing.
Bad apple analogy...
those guys are Declaring a barrel of apples bad due to the fact that they found a handful rotten while the rest are still fresh….
jeez… very discriminating for the whole group of people for a few who are REALLY the cause of the problem….
I say… They are TOO LAZY to do investigation and wanted to clean up in the fastest way possible with greatest rist in it… -_-
Maybe they need to change their Motto
From “We Serve and Protect.” to We surveil and harass
I'm reminded of a TED talk
I forget who it was that gave the talk, but it centered around his being a victim of a hit & run by a drunk driver (who hit a telephone pole shortly after his car, and so was apprehended on the scene).
The police report was terribly confused, mixing up the two people and cars in a way that obviously made no sense. However, the cop made a single mistake that screwed the speaker: he checked the wrong box for who was at fault, and so the speaker was facing a bill of around $10,000 to repair damages to the drunk driver’s car.
The speaker obsessed over this and tried to rectify it — to no avail. His insurance company said nothing could be done unless the report was corrected, and the police were totally unwilling to revisit the issue, telling him “do the right thing and pay for the guy’s car.”
If there are such extreme problems with correctly handling crimes that have already been committed, just imagine how accurately predicting crimes that haven’t been committed will work. It’s all coming from the same data, after all.
Re: I'm reminded of a TED talk
It was Mike Birbiglia who had that happen to him. I heard the same talk – it was infuriating, yet hilarious because he is so funny.
Here is that entire story in his own words: http://www.thisamericanlife.org/radio-archives/episode/379/return-to-the-scene-of-the-crime
Re: Re: I'm reminded of a TED talk
Yes, that was it! Although you should really watch the video of the talk — he shows the actual police report, which is hilarious and disturbing all by itself.
“Another concern bubbling up in Chicago is that the programs are effectively using racial profiling to target already-troubled areas where crime naturally would be greater due to poverty, without anybody bothering to perform a deeper analysis of why those areas might be having problems (aka targeting symptoms, not disease)”
Well, it’s not the police’s job to prevent poverty (or the disease as we use in the analogy), their job is supposed to be treatment of the symptoms. Their local legislature is supposed to “regulate” away the disease through social programs.
Re: Re:
that’s great, nick at nite, but you are missing the forest for the trees: whites use -say- illegal drugs at a slightly HIGHER rate than blacks, and yet blacks are FAR MORE likely to be arrested and jailed for possession of said illegal drugs…
one might be -if one were not obeisant to authority- curious as to why that is so…
well, i’ll tell you why: IF the piggies went no-knocking in more affluent and/or whiter neighborhoods and picked up all the white people with coke/etc, there would be such screaming and kvetching that you would have thought some sort of real injustice was going on…
so, IF the piggies were actually doing their jobs without fear or favor, instead of picking on the poor and benighted who can’t get out of the way of piggies, you would find equal numbers of drug crimes in each of the neighborhoods, and thus their ‘predictions’ should present them each as being equivalent…
they do not…
they will not…
they dare not…
they are picking on the poor and weak because they can; EVEN IF the donut eaters wanted to go after wealthier/whiter people, they will not do so for fear of losing their jobs…
OBJECTION! No Foundation!
Where’s Harry Seldon when you need him?
“Create an environment where police can show up at anyone’s door at any time for any reason.”
Don’t want to burst this guys bubble.. But as far as I know the police can show up at anyone’s door any time for any reason.. You don’t have to answer, but I’m pretty sure they are allowed to knock
Re: Re:
Thing is though, with a system like this in place, if you don’t answer for any reason, whether you’re distracted/occupied by something else, or you don’t feel like talking to a group of thugs doing house calls to tell people how bad they are, since the system has tagged you as a ‘likely threat’, odds are they’d feel fully justified in breaking in to stop whatever nefarious crime you were obviously planning.
Re: Re:
Precisely. Unless they are using this data/algorithm as evidence for a search/arrest/prosecution, there’s nothing really problematic.
Re: Re: Re:
Exactly. Unless they are using the charge of “resisting arrest” to arrest you for, well… “resisting arrest”, there’s nothing really wrong here.
Move along, move along. Nothing to see, move along.
Please explain how this is different from Bull Connor approach.
Garbage in, garbage out... but what about the algorithm itself?
Are you assuming the algorithm itself is scientifically sound? Even if it is, what if the police have more confidence in the algorithm’s output than is statistically warranted?
Consider how such a system would work if it were fed all the metadata NSA is routinely collecting and search warrants were issued based on its output. We’d end up with a police state propped up by questionable science. We’d find a lot of bad guys, to be sure, but we’d also harass and invade the privacy of many innocents.
Even if the system were 99% effective, its output should not be taken as evidence of future wrongdoing. Being considered likely to commit a crime in the future is not probable cause for anything.
Re: Garbage in, garbage out... but what about the algorithm itself?
“Being considered likely to commit a crime in the future is not probable cause for anything.”
This should read “likely to commit some crime in the future”. If there’s reason to believe a specific crime is about to be committed in the very near future, greater suspicion is warranted. If it’s merely likely that some kind of crime is likely to be committed in the future, then the confidence is poor, no matter the statistics.
Re: Garbage in, garbage out... but what about the algorithm itself?
It is now. Take it from some one who knows. LEO does not need probable cause anymore. Just a hunch is good enough for them.
Re: Re: Garbage in, garbage out... but what about the algorithm itself?
But it’s a hunch from a computer! Totally different. /s
“Major Strasser’s been shot…Round up the usual suspects.”
-Captain Renault, Casablanca
No-Fly List?
Is this how some people get on the no-fly list too?
Furthermore...
Errors are not the only problem. What about the person who maliciously gets added to the list simply because the police want an excuse to harass him or the police simply blaming the harassment of a citizen on a predictive error of the computer that never actually existed? Oh wait. What am I thinking. Law enforcement would never do anything so unethical, malicious, and manipulative as that. Would they?
Sue the algorithm for false arrests, easy /s
They've already stopped 12 active shooters....
What they don’t bother to mention is all the false positives that are bound to occur. Like this guy with a mortar.
I’m sure he is on the terrorist watchlist and no-fly list with no way to get off (because “government secrets”) by now.
They've already stopped 12 active shooters....
How many of those shooters were patsies or government plants who only exist to make them look good?
Okay, let’s just test it out on them first, or are they afraid that all the potential dog-killers and rapists will thin out their roster?
I don’t see the problem in a law enforcement officer showing up and reminding someone to follow the law.
A program like this shouldn’t be used as evidence in a prosecution or justification for a search, but I don’t think there’s anything preventing police from randomly talking to people (much less talking to people based on data analysis).
Another Biased Journalist?
“…your data is wrong (which it often will be)…”
Listen, I’m not advocating for this system. But don’t tell me that the data the police have is wrong unless you have some proof and can show it. Otherwise you are just one more biased journalist.
Re: Another Biased Journalist?
Wait just a second. Are you calling a Techdirt writer biased!!???
mind blown
Re: Re: Another Biased Journalist?
Is a certain someone with a certain agenda whining about Techdirt being biased? mind blown
Re: Re: Re: Another Biased Journalist?
Um…does “certain someone” refer to anybody who controverts the dominant Techdirt groupthink? If so, then yes. If not, then you probably have me confused with some other certain someone.
Re: Re: Re:2 Another Biased Journalist?
Because this the first time you’ve ever whined about Techdirt on Techdirt, right?
Re: Re: Re:3 Another Biased Journalist?
Not at all. But it’s been a few months. I don’t think I’m that memorable.
Re: Another Biased Journalist?
2. Do you understand what is meant by “hypothetical proposition”?
Re: Another Biased Journalist?
Look, all algorithmic predictions will come up with wrong conclusions at least some of the time. Look at weather models predicting things. Google is often wrong about what people are looking for. It’s the nature of algorithmic predictions. The difference though is that you don’t have the police making judgements about whether or not a person needs to be arrested or not based on a computer program that sometimes gets things wrong.
Re: Re: Another Biased Journalist?
And if you look at the response given by the person in charge of the system, it is quite obvious that he doesn’t understand (or doesn’t care about) this:
“If you end up on that list, there’s a reason you’re there.”
He believes that a computer system, that by it’s very nature cannot be perfect, is in fact perfect. Either that or he doesn’t care about when it is not and an innocent person is affected.
That is a BIG problem.
Re: Re: Re: Another Biased Journalist?
This is a well-known problem with the output of systems like this. People trust it, in large part because it uses precise-seeming numbers. “Such and such is 86.25432% likely” sounds trustworthy, as if the system is so precise that 5 decimal places is worth mentioning. It never is.
Re: Re: Another Biased Journalist?
It’s my understanding they aren’t arresting anybody based on their presence on the list. If they were, that would be a problem.
Re: Re: Re: Another Biased Journalist?
True, it would be worse if they were arresting people. But putting them on a watch list is pretty bad, too.
Re: Re: Re:2 Another Biased Journalist?
Why?
Re: Re: Re:3 Another Biased Journalist?
Is that a serious question??
For a lot of reasons, but let’s just start with all the people who mistakenly assume that if they’re watching you, you must be guilty…
Re: Re: Re: Another Biased Journalist?
Of course they aren’t. NYPD isn’t arresting people with Stop and Frisk either, are they? They are just harassing people based on profiling. This is profiling that they’ve just automated with a computer with the convenient excuse to say, “Oh it’s not our fault, the computer must have made a mistake.” That’s how it always happens to. They say say how perfect and infallible the system is when they are trying to sell it to the public, but conveniently forget all of that when innocent people get wronged later once it’s implemented and they want to avoid accountability.
Re: Re: Re:2 Another Biased Journalist?
What does stop and frisk have to do with this?
Of course this system could be used poorly, but is there any evidence that is is being used poorly?
Talking to people is not an arrest/search/charge. Talking to people is not a violation of their rights.
Re: Re: Re:3 Another Biased Journalist?
Let’s say that Stop and Frisk was instead Stop and Talk. Then MAYBE that would be ok. That is a cop walking down the street and happening upon someone in public and addressing them. It’s completely different if they decide to go to someone’s house to “talk” to them based on the fact that a computer gave them some statistical analysis data. If you or I did something like that to someone more than once, we would be quickly slapped with a restraining order for stalking them.
Re: Re: Re:4 Another Biased Journalist?
You keep mentioning Stop and Frisk, as if it had some relevance to this other program in another city.
Re: Re: Re:5 Another Biased Journalist?
As I already said, Stop and Frisk is profiling. This is profiling with a computer. Neither are based on any evidence that the person is guilty of committing any crime.
Re: Re: Re:6 Another Biased Journalist?
There is nothing inherently wrong with “profiling”, especially if it is just used to talk with someone (not a stop, not a search, not an arrest).
If you fit the “profile” of a guy running down the street with a gun and one hand and a bank bag in the other, it’s completely reasonable to stop you on suspicion that you may have committed a crime.
If you are “profiling” someone based on a prohibited bases (e.g., race), that’s a different question. Even then, though, it still depends on that actions the “profiling” is used for. You want to talk to “community leaders” on an issue, maybe you want to make sure you talk to leaders that racially/ethnically represent the community, for example.
Re: Re: Re:5 Another Biased Journalist?
Did you also.stop to think where they might get some of this data that they plug into the algorithm? Given the reported problems of warrantless spying generating large amounts of data on individuals, and effectively no oversight on this collection or how the data is used,it is not a stretch that this sort of information will be fed into such programs, which will then point law enforcement to profile people based on such pointers. Then take into account that law enforcement seems to have no problem with the practice of evidence laundering which is what they euphemistically call parallel construction, and the whole thing becomes an efficient gutting of the fourth amendment.
Re: Re: Re:6 Another Biased Journalist?
These hypotehtical other things you are mentioning might be problematic. But by that logic (X might not be bad, but X might lead to or be correlated with Y, which is bad), we should get rid of police departments (since they might lead to police abuse).
Re: Re: Re:7 Another Biased Journalist?
But by that logic (X might not be bad, but X might lead to or be correlated with Y, which is bad), we should get rid of police departments (since they might lead to police abuse).
Nice argument, but like a bucket with a hole in the bottom, it fails to hold water.
With law enforcement sharing access to all kinds of databases these days, it’s not so hypothetical after all:
Parallel Construction Revealed: How The DEA Is Trained To Launder Classified Surveillance Info
Re: Re: Re: Another Biased Journalist?
And I didn’t say that they were arresting them based on that. I said that they were deciding whether the person NEEDS to be arrested based on that. A police officer may decide that someone committed a crime but not have enough evidence to make a strong case so they don’t arrest them. They are still making the judgement that the person must be guilty beforehand based on limited information that could be faulty even if they choose not to arrest them.
Re: Re: Re:2 Another Biased Journalist?
So…that’s just your hypothetical conjecture, right? I mean, you are assuming the cops are deciding that, right? Evidence? Bah! Who needs evidence. Their cops! We know what they’re like!
pot, kettle, etc.
Re: Re: Re:3 Another Biased Journalist?
*They’re
Re: Re: Re:3 Another Biased Journalist?
His attitude from the quote says that he’s decided that these people are guilty. He’s essentially saying that these people wouldn’t be on this list if they weren’t guilty of something.
Re: Re: Re:4 Another Biased Journalist?
Maybe he’s right. Maybe past convictions is a data point used.
But he doesn’t actually say that. He says they must have some “reason” to be on the list. That’s not necessarily a criminal act.
Re: Another Biased Journalist?
I don’t think you’ve been paying attention to the hundreds of reports of police raiding the wrong house, killing the wrong people, arresting the wrong people, etc., etc.
Hell, we even have innocent people on various death rows across the nation.
What more proof do you need?
They've already stopped 12 active shooters....
1. in spite of my criticism to follow, thank you for your post and point…
2. um, i call ‘bullshit’ on the 12 ‘active shooters’ (whatever, when LEO start in on their own lingo, perverting and changing the meanings of normal words, you know they are hiding shit they are up to)…
and the feebs have ‘foiled’ a dozen ‘terrorist’ plots, too, doncha know ! ! ! /sarc off
3. simply put: an academic who is ‘selling’ (whether literally or not) a system (‘his’ system?) which predicts stuff and he says others say it did and stopped 12 shooters ? ? ? well, by gosh and by golly that surely must be true, then…
4. not only does LEO and the (in)justice department have a license to kill, they have a license to lie; i don’t trust ANY OF THEM, EVER… (they have earned that distrust)
5. you know what, i spy on 100% of the pre-criminal citizens, maybe i can stop a few crimes too…
BUT AT WHAT COST ? ? ?
they lose their souls, and we lose our freedom…
NOT a ‘trade-off’ i want to make…
even remotely…
our brief *SHOULD* be to maximize freedom, not minimize it… we are going the wrong way…
I think they should first apply this technology to predict which politicians are most likely to accept bribes, kickbacks and other corrupt forms of compensation.
They can follow that up by using to it to predict which police officers, prosecutors and other members of the justice department are most likely to accept bribes, manufacute evidence, torture suspects, etc…
After that, if there is anyone left to manage such a program they can direct it at “potential criminals”
They've already stopped 12 active shooters....
The claim that this already stopped “12 shooters” seems very specious at best, for a number of reasons — the most important one being that you’ve defined “active shooter” as someone who is not, in fact, an active shooter:
In other words, the definition is: an active shooter is one that has been designated as such by this system. It’d be more impressive if they were, you know, actually active shooters.
This is a mischaracterization. I doubt that a sizable percentage of people using social media are voluntarily donating their data for this use. It’s more likely that they don’t really understand that they’re being actively spied on.
By the way, this reads like you’re saying that everyone who uses the internet is voluntarily donating data for this use. That can’t be what you mean, but I wanted to double-check.
You and I are diametrically opposed in our reactions to this. I am the exact opposite of comforted by this effort. In fact, it’s pretty terrifying. Certainly scarier than terrorism or random nutcases shooting up schools.
Re: They've already stopped 12 active shooters....
Active shooters appears to be their term, not the poster’s definition. I’m getting the vibe they mean people who clearly were in the stages of putting together some terrible event and just short of acting the event out.
As to people ‘voluntarily’ providing information, it is considered open source in that when we comment – like now, there is no expectation of privacy and we voluntarily participate on the internet in so many ways. No one is forcing us to to do. That the majority don’t understand the lack of privacy doesn’t change the voluntary nature of the participation.
The video tapes mention the open source intelligence as social media. I also read the post and the comment ‘comforted’ was part of the response. The other part was the author was less than comforted by the reality it could go too far.
Re: Re: They've already stopped 12 active shooters....
“Active shooters appears to be their term, not the poster’s definition.”
Yes, but one the commenter accepted and embraced in her essay.
“I’m getting the vibe they mean people who clearly were in the stages of putting together some terrible event and just short of acting the event out.”
You don’t need to go with a “vibe”. They specifically defined the term, and I quoted their definition.
” it is considered open source in that when we comment – like now, there is no expectation of privacy and we voluntarily participate on the internet in so many ways.”
In a venue like this, yes. But not on the internet as a whole, which is what it sounded like she was saying — thus my request for clarification.
“That the majority don’t understand the lack of privacy doesn’t change the voluntary nature of the participation.”
That’s a different thing. Not understanding that you’re being spied on absolutely does change the voluntary nature of “donating your data” to the spies. Nobody is volunteering to donate their data to the NSA, for example.
“The video tapes mention the open source intelligence as social media”
Yes, my question was about the commenter’s inclusion of “the internet” along with social media. “The internet” is a much larger sphere and includes lots of things that are unambiguously private.
“If you end up on that list, there’s a reason you’re there.”
Ask Rahinah Ibrahim about that. Sure there was a reason she was on that list. Someone fucked up and everyone believed that she was there for a good reason. And when it finally became obvious that it was because someone fucked up, they did everything they could to avoid accountability.
Start now!
Based on past history the city of Chicago politicians and justice system is rife with corruption. We need to imprison them NOW to end this scourge.
If they can see into the future in Chicago,
why didn’t they stop Obama, Holder, and Rahm Emmanuel?
What if they started feeding the personal information and entire employment history including every case they were involved with, partners they had, training, etc. into a computer model that was designed to predict the likelihood of officers to engage in corruption and/or abuse their authority with regards to the public for the purposes of flagging potentially bad officers that needed to be watched for those sorts of things. Do you think the Chicago PD would support such a program if they were assured that the algorithm was really good, and “if you end up on that list, there’s a reason you’re there.” If not, then why would it be ok for such a program to be used to analyze the public in such a way?
Why is everyone in a position of power trying desperately to make the world a dystopia?! It’s not bad enough we’ve got massive surveillance and governments trying to grant sovereignty to corporations, now we’ve got police departments forming “pre-crime” divisions?!
More proof that we do indeed live in a police state.
I wonder what the algorithm would say
Violation of civil, constitutional and statutory rights is a crime. So is conspiracy to commit those crimes. I wonder how many police officers would end up on the list if the algorithm were expanded to include those sorts of crimes?
serious question
Do they plug into the algorithm that they visit and tell the “could be” perp to behave? What does the model indicate the impact of this conversation has on the likelihood of the future crime?
Having this list is not the problem, but using it for sending people to intimidate people on the list is. Instead they should those people free passes to libraries, cultural places, sport events, support groups, or something like that.
Crime prevention doesn’t work through threats and punishment.
Re: Re:
“Crime prevention doesn’t work through threats and punishment.”
I really, really wish that law enforcement officers and legislators understood this truth.
Speaking as someone who works with cognitive models of human thinking and is quite familiar with the benefits and detriments of computer-run algorithms…
Whoever is championing the code should be taken out back and given some wall-to-wall therapy before being tied to a chair at the university’s “Programmer Ethics” equivalent. And probably a basic class on algorithms, heuristics and the like.
And if their university doesn’t cover such material… then that explains the problems, I suppose.
Computer decision-making should only be an *adjunct* to human decision-making in non-trivial circumstances. An algorithm is only as good as the worst of the people who created it AND the people who implemented it.
And while human decision-making, with on-the-fly application and development of new heuristics as needed, can generally adapt to situations with incomplete or incorrect information at least reasonably well (all things considered), computer decision-making just sits there and craps itself.
Re: Re:
Al algorithm that works as described (identifies individuals at risk of committing a crime) is working with extremely incomplete information for its modeling.
Now, using it to predict future lawbreaking and then comparing to future data (ie, do they commit violent crimes) could lead (assuming the model is clean) to pinning down innocuous causal factors that can be fixed without human risk or cost, but the way they’re using it is most certainly Doing It Wrong.
Why don’t they just arrest everyone on the list and charge them with crimes they might commit in the future? Yeah! Great idea!
This would be a violation of the 4th amendment, as it is illegal search and seizure. Collecting information and storing it before a crime has been committed. This has already been tested in the Supreme Court on mass collections of DNA for a database that would quickly pinpoint the perp of a crime. The Supreme Court says, not no, but hell no.
Poverty = Crime?
“…where crime naturally would be greater due to poverty?”
Has this link really been established? If so, please show me the research that proves it.
"Slippery Slope" is usually a fallacy...BUT
But in every instance of government action, it’s a valid predictive method. I mean, people are worried the algorithm might not be accurate? What the hell, they use dogs ‘alerting’ as evidence to warrant sometimes destructive searches, and just suppress studies that place the accuracy of dope-sniffing dogs at around 60%.
So yeah, everything about this “It’s just to give those people a warning.” system is just teetering on a steep, greased hill. Combined with militarization of PDs across the country and an ever intensifying ‘us against them’ mentality, I don’t think the citizenry will be well-served (or protected) by these pre-crime efforts.
That said, a couple of points:
a) I very much doubt a single person on the list is without previous violent violations. They are NOT going to anticipate someone’s first crime.
b) IF we were to allow for the possibility of good intentions from the CPD (not a given), whatever the underlying causes (or ‘disease’ in the analogy) of crime, if the ‘symptom’ is murder or violent assault, the police are obliged to deal with it, without any obligation to make sure every demographic is happy, prosperous, well-educated and non-violent.
Just sayin’…
cops and robbers
I always thought that cops loved crime. Don’t firemen love a fire? Don’t painters love a peeling or unpainted wall? Doesn’t a seamstress love a fraying seam of cloth?
Cops are the purported repairmen of societal fray. But since all municipal police departments ultimately become unionized, they are also intent on making sure that their livelihood is maintained and expanded. They do this in tandem with the bureaucratic/political unions, whom also use crime as a platform for the justification of their existence, to a partial extent.
Show me one police officer that wants crime to end, and I’ll then show you an axeman that wants the forest to die.
algorithm
Algorithm you say….well that just working out so damn well for Facebook and their report button algorithm