Latest / Stuff They Don't Want You To Know / CLASSIC: At What Price Safety: The Conundrum of Clearview AI
Transcript
- 0:00Speaker 1: Friends and neighbors, fellow conspiracy realist, It is a government
- 0:04Speaker 1: and AI week here on stuff they don't want you
- 0:08Speaker 1: to know. So for Tonight's classic, we looked back through
- 0:12Speaker 1: the archives and we found some stuff that we had
- 0:16Speaker 1: been sort of flum mixed about for a while, for
- 0:18Speaker 1: more than five years now.
- 0:20Speaker 2: Oh yeah, we talk a lot about the technology that
- 0:22Speaker 2: was shown off in that certain Tom Cruise movie. You
- 0:26Speaker 2: remember the one, the one with the psychics, uh huh,
- 0:32Speaker 2: the ones that could pre crime everybody. Well, one of
- 0:35Speaker 2: the big things that was shown off in that movie
- 0:37Speaker 2: that struck all of us at the time when it
- 0:39Speaker 2: came out is this concept of facial recognition everywhere you
- 0:43Speaker 2: go in a city. Well, there's a company that's been
- 0:47Speaker 2: trying to do this for a long time, and some
- 0:51Speaker 2: stuff went down in twenty twenty, and that's when we
- 0:53Speaker 2: made this episode.
- 0:55Speaker 1: Yes, this is where we introduce you to the concept
- 0:57Speaker 1: of clearview AI. The argument of supporters would be this
- 1:02Speaker 1: can prevent crime, This can make the world a safer
- 1:05Speaker 1: place through the power of pattern recognition and super juiced
- 1:11Speaker 1: up algorithms. But then, as we learn there are a
- 1:15Speaker 1: lot of detractors for technology like this being rolled out
- 1:19Speaker 1: and clear view stories they originally tell. It doesn't always
- 1:23Speaker 1: match the things people found out later.
- 1:26Speaker 3: Yeah, I mean this is a story that is more
- 1:28Speaker 3: relevant now than ever. So why don't we take a
- 1:30Speaker 3: quick break here we're from our sponsor and then jump
- 1:32Speaker 3: into this classic.
- 1:34Speaker 4: From UFOs to psychic powers and government conspiracies. History is
- 1:39Speaker 4: riddled with unexplained events. You can turn back now or
- 1:43Speaker 4: learn the stuff they don't want you to know. A
- 1:46Speaker 4: production of iHeartRadio.
- 1:58Speaker 2: Hello, welcome back to the show. My name is Met,
- 2:01Speaker 2: my name is Noah.
- 2:02Speaker 3: They call me Ben.
- 2:03Speaker 1: We're joined as always with our super producer Paul, Mission
- 2:06Speaker 1: control decand most importantly, you are you. You are here,
- 2:10Speaker 1: and that makes this stuff they don't want you to know.
- 2:15Speaker 3: Here in the.
- 2:15Speaker 1: Age of social media, right, we're recording this as longtime
- 2:20Speaker 1: listeners know on a virtual interface. Right, we're mentally and
- 2:28Speaker 1: soulfully together, but not physically together, and that may be
- 2:31Speaker 1: the case for some time. However, technology is allowing us
- 2:35Speaker 1: to feel closer.
- 2:37Speaker 3: Right.
- 2:38Speaker 1: And the crazy thing is this stuff is so cheap.
- 2:41Speaker 1: Now we're talking about the proliferation of cheap imaging technology.
- 2:46Speaker 1: You guys, remember when not everyone could afford a camera
- 2:51Speaker 1: editing software and stuff like that.
- 2:54Speaker 3: I mean, hell, I remember when not every phone had
- 2:56Speaker 3: a camera on it or it was like, who's got
- 2:58Speaker 3: a digital camera? Find one? You know, Now it's like
- 3:01Speaker 3: a non issue.
- 3:02Speaker 2: You know, Well, if you were going to take a
- 3:04Speaker 2: video with one of those, it was this tiny little
- 3:07Speaker 2: MPEG file that you'd end up getting. Even though you
- 3:11Speaker 2: had this powerful camera, the digital camera you could use,
- 3:14Speaker 2: your video was so terrible.
- 3:15Speaker 1: Yeah, yeah, and you just had a nice curated set
- 3:20Speaker 1: of pixels that would move and.
- 3:21Speaker 3: What appeared to look like a face exactly.
- 3:25Speaker 1: But but now we have you know, now it's it's
- 3:28Speaker 1: so affordable to buy these things, and and on balance
- 3:32Speaker 1: that's a good thing. But combine that with the massive
- 3:35Speaker 1: growth of social media and the massive growth there's some
- 3:38Speaker 1: of you people don't talk about as often, the massive
- 3:40Speaker 1: growth of archival capacity.
- 3:43Speaker 3: This means, if you.
- 3:44Speaker 1: Want to be sweet nostalgic about it, that are days
- 3:47Speaker 1: of cherishing a few like glamour photographs from your from
- 3:51Speaker 1: your local photography studio, or a few expensive polaroids. Those
- 3:56Speaker 1: days are all but gone. Now you can post photos
- 3:59Speaker 1: of yourself, your friends, or strangers, and you can post
- 4:02Speaker 1: them anywhere you please. There are some laws about it,
- 4:07Speaker 1: but they're they're not really good because legislation has always
- 4:10Speaker 1: been technology's slower sibling, and so those those laws often
- 4:15Speaker 1: don't have a lot of teeth to them. And with
- 4:19Speaker 1: the rise of all this technology and the rise of
- 4:21Speaker 1: all these capabilities, we also see the rise of technology
- 4:24Speaker 1: that is exclusively meant to exploit the vast amount of
- 4:28Speaker 1: stuff that we've been making just from like live journal, Tumblr, MySpace, friends,
- 4:33Speaker 1: their Facebook, all, you know, all the hits, all that jazz.
- 4:37Speaker 1: So what do you guys think, I mean, optimists and
- 4:40Speaker 1: pessimists have a different idea of what this means for
- 4:42Speaker 1: the world.
- 4:43Speaker 2: Well, yes, on the optimist side, I would just say
- 4:47Speaker 2: the number of photographs and captured memories I have of
- 4:51Speaker 2: my son is probably more, you know, it's certainly thousands
- 4:58Speaker 2: of times greater than that which exist for me as
- 5:01Speaker 2: a child in my growth, my you know, my parents,
- 5:04Speaker 2: and then going back and back and back and back,
- 5:06Speaker 2: it's just exponentially a larger amount of memories that are
- 5:10Speaker 2: captured throughout his life, my son. So that's great, that's wonderful.
- 5:15Speaker 2: We can look back on those, you know, until the
- 5:19Speaker 2: machines all turn off or we lose electricity, those those
- 5:23Speaker 2: images will exist for my family and people that care
- 5:25Speaker 2: about me and my son.
- 5:28Speaker 3: But on the flip side of that, Matt, you know,
- 5:30Speaker 3: like folks like my mom, for example, she doesn't consider
- 5:33Speaker 3: those real Like she's like, if it's not printed out
- 5:37Speaker 3: and it's not in a frame on my mantle, then
- 5:40Speaker 3: that is not a real memory. Which is interesting because
- 5:44Speaker 3: to your point about like the machines turning off, there's
- 5:47Speaker 3: all these companies now that will take your Instagram photos
- 5:49Speaker 3: and print them out and send you framed copies of them,
- 5:52Speaker 3: and the proliferation of like instacs, cameras and like more
- 5:56Speaker 3: analog technology is sort of a response, almost a backlas
- 6:00Speaker 3: to this over digitization of like our memories and our
- 6:04Speaker 3: collective kind of you know, unconscious really in a way
- 6:08Speaker 3: with all the stuff that's out there, it's like there's
- 6:10Speaker 3: this sense of is it really ours once it's out
- 6:13Speaker 3: there in the ether. And I think that's a big
- 6:15Speaker 3: part of today's question.
- 6:16Speaker 1: Yeah, who owns the art? The artist or the audience.
- 6:20Speaker 1: It's a very old question, but it's one that has
- 6:23Speaker 1: continually been relevant.
- 6:25Speaker 2: Right And yeah, but but the other question, Ben, is
- 6:28Speaker 2: is it art or is it just data?
- 6:32Speaker 3: Right?
- 6:32Speaker 1: Right there we go. That's a good distinction. It's a
- 6:36Speaker 1: tough distinction to make. Optimists are saying this is great,
- 6:40Speaker 1: you guys, the world's going to be a safer place,
- 6:43Speaker 1: and think how convenient everything's gonna be. You'll never lose
- 6:46Speaker 1: your favorite photo again, you know, and criminals can no
- 6:50Speaker 1: longer disappear into a crowd. And then pessimists say, you know,
- 6:54Speaker 1: in a world beyond privacy, every person is monitored continually
- 6:59Speaker 1: throughout their lives. Your image is logged countless times, and
- 7:02Speaker 1: all the data from that image and all the other
- 7:05Speaker 1: images of you, it's analyzed. It's bought and sold without
- 7:09Speaker 1: your knowledge or your permission. A world careening towards sci
- 7:12Speaker 1: fi dystopia, pre crime, pre existing medical conditions. Advertising so
- 7:18Speaker 1: invasive and so bespoke that it might as well be
- 7:21Speaker 1: reading your mind, because it's doing something very close to that.
- 7:25Speaker 1: But before you get all blade round round minority report,
- 7:29Speaker 1: let's start closer to the beginning. There is a revolution
- 7:33Speaker 1: happening in the world of facial recognition right now, and
- 7:36Speaker 1: its name is Clearview AI.
- 7:39Speaker 3: Here are the facts. Yeah, I mean, the first time
- 7:42Speaker 3: I heard of Clearview AI was on a recent in
- 7:45Speaker 3: PR piece where the founder was interviewed. So this is
- 7:48Speaker 3: a very relatively new company, at least to me. But
- 7:51Speaker 3: as far as the background, it is a research tool
- 7:56Speaker 3: that is being used by law enforcement, specifically in the company.
- 8:00Speaker 3: It makes it very clear that this is something that's
- 8:02Speaker 3: designed exclusively for use by law enforcement in order to
- 8:07Speaker 3: identify perpetrators and victims of crimes. It's a little bit
- 8:13Speaker 3: pretty very vague side I would say, yeah.
- 8:16Speaker 1: Yeah, I mean that's you know, we have to be honest.
- 8:19Speaker 1: That's that's the high level like logline or pitch from
- 8:25Speaker 1: the company on its own website. So we can't blame
- 8:28Speaker 1: them for not wanting to get too into the weeds
- 8:30Speaker 1: about algorithms. But if we look at how it actually works,
- 8:33Speaker 1: we can we can get a high level understanding.
- 8:36Speaker 3: It's a startup.
- 8:37Speaker 1: It has a massive database of faces, just specifically faces,
- 8:43Speaker 1: somewhere north of three billion images and as we're as
- 8:48Speaker 1: we're recording this, that number is likely still growing. So
- 8:52Speaker 1: that number is going to be a little bit higher
- 8:54Speaker 1: by the time you get to the end of this episode.
- 8:57Speaker 3: Yeah, and there was the gentleman who found the startup. Hann
- 9:01Speaker 3: thought was who I heard interviewed on I believe it
- 9:04Speaker 3: was The Daily. Actually, so if you want to hear
- 9:07Speaker 3: directly from this gentleman's mouth what he believes the mission
- 9:11Speaker 3: of this technology in this company is it a very
- 9:14Speaker 3: fascinating interview that kind of delves into some of the
- 9:17Speaker 3: slippery slope sides of this technology that we're going to
- 9:20Speaker 3: get into. But as far as the end users concerned,
- 9:23Speaker 3: which in this case would be law enforcement, because currently
- 9:26Speaker 3: this is not available to civilians. You take a picture
- 9:31Speaker 3: and it identifies who that person is, simple as that,
- 9:35Speaker 3: by matching it with all of this information that they've collected,
- 9:39Speaker 3: and they've collected it, you know, not through any really
- 9:42Speaker 3: nefarious means. It's all that information that we talked about
- 9:45Speaker 3: that's already out there. All they've got to do is
- 9:47Speaker 3: reach out and grab.
- 9:48Speaker 1: It's a it's a search engine for faces, and it
- 9:51Speaker 1: sounds innocuous at first. I also, if it was The Daily,
- 9:56Speaker 1: I think I listened to that interview, but I also
- 10:00Speaker 1: listen to an excellent interview you sent Matt an extended
- 10:02Speaker 1: CNN business conversation. So if you want to see them,
- 10:07Speaker 1: like if you want to see the founder using the
- 10:10Speaker 1: app a couple of different times, I think he does
- 10:12Speaker 1: it on himself, he does it on the interviewer, and
- 10:16Speaker 1: then he does it on one of their producers, and
- 10:19Speaker 1: it seems to work. It's not a sensationalized thing either,
- 10:24Speaker 1: at least the way they're presenting it.
- 10:26Speaker 2: Yeah, but we're gonna get it. We're gonna maybe talk
- 10:29Speaker 2: about that a little more. Just if you do want
- 10:32Speaker 2: to look at that video. It's an interview with Donnie
- 10:36Speaker 2: O'Sullivan with CNN Business and it is where we're going
- 10:39Speaker 2: to reference it several times in this episode.
- 10:41Speaker 3: Yeah.
- 10:42Speaker 1: So, like you said, Noel Clearview has taken this from
- 10:46Speaker 1: publicly available sources, your Facebook's, your YouTube's, your venmos, which yes,
- 10:52Speaker 1: are publicly available, and the Federals and local law enforcement
- 11:00Speaker 1: admit that they only have a limited knowledge of how
- 11:03Speaker 1: the app actually works mechanically, the nuts and bolts of it,
- 11:07Speaker 1: but they do confirm they've already used it to help
- 11:09Speaker 1: solve shoplifting cases, identity theft cases, credit card fraud, child abuse,
- 11:16Speaker 1: even and even several homicides. This technology goes, by the way,
- 11:21Speaker 1: far beyond anything constructed by Silicon Valley giants or anything
- 11:26Speaker 1: officially created or used by the US government.
- 11:30Speaker 2: Yeah, in that interview that we just mentioned. Juan mentions
- 11:35Speaker 2: that they had this image or this video file of
- 11:41Speaker 2: a child abuse incident and the person of interest walks
- 11:48Speaker 2: past in the background at one point, and there were
- 11:51Speaker 2: only two usable frames of this person's face. And again,
- 11:55Speaker 2: like imagine in the background of a video, they were
- 11:59Speaker 2: able to take still from that video of this person's
- 12:03Speaker 2: face and identify him through other social media websites where
- 12:07Speaker 2: they've got all this data from and make an arrest
- 12:12Speaker 2: on this person like that. That is it's mind blowing
- 12:17Speaker 2: that you could do that, and thank goodness that they
- 12:20Speaker 2: were able to do that. But it's the implication of
- 12:23Speaker 2: being able to do that in that one instance and
- 12:26Speaker 2: then applying it across everything is what makes it feel
- 12:29Speaker 2: a little scary because.
- 12:31Speaker 3: For the critics, for the critics, well.
- 12:34Speaker 2: Yeah, for the critics, well for the people who are
- 12:37Speaker 2: looking at it big picture. So if you're you're thinking
- 12:40Speaker 2: about it just in a law enforcement application, then it's amazing, right,
- 12:46Speaker 2: and it's the best thing ever. But if you apply
- 12:49Speaker 2: it to everything, to everyone at all times outside of
- 12:53Speaker 2: law enforcement, then well.
- 12:55Speaker 3: And just strictly on like a technological level, it's fascinating
- 12:59Speaker 3: and clearly a huge leap forward in this kind of
- 13:04Speaker 3: technology because I mean we've seen these videos they're talking about,
- 13:07Speaker 3: these grainy surveillance videos or ATM camera videos. I mean
- 13:11Speaker 3: it's hard to as a human being, even if you
- 13:13Speaker 3: knew that person, it'd be difficult for you to identify them.
- 13:17Speaker 3: So this using this algorithm and you know, analyzing and
- 13:21Speaker 3: comparing it to all these different subjects I imagine using
- 13:23Speaker 3: points of articulation like on the faces and the structures
- 13:26Speaker 3: or whatever, it's able to come back with positive matches.
- 13:30Speaker 3: I think that's really fascinating and the really, you know,
- 13:33Speaker 3: pretty innovative use of technology. For sure, as a nerd
- 13:37Speaker 3: speaking you know, exclusively on that level. But you're right, Matt,
- 13:40Speaker 3: when you start to apply on a larger scale, it
- 13:45Speaker 3: gets really nineteen eighty four ish and kind of scary.
- 13:49Speaker 3: I appreciate that point, Noel.
- 13:50Speaker 1: It's something that's going to come into play later and
- 13:53Speaker 1: it may surprise some of our fellow listeners to learn
- 13:58Speaker 1: a little bit more about this technology and some of
- 14:01Speaker 1: the other topics that you raise there. I want to
- 14:04Speaker 1: say this, their pitch is very law enforcement based on
- 14:09Speaker 1: the website. They're quick to point out, they're quick to
- 14:13Speaker 1: claim I should say that they've helped law enforcement track
- 14:16Speaker 1: down hundreds of at large criminals. They mention the real
- 14:21Speaker 1: the things that make people very emotional, like child abusers, terrorists,
- 14:26Speaker 1: sex traffickers, but they also say this technology is used
- 14:30Speaker 1: to help exonerate the innocent and identify victims. And it
- 14:35Speaker 1: seems that they're they might be embellishing a little bit
- 14:39Speaker 1: in terms of the degree of help they're providing, but
- 14:42Speaker 1: they are providing help. The authorities have used this, it's
- 14:46Speaker 1: happening now. They've done so without much public scrutiny, or
- 14:50Speaker 1: they were for most of twenty nineteen doing so without
- 14:53Speaker 1: much public scrutiny. In early twenty twenty, Clearview said more
- 14:58Speaker 1: than six hundred different law enforcem agencies started using our
- 15:01Speaker 1: app just in the last year, but they refuse to
- 15:05Speaker 1: provide more specifics because they're very protective of their client base.
- 15:10Speaker 1: So the badger is out of the bag. Metaphorically, this
- 15:15Speaker 1: is not a what if scenario.
- 15:17Speaker 3: This exists.
- 15:19Speaker 1: This is in use right now, and that means it
- 15:22Speaker 1: can therefore be used on you and every single person
- 15:25Speaker 1: you know.
- 15:26Speaker 3: Is that a bad thing?
- 15:28Speaker 2: Ah?
- 15:29Speaker 1: Here's the interesting part. This technology is maybe innovative in application,
- 15:35Speaker 1: but very much not innovative in theoretical terms. It has
- 15:39Speaker 1: been possible for a while. People in the tech world
- 15:44Speaker 1: knew about this possibility for a long long time, and
- 15:47Speaker 1: some tech companies this is so weird. Tech companies barely
- 15:50Speaker 1: ever do this. Some tech companies like Giants said, you know,
- 15:55Speaker 1: we can build something like this. They realized it years ago,
- 15:58Speaker 1: and they actually decide to stop researching it. They decided
- 16:02Speaker 1: not to deploy this because of the massive potential privacy concerns.
- 16:07Speaker 1: Even Google wouldn't do it.
- 16:10Speaker 3: It's like that thing in The Batman, which which one
- 16:14Speaker 3: is it. It's the one with Keith Keith Fledger, where
- 16:15Speaker 3: there's that technology that turns every cell phone in the
- 16:18Speaker 3: world into a listening device, and you know he has
- 16:21Speaker 3: to destroy it after using it once because it's just
- 16:24Speaker 3: such problematic tech that, you know, Morgan Freeman's character basically
- 16:29Speaker 3: builds a safeguard device that causes it to you destroy itself,
- 16:32Speaker 3: because it's that's the thing, Ben about the badger being
- 16:35Speaker 3: out of the bag. Once it's out, you can say
- 16:38Speaker 3: all day long, this is just for law enforcement. But
- 16:40Speaker 3: like you said, Ben, the capability has been there, so
- 16:43Speaker 3: who's to stop somebody from doing a kind of copycat
- 16:47Speaker 3: version of this once it's out there and available to
- 16:49Speaker 3: check out, even if it's only quote unquote just for
- 16:52Speaker 3: law enforcement, you know. I mean, so were certain kinds
- 16:55Speaker 3: of radios and tasers and things like that, you know,
- 16:58Speaker 3: and people in the public of those now too, don't they.
- 17:01Speaker 3: It's a very good point.
- 17:02Speaker 1: Yeah, I mean back in twenty eleven, think about how
- 17:05Speaker 1: long ago.
- 17:06Speaker 3: That was.
- 17:06Speaker 1: Almost ten years ago, the chairman of Google said, facial
- 17:11Speaker 1: recognition technology, this kind of stuff that Clearview is doing,
- 17:14Speaker 1: is the one technology that Google is holding back on
- 17:18Speaker 1: because we feel like it could be used in a
- 17:20Speaker 1: very bad way. And now this is a fascinating point here.
- 17:24Speaker 1: Some cities have even preemptively banned police from using any
- 17:29Speaker 1: kind of facial recognition technology. And one of the most
- 17:33Speaker 1: notable examples of a city that's banned this San Francisco,
- 17:37Speaker 1: at the very heart of technological innovation here in the US,
- 17:41Speaker 1: so they definitely knew what was coming. Their lawmakers, their
- 17:45Speaker 1: constituents gave it a big not nope. And here's the
- 17:50Speaker 1: other thing, Clearview AI. If we think about the future
- 17:52Speaker 1: of it, the code behind it is fascinating.
- 17:56Speaker 2: If anyone out there has been watching the latest season
- 18:00Speaker 2: of Westworld. You may be familiar with the glasses that
- 18:05Speaker 2: several characters wear, and just think about that. If you
- 18:09Speaker 2: are familiar with that, think about that as we describe
- 18:13Speaker 2: what this thing can do.
- 18:15Speaker 1: Yeah, think about it carefully, because you see Clearview AI.
- 18:21Speaker 1: The code behind it isn't just facial recognition. It also
- 18:27Speaker 1: includes programming language that would allow this tool to pair
- 18:33Speaker 1: with AR or augmented reality glasses.
- 18:35Speaker 3: What does this mean.
- 18:37Speaker 1: This means that if the sunglasses I'm wearing now, we're
- 18:42Speaker 1: functioning with Clearview AI, I could walk down the street
- 18:45Speaker 1: and I could identify almost every person I saw, and
- 18:49Speaker 1: not just other pictures of them, but breadcrumbs that would
- 18:54Speaker 1: tell me where they lived, you know, what their children's
- 18:57Speaker 1: identities were, what their name was, they did for a living,
- 19:01Speaker 1: and so it would automatically play the six Degrees of
- 19:04Speaker 1: Kevin Bacon. Actually, that's a really great idea for an
- 19:08Speaker 1: add on game for these classes. You can see how
- 19:11Speaker 1: close people are to Kevin Bacon.
- 19:13Speaker 3: You heard it here.
- 19:14Speaker 1: First, But but it's scary because then you would like
- 19:16Speaker 1: you would see you would see Matt Frederick, and then
- 19:19Speaker 1: you would you would know his friend's Paul. Mission Control Deckett,
- 19:24Speaker 1: Noel Brown, Ben Bollen, and you would know everyone he
- 19:26Speaker 1: worked with. It just gets real sticky really quickly. And
- 19:32Speaker 1: can you hear the FEDS drooling in the background? Is
- 19:36Speaker 1: that just my cat? I don't know at this point,
- 19:39Speaker 1: but that's the high level look.
- 19:41Speaker 2: Or spies, I mean, just any spy. Imagine that, well,
- 19:45Speaker 2: you guys, I mean.
- 19:46Speaker 3: I think we're forgetting. An even more classic sci fi
- 19:49Speaker 3: example of this, that the kind of heads up display
- 19:52Speaker 3: you'd see in a movie like The Terminator. You know,
- 19:55Speaker 3: where you're seeing from the perspective of this, you know,
- 19:58Speaker 3: assassin robot who's looking for a target and being able
- 20:02Speaker 3: to see who everybody is and get all these stats
- 20:05Speaker 3: and metrics. You could apply this to any number of
- 20:09Speaker 3: you know, let's call them parameters surrounding an individual. You
- 20:13Speaker 3: could then go deeper with this and start pulling stats
- 20:15Speaker 3: from their social media and be served up with information
- 20:18Speaker 3: about their height and or you know whatever, like any
- 20:21Speaker 3: like let's say they have pre existing condition or something
- 20:26Speaker 3: like that, or they have some kind of you know
- 20:29Speaker 3: what's the word I'm looking for, like compromand on them.
- 20:32Speaker 3: You could even go deeper and find pictures of them,
- 20:34Speaker 3: like maybe smoking when they've claimed that they don't smoke,
- 20:38Speaker 3: and they are, you know, committing insurance fraud. I don't know.
- 20:41Speaker 3: I'm just saying, there's all kinds of ways you could
- 20:44Speaker 3: deep dive scrape this data and apply these points to
- 20:47Speaker 3: an individual and then get that information served up to
- 20:50Speaker 3: you very quickly in this display.
- 20:53Speaker 1: And yeah, an optimists are so quick to point out
- 20:56Speaker 1: the potential benefits of clearview AI recognition in general. It's
- 21:01Speaker 1: true there are benefits. Pessimists, however, claim the disadvantages here
- 21:07Speaker 1: in this specific case, far outweigh any of those benefits,
- 21:11Speaker 1: outweigh them, in fact, by a vast and disturbing margin.
- 21:16Speaker 1: We'll pause for word from our sponsor and return to
- 21:20Speaker 1: dive down the rabbit hole.
- 21:27Speaker 3: Here's where it gets crazy.
- 21:30Speaker 1: There are a ton of problems we've you know, we've
- 21:35Speaker 1: done some foreshadowing on these, but let's talk a little
- 21:39Speaker 1: bit more about them first. You know, you can categorize
- 21:43Speaker 1: these right. So first, there are not many strong operational
- 21:48Speaker 1: restrictions on how this technology is actually used. That means
- 21:53Speaker 1: that it can not only be easily misused, but that
- 21:57Speaker 1: there will probably not be repercussions for people, at least
- 22:01Speaker 1: not not legal repercussions.
- 22:03Speaker 3: It's like a point you made earlier, been off off, Mike.
- 22:05Speaker 3: It's like when you used to go into those tobacco
- 22:08Speaker 3: shops and uh, you know, you are not allowed to
- 22:11Speaker 3: refer to certain smoking apparatuses as bongs. They are water
- 22:17Speaker 3: pipes exclusively because that absolves the distributor, the retailer of
- 22:23Speaker 3: any responsibility if you choose to use them in any
- 22:26Speaker 3: illegal manner. It's the same thing. It's like saying, hey,
- 22:31Speaker 3: whatever you do with my thing is up to you.
- 22:32Speaker 3: I know, it's designed exclusively for a legal purpose. It's
- 22:37Speaker 3: a real sticky situation, isn't it, Because I mean it's
- 22:40Speaker 3: it not only puts all of the burden on the user,
- 22:44Speaker 3: it absolves the creator of any kind of responsibility at all.
- 22:48Speaker 1: Or does he do this ta hey, hey buddy, that's
- 22:51Speaker 1: a tobacco water pipe, all right?
- 22:54Speaker 2: Okay, yeah, yeah, say, you know, saying something like the
- 23:00Speaker 2: way the founder did in that interview with CNN where
- 23:03Speaker 2: he was just saying, you know, the thing says, just
- 23:06Speaker 2: use it in this way. It's it's we've seen it
- 23:09Speaker 2: over and over and over again. There there are a
- 23:11Speaker 2: couple of good things that he does speak about in
- 23:14Speaker 2: that CNN Business Interview, and one of the major things,
- 23:19Speaker 2: you know, if you put it at the top, it's
- 23:21Speaker 2: that the software costs around fifty thousand dollars to have
- 23:25Speaker 2: a license for a you know, for a department or
- 23:28Speaker 2: something for a year, years for two years, and so
- 23:31Speaker 2: at least there's a buy in level there that just
- 23:35Speaker 2: your average person if they're getting if they're obtaining it legally,
- 23:38Speaker 2: wouldn't be able to use. And then there are you know,
- 23:41Speaker 2: ways to verify if you know an end user is
- 23:44Speaker 2: actually if that license is is actually that person they
- 23:49Speaker 2: should be using it? That's good. And the other thing
- 23:53Speaker 2: is if you're a manager within the software of a
- 23:58Speaker 2: you know, a licensee of this software, you get audits
- 24:01Speaker 2: of every time the software has been used, which is
- 24:04Speaker 2: really nice. So at least at least there could be
- 24:07Speaker 2: some oversight to it and they could check to see
- 24:10Speaker 2: if you are quote, using it in a way they're
- 24:12Speaker 2: not supposed to.
- 24:13Speaker 1: Yeah, yeah, there's any there's an example of that. I
- 24:16Speaker 1: think they'll come into play here later, which is kind
- 24:19Speaker 1: of it's pretty weird.
- 24:21Speaker 3: It's it's a little it's a.
- 24:22Speaker 1: Little spooky, but I want to go back to that warning.
- 24:26Speaker 1: It's it's literally something. It's it's like that old FBI
- 24:29Speaker 1: VHS tape warning that we mentioned in earlier episode. You
- 24:33Speaker 1: go on to the homepage essentially of the app, which
- 24:36Speaker 1: you can use on desktop or you can use on
- 24:38Speaker 1: your phone, and basically all it says is like, don't
- 24:42Speaker 1: use this in a bad way.
- 24:45Speaker 3: And that's it's.
- 24:46Speaker 1: It's like as effective as those tags on mattresses that
- 24:51Speaker 1: say don't remove this tag, or or the the little
- 24:55Speaker 1: note on your boxy Q tips that's like, hey, don't
- 24:58Speaker 1: put these in your ear. You know it feels awesome. Uh,
- 25:02Speaker 1: and I know that's what everybody does with it, but
- 25:04Speaker 1: don't don't put them in your ear.
- 25:06Speaker 3: Uh. And then like you said, no that absolves them.
- 25:09Speaker 3: Or the warning about seizures at the beginning of when
- 25:11Speaker 3: you start up your PlayStation, you know what I mean,
- 25:13Speaker 3: it's like, yeah, okay, maybe I get seizures, but I'm
- 25:16Speaker 3: gonna roll the dice because I sure want to play
- 25:18Speaker 3: me some Borderlands three, you know.
- 25:20Speaker 1: Yeah, because you know, now that I think about it,
- 25:23Speaker 1: we might be at a bit of a glasshouse situation here, uh,
- 25:27Speaker 1: my friends, because our show literally starts with the warning
- 25:30Speaker 1: that says you can turn back now we're not We're
- 25:33Speaker 1: not as bad as Q tips.
- 25:35Speaker 3: You can you can turn back. Now that's definitely as
- 25:38Speaker 3: as is our prerogative to follow said warnings and not
- 25:41Speaker 3: stick Q tips in our ears. But who wants to
- 25:43Speaker 3: do that? What are you gonna use them for? Like
- 25:45Speaker 3: putting on eyeshadowers? I mean, I don't know the back
- 25:49Speaker 3: of your ear. Really do you need it to do?
- 25:51Speaker 1: You a C tip for that?
- 25:52Speaker 3: Can't you just use like a like a like a tissue.
- 25:55Speaker 3: Some people are living wild man. I don't know what
- 25:57Speaker 3: to tell you. You tips are literally probes. I don't
- 26:00Speaker 3: know anyway. I just feel like they're very clearly designed
- 26:03Speaker 3: to shove in your ear and it does feel great,
- 26:05Speaker 3: but yeah, it can. It's pretty scary when you pull
- 26:08Speaker 3: out some blood, you know, just saying.
- 26:10Speaker 2: So, yes, I've never done that. Actually I'm gonna I'm
- 26:14Speaker 2: gonna start trying. Uh no, we get gotten blood like
- 26:18Speaker 2: gone in for wax out with blood. But you know,
- 26:21Speaker 2: maybe I'm just fortunate. I want to bring up something
- 26:24Speaker 2: that is possible with this software and just how it
- 26:29Speaker 2: uses it before we jump into the next thing we're
- 26:31Speaker 2: going to talk about. And that's just that when your
- 26:35Speaker 2: face ends up on a social media website somewhere, if
- 26:39Speaker 2: you even if you did not take that picture, even
- 26:42Speaker 2: if you were in the background of a selfie that
- 26:44Speaker 2: someone took at Bonaru, let's say, or something like that,
- 26:48Speaker 2: your face is still searchable with facial recognition software if
- 26:54Speaker 2: it is visible in the background of an image, if
- 26:56Speaker 2: you're in a crowd, and if anywhere you are anywhere
- 26:59Speaker 2: you've been in where a photograph has been taken where
- 27:02Speaker 2: you can are visible somewhere in the background. This thing
- 27:07Speaker 2: clearview AI can may very much likely has that image
- 27:12Speaker 2: within its database and your face will be recognizable to it.
- 27:16Speaker 2: Just let's put that.
- 27:17Speaker 1: There, so it doesn't matter if you're on social media exactly.
- 27:21Speaker 3: Guys Devil's advocate here. Isn't that one of the situations
- 27:24Speaker 3: where if you're in public you're sort of giving up
- 27:28Speaker 3: that right to privacy, Or if you go to a
- 27:31Speaker 3: bon Aroo, there's usually signs that said when you pass
- 27:34Speaker 3: these gates, you are subject to being filmed, etc. And
- 27:37Speaker 3: that's usually for more like documentary purposes of the festival,
- 27:41Speaker 3: but I would think it would also apply if you're
- 27:42Speaker 3: in public and you're caught on you know, one of
- 27:46Speaker 3: these millions of little surveillance devices that people carry around
- 27:49Speaker 3: with them. Is that fair game? Can you sue over
- 27:54Speaker 3: use of your of misuse of your image? I'm gonna
- 27:57Speaker 3: I wouldn't have think. I wouldn't have thought.
- 27:58Speaker 2: So I'm certainly not arguing about, you know, the legality
- 28:01Speaker 2: of it. I'm just saying the reality of it is
- 28:04Speaker 2: that wherever you go, if there is a camera taking
- 28:08Speaker 2: a picture, whether it's cct well, CCTV generally wouldn't show
- 28:11Speaker 2: up in one of these things unless it was placed
- 28:13Speaker 2: into the system.
- 28:14Speaker 1: At the At the risk of sounding to I don't know,
- 28:19Speaker 1: to extreme about this, maybe too prescient, I think the
- 28:23Speaker 1: best for people concerned about this, the best way to
- 28:27Speaker 1: handle it, which is maddening, is to is to apply
- 28:31Speaker 1: some of the rules that some of the rules that
- 28:33Speaker 1: we apply to guns to cameras. So you always assume
- 28:38Speaker 1: a gun is loaded, right, you never want to point
- 28:41Speaker 1: it at you, So always assume a camera is recording.
- 28:44Speaker 1: And if you don't want to be with that assumption,
- 28:47Speaker 1: if you don't want to be in the footage or
- 28:51Speaker 1: in the photo, then don't have it pointed at you.
- 28:54Speaker 1: You guys, remember, I was very uncomfortable with Pokemon Go
- 28:59Speaker 1: when it came out, it was like, no, they're putting poke,
- 29:02Speaker 1: They're putting these ar little cute things to collect because
- 29:06Speaker 1: you're filming.
- 29:07Speaker 3: You're filming for them.
- 29:08Speaker 1: You guys are like Ben big Brother does not give
- 29:12Speaker 1: a like the faintest hint of a about Charis Arts
- 29:18Speaker 1: or whatever. Right, But still that stuff freaks me out yet,
- 29:22Speaker 1: you know, But we did.
- 29:23Speaker 3: A whole story about that company, Nantek, and like how
- 29:26Speaker 3: it had roots and intelligence gathering. I mean, you're not wrong, Ben,
- 29:31Speaker 3: And I remember that we'd be out hanging out and
- 29:34Speaker 3: people would be doing their little capturing their pokemons and
- 29:37Speaker 3: they'd be on your cell phone or something, and you
- 29:38Speaker 3: would like cover your face, and that is your prerogative
- 29:41Speaker 3: to do. And I completely understand where you're coming from,
- 29:43Speaker 3: because it's like with the whole zoom phenomenon that we're
- 29:46Speaker 3: obviously we're using it right now. You can turn off
- 29:49Speaker 3: that camera as far as you're concerned where it doesn't
- 29:52Speaker 3: show it to the group, but I guarandamn to you,
- 29:56Speaker 3: it's still capturing what's going on and storing it somewhere.
- 30:00Speaker 3: So you know, it's a great technology. It works real well,
- 30:04Speaker 3: which is why it's having this moment right now. But
- 30:07Speaker 3: I just would wouldn't trust it too implicitly.
- 30:10Speaker 2: You know, you know, the best technology it's what ben
- 30:13Speaker 2: ben and it's a sticker or some tape just go
- 30:19Speaker 2: at all times.
- 30:21Speaker 1: Yeah, but that's that's that's still that's that's a low
- 30:25Speaker 1: tech hack. But you have to do it because this
- 30:27Speaker 1: stuff can be weaponized. And uh, there's a guy named
- 30:31Speaker 1: Eric Goldman at the High Tech Law Institute at Santa
- 30:34Speaker 1: Clara University.
- 30:37Speaker 3: I like the way he put it.
- 30:38Speaker 1: He said, the weaponization possibilities of this are endless. Imagine
- 30:43Speaker 1: a rogue law enforcement officer who wants to stalk someone
- 30:47Speaker 1: that they want to hit on later, or maybe a
- 30:50Speaker 1: foreign government is doing similar to what you described. No,
- 30:53Speaker 1: they're digging up secrets about people to blackmail them or
- 30:56Speaker 1: to throw members of an opposition party in jail.
- 30:59Speaker 3: This stuff can happen.
- 31:01Speaker 1: And you know, Clearview is also notoriously secretive. Even now
- 31:06Speaker 1: that's the second big issue. It leaves a lot of
- 31:09Speaker 1: critics ill at ease. We know some stuff because of
- 31:14Speaker 1: some excellent research from earlier reporters. Let me tell you, guys,
- 31:19Speaker 1: this weird story. So there's a there was a New
- 31:22Speaker 1: York Times journalist who was working on a story about
- 31:25Speaker 1: Clearview that really brought it to international attention. And while
- 31:30Speaker 1: this journalist is working with clear is trying to do
- 31:34Speaker 1: this story, Clearview isn't really answering him. They're like ducking
- 31:38Speaker 1: attempts to contact via text, email, phone call.
- 31:41Speaker 3: And so this reporter.
- 31:44Speaker 1: Knows that a lot of law enforcement agencies have Clearview access,
- 31:48Speaker 1: and so this reporter asks some contacts in the police
- 31:52Speaker 1: department to run his photo through the app, and they
- 31:57Speaker 1: do a couple times. So the first approach, the first
- 32:02Speaker 1: thing he learns about Clearview approaching outside forces is not
- 32:07Speaker 1: when they come to him. It's when they go to
- 32:09Speaker 1: the police that he worked with and they ask the police, Hey,
- 32:14Speaker 1: are you talking to the media.
- 32:16Speaker 3: How shady is that?
- 32:17Speaker 2: Wow? Yes, that is whoa.
- 32:21Speaker 3: They have an explanation right, oh absolutely, And just really
- 32:24Speaker 3: quickly that's the piece that was on the Daily because
- 32:26Speaker 3: the Daily is affiliate with the New York Times, So
- 32:28Speaker 3: this was after the fact of this reporter whose name
- 32:31Speaker 3: I'm trying to find, had done all this research and
- 32:33Speaker 3: gone through this whole run around to even get a
- 32:37Speaker 3: sit down. The reporter eventually found the office and it
- 32:40Speaker 3: was like very shady like there weren't with that many
- 32:42Speaker 3: people there. It was a skeleton crew and finally got
- 32:45Speaker 3: access to the founder. But that is all, you know,
- 32:49Speaker 3: described in great detail on that piece by The Daily.
- 32:51Speaker 3: I highly recommend checking it out.
- 32:53Speaker 1: Yeah, Kashmir Hill, the secretive company that might end privacy
- 32:57Speaker 1: as we know it is the is the Times article. Yeah,
- 33:00Speaker 1: I like what you're saying about. When when when Hill
- 33:03Speaker 1: finally figured out where the company was originally on LinkedIn,
- 33:08Speaker 1: they just had like one employee listed and it turned
- 33:11Speaker 1: out to be a pseudonym or a fake name that
- 33:14Speaker 1: the founder was using. So they're they're aware of the
- 33:18Speaker 1: dangers here. But but when they did eventually speak to
- 33:22Speaker 1: the reporter, they had and they had reasonable explanations for
- 33:27Speaker 1: both of these things. First, the founder said, look, we
- 33:31Speaker 1: were avoiding speaking to the media, not because we're like
- 33:35Speaker 1: super villains or something. It's because we are a tiny
- 33:38Speaker 1: startup and a lot of times when you're a tiny startup,
- 33:41Speaker 1: you're in you're in stealth mode, right, And that makes
- 33:43Speaker 1: sense because the NDAs on those things are pretty ironclad.
- 33:47Speaker 1: You don't want anyone to scoop you. And then he
- 33:49Speaker 1: additionally said, here's why we talked to the police about
- 33:52Speaker 1: your search terms. We actually monitor all the searches that
- 33:57Speaker 1: are done because we want to make sure that they're
- 34:00Speaker 1: being used correctly. In other words, you know, we want
- 34:03Speaker 1: to make sure nothing like like there's never a situation
- 34:07Speaker 1: where a police officer is jilted, is jilted by a
- 34:11Speaker 1: romantic partner and then they find out their X has
- 34:14Speaker 1: a new lover and they're like.
- 34:15Speaker 3: Well, I'm gonna find out about this.
- 34:19Speaker 1: God, I'm even afraid to make up a name because
- 34:21Speaker 1: I don't want to put it out there. Uh, Marcus
- 34:26Speaker 1: mcgilly cutty. Oh that's swinging a miss.
- 34:28Speaker 3: Funny thing about that, though, Ben is like, you don't
- 34:31Speaker 3: even need technology this robust to dig into somebody that
- 34:35Speaker 3: closely removed from you know, a person you really know
- 34:38Speaker 3: they're probably already like posting tagging each other on Instagram
- 34:42Speaker 3: posts and you just you know, go and just do
- 34:43Speaker 3: a couple of clicks and you're there. But no, it's
- 34:47Speaker 3: just a really good point and that's what what's set
- 34:48Speaker 3: off those alarm bells. But yeah, the way it's described
- 34:51Speaker 3: in the in the in the Daily piece, it was
- 34:54Speaker 3: very it felt very like okay, a they know they're
- 34:57Speaker 3: onto something and be they don't want to be found.
- 35:04Speaker 3: And I can see to your point been the whole
- 35:05Speaker 3: startup mode. They don't want to get scooped. They don't
- 35:07Speaker 3: want anybody stealing their proprietary in their you know, technology
- 35:11Speaker 3: and trying to misuse it or you know, create a
- 35:15Speaker 3: copycat kind of version of it. So that does make sense,
- 35:17Speaker 3: but just it's the whole thing is a little strange.
- 35:19Speaker 3: But this app clearview claims you know, is is up
- 35:24Speaker 3: to security industry standards in terms of I guess in
- 35:28Speaker 3: terms of you know, hackability or like firewalls security or
- 35:33Speaker 3: you know, being not being vulnerable to being infiltrated, but
- 35:37Speaker 3: who is actually watching them? Who is actually monitoring them
- 35:42Speaker 3: because it's it's essentially untested and already being put out
- 35:46Speaker 3: into the wild in cases that can have huge effects
- 35:50Speaker 3: on people's lives. We'll talk more about that after a
- 35:53Speaker 3: quick word from our sponsor. Yes, you know that was
- 36:03Speaker 3: a perfect ad break.
- 36:04Speaker 1: Bill also came be time to get this cat out
- 36:08Speaker 1: of out of my recording studio.
- 36:10Speaker 3: I heard you had a little me hour.
- 36:13Speaker 1: Yeah uh yeah. Well we're all dealing with new coworkers, right.
- 36:19Speaker 3: Can I just do a quick aside, a quick quarantine
- 36:21Speaker 3: aside had a crisis yesterday. We I don't know if
- 36:24Speaker 3: I've mentioned the show, that my daughter got a hamster
- 36:27Speaker 3: right before quarantine kicked in, and I had to inherit
- 36:30Speaker 3: it because her mom was allergic. She found out hamster's
- 36:33Speaker 3: name is Honakah, which is like an anime character. But anyway,
- 36:36Speaker 3: we realized yesterday afternoon that Honakoh was not in her cage.
- 36:41Speaker 3: So we had a hamster search party all up in
- 36:45Speaker 3: this house, freaking out, and I had just these horrible
- 36:48Speaker 3: visions of like finding a dead, decaying hamster, you know,
- 36:51Speaker 3: two weeks later in my sock drawer or something. So luckily,
- 36:55Speaker 3: the little the little critter found herself back to where
- 36:58Speaker 3: she belonged on her own, which was great. What great? Yeah,
- 37:02Speaker 3: she just popped out and was right there, even thought
- 37:04Speaker 3: it was a sock and it was turns out it
- 37:05Speaker 3: was a hamster and it wasn't long. I read all
- 37:08Speaker 3: this horror stories about how you can only search late
- 37:10Speaker 3: at night and you gotta bait it with peanut butter
- 37:13Speaker 3: and all this stuff, and it she just came came
- 37:15Speaker 3: right back. It was great.
- 37:16Speaker 1: That's fantastic. Yeah, I had a I had a gerbil
- 37:21Speaker 1: and in my younger days that escaped and lived and
- 37:25Speaker 1: died in the walls of a house in my tried
- 37:28Speaker 1: to tried to move the oven and appliances to get
- 37:31Speaker 1: to it, but then we couldn't, and my father thought
- 37:36Speaker 1: it would be a good lesson in mortality.
- 37:39Speaker 3: Oh boy, I was hoping to avoid that particular lesson
- 37:42Speaker 3: for my kid, especially during these trying times when things
- 37:46Speaker 3: have been mainly pretty positive for us here at the house.
- 37:49Speaker 3: I'm so glad. Yeah, thanks guys, but no, you're right, Ben,
- 37:53Speaker 3: tell us a little bit about the security of this
- 37:56Speaker 3: app and the idea of it being tested for.
- 38:02Speaker 1: Vulnerabilities, right, Yeah, this is a very interesting point. So
- 38:06Speaker 1: a lot of the reporting that came up about Clearview
- 38:10Speaker 1: Ai came before late February of this year. In late February,
- 38:18Speaker 1: clearview app security was actually tested in the form of
- 38:22Speaker 1: a hack. On February twenty six, the public learned that
- 38:26Speaker 1: the company had been breached, their security been compromised. Hackers
- 38:30Speaker 1: had stolen Clearview AI's entire customer list, which was coveted
- 38:35Speaker 1: and very much secret right the customer list, adding to
- 38:40Speaker 1: the troubling facts about this company, The customer list did
- 38:43Speaker 1: not jibe with what Clearview had said earlier. It spans
- 38:47Speaker 1: thousands of government entities and private businesses across the world,
- 38:52Speaker 1: so the US Department of Justice, Immigration, and Customs. But
- 38:56Speaker 1: it also includes banks, and he does allude to banks
- 38:59Speaker 1: in a couple of different and interviews the founders. It
- 39:01Speaker 1: also includes Macy's and best Buy and Interpool. And then
- 39:06Speaker 1: on the list you also see credentialed users at the FBI.
- 39:11Speaker 1: You know, we said hundreds of police departments. But here's
- 39:14Speaker 1: something very interesting. It also includes users in Saudi Arabia
- 39:18Speaker 1: and the United Arab Emirates. These are countries that are,
- 39:23Speaker 1: you know, not particularly known for their progressive policies or
- 39:27Speaker 1: their their First Amendment rights. The founder, by the way,
- 39:30Speaker 1: says that they're exercising a First Amendment right when they
- 39:34Speaker 1: scrape all these images.
- 39:36Speaker 2: Yeah, they're they're exercising a First Amendment right because they're
- 39:39Speaker 2: all publicly available and out there. And if the argument
- 39:43Speaker 2: that they make generally is that if Google can scrape
- 39:46Speaker 2: all the websites and create search terms for all those
- 39:50Speaker 2: things and then allow you to search that information. If
- 39:54Speaker 2: you know, YouTube has you know, when you upload something
- 39:57Speaker 2: to YouTube, it's all publicly available, they can scrap all
- 40:00Speaker 2: the information for all the data, and then the other
- 40:02Speaker 2: companies intertwine and use that data to help each other
- 40:05Speaker 2: find other things. Well, hey, why can't we do that too?
- 40:09Speaker 2: So we're going to However, a lot of the scrutiny
- 40:11Speaker 2: that's been happening because of that hack that was reported,
- 40:15Speaker 2: as well as just the tech in general and how
- 40:18Speaker 2: these other giant tech companies how their data is being used.
- 40:21Speaker 2: Tech giants like Twitter, Google, and YouTube same thing have
- 40:25Speaker 2: sent cease and desist letters to Clearview orders saying do
- 40:30Speaker 2: not do this anymore. And even like because of this,
- 40:34Speaker 2: some police departments and other users have just discontinued use
- 40:38Speaker 2: of the software. And when the founder, you know, responds
- 40:42Speaker 2: to this kind of thing, he says, just you know,
- 40:45Speaker 2: our legal team is on it. They're working right now
- 40:48Speaker 2: to fight that battle using the First Amendment as our
- 40:53Speaker 2: major argument.
- 40:54Speaker 3: Yeah.
- 40:54Speaker 1: And then importantly we should note that they said the Sstomer
- 41:00Speaker 1: lists was the only thing compromised, so the hackers don't
- 41:03Speaker 1: have that three billion plus and growing image archive. But
- 41:08Speaker 1: you're right, Matt, legal troubles gather like storm clouds on
- 41:12Speaker 1: the horizon. Clearview has said it's complied with things like
- 41:17Speaker 1: policies proposed by the ACLU. The ACLU has said that
- 41:21Speaker 1: they don't comply. And I think there's a case coming
- 41:25Speaker 1: up against Clearview in Illinois, and it's one of what
- 41:29Speaker 1: may be many arguably legal troubles should be gathering on
- 41:34Speaker 1: the horizon and the storm should hit, especially when we
- 41:37Speaker 1: look at the implications of this technology, and before we continue,
- 41:41Speaker 1: we want to be like cartoonishly transparent here. We're not
- 41:46Speaker 1: saying that Clearview or the people in the startup are
- 41:52Speaker 1: bad people, and we're not saying even that their intentions
- 41:54Speaker 1: are bad. We're saying that this technology, just like any
- 41:57Speaker 1: other technology, has inherent implications.
- 42:01Speaker 3: It hasn't just like fire.
- 42:04Speaker 1: It has it has some positive benefits, and it has
- 42:07Speaker 1: some very dangerous possible consequences, like how could it be misused?
- 42:13Speaker 1: The first one just off the top here. The first
- 42:15Speaker 1: one is the one we talked about a couple times
- 42:17Speaker 1: in this episode. Somebody with access could use it for
- 42:20Speaker 1: personal reasons. We did that hypothetical cop who's you know,
- 42:25Speaker 1: mad at his ex for dating Mark McGillicutty or whatever.
- 42:28Speaker 1: But we can go beyond the hypothetical examples. There's a
- 42:32Speaker 1: real life example. There's a billionaire Clearview investor who used
- 42:36Speaker 1: the app to stalk his daughter's newest bow. He wanted
- 42:39Speaker 1: to find out about this guy. I imagine, you know,
- 42:43Speaker 1: whether you're a millionaire or a pauper. Of course you
- 42:48Speaker 1: want to know who your kid's dating. You want to
- 42:49Speaker 1: know about them. But it's crazy that this guy was
- 42:53Speaker 1: able to use it and he found out all the
- 42:57Speaker 1: stuff that we just mentioned. Where the guy lives, what
- 43:00Speaker 1: his job is, you know, his known connections, his social
- 43:03Speaker 1: media and stuff. This example is a little weird though,
- 43:06Speaker 1: because I imagine being a billionaire, you could just hire a
- 43:09Speaker 1: person to dedicate their time to doing that.
- 43:13Speaker 3: Just hire a PI.
- 43:15Speaker 2: This is a This is a really important thing about
- 43:17Speaker 2: this that I maybe we mentioned but we didn't get
- 43:19Speaker 2: into it. The reason why he was able to find
- 43:22Speaker 2: all that stuff about this person. The way I've seen
- 43:26Speaker 2: the software work, it's not like you identify this person,
- 43:29Speaker 2: then it gives you a giant readout of all of
- 43:32Speaker 2: this person's information. You you identify that that person's face,
- 43:38Speaker 2: then all of the other publicly publicly available on the
- 43:41Speaker 2: internet faces show up and you can click on it
- 43:45Speaker 2: and it takes you to the website where that face is,
- 43:48Speaker 2: where that image is located, so that means you're then
- 43:51Speaker 2: directly connected to their social media to anywhere that that
- 43:54Speaker 2: photo has been posted, and then through those other websites
- 43:57Speaker 2: you can gather all that other information. At least that's
- 44:00Speaker 2: the way it was demonstrated in that CNN Business interview
- 44:03Speaker 2: in a couple of other places.
- 44:05Speaker 1: That's the way it works for now, But we know
- 44:08Speaker 1: that that UI could easily be upgraded to maybe pull
- 44:13Speaker 1: to scrape that information if it's publicly available, right, I mean,
- 44:17Speaker 1: think about how many things are we've talked about before,
- 44:20Speaker 1: how easy it is to find a phone number, how
- 44:23Speaker 1: easy it is to find not just your address, but
- 44:26Speaker 1: the last few addresses you lived at over the years.
- 44:29Speaker 1: This is where we should also mention people don't trust
- 44:33Speaker 1: clearview AI is a little bit of stereotyping, but they
- 44:36Speaker 1: don't trust it because one of the investors, Peter Thiel,
- 44:39Speaker 1: is also big in Facebook.
- 44:42Speaker 3: So Facebook.
- 44:46Speaker 1: Facebook is good at a lot of things, and users
- 44:49Speaker 1: privacy has never been one that is not a bug,
- 44:52Speaker 1: that's a feature from their opinion, and.
- 44:55Speaker 2: This is just a dovetail on all of that. In
- 45:00Speaker 2: that interview, Donnia Sullivan does a search of his face
- 45:04Speaker 2: and it's the official profile picture on the website where
- 45:09Speaker 2: he works for the CNN business site. And when it scrolled,
- 45:13Speaker 2: when they scrolled all the way down, there was an
- 45:16Speaker 2: image of him because he's in his I guess thirties,
- 45:19Speaker 2: late thirties or something to that. I don't know how
- 45:21Speaker 2: old he is, but he's a person about our age.
- 45:24Speaker 2: Ben and Nola and I. They scrolled the way down
- 45:27Speaker 2: and there's an image of him from a local newspaper
- 45:30Speaker 2: where he was in Ireland, right when he was sixteen
- 45:35Speaker 2: years old. And it's a group of you know, teenagers
- 45:38Speaker 2: holding up a sign and you can just see his
- 45:40Speaker 2: face and he looks absolutely different. I mean, I would
- 45:46Speaker 2: not have been able to tell that that was him,
- 45:48Speaker 2: but he recognized the photo and he knew it was him.
- 45:51Speaker 2: And this is the kind of thing where you know,
- 45:54Speaker 2: this is not an allegation on my part. This is
- 45:58Speaker 2: just me wondering and ruminating. I'm wondering how they connected
- 46:03Speaker 2: that up, How this face recognition software inserted that picture
- 46:08Speaker 2: of him where he looks so different. And the founder says, well,
- 46:13Speaker 2: you know that there's still parts of your face, the
- 46:15Speaker 2: geometry of it that remained the same over the years,
- 46:18Speaker 2: even if you put on weight, even if you're wearing glasses,
- 46:20Speaker 2: even if you're covering your face. I get that argument,
- 46:25Speaker 2: But to me, it makes me wonder if there is
- 46:31Speaker 2: some kind of added like Spochio search that's going on
- 46:36Speaker 2: within this system where you're looking at that's just another
- 46:40Speaker 2: company that you can search social media, any social media
- 46:45Speaker 2: accounts that somebody has using an email or a phone
- 46:47Speaker 2: number or a name, And it makes me wonder if
- 46:50Speaker 2: they're connecting things up there as an extra layer of like.
- 46:54Speaker 1: A name, like prioritizing an image by other information.
- 46:59Speaker 2: Yeah, it makes me wonder if that's happening. I guess not,
- 47:02Speaker 2: but they I would say he didn't sufficiently, the founder
- 47:06Speaker 2: didn't sufficiently explain how that image got in there, and
- 47:10Speaker 2: the interviewer is clearly a little weirded out and disturbed
- 47:14Speaker 2: by it.
- 47:15Speaker 3: Well, another thing too, is I mean, you know, in
- 47:17Speaker 3: the United States, presumably you you know you might be
- 47:21Speaker 3: identified using this technology, but doesn't mean you are immediately
- 47:24Speaker 3: going to be convicted of something without some kind of
- 47:27Speaker 3: physical evidence or without a trial. But in other countries
- 47:31Speaker 3: where stuff like that isn't as much of a thing,
- 47:33Speaker 3: it could be used to identify social dissidents or like
- 47:37Speaker 3: people that are speaking out against the totalitarian government and
- 47:40Speaker 3: flag them and throw them in the gulag or whatever,
- 47:43Speaker 3: you know, without ever you know, batting an eye. And
- 47:46Speaker 3: that is sort of the crux of the problem with
- 47:49Speaker 3: this software is it depends on the notion of people
- 47:53Speaker 3: that use it being good actors and and being good
- 47:57Speaker 3: stewards of this power that they have. And as we know,
- 48:01Speaker 3: power corrupts, and we've got a lot of governments that
- 48:05Speaker 3: are kind of corrupt and out to shut down any
- 48:08Speaker 3: kind of criticism of what they do. And this would
- 48:13Speaker 3: be a great way to identify those people and round
- 48:16Speaker 3: them up and put them away.
- 48:19Speaker 1: Yeah, and it's it's it's dangerous for that reason. You know,
- 48:23Speaker 1: you could you could detain people, and you could cut
- 48:27Speaker 1: around some of their legal protections like what if this,
- 48:31Speaker 1: what if this counted in a court of law as
- 48:34Speaker 1: definitive identification, then you would say, well, we don't need
- 48:37Speaker 1: to figure out if you're the person, right, It could
- 48:40Speaker 1: get so dangerous so quickly. There's another point we have
- 48:43Speaker 1: to add. You know, this sounds like some kind of
- 48:46Speaker 1: Skynet Big Brother stuff. It can also be used inaccurately.
- 48:51Speaker 1: It is far from perfect. There's It's one thing that
- 48:54Speaker 1: I've been talking a lot about with some people online about, uh,
- 48:59Speaker 1: facial recognition has an inherent racial bias that I wish
- 49:04Speaker 1: more people talked about for non white people, for persons
- 49:08Speaker 1: of color. This stuff lags in accuracy, not just clearview,
- 49:12Speaker 1: but in general. Facial recognition is known for this. And
- 49:16Speaker 1: this means that you could be arrested due to a
- 49:19Speaker 1: computer error because the software decided that you look like
- 49:24Speaker 1: a guy you never met, who lives in a state
- 49:27Speaker 1: you've never been to, who committed a crime that occurred
- 49:30Speaker 1: while you were not alive, and it could happen. It
- 49:33Speaker 1: could happen here.
- 49:34Speaker 2: And to continue with that, we mentioned, you know, the
- 49:38Speaker 2: people that were known to be using this software, the organizations,
- 49:42Speaker 2: and two of them or three of them, let's just
- 49:44Speaker 2: point these these out Department of Justice, US Immigration and
- 49:49Speaker 2: Customs Enforcement. Let's just leave it to those two. So
- 49:53Speaker 2: take the problems you're talking about ben that are inherent
- 49:56Speaker 2: to this system. Apply that then to let's say, Immigration
- 50:00Speaker 2: and custom looking for people who are in the United
- 50:04Speaker 2: States illegally, and imagine that you're attempting to round everyone up.
- 50:10Speaker 2: The implications of this software would be extremely effective, I
- 50:15Speaker 2: would say, or completely ineffective because they're detaining and rounding
- 50:21Speaker 2: up people incorrectly because of the problems with the software.
- 50:25Speaker 1: Well think about this too, the problems with the software.
- 50:31Speaker 1: So they can pound on the state level, but they
- 50:33Speaker 1: can pound on the private level too, And this is
- 50:36Speaker 1: one that I think for the immediate future is at
- 50:41Speaker 1: least as dangerous as the Orwellian stuff we're talking about
- 50:44Speaker 1: right now. So that data, when it's mined right, just
- 50:48Speaker 1: like your data on any other social media, it can
- 50:50Speaker 1: be sold to third parties, insurance companies, financial institutions, and
- 50:55Speaker 1: so on. This goes way past ads the phrase. I
- 51:00Speaker 1: know I'm kind of churchifying here, but the phrase I
- 51:02Speaker 1: think of is a longitudinal profile of your face and
- 51:07Speaker 1: your behavior over time. So it's like watching a real
- 51:12Speaker 1: life progression of you aging. And let's say that an
- 51:17Speaker 1: insurance company or an algorithm could start to pick up
- 51:20Speaker 1: on certain medical conditions that are manifesting in your face
- 51:24Speaker 1: in very minuscule ways ways of human being, and even
- 51:28Speaker 1: a human doctor wouldn't sense. The computer picks this up
- 51:32Speaker 1: before you know anything's wrong, sort of like the way
- 51:35Speaker 1: targets algorithms figured out that poor girl was pregnant before
- 51:39Speaker 1: her family knew.
- 51:42Speaker 3: So here's the question.
- 51:44Speaker 1: Would the insurance company legally be required to notify you
- 51:49Speaker 1: of this possible health condition? Say, like, there's clearly something happening.
- 51:54Speaker 1: Maybe if they have video to in your behavior that
- 51:57Speaker 1: is indicative of an early on set terminal condition, would
- 52:03Speaker 1: they be legally required to tell you that they knew
- 52:06Speaker 1: you had a let's say, eighty percent chance of dying
- 52:10Speaker 1: in the next two years and give you the assistance
- 52:12Speaker 1: you need right now. That's a nope.
- 52:15Speaker 3: That is a big no. Chief. They don't have to
- 52:18Speaker 3: do a damn thing for you.
- 52:19Speaker 1: What they can do with absolutely no repercussions is say, hey,
- 52:23Speaker 1: this person's probably going to be dead in two years
- 52:25Speaker 1: and it's going to be expensive to get them to
- 52:27Speaker 1: a year two.
- 52:28Speaker 3: So let's just drop them. You, let's just drop them.
- 52:31Speaker 1: There's no law that, there's no like ethical legal requirement
- 52:36Speaker 1: for them to tell you what they know.
- 52:38Speaker 3: And that's that's.
- 52:40Speaker 1: Mind boggling, that is reprehensible. That is like I need
- 52:44Speaker 1: I need to get my good non abridged the source
- 52:47Speaker 1: to come up with all the different words for how
- 52:50Speaker 1: bad and disturbing that is.
- 52:51Speaker 3: Oh, ben, I would I love it when you do that,
- 52:53Speaker 3: because I always learn a new word.
- 52:57Speaker 1: Uh, did you guys ever just light in the mood?
- 53:00Speaker 1: Did you guys ever hear that old joke about the
- 53:03Speaker 1: abridged thesaurus. It's like I have an abridged thesaurus. Not
- 53:06Speaker 1: only is it terrible, it's terrible. So I didn't write that,
- 53:12Speaker 1: but you but you can. The good news is right now,
- 53:16Speaker 1: depending on where you live, you can opt out of Clearview.
- 53:20Speaker 3: Oh my gosh.
- 53:21Speaker 5: But even the way you opt out isn't like it's
- 53:25Speaker 5: such a burden. It's such a burdensome process. You can't
- 53:28Speaker 5: just click a box in an email, right you. All
- 53:31Speaker 5: you have to do is send a headshot. We all
- 53:33Speaker 5: have those and an image of your government issued ID.
- 53:38Speaker 5: And this only applies also to residents of California or
- 53:42Speaker 5: the EU.
- 53:43Speaker 3: For whatever reason, sorry, rest of the world.
- 53:46Speaker 2: Right because of privacy, because of privacy laws that exists there.
- 53:52Speaker 1: Yeah, we shall also point out, going back to this
- 53:55Speaker 1: accuracy thing there, there are serious questions about how accurate
- 53:59Speaker 1: Clearview is in general. So BuzzFeed, according to marketing literature
- 54:04Speaker 1: they found from Clearview, says that the company the company
- 54:08Speaker 1: touts the ability to find a match out of one
- 54:11Speaker 1: million faces ninety eight point six percent of the time.
- 54:14Speaker 1: But when Clearview did finally start talking to the New
- 54:16Speaker 1: York Times, they said, the tool produces a match up
- 54:19Speaker 1: to seventy five percent of the time, and we don't
- 54:22Speaker 1: know how many of those are quote unquote true matches.
- 54:25Speaker 1: So there's some contradictory informations coming out which can make
- 54:29Speaker 1: people feel uncomfortable, obviously, and that's that's where we leave today.
- 54:34Speaker 1: The battle lines are firming up. On one side, you
- 54:37Speaker 1: have Clearview, its investors, it's clients, and what I would
- 54:41Speaker 1: call the techno optimists. Right on the other side, you
- 54:45Speaker 1: have privacy advocates, you have tech giants like Microsoft IBM,
- 54:49Speaker 1: you even have the Pope. The Pope came out against
- 54:51Speaker 1: facial recognition and he summed it up pretty nicely. He said,
- 54:55Speaker 1: quote this asymmetry by which a select view know everything
- 54:59Speaker 1: about us while you know nothing about them Doll's critical
- 55:02Speaker 1: thought and the conscious exercise of freedom.
- 55:05Speaker 2: Imagine applying that to a priest who hears the confessions
- 55:10Speaker 2: from everyone every week when they go and tell their
- 55:16Speaker 2: dirty secrets. The priest is then the one who everybody
- 55:20Speaker 2: knows nothing about, but he knows everything about them, just
- 55:24Speaker 2: putting it out there, especially hey, and also this thing
- 55:28Speaker 2: is targeting child sex abuse and abusers. Just also leaving
- 55:33Speaker 2: that there.
- 55:34Speaker 1: Yeah, one other thing for anyone listening along and thinking,
- 55:39Speaker 1: good thing.
- 55:40Speaker 3: I made my.
- 55:40Speaker 1: Profile on insert social media here private years ago.
- 55:44Speaker 3: Sorry, homie.
- 55:45Speaker 1: Anything that was public at some point is pretty much
- 55:48Speaker 1: in the system, even if you later made it private.
- 55:50Speaker 3: So check your MySpace.
- 55:52Speaker 2: In that CNN Business interview, as you said, Ben, they
- 55:55Speaker 2: test out the software on the producer who has a
- 55:58Speaker 2: private Instagram account, and images from that account show up
- 56:03Speaker 2: in the search and it is because it was public
- 56:06Speaker 2: at one point.
- 56:07Speaker 1: And that's where we leave off today. What do you think,
- 56:11Speaker 1: fellow conspiracy realists? Do the benefits outweigh the potential consequences
- 56:16Speaker 1: of this? Do the consequences outweigh the potential benefits?
- 56:21Speaker 3: Let us know.
- 56:22Speaker 1: You can find us on Facebook, you can find us
- 56:23Speaker 1: on Instagram, you can find us on Twitter. We always
- 56:26Speaker 1: like to recommend our Facebook community page.
- 56:29Speaker 3: Here's where it gets crazy. Now.
- 56:32Speaker 2: You can jump on there. You can talk about this
- 56:34Speaker 2: episode or any other in the past. You can post
- 56:37Speaker 2: some dank memes, let's say, some conspiracy memes, or facial
- 56:40Speaker 2: recognition stuff. Maybe you've had access to this before, to clearview,
- 56:47Speaker 2: and you want to talk to us about it. Don't
- 56:48Speaker 2: do that on Facebook. Don't do that. But if you
- 56:50Speaker 2: do want to tell us about that, you can call
- 56:52Speaker 2: our number. We are one eight three three std WYTK.
- 56:58Speaker 2: You can leave a message there, tell about it, let
- 57:00Speaker 2: us know. If you don't want to be identified, if
- 57:02Speaker 2: you don't want us to know or talk about anything
- 57:06Speaker 2: on air, just just let us know, give us the information.
- 57:11Speaker 2: We'd love to hear from you.
- 57:12Speaker 1: And I want to add to this to say that
- 57:15Speaker 1: I finally start despite my strange phobia regarding phones, I
- 57:22Speaker 1: started diving in Matt You're doing massive, amazing work there,
- 57:26Speaker 1: and I've started listening to these calls as well. Thank
- 57:30Speaker 1: you so much to everyone who calls in. It's inspiring
- 57:34Speaker 1: and I don't know about you. I don't know about
- 57:36Speaker 1: you guys, but it makes me feel like what we're
- 57:39Speaker 1: doing is worthwhile, if that makes sense.
- 57:42Speaker 2: Oh, definitely, definitely. I've had you know. Hopefully you're gonna
- 57:47Speaker 2: if you could get past the phobia, Ben, hopefully you
- 57:49Speaker 2: can call a few people because you can use that
- 57:52Speaker 2: app and you should be good to go actually speaking
- 57:56Speaker 2: to to you know you know who you are. If
- 58:00Speaker 2: we've talked on the phone, it means a great deal,
- 58:04Speaker 2: as Ben is saying, just to us, to know that
- 58:07Speaker 2: we're not just talking in a darkened room and that
- 58:11Speaker 2: you guys, you guys care about us as much as
- 58:13Speaker 2: we care about you. So this is a great relationship.
- 58:16Speaker 1: Let's keep doing it well, said Matt Well said, And
- 58:21Speaker 1: we'll be following up with some of those messages in
- 58:24Speaker 1: the future, so keep them coming. One last thing, if
- 58:28Speaker 1: you say, look, Ben's right, phones are terrifying and weird,
- 58:33Speaker 1: but also social media. You guys just told me how
- 58:36Speaker 1: dangerous that is.
- 58:37Speaker 3: Why Why on earth? What gives how?
- 58:40Speaker 1: I have a story that my fellow listeners need to
- 58:43Speaker 1: know about. I have some I have some experience. I
- 58:46Speaker 1: have some terrible jokes to tell you, but I don't
- 58:49Speaker 1: know how to contact you.
- 58:50Speaker 3: Well, we have good news for you.
- 58:53Speaker 1: You can always twenty four to seven contact us at
- 58:56Speaker 1: our good old fashioned email where we.
- 58:58Speaker 2: Are conspiracy see at iHeartRadio dot com. Stuff they Don't
- 59:21Speaker 2: want you to Know is a production of iHeartRadio. For
- 59:24Speaker 2: more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,
- 59:28Speaker 2: or wherever you listen to your favorite shows,