For this system there is no requirement that criminals faces are used to test it.
At first it's a trial, experimental. Then it's standard, but optional and with incentives to do it. Then it's standard but you can opt out and there's lots of signs to tell you how. Slowly they make it more uncomfortable to opt out, and the signs disappear, and they berate you for trying it. Eventually it is no longer legal to opt out.
Any one step is hard to fight back against without looking like a loon to people who don't care about this kind of thing, and over the course of a few years you can take people gently through a process that, had it happened suddenly, would never have succeeded due to pushback.
Opposition can now be met with "it's just a trial, relax". Then you expand the trial a bit. Keep running it until the discourse moves on. And then a few years down the line, when you roll out the full system and people complain, you can counter with "these people are exaggerating their concerns, we've trialled this for yeaaars and there were very few problems!"
It’s especially disappointing when the entire system is foiled by the balaclavas worn by so many of the country’s criminals every day anyway.
https://en.wikipedia.org/wiki/Killing_of_Jean_Charles_de_Men...
Someone who's face resembles that of a known terrorist was detected entering the station 8 seconds ago. He's carrying a backpack and will be on a crowded platform in 37 seconds time. There are armed police in the ticket hall. What do you do? Only 31 seconds now...
Walk up and place him under arrest. He might have a bomb. He might have a trigger for it, or he might have a dead-man's switch. It might detonate and kill you along with bystanders. That's a risk you take.
Unless you know with certainty that he has a bomb and does not have a dead man's switch, you don't just take his life because you're concerned about what-ifs.
It's no different than cops killing a man because he was holding a wallet, which they imagined might be a gun (e.g. Amadou Diallo).
Police need to stop pretending that they exist within fiction, where the outcome is known by the author. Just because you're afraid of something doesn't mean it's an actual threat.
Since 2001, there's been a real uptick in confusing fear and suspicion for evidence and knowledge.
Is the suspect black?
(I'm asking this because there is evidence suggesting current facial recognition technology is less accurate at matching faces from some demographics, including black people. Why, what did you think I meant?)
https://www.btp.police.uk/SysSiteAssets/media/images/british...
It's one thing to identify a person from a database, another from camera 1, then camera 2 etc. My guess is the latter has a much lower false positive rate
Also, this seems to just be a trial expanding existing time and location scoped deployments of live facial recognition into a new environment, rather than a trial run for always-on live facial recognition across the tube network.
The real issue is not (just) this specific use case, rather it's the govt generally knowing where people are going and where they have historically been.
It’s already easy for the police to prevent protests in a practical sense. It’s a political and legal issue more than a technical one.
If you need to arrest someone in London (someone who might be about to cause harm) then there's a very good chance they will go through a tube station, and they tend to have relatively few, easily controlled exits. Most tube stations have two or less.
A few high profile tube stations occasionally have armed police presence.
But I think the idea that face recognition is being used to suppress protest is pretty overcooked; apart from the utter silliness around the banning of Palestine Action, we have a solid right to protest and we do.
The main problem is London has far too few police officers, and it has a transport system that is under consistent terror threat.
The latter is legitimate, isn't it? Do we not anticipate trying to arrest people who might be in the process of committing a terror attack, on the balance of evidence about their conduct? Or are we just going to wait?
Not to say that hasn't gone wrong (infamously and tragically, although that was largely a failure of conventional surveillance) but one key part of mass surveillance policing is arresting people who are clearly about to do a thing.
That might indeed be disruptive protest where it is illegal — for example if you know someone is intending to deface a painting or a monument, or has been planning to attend a protest to disrupt it.
I think people look at face detection policing in this context and imagine thoughtcrime, when the reality is e.g. finding a disturbed person who has been doing a dry run on pushing a stranger off a tube train platform (a real problem that seems to happen every couple of years), catching someone who has been reported sexually assaulting fellow passengers or who has violently assaulted another passenger or someone above ground (a regular problem), trying to disrupt county-lines drug distribution etc. (absolutely constant), or, heaven help us, finding someone who has been spotted at a national railway station with someone on terror watchlists who has then disappeared into the underground.
London Underground isn't a thing that can be effectively managed with ordinary policing. It has 272 stations, over 70 in central London alone. During rush hour there are up to half a million people in that system at any one time. Tube exits are choke points.
No, arresting people for crimes they haven’t committed is horrifying and bizarre.
We're talking about preparation, and the equivalent of reasonable suspicion for a stop or probable cause for an arrest in the USA, where people absolutely are arrested every day for things an officer has seen enough evidence to believe are about to happen.
e.g. when they do things like holding illegal employment blacklist[1][2][3] of innocent people.
I don't want to live in a country where, for example, the police can punish you just because they don't like trade unions.
What's I find ironic is that all this police state bullshit never seems to actually deal with antisocial behaviour or low level crime, it's always for political infractions.
We have more CCTV in London than almost anywhere else on the planet and you still get harassed outside tube stations by organised gangs of soliciting very forcefully for "charity" collections.
1. https://www.parliament.uk/external/committees/committee-news...
2. https://ico.org.uk/for-the-public/ico-40/construction-employ...
3. https://www.london.gov.uk/who-we-are/what-london-assembly-do...
There's a gray area for sure, and in the US you can, for example, be charged with conspiracy to commit murder without actually getting to that end goal. I'm always uncomfortable with how a court decides to allow a trial in such a case, though if someone has actively been surveiling their target, collecting materials or weapons, and ideally documenting their intentions somehow then I do understand why most would consider that itself punishable.
If a person without a warrant gets off the tube and walks in a general direction of a protest, I wouldn't see any grounds for arrest. I'm sure there's still gray area there, say they're carrying smoke grenades or flash bangs or something - though in that case I'd assume the potential crime is possessing those items rather than walking in a direction.
The police don't manage mass protests this way either. If they want to stop you joining a disruptive protest they just cordon off streets, close tube exits, stand in the way and tell you to turn back.
But severely disruptive micro-protests are now a significant problem in London; I am quite relaxed about the idea that someone known to be planning to seriously disrupt might get tracked to their destination, and I suspect so are they, or they aren't very serious about it.
I think people from the USA really misunderstand how policing works here. When we say we are "policed by consent" it is taken pretty seriously.
Back when I was a student, this idea of facial recognition was only a glimmer of possibility and the idea of all of this would have got my freedom fighter hackles up.
But if you've ever been in a tube station, one of the deep lines, when they've had to close exits and you're one of thousands of people who need to get out, you realise that these are high-pressure environments where even British reserve and fairness begin to fray. They are very vulnerable places. So I look at this plan and I am like… nope, doesn't bother me. Getting in and getting out at rush hour bothers me.
I am not without my concerns about the mobile recognition vans; I think there is a risk that they have a chilling effect on gatherings that is disproportionate. I dislike that concept in much the same way that I dislike our regional police force's tendency to get the helicopter out when a bit more road policing might do the job.
But inside the tube? Nope, facial recognition seems like a pretty worthwhile tool, not least because it could be so efficiently deployed.
Every passenger taking the "nope" lane.
If you follow the links there is a lot of information about the trial, including a whole series of documents identifying the legal framework that authorises LFR, including rationale under GDPR, Data Protection etc.
Of course, this technology could be used for bad purposes. But within a democracy, policing happens with the consent of the people, expressed through various channels including legal (parliament and local government) and police commissioners. What is interesting is the way that these frameworks are used to constrain the purpose and extent of this processing. For example, at the moment the images on the watchlist can only be drawn from custody images.
The police already have powers to seize CCTV and door-cam imagery in order to investigate crime. They need a warrant from a judge - so it isn't justified without good reason such as serious crimes. This is the way you constrain these things - you put the regulatory framework around them. You can't put the technology back in the box, so you put rules and expectations around how it is used.
Antiterrorism and anti-disorder laws sold to the public as needed to prevent serious threats have been used extensively outside of what I think most would consider reasonable to surveil people, restrict freedoms and harass. The prime example are the stop and search laws, used to search a large numbers of people, predominantly based on racial stereotype, with zero impact on terrorism.
Remember that the Met police is known to be institutionally racist & sexist. It is not an organisation to be trusted.
In the UK if you keep your mouth shut during an investigation, it may harm your defense. There is zero right against self-incrimination. It is also the place where you can be jailed for not providing the PIN to your phone. I could go on.
What we can do, and should do as a society, is decide on how that technology should be used. And that means regulating it, in the same way that we regulate other technologies. For example, we regulate cars. Not just who can use them, but how they have to be constructed too. And we also place limits on how the authorities can use them, for example the rules that the police must obey when in an emergency response situation.
We are a democracy, so these rules are made by the people we have elected. so when it comes down to it, yes, this is what we as a people have asked for. And we are responsible for the conditions that the police can use this technology under.
FWIW, the capabilities and resources that the police have to track us 24/7 are an order of magnitude less than Google or Meta has. And they are an awful lot less accountable.
This will be a good argument when Google or Meta are able to arrest you.
And the thing is that they're not necessarily wrong to want some of those things, but they don't understand or don't care about collateral damage or what you might perceive as a lack of freedom. So they will happily vote for it.
Democracies vote to shoot themselves in the foot all the time. Look around you.
By "may harm your defence" what is meant is that Miranda was modified to inform people who are arrested that magistrates and juries are fully allowed to infer that you might be seeking to mislead the court if you didn't tell the police something significant that you later rely on as a defence in court. You're still allowed to do it, and they are allowed to conclude it is sketchy. That's the point of the modification.
Juries do and did this anyway.
The modified Miranda doesn't say you have no right against self-incrimination; it now warns you that if something is crucial to your defence it might be in your interest to mention it so the jury doesn't think you are trying to shine them on.
It's important to understand that this happens in the context of a country that does not have a plea bargaining mechanism. Cases go to court if it is in the public interest that they do (and the police don't decide this). But you can still no-comment your way through an interview and you aren't going to be bullied by a prosecutor into accepting a plea deal.
UK suspects on arrest are, nevertheless, read their rights in what is colloquially understood here as a "Miranda warning". It's my understanding that "Miranda rights" are a sort of colloquial shorthand in lots of countries, because we all see US crime procedurals.
(It's actually called "The Caution", and US Miranda rights were inspired by an earlier version, though AFAIR we borrowed back some changes that were influenced by the Miranda wording, before the contentious modification about "may harm your defence")
The subsequent "they modified Miranda" fuss made quite a bit of international news when it was introduced, with Americans sniping at us about changing it, arguing we'd taken away the right not to self-incriminate, etc., that we'd sullied the grand tradition of English Common Law, all that rather ironic snarking from the country that pioneered the abusive plea bargain system.
The Caution read at arrest is actually shorter in the UK; it doesn't mention that the cautioned have a right to an attorney or one will be appointed for them. They do, but the duty sergeant at the police station tells them that when they are booked.
The only realistic protection here is a “decoy” partition, a partition that provides access to a separate installation of the OS. That gives the device owner plausible deniability that their real system even exists.
Again, no, not just "in an investigation".
It has to be an investigation that has reached the point where they have gone to a judge and got permission to serve written notice. The judge will ask them if there is a necessary and proportionate reason. There is then a timeframe in which you have to respond.
Police can't simply demand your PIN and charge you if you don't comply.