500K Facial Recognition Scans at UK Stations Yield Zero Arrests, One False Positive (theguardian.com) 48
A six-month British Transport Police trial of live facial recognition at London railway stations scanned more than 500,000 faces, cost over 320,000 pounds, and produced just one alert with no arrests directly resulting from the technology. That one alert ended up being a false positive.
Despite those results, BTP has extended the pilot through November and expanded it to selected Underground stations. The Guardian reports: British Transport Police said that since the extension there had been three positive confirmed alerts of people who were subsequently confirmed to be complying with sexual harm prevention orders or other imposed court conditions. When Transport for London announced its support for the extension of the trial, it said LFR technology "will target and identify people on police watchlists at key stations chosen for maximum impact" and would "specifically tackle violence against women and girls, whose travel behaviors are shaped by experiences of sexual harassment and sexual offenses."
Fraser Sampson, a former biometrics and surveillance camera commissioner for the UK, said the police needed to show that the use of the technology was proportionate. "There many variables and all need to align: the choice of location, times of day, make up of the watchlist and likelihood of people being present and many more." Sampson, who is now a non-executive director of Facewatch, a company that operates facial recognition systems in shops, added: "We know the technology works, but using it in supermarkets to deter shoplifters and prevent attacks on staff is very different from deploying it to catch people on a public transport network."
"Success in a shop means no such people coming in. Success for the police trying to catch people means people being caught. In that respect the trial doesn't appear to have been very fruitful. Everything depends on the police watchlists; if you're not on a watchlist, the live FRT cameras do not "see" you and they don't retain your image in the same way as CCTV cameras. However, they are processing special category personal data and the technology must be used in a way that is appropriate, proportionate and necessary. It's for the police to show how any deployment meets those criteria and -- as the ICO [information commissioner's office] has recently reported -- for their oversight boards to hold them firmly to account in doing so."
Despite those results, BTP has extended the pilot through November and expanded it to selected Underground stations. The Guardian reports: British Transport Police said that since the extension there had been three positive confirmed alerts of people who were subsequently confirmed to be complying with sexual harm prevention orders or other imposed court conditions. When Transport for London announced its support for the extension of the trial, it said LFR technology "will target and identify people on police watchlists at key stations chosen for maximum impact" and would "specifically tackle violence against women and girls, whose travel behaviors are shaped by experiences of sexual harassment and sexual offenses."
Fraser Sampson, a former biometrics and surveillance camera commissioner for the UK, said the police needed to show that the use of the technology was proportionate. "There many variables and all need to align: the choice of location, times of day, make up of the watchlist and likelihood of people being present and many more." Sampson, who is now a non-executive director of Facewatch, a company that operates facial recognition systems in shops, added: "We know the technology works, but using it in supermarkets to deter shoplifters and prevent attacks on staff is very different from deploying it to catch people on a public transport network."
"Success in a shop means no such people coming in. Success for the police trying to catch people means people being caught. In that respect the trial doesn't appear to have been very fruitful. Everything depends on the police watchlists; if you're not on a watchlist, the live FRT cameras do not "see" you and they don't retain your image in the same way as CCTV cameras. However, they are processing special category personal data and the technology must be used in a way that is appropriate, proportionate and necessary. It's for the police to show how any deployment meets those criteria and -- as the ICO [information commissioner's office] has recently reported -- for their oversight boards to hold them firmly to account in doing so."
The scientific method (Score:5, Insightful)
Re:The scientific method (Score:4, Interesting)
Re: (Score:1, Informative)
A trail failing is actually a success.
To be completely fair to the system in place, we don't have the stats to know if it failed or not at its stated goal.
The cameras were only designed to detect faces and identify matches, then send that info in near-real-time to the police.
Out of half a million scans, how many matched correctly? They do not give that stat.
That the police acted on zero of them is not only expected but not the fault of the facial recognition system.
Never forget, the UK has had hundreds of stabbings, beheadings, and gang rapes
Re: (Score:2)
320K pounds is a pretty cheap test of the efficacy of the system. There's a lot to learn from this trial regardless if there were a thousand arrests or zero arrests.
Re: The scientific method (Score:2)
Re: (Score:2)
"We have NOT failed. We've merely found half a million ways that do not work"
Re: (Score:2)
A trail failing is actually a success.
It worked perfectly. Not a sing bad person on the watchlist wanted to take the risk of getting caught and stayed away. Keeping everybody else safer as a result.
Police are the dumbest of humanity (Score:3, Informative)
Re: (Score:2)
You can't reason with them. Don't even try it.
That's what political incompetence counts on (in order to preserve the status quo.)
Politics ain't cheap nor a clean business. Also, it's not about reasoning with them. It's about making it politically unprofitable to continue in a current path. Almost every single right you have today that didn't involve bloodshed in a rebellion or revolution was obtained like that.
The point isn't for arrests (Score:3, Insightful)
Nobody gives a shit when your car gets stolen. But you better believe that if you upset the Epstein class there are systems in place to deal with you.
If you want to see how fast you can get canceled for real as opposed to getting your third Netflix special tell some jokes about Israel's occupation of Palestine and let me know how that goes for you... Or crack a few jokes about how netanyahu knew about October 6th well in advance.
Re:The point isn't for arrests (Score:5, Insightful)
Re: (Score:2)
When they say its cost £500000, is that before or after selling the scan data to Palantir and others?
Re: (Score:2)
The system scanned 500,000 faces. It cost £320,000.
Re: (Score:2)
The system scanned 500,000 faces. It cost £320,000.
And after the sunk cost of the hardware and installation. The next 500k will be £20 for the electricity to run them.
Re: (Score:2)
sure, because it needs no servers, no dedicated personal, no maintenance... and the electricity cost rising, for sure it costs more than £20 :D
Re: The point isn't for arrests (Score:2, Flamebait)
No like Trump and... (checks notes) Fergie? Weird.
Re: The point isn't for arrests (Score:2)
Lol denying that Trump is literally the single best example of the Epstein class is like denying that supporting him is supporting child rape. Only those who wish they were raping children themselves are confused about either thing.
Re: (Score:2)
muhammad is the ultimate pedo supporter and child rapist enabler. But you have no problem supporting that...
No one familiar with me would say that, because of the things I've said about Islam and Mohammad here in the past. You've just completely destroyed all of the child rape support you were working for.
Re: (Score:2)
You support "palestine"
Wrong. I oppose genocide and there's no need for quotation marks there since it clearly is one.
you're all in on Iran
I'm all in on the problems with Iran being literally caused by our doing a coup there, and creating more martyrs there not being the way to solve them.
And Israel is your #1 enemy
No, the orange child rapist is my #1 enemy. He wanted bibi to hurry his genocide up, so he's a big part of the problem there too.
You hate Christians
I hate all religion, which is all for controlling people, which Christians love to do — but they hate to walk their talk.
Re: The point isn't for arrests (Score:2)
You can't even write "Jews" and you think I'm the problem? I don't hate Jews, I hate genocidists like you.
For now, that is... (Score:5, Insightful)
... if you're not on a watchlist, the live FRT cameras do not "see" you and they don't retain your image in the same way as CCTV cameras.
This whole thing strikes me as one of those "bargains" likely to be altered, with a "pray I don't alter it further" thrown in for bad measure.
Re: (Score:3)
Do you actually believe they aren't retaining the data? Why?
Re: (Score:2)
For me it's because there have been many, many examples of companies and organizations saying that they are not retaining the data, and then it is revealed that they are, in fact, retaining the data.
It's never easy to tell if it happens in any given case because of maliciousness or incompetence, but it happens repeatedly.
Re: (Score:2)
For me it's because there have been many, many examples of companies and organizations saying that they are not retaining the data, and then it is revealed that they are, in fact, retaining the data.
It's never easy to tell if it happens in any given case because of maliciousness or incompetence, but it happens repeatedly.
At this point, you'd have to be completely and utterly disconnected from any form of our current reality to ever believe any entity saying that they aren't retaining any data. They're either retaining it themselves, or passing it on to some other service to retain it so that they can "legally" claim they aren't retaining it.
Re: (Score:2)
considering how people who have only their name and ethnicity in common have been singled out for...."enhanced scrutiny", i'm not inclined to trust the watchers even a little bit.
Must be blithering idiots in charge (Score:2)
They should be charged the full cost of their follow-up failures and fired.
Re: If only... (Score:2)
My guess is (Score:2)
Wanted people drive cars or bikes, where a face-mask doesn't raise suspicions.
Rule #1 (Score:2)
Unsurprising, no face recognition needed... (Score:2)
Sorry (Score:1)
Sailing close to the wind (Score:3)
The UK has a cultural sensitivity about having a right to privacy. But I think we make a mistake by considering the rules in this area some sort of a fixed "red line", rather than a spectrum.
The law, by nature tends to set "red lines", so that we can rule that this case was illegal, while that case was legal". But with an issue like privacy, the reality is that there is not a clear red line as such. As privacy is gradually encroached upon, we feel increasingly worse about it.
Having an army of cameras automatically detecting people who look like those on watchlists, and being used in a "proportionate" manner, might be legal, but it certainly sails close to the wind.
success (Score:2)
and produced just one alert with no arrests directly resulting from the technology.
A resounding success. No large-scale demonstrations, damaged cameras, lawsuits or political suicide for those involved. So we can expect this to be rolled out fully.
Crime prevention? Who said this has anything to do with crime?
Let's use human intelligence (Score:2)