Latest / 7am / Police trial live face-scanning technology: what could possibly go wrong?
Transcript
- 0:01Speaker 1: I'm Daniel James and you're listening to 7am. West Australian
- 0:10Speaker 1: police recently finished trialling a technology that can identify people
- 0:14Speaker 1: as they walk past, revealing they scanned nearly 1.4 million
- 0:18Speaker 1: faces over a period of five months. Police say the
- 0:21Speaker 1: cameras helped them make 122 arrests and are now considering
- 0:25Speaker 1: whether to use the technology more widely. But is catching
- 0:29Speaker 1: 122 people worth putting nearly 1.4 million others under surveillance?
- 0:35Speaker 1: And how much privacy should Australians be expected to give
- 0:37Speaker 1: up in the name of public safety? Today, we revisit
- 0:41Speaker 1: our conversation with University of New South Wales cyber security expert,
- 0:44Speaker 1: Professor Richard Buckland. Recorded at the start of the trial
- 0:48Speaker 1: on the dangers of live facial recognition and what happens
- 0:52Speaker 1: when police surveillance powers expand. It's Sunday, November 8th, and
- 1:03Speaker 1: this episode first aired in June 2026. This is.
- 1:11Speaker 2: Our custom-built live facial recognition van, which is going to
- 1:15Speaker 2: be used to assist police operations. The van has a
- 1:18Speaker 2: number of features inside it, including a computer system, which
- 1:22Speaker 2: processes images from high-quality cameras, which then compares those images
- 1:26Speaker 2: against a database.
- 1:27Speaker 1: Richard, WA Police have announced a trial of live facial
- 1:30Speaker 1: recognition technology. What can you tell us about the technology
- 1:34Speaker 1: and how will this trial work?
- 1:36Speaker 3: So this is a very limited trial they've announced. It's
- 1:39Speaker 3: just in one van which will be marked saying, according
- 1:43Speaker 3: to the images we've seen, that there's facial recognition running.
- 1:47Speaker 3: And how it works is they are looking for particular
- 1:51Speaker 3: people in this case.
- 1:52Speaker 4: So they've got presumably a.
- 1:54Speaker 3: Database somewhere of people faces of people that fit a
- 1:58Speaker 3: certain category and they're trying to see if any of
- 1:59Speaker 3: those people walk past the van.
- 2:01Speaker 5: The important thing for us in our police force working
- 2:04Speaker 5: with our community is transparency and public safety. That's why
- 2:09Speaker 5: the van has police written on the side. It has
- 2:12Speaker 5: very clear signage about what it's doing.
- 2:14Speaker 3: And the categories they've mentioned are people who are wanted
- 2:18Speaker 3: for domestic violence, known sex offenders. So pretty reprehensible people
- 2:23Speaker 3: and the sort of people you'd really want the police
- 2:25Speaker 3: to be able to catch. And it would seem reasonable
- 2:27Speaker 3: that they'd be out in public looking for these people.
- 2:29Speaker 5: This is not about mass surveillance. This is about specifying
- 2:32Speaker 5: those in our community who are wanted by police because
- 2:37Speaker 5: there is an arrest warrant out for them, who are
- 2:40Speaker 5: on sex offender lists, who should not be in the vicinity,
- 2:43Speaker 5: particularly of young children. But it's also about helping missing people.
- 2:48Speaker 3: As we understand it, they're obviously a bit coy about
- 2:50Speaker 3: the details, but the images are scanned in real time.
- 2:54Speaker 3: If there's a match, then the recording is left showing
- 2:58Speaker 3: those faces and the police are alerted in real time,
- 3:00Speaker 3: presumably in the van. And then the other faces that
- 3:04Speaker 3: don't match anyone on this list are pixelated.
- 3:07Speaker 5: This is actually a way that we can increase the
- 3:10Speaker 5: freedoms and the privacy of our community.
- 3:12Speaker 1: The police commissioner says this is actually a less intrusive
- 3:15Speaker 1: option than what police currently rely on, which is CCTV
- 3:18Speaker 1: footage from private citizens.
- 3:20Speaker 6: Look, I've always been concerned about the spread of CCTV
- 3:25Speaker 6: around our community, particularly the ownership of the cameras, where
- 3:29Speaker 6: the information's stored, how long they're stored. We don't even
- 3:32Speaker 6: know half these cameras and who's using them.
- 3:35Speaker 4: What do you make of that argument?
- 3:36Speaker 3: Yeah, I mean, it's the argument that we always hear
- 3:39Speaker 3: with privacy-related things. I'm sure the police will continue using
- 3:43Speaker 3: the methods that they currently have. So there's in no
- 3:45Speaker 3: sense is adding this into the mix less intrusive. It's
- 3:48Speaker 3: actually widening the range of.
- 3:49Speaker 4: Surveillance powers they have.
- 3:51Speaker 3: The argument's not going to the heart of things for
- 3:54Speaker 3: a number of reasons. It is certainly the case that
- 3:57Speaker 3: the police will use this to catch people that everyone
- 3:59Speaker 3: agrees should be caught. The problem isn't with the software
- 4:04Speaker 3: doing its job. The problem is what are the consequences
- 4:08Speaker 3: to everyone else? What's the consequence to a society? of
- 4:12Speaker 3: having cameras everywhere. And if this goes, continues on this trajectory,
- 4:16Speaker 3: then this software now could be used once the trials
- 4:20Speaker 3: run and everyone's happy with the trial, which achieved its no-blame.
- 4:23Speaker 3: It could then be used to monitor everyone everywhere in
- 4:27Speaker 3: a way we've never had before. Once you've got technology
- 4:30Speaker 3: able to do something, then everyone sits around thinking of
- 4:33Speaker 3: reasons and ways you can use it to do other things.
- 4:36Speaker 3: And this is scope creep. And we see this over
- 4:38Speaker 3: and over again with privacy. So the Example I'm really
- 4:42Speaker 3: familiar with is in New South Wales, we had a
- 4:45Speaker 3: trial of cameras to detect people using mobile phones because
- 4:49Speaker 3: a lot of road safety data, as I'm sure everyone knows,
- 4:52Speaker 3: shows that mobile phone usage by the driver was involved
- 4:55Speaker 3: in a frightening number of accidents that caused harm to
- 4:58Speaker 3: lots of other people, innocent people, when the driver was
- 5:01Speaker 3: using it. So they put these cameras in locations and
- 5:04Speaker 3: they're very high resolution that can detect if someone's using
- 5:08Speaker 3: a mobile phone.
- 5:09Speaker 4: It was a success.
- 5:10Speaker 3: It identified lots of people using the phone as on
- 5:14Speaker 3: the basis of that successful identification, those cameras have been
- 5:16Speaker 3: rolled out and they're increasingly appearing around New South Wales.
- 5:20Speaker 3: But the scope creep happened when someone said, we've got
- 5:22Speaker 3: this footage, what else can we use it for? We
- 5:24Speaker 3: can easily see if someone's wearing a seatbelt using this.
- 5:27Speaker 3: So now those cameras are used for seatbelt offences, not
- 5:30Speaker 3: wearing a seatbelt. Sure, it's still a good thing to do,
- 5:33Speaker 3: I guess, but the difference is with a mobile phone,
- 5:37Speaker 3: the argument in the trial was, You can kill innocent people.
- 5:41Speaker 3: Other people suffer from you breaking this law. If you're
- 5:44Speaker 3: not wearing a seatbelt, you're probably the only one that's
- 5:47Speaker 3: going to suffer from not wearing the seatbelt. It won't
- 5:49Speaker 3: go back to the public and it won't be tested again.
- 5:51Speaker 4: Now they've got that capability, it can be used for
- 5:54Speaker 4: other things.
- 5:54Speaker 3: And this is, I guess, what's scary and why it's
- 5:56Speaker 3: always good to check back after a short period of
- 5:58Speaker 3: time what the thing's being used for and by someone independent,
- 6:02Speaker 3: is this in line with our intents and values?
- 6:04Speaker 1: Right now they're using for this trial a very clearly
- 6:08Speaker 1: marked police fan, but police haven't ruled out using this
- 6:11Speaker 1: in a more covert form in future, including at protests.
- 6:15Speaker 1: How does that change the equation compared to what was
- 6:18Speaker 1: originally being trialled?
- 6:19Speaker 4: You're spot on.
- 6:21Speaker 3: If it's used at protests, for example, imagine we're back
- 6:25Speaker 3: in the old days where it was illegal to be homosexual,
- 6:27Speaker 3: a criminal offence. Then if there was a protest for
- 6:30Speaker 3: gay rights, having a camera there identifying people in real
- 6:34Speaker 3: time obviously means that a database can be built of people.
- 6:37Speaker 4: Now, you could argue, well, we could have.
- 6:39Speaker 3: The camera and not record it in real time. We
- 6:42Speaker 3: could have the camera and analyze this later. But that's
- 6:44Speaker 3: a much bigger job once you've got it. You probably
- 6:46Speaker 3: know yourself, if you get a book and read it,
- 6:48Speaker 3: it's easy. But if you put it in that pile
- 6:49Speaker 3: of books and that pile just gets bigger and bigger
- 6:51Speaker 3: and bigger, it's much harder to search. So giving people
- 6:54Speaker 3: real-time capability to update databases means you could identify people
- 6:59Speaker 3: who are violating the law, even when perhaps it's not
- 7:03Speaker 3: a law we're totally happy with. And it's a chilling effect, really,
- 7:06Speaker 3: on your ability to protest, like we saw in China
- 7:08Speaker 3: and Hong Kong.
- 7:10Speaker 1: Coming up, how facial recognition technology is being used elsewhere
- 7:14Speaker 1: in Australia and the world. Richard, what can you tell
- 7:23Speaker 1: us about how this technology is being used by police
- 7:25Speaker 1: in other parts of the world?
- 7:27Speaker 3: Yeah, so in the States, they do a lot of
- 7:30Speaker 3: facial recognition stuff. And it's being used to identify people
- 7:35Speaker 3: that the police are interested in because they've had a
- 7:38Speaker 3: long history of doing that. But I do think the
- 7:40Speaker 3: big problem in the States is the chilling effect, that
- 7:42Speaker 3: you would now be frightened to protest against things you
- 7:45Speaker 3: might think are egregious.
- 7:47Speaker 4: And then there's this other question, which.
- 7:49Speaker 3: Is once your movements are tracked and known and that
- 7:52Speaker 3: data is stored somewhere, who has.
- 7:54Speaker 4: Access to this?
- 7:55Speaker 3: So in the States, there are examples where the police individuals,
- 7:59Speaker 3: insiders inside the police, use it for tracking their spouse,
- 8:03Speaker 3: use it for finding spouses who've escaped them for family
- 8:06Speaker 3: violence reasons.
- 8:07Speaker 7: Orange County DA's office says this former Costa Mesa police
- 8:10Speaker 7: officer not only used these flock cameras to track his mistress,
- 8:13Speaker 7: her new boyfriends, as well as his current wife, but
- 8:15Speaker 7: also used confidential law enforcement systems to be able to
- 8:18Speaker 7: run people a number of times that had nothing to
- 8:21Speaker 7: do with his work.
- 8:22Speaker 3: One police officer used it to track down and date
- 8:24Speaker 3: someone he thought was interesting and work out where they
- 8:26Speaker 3: lived and where they went and all sorts of things
- 8:28Speaker 3: like that.
- 8:28Speaker 8: I have a former Milwaukee police officer pleading guilty today
- 8:31Speaker 8: to a misdemeanor. It comes after court records show Josue
- 8:34Speaker 8: Ayala searched the locations of his girlfriend and her boyfriend
- 8:38Speaker 8: more than 170 times using Milwaukee Police Department's license plate
- 8:42Speaker 8: reading technology.
- 8:44Speaker 3: So even if the government and the police as an
- 8:47Speaker 3: organization say they're going to look after the data and
- 8:50Speaker 3: it's been used for good purposes, of course, individuals have
- 8:52Speaker 3: access to the data. And then the extension of that
- 8:55Speaker 3: is if they break into this database, so do criminals.
- 8:59Speaker 1: Richard, now a Florida man is suing police after AI
- 9:02Speaker 1: facial recognition technology led to his wrongful arrest. How likely
- 9:07Speaker 1: are situations like that to arise with this technology?
- 9:10Speaker 4: Technology always makes mistakes.
- 9:12Speaker 3: The only way of stopping and making mistakes is spending
- 9:14Speaker 3: more money and often to reduce mistakes by a lot
- 9:17Speaker 3: costs a huge amount more. So usually we solve mistakes
- 9:21Speaker 3: by shifting them from one type of mistake to the other.
- 9:23Speaker 3: So there's type one and type two mistakes that this
- 9:26Speaker 3: software can make. One is where it wrongfully identifies an
- 9:31Speaker 3: innocent person as being a bad person. And the other
- 9:34Speaker 3: mistake is where a bad person doesn't get properly identified.
- 9:38Speaker 3: And really there's a null almost on the software. You
- 9:41Speaker 3: can choose where you set it. If you set it
- 9:44Speaker 3: too far, if you're too determined to catch all the
- 9:46Speaker 3: criminals and not let anyone through, then it will wrongfully
- 9:48Speaker 3: identify a range of people.
- 9:50Speaker 1: And with this technology, Richard, what about racial bias? Because
- 9:53Speaker 1: there's been an issue with technology like this in the past,
- 9:56Speaker 1: not really taking into account the diversity with communities.
- 10:01Speaker 4: What can you say to that?
- 10:02Speaker 3: Yeah, most of the data sets that things are being
- 10:04Speaker 3: trained on are white and white. middle-class white males and,
- 10:08Speaker 3: you know, high-profile people in disadvantaged groups, minority groups in America,
- 10:13Speaker 3: people with perhaps disabilities. You know, you can just imagine,
- 10:16Speaker 3: I haven't trained that on everyone in the world. If
- 10:18Speaker 3: your training data isn't good enough, it's making a guess
- 10:21Speaker 3: based on what it's saying. So, yeah, absolutely.
- 10:23Speaker 4: There's all these sort of biases in the software.
- 10:27Speaker 3: Even facial recognition with human makes mistakes.
- 10:29Speaker 4: It's not a very good way of identifying people.
- 10:31Speaker 1: Another one of the police commissioner's arguments was that This
- 10:34Speaker 1: technology is already being used out in the community. How
- 10:37Speaker 1: widespread is this both inside and outside of Puliso?
- 10:40Speaker 4: Yeah, I mean, it is.
- 10:42Speaker 3: Can I just say that's not a brilliant argument.
- 10:44Speaker 4: My children try that one.
- 10:46Speaker 1: Everyone's doing it is the argument.
- 10:47Speaker 3: Yeah, everyone's doing it so I can. I mean, this is,
- 10:49Speaker 3: I guess, just taking a quick diversion, this is really
- 10:53Speaker 3: the problem with the cameras, you put your finger on
- 10:55Speaker 3: it in a nutshell, that we're slowly drifting by a
- 10:59Speaker 3: series of incremental steps towards And at every point, we
- 11:02Speaker 3: can make an argument that if you don't look at
- 11:03Speaker 3: it too carefully, you think, oh, yeah, that's reasonable. But
- 11:06Speaker 3: as we make step by step by step, people's privacy
- 11:09Speaker 3: and ability to live a life not being.
- 11:12Speaker 4: Surveilled is diminished.
- 11:14Speaker 3: And arguments then are, well, you've already lost it, so
- 11:16Speaker 3: we can just keep doing it. So this incremental drift
- 11:19Speaker 3: is actually quite scary. Visual capture is very widespread. There's
- 11:23Speaker 3: cameras everywhere, if we take it step by step, including
- 11:26Speaker 3: lots of private cameras pointing onto the street.
- 11:28Speaker 4: It's very hard to walk around.
- 11:29Speaker 3: In Sydney, if you drive down the road to the
- 11:31Speaker 3: Blue Mountains. There used to be one surveillance camera at
- 11:34Speaker 3: the beginning. There's now one every 200 metres. I don't
- 11:36Speaker 3: know how they got the budget to do that. So
- 11:39Speaker 3: then there's facial recognition applied to recordings. And that is
- 11:44Speaker 3: not so widely done by people. That's more forensic people.
- 11:47Speaker 3: But real-time facial recognition, that has commercial advantages.
- 11:50Speaker 4: And people do use that.
- 11:52Speaker 3: And Bunnings and Kmart is a famous case that happened
- 11:57Speaker 3: recently where they were using it to identify people that
- 12:00Speaker 3: would be abusive to their staff or people that committed
- 12:04Speaker 3: various sorts of fraud, either checkout fraud, not scanning things,
- 12:07Speaker 3: or people who made multiple fraudulent returns and claimed money
- 12:10Speaker 3: back on items they hadn't bought in the store.
- 12:12Speaker 9: Some of our most popular retailers are being accused of
- 12:14Speaker 9: invading customers' privacy by installing facial recognition cameras in their stores. Kmart,
- 12:21Speaker 9: Bunnings and the good guys have been blasted by.
- 12:23Speaker 3: This was met with a lot of public shock that
- 12:27Speaker 3: they were recording people and identifying people without telling them.
- 12:30Speaker 9: Kmart has been found to have breached privacy laws in
- 12:34Speaker 9: its use of facial recognition technology in stores across the country.
- 12:37Speaker 10: Bunnings has been given the green light to use facial
- 12:40Speaker 10: recognition technology across its stores. The hardware giant maintains it
- 12:44Speaker 10: will only use the technology to prevent shoplifting and protect staff.
- 12:49Speaker 10: It won an appeal in the Administrative Review Tribunal following
- 12:53Speaker 10: an earlier ruling that it breached customer privacy.
- 12:56Speaker 3: So yeah, it is being used and I would expect,
- 12:58Speaker 3: unless there is a stop to it, if we just
- 13:00Speaker 3: keep saying, yeah, that sounds reasonable, we could eventually live
- 13:03Speaker 3: in an Orwellian world. Wherever you go, you're recognised and noticed,
- 13:07Speaker 3: your movements are known. And that's an obvious problem for
- 13:10Speaker 3: some people, witness protection plans, people who are trying to
- 13:13Speaker 3: hide from abusive partners, people trying to whistleblow. We've seen
- 13:17Speaker 3: some whistleblowing happening recently, the KPMG case, they're now saying
- 13:21Speaker 3: they wouldn't whistleblow again, that the fact of surveillance is
- 13:24Speaker 3: actually enough to stop people wanting to speak out. So yeah,
- 13:28Speaker 3: so the world we're moving into, if this is allowed,
- 13:31Speaker 3: would be a very different world. And it'd be good
- 13:33Speaker 3: for us to stop and think, is this what we want?
- 13:35Speaker 1: To that effect, Richard, this is a trial, but history
- 13:39Speaker 1: shows us that when police are given powers, that those
- 13:42Speaker 1: powers are usually extended. So What checks and balances are
- 13:46Speaker 1: currently in place to show that there isn't overreach with
- 13:50Speaker 1: this technology?
- 13:51Speaker 3: My understanding, and I could be wrong, is very little.
- 13:54Speaker 3: We've got some words saying this is what we're going
- 13:56Speaker 3: to use it for now. And credit to the police
- 13:59Speaker 3: commissioner for saying this. Most commissioners of police in most
- 14:03Speaker 3: states do this stuff under the radar as much as
- 14:05Speaker 3: they can. It's really good that he spoke about what
- 14:08Speaker 3: they're doing and he has spoken about their intentions, but
- 14:11Speaker 3: he's refused.
- 14:11Speaker 4: To rule out other ways other intentions.
- 14:14Speaker 3: I mean, the things you'd like to see, an independent
- 14:17Speaker 3: review after a short period of time, constraints on what
- 14:20Speaker 3: the surveillance could involve, the sorts of things we saw
- 14:22Speaker 3: in COVID when the government tried to bring out the
- 14:25Speaker 3: contact tracing app and there was a lot of alarm
- 14:27Speaker 3: that the government would get everyone's location data in real time.
- 14:30Speaker 3: The government then put a whole lot of quite good
- 14:33Speaker 3: restrictions on how it could be used with the sunset clause.
- 14:36Speaker 3: It would be good to see this sort of thing,
- 14:38Speaker 3: but it's very hard to get anyone to put constraints
- 14:41Speaker 3: on the police. Really, both sides of politics are almost
- 14:43Speaker 3: outdo themselves for giving the police extra powers. I would
- 14:47Speaker 3: hope that there will be a review, an independent review.
- 14:50Speaker 3: I would hope there'll be restrictions on who can see
- 14:53Speaker 3: the data. I would hope there'd be requirements about the
- 14:55Speaker 3: data being stored in a way that inside is inside
- 14:58Speaker 3: the police can't access it and that accessing get reviewed.
- 15:03Speaker 3: I would hope that the data be encrypted in a
- 15:05Speaker 3: reasonable way. So if hackers break in, they can't access
- 15:08Speaker 3: the data. I would hope recordings would be destroyed and
- 15:11Speaker 3: no one ever likes destroying data. So they won't unless
- 15:14Speaker 3: forced to destroy it.
- 15:15Speaker 4: They'll keep it. And then the last thing is the pixelation.
- 15:18Speaker 3: There should be legislative restrictions, not just down to the
- 15:21Speaker 3: current way it's done, saying this pixelation has to happen
- 15:26Speaker 3: either on the camera or on device, but certainly before
- 15:28Speaker 3: any recording is made.
- 15:30Speaker 4: So there is no way someone can ever get hold
- 15:32Speaker 4: of the unpixelated version. It's not good enough to say, oh, we.
- 15:35Speaker 3: Promise we never will, or we have checks and balances
- 15:37Speaker 3: in place to show it never will. There should be
- 15:39Speaker 3: an independent audit to make sure you can't get anyone
- 15:41Speaker 3: else's ID in.
- 15:44Speaker 1: Well, thank you very much, Richard. We should note that
- 15:46Speaker 1: the police say that they won't store the data of
- 15:48Speaker 1: Australians who aren't on their wanted list. But in the meantime,
- 15:52Speaker 1: what could possibly go wrong? Richard, thank you so much
- 15:55Speaker 1: for your time.
- 15:56Speaker 11: Thanks, Daniel.
- 16:09Speaker 1: 7am, we'll be back tomorrow with Midnight Oil frontman and
- 16:12Speaker 1: former Federal Environment Minister, Peter Garrett. He's led an independent
- 16:16Speaker 1: inquiry that's calling for Australia's $ 368 billion Orca submarine deal
- 16:21Speaker 1: to be scrapped and says Australians have been deceived about
- 16:24Speaker 1: what they're signing up for.
- 16:26Speaker 12: The deception, Daniel, is that the majority of Australians were
- 16:29Speaker 12: not aware when Scott Morrison announced the Orca steel five
- 16:33Speaker 12: years ago that they would have a bill for $ 33
- 16:37Speaker 12: million a day for the next 30 years for submarines
- 16:40Speaker 12: that may or may not arrive. This is something which
- 16:43Speaker 12: we came to realise many, many people were unaware of.
- 16:46Speaker 4: It was a deception.
- 16:48Speaker 1: That'll be in your feed on Monday. I'm Daniel James.
- 16:51Speaker 1: Thanks for listening.