Scammers Use AI to Make Lost Pet Scams More Believable

(MUSIC INTRO)

[00:00:01] Bob: This week on The Perfect Scam.

[00:00:04] Bill Cosens: About 12:30 we get a phone call that says, “We think we may have found your dog. He's been hit by a car and they need to do emergency surgery on him.” And I said, “Can we just get a picture to make sure it's him. And in about 30 seconds we have a picture, and at first glance it's Archer. He's laying on a operating table. There's two vets standing over him in surgical masks, ready to operate.

(MUSIC SEGUE)

[00:00:40] Bob: Welcome back to The Perfect Scam. I'm your host, Bob Sullivan.

(MUSIC SEGUE)

[00:00:44] Bob: We keep telling you that artificial intelligence will be a boon for criminals. And it seems like each week brings a new hint that this future is, well, it's arriving fast. In a little while we'll talk to a Harvard scholar who tracks AI incidents in a kind of police blotter for artificial intelligence. But first, you'll hear an incredible, dark story of a scam that seems tailor-made for AI. And as our Harvard scholar says, when you hear what happened, you'll say, of course, now that's another thing we have to worry about. Our story begins with Archer.

[00:01:23] Bill Cosens: Yeah, he's 11. We got him from the Volusia County Human Society over in New Smyrna Beach 10 years ago, a little over 10 years ago.

[00:01:31] Bob: Oh wow.

[00:01:32] Bill Cosens: Yeah.

[00:01:34] Bob: Archer is a beagle mix, a dog, and that's the voice of Bill Cosens who lives in Florida about halfway between Orlando and Daytona Beach.

[00:01:44] Bill Cosens: So and yeah, and they thought, they thought he was about 1 years old when we got him. We got him and Hazel, like I said, the smart one that went back in the house, we got the two of them at the same time. Archer was for my son, and Hazel was for my daughter, and we, when we had moved into our house, that was a housewarming. We'd gone from, and that's a whole other story, different places where we lived where they weren't able to have animals, and so we got these two right after we moved into the house. So we moved into the house just over 10 years ago.

[00:02:18] Bob: So this, this dog has all sorts of extra significance for you.

[00:02:21] Bill Cosens: He's our happy-go-lucky guy. He's just always happy and always smiling. He's the one that you just say his name and he starts wagging his tail. And he wags it so hard it's like he's wagging his whole butt, and it's funny. We call him his helicopter tail because he doesn't just go side to side, it goes like in a big circle he's wagging so hard. And it looks like he's trying to take off from his rear end with a helicopter.

[00:02:46] Bob: (laughs)

[00:02:46] Bill Cosens: So he's the sweetest little guy. He's just obviously and understandably just not been himself since this whole thing happened.

[00:02:58] Bob: When you have a, a happy dog who suddenly is sad, it's this, it's such a crushing thing just to witness, right?

[00:03:06] Bill Cosens: It is. It's terrible.

[00:03:09] Bob: Why is Archer not wagging his butt quite so much? Why isn't he himself? It all started on a Saturday night a few weeks ago.

[00:03:17] Bill Cosens: We came home from a Solar Bears hockey game. Got home a little after 10:30. Normally we're greeted by our three big dogs at the door and only had one at the door. So we uh, thought, okay, maybe the other two are in the backyard doing their business, and so we go into the backyard, look, calling for them. No, no joy. We're not seeing anybody. We're not seeing them at all, and we noticed that three of the fence panels have been pushed out. We have one of the vinyl fences, and three of the panels had been pushed out.

[00:03:49] Bob: God, that's a horrible feeling, right?

[00:03:51] Bill Cosens: Right, so we think, think goodness, they've gotten out. Why would they push the panels out? They've never pushed them out before. We've never had an issue like this.

[00:04:01] Bob: It's pretty obvious what's happened. The dogs got out, and now they are, well who knows where.

[00:04:09] Bill Cosens: So we go into the front yard, side yard where the panels got pushed, and we're calling for them, and the two got out. We had Lyric and Archer, and we're calling and Lyric comes running after about 30, 40 seconds; she's all scratched up, you know, her belly, her sides are all scratched up. And she's all out of breath like she's been out for hours. We bring her in, she drinks like a gallon of water, but what's weird is Archer is a follower. He is not an alpha dog, and so would usually, we expected if Lyric's there, Archer's going to be there. And there's no Archer.

[00:04:43] Bob: There's no Archer.

[00:04:46] Bill Cosens: We go back out and calling for Archer. We start walking the neighborhood, to walk all the way around the loop that we normally would walk the dogs, calling his name. We see a few people that are out. Nobody's seen him, and that's weird. So we get in the car, we drive around for probably another two hours, so at this point we put about 2½ hours into looking for him. We probably drove every single side street and every single major road in about a 2-mile radius from our house. Nobody's seen him. There's no sign of him. Just worried that he's gotten out. There are some woods and some swamp areas around us, so we're worried that maybe he's gotten trapped somewhere, 'cause the only reason why he wouldn't come if we call is if he couldn't.

[00:05:28] Bob: Yeah, yeah.

[00:05:31] Bob: As they're driving around reality starts to set in. Archer is missing. So Bill's wife writes a lost dog post and puts it on social media. Hours later they still heard nothing. So they head home which is eerily quiet.

[00:05:48] Bill Cosens: The dogs were even moping, especially Lyric, which is the big one that pushed through the fence. Um, she's very sympathetic and empathetic. And she, the whole time he was gone, is acting like she's at fault. She's just not herself. She's not eating like she normally does.

[00:06:08] Bob: And now they start to imagine the worst.

[00:06:13] Bill Cosens: But you know our first worry is that they're housedogs. They don't know anything about cars. Luckily, we didn't see him hit by a car anywhere in, in our driving. So that was our first, at least somewhat consolation is he doesn't appear to be hit by a car. Second one is worried because they do have some bears in the woods that are around us, hopefully he didn't get taken by a bear or an alligator in the swampy areas and stuff. And so anyway, we finally make it to bed, don't sleep much. Wake up in the morning, search for another hour, an hour and a half. Can't find him.

[00:06:48] Bob: Can't find him. It's all starting to feel hopeless.

[00:06:53] Bill Cosens: And then about 12:30 we get a phone call that says, "We see that, that you have lost a dog. We think we may have found your dog." So we asked them, "Okay, what does the dog look like?" And they describe Archer to a T.

[00:07:09] Bob: Archer has been found. That's the wonderful news. But...

[00:07:15] Bill Cosens: And so now we're just ecstatic. Great, somebody found our dog, awesome. We're super happy about it. And then they say, "He's, he apparently has been hit by a car." So now we go from elation to dread. Is he dead? Is he injured? What's going on? We were asking questions on where did this happen, and they were telling us this happened on Laredo Drive, which is right around the corner from our house. There's a long stretch. "It looked like somebody hit him, hit him with a car about three hours ago. A woman found him about a half hour ago and brought him to the vet facility here and so that's why we're working on him."

[00:07:58] Bob: We're working on him? Who is we?

[00:08:02] Bill Cosens: They tell us that they're a veterinarian that's, it's like a veterinarian team that's been contracted by the state. That they come in and they're assigned to this area and whenever there's a vet office or a animal hospital that's about to close, these people get called in if there's surgery or something that needs to be done.

[00:08:23] Bob: In this case, something needs to be done.

[00:08:27] Bill Cosens: And that he's been hit by a car, he's got a shoulder injury, and he's got a broken femur. And there's some sort of pinched nerve that's causing partial paralysis. And they need to do emergency surgery on him, and there's about a 65% chance that they can restore all of his mobility.

[00:08:49] Bob: It's all so much to take in. But Bill is ready to spring into action.

[00:08:54] Bill Cosens: So again, now upset, and we're like, tell us where you are. They give us an address. My wife writes it down. It's like, okay, we're coming to your right now. They said, "We just need you to authorize the surgery." "Okay, great, we're, we're going to come down there. We want to see him." "Unfortunately, you can't see him before the surgery because he's already prepped for surgery. Time is of the essence; we don't want him to lose mobility in his legs. He's like the lower half of him is paralyzed right now. We, we are pretty certain that we can restore the mobility, but we've got to act quick. So will you authorize the surgery?" "Absolutely! Give him the surgery." So, "Okay, great." So they said, "But you're welcome to come down here and wait in the lobby." And so we said, "Okay, great. That's where we're headed." So they said, "Let me transfer you over to Miss Francis who's in our accounting department so you can get the payment all set up." So we asked him "How, how much is it going to be for the surgery?" They said, "$2,746." And we said, "Okay, fine."

[00:09:53] Bob: That's a lot of money, but of course they will do anything for Archer. But Bill's wife says, let's make sure it's Archer.

[00:10:02] Bill Cosens: We said, "Can we just pay that now. We, if we can't see him, can we at least get, get a, a picture to make sure it's him?" So they said, "Sure, no problem." And in about 30 seconds we have a picture, and at first glance it's Archer. He's laying on a operating table. There's two vets standing over him in surgical masks, ready to operate. There's background; you have some x-rays up there and so it's Archer's face for sure. There's tubes coming out of his mouth. He's knocked out, you see an open wound on his front right shoulder. And we're, my wife sees this and she's hysterical. She's crying. "Oh my gosh, it's Archer. Look at the poor baby," and, and how, "Okay, let's go ahead and pay it now. We'll be down there in a minute."

[00:10:50] Bob: But they won't be able to see Archer right away. So they ask how long the surgery is going to take.

[00:10:56] Bill Cosens: "It's going to be a 2-hour surgery and about an hour to recover from the anesthesia. He's already knocked out now, but you're welcome to come and wait in the lobby. And as soon as he's awake, then you can come back and get him." And great. She says, "How, how can we pay you?" They say, "We'll take," you know, "credit card, debit card. You can pay with PayPal, you can pay with Cash App, you can pay with Zelle, you can pay with Venmo." And my wife just freezes. And I said, "Well just give them the Apple card." And she just says, "Uh, uh..." and I said, "Just give them the Apple card." And she says, "Can I call you right back at this number?" And he says, "Oh sure, you can call right back at this number, but we're going to go ahead and start the operation since you said you were authorizing it." She said, "Okay, we'll call you right back."

[00:11:43] Bob: So she hangs up and Bill is confused.

[00:11:48] Bill Cosens: And she hangs up on them. I'm like, "What's the problem? Just give them the Apple card number." And she looks at me and says, "There was something off by the photo," which I had not seen at this point. And I said, "Okay." and she said, "What type of vet takes Zelle?"

[00:12:03] Bob: What type of vet takes Zelle? It's a good question, and it's enough to spur the desperate dog owners to do just a little more research.

[00:12:14] Bill Cosens: And she said, "Let's look up the address that they gave us." So we do, and it turns out it's City Hall. And we're like, there's no vet there at City Hall, that's, and she had asked the name and they said it was Volusia County Animal Control Vet. And we, it just so happened, I had talked to Volusia County Animal Control about 20 minutes before the call looking to see if somebody had caught him while he was out and didn't bother to call us. "But let me just do some due diligence. We're going to check with Volusia County to see if maybe they were brought into the county and something was done with that." He put us on hold, comes back about 5 minutes later, and says, "You're still there?" "Yes, we're still here." He said, "Yeah, we just talked to the Volusia County. They don't have anybody; so they've got one guy on. His kennels are empty. They don't have a vet. And both of us were talking about it and we said that had somebody brought in an injured dog, the first thing we would do is send him to one of the 24/7 type of animal hospitals. So there's no way we would have them here, and we've never heard of any sort of contracted state vet. It sounds like you're being scammed."

[00:13:24] Bob: It sounds like you're being scammed? So they hang up with animal control and Bill turns to his wife.

[00:13:31] Bill Cosens: And I said, "You've got to be kidding me! Who would do this?" It never occurred to me that somebody would prey upon a person who's already distraught that they had an animal missing for about 12, a little over 12 hours now at this point, that she would then make up a story about them being injured and try to scam them out of money.

[00:13:54] Bob: But it seems undeniable. These people who called are lying. They're just asking for money. And worse yet, Archer is still missing.

[00:14:04] Bill Cosens: Yeah, we're just upset at this point, but we had gone through, and not trying to make my wife feel bad or anything. She was very depressed when he was gone. Moping. Just didn't want to eat, couldn't sleep. And so the only positive, if you will, of somebody trying to scam us was she went from depressed to mad. And you can function mad.

[00:14:28] Bob: How interesting, huh, yeah.

[00:14:29] Bill Cosens: You can't really function depressed, but you can function mad. So now she's ticked off, and you don't want to mess with her when she's ticked off.

[00:14:37] Bob: So the first thing she does is enlist the help of the kids. And then the story really starts to fall apart.

[00:14:45] Bill Cosens: So... she's now armed with, uh, she's taken the picture that they sent us, had sent it to both of our kids and both of our kids come back and go, "This is definitely AI. If you look here," and there were some different, my daughter sent a picture of Archer that was recent, and said, "Look, the coloring on the back of his legs is wrong. The coloring on his stomach, he doesn't have fur on his stomach, it's skin, and this has the fur all the way there." And just some other issues that were not right with the photo. And my son has said, "Yeah, everywhere that you've got text and you try to zoom in on the text, it's gibberish. It's not, it, it's like you're looking at Klingon or some alien language, it's not English." And he said, "That's the first tell-tale sign that somebody's used AI on a photo."

[00:15:36] Bob: And yet the, the face looked really real, right?

[00:15:39] Bill Cosens: And so we're like, "How would they have the face?" But again, my wife had posted when he went missing, while we were driving the neighborhood, and I'm driving and she's yelling, and I'm yelling, and she's gone on all of the social media area, local dog, missing dog, social media accounts and has posted pictures of him because obviously saying we're missing a dog and describing him is nowhere near as accurate as here's a picture of him. So she's put pictures of Archer on 7 or 8 of the local social media outlets saying that, that are designed for lost dogs.

[00:16:18] Bob: So they are still desperate to find out anything they can about Archer, and well we don't recommend this, but they call the criminals again.

[00:16:27] Bill Cosens: She's asking him questions and then repeating the question, and then will ask four or five more questions, and come back to the first question and ask again. And what she's doing is seeing if he's consistent with his story. And what we see is the pattern of after asking five or six questions then going back to a question that you'd asked before those five or six questions, they were giving differing answers. The guy started getting really frustrated. We also noticed that he was very good at telling us exactly how we could pay, but wasn't very good at dog anatomy. So my wife asked, "Well, if he's up and around, can we get a picture of him?" And they say, "Sure." And they clearly were ready to give us a picture when they said we couldn't go see him before the surgery, because they had that picture ready to go. This time they're scrambling to get another picture, and they send us a picture that is identical to the first one except now there's two fake bandages on Archer. They've moved where the vets were, the two vet surgeons have moved, and the towel that was underneath Archer has been moved and it's just the stainless steel table, and I guess they had taken this AI from an actual picture, like they took a picture of a vet's office and a picture of Archer and used AI to combine it to make a picture. And they forgot to get rid of actual text on the x-rays. So when we zoom in on the x-rays that now have text on that didn't before, you can read "Post-operation 12/12/2022." So this operation that supposedly just happened, according to the x-ray, happened almost four years ago. And, and their second attempt at fooling us with AI was just ridiculous. And that just, the more my wife asked, the more they get ridiculous and the answers start getting ridiculous. The guy, you can tell, is starting to get frustrated because he's been found out. And he starts getting snappy with us and ruder with us, and then tells us that we cannot hang up the phone. Then she got him to say, "We can take payment at CVS, and we can take payment at Walmart. There's a CVS right around the corner from your house, but Walmart's a bit farther, so you might want to just go to the CVS, 'cause you've got to stay on the phone while we're doing this. If you hang up, then what's going to happen is you lose your rights to your dog and you're going to have to go through the state to get him through auction." We're like, that's so ridiculous. It's, are you even listening to the stuff you're telling us?

[00:19:25] Bob: Eventually the call ends, and thankfully Bill and his wife never send any money to these veterinarians, but they are no closer to being reunited with Archer, so they retrace their steps.

[00:19:37] Bill Cosens: We go back and call the neighborhood again, calling for his name, driving around the neighborhood and still nothing. So we just go back to the house just still mad, still upset, figuring we've probably have lost our dog. Hopefully he's not hurt, hopefully somebody found him, maybe somehow his collar's off, and so he was found by somebody who had no idea to look on the lost dog social media sites. And we're just hoping that the positive of, okay, we're not going to get him back, but hopefully he's safe somewhere.

[00:20:16] Bob: A few days later with the help of a neighbor and some camera sleuthing, they being to suspect maybe another neighbor has taken their dog. And then a huge surprise. That helpful neighbor, well he calls to say...

[00:20:29] Bill Cosens: He says, "He's here in your side yard." "What?!" So I go running around the side yard, and there's Archer trying to get in, obviously that night when I saw that the fence panels had been pushed through, I fixed the fence, and he's trying to push back through.

[00:20:46] Bob: Archer is okay, but he's certainly had a bad couple of days.

[00:20:51] Bill Cosens: And he looks like he hasn't eaten, he's barely walking. His back legs are, he's limping really bad. He's walking like he's twice as old as he is, and he was already 11, and so I just go and pick him up and he wags his tail, and he's a happy-go-lucky dog. He's always wagged his tail, and he, he gives me a little half wag, and I pick him up. And so he hadn't eaten in five days, he drank a bunch of water. When we gave him food, he wouldn't even stand up on his hind legs. He would only prop himself up and eat while laying down, which he's never done. So we call the vet to get, he just seemed like he was exhausted. As soon as he finished eating, he just went to sleep.

[00:21:34] Bob: Bill is still working with the police to figure out what really happened. But in the meantime...

[00:21:40] Bob: And I sure hope all this works out, and I'm really, I hope that Archer gets back to wagging that tail as much as he always did.

[00:21:46] Bill Cosens: Yeah, he, he just walked in. He was able to get through the doggy door again. Still a little ginger... he used to just hop right through without thinking about it. Now it's a gradual walk through and but he came in and I looked up at him, and he saw me looking at him, and he just wags his tail again. So...

[00:22:01] Bob: There you go.

[00:22:02] Bill Cosens: He's getting there.

[00:22:03] Bob: And in the meantime, we wanted to bring this story to you because it's a powerful example of so many things we talk about on The Perfect Scam, getting your emotions to 11, that amygdala hijack we talked about before, and then piling on the pressure.

[00:22:19] Bill Cosens: Oh yeah, and hindsight being 20/20, this was exactly what the scammers want. They want you to not be thinking clearly. They want you to be distraught, they want you to be elated. They want you to be feeling all the feelings all at once because if you're feeling all these emotions, and your emotions are overwhelming you, you're not going to use logic. You're not going to sit down and think. It's very easy now that I've told the story probably 20 times, to look back and see all of the red flags that we weren't seeing when we were ready to just pay it.

[00:22:56] Bob: It so easy, yeah, when, when you know the punchline, the joke isn't funny anymore.

[00:22:59] Bill Cosens: Right.

[00:23:00] Bob: But yeah, yeah.

[00:23:01] Bill Cosens: Exactly. And, and looking at it, not only did they want you to have all these emotions, but then it was the timeline. You've got to do it now. There's emergency, push, push to get it done. First, it was he's going to be paralyzed, he's only got a 65% chance of having full mobility if you act now. So there was this sense of urgency and they're, they're trying to combine emotions and get you all worked up and then tell you exactly what you have to do and give you a very tight deadline, because when you're distraught as much as you think, oh, I'm distraught, and I'm going to think through this, and I'm going to, I'm going to weigh my options. When you're distraught and all your emotions are going, really what you want is somebody to tell you exactly what to do and when to do it. And that's exactly what they're doing, is they're purposely getting all the emotions going. And then giving you, this is the solution to all your problems, and you have to do it now, and do exactly what I'm telling you to do. So it's very easy to get scammed, it's very easy to have that.

[00:24:08] Bob: Now this is very important. As you listen to their story, there was one very critical moments.

[00:24:15] Bill Cosens: Try to, when you're distraught, take it, call back. That was the one thing that helped us was, can I call you right back? I can't think straight right now. Can I call you right back?

[00:24:26] Bob: Yeah, for sure that was a big step, yeah, yeah.

[00:24:28] Bill Cosens: And by getting off the phone and sure, you can call back at this number. Great. We were able to sit down and go, okay, and then talk to somebody else at that point. Does this sound legit? What, what's going on? Why does this maybe not sound legit? What questions do you have? And try to think clearly because maybe you're not thinking clearly and if you're not thinking clearly and the other person you're with is not thinking clearly, but you're thinking clearly on certain things but others aren't, that may be different things than what they're thinking clearly. So usually when you can sound it back and forth and talk, it'll help clearer heads prevail.

[00:25:07] Bob: Hang up, talk it over. I hope that sounds vaguely familiar, because it sounds a lot like AARP’s new campaign called “Pause, Reflect, Protect.” You might have seen the TV ads about it recently.  It’s a simple formula to remember whenever you get an unexpected email, or phone call, or any kind of communication.  Pause, Reflect, Protect.  You can see why Bill hanging up and talking this over was the best part of this story. But the worst part is the way tech tools enabled those criminals.

[00:25:43] Bob: Believe it or not, we've actually done a, a story a couple of these different kinds of stories, but one about a Florida incident where a woman posted that her dog was lost and got a phone call very similar to this. But what's unique about your situation to, to us is this AI photograph that takes this crime to me to another level. I mean seeing that photograph must have been a really big part of why this was all so convincing for you, right?

[00:26:09] Bill Cosens: Oh, absolutely. I, that was the first thing that we thought was it was a little odd that we can't come see our dog before we make a payment. Why can't we see him? And they had a very legitimate, in their mind, at least, answer is he's already prepped for surgery. We've got to have this surgery done now, so that he has retains the mobility but and then they have this picture ready showing him on the operating table. So I mean it made a lot of sense as to why we couldn't see him ahead of time, but they were obvious, whether it was they had done this to multiple people and that was the first thing they said was, if I can't see him, then you give me a picture. And then, we don’t have a picture of him. You said he's right there in front of you and you're on a cellphone. Take a picture with your cellphone and send it, And it's almost like through trial and error these people are perfecting their craft of swindling people.

[00:27:12] Bob: Through trial and error, they are perfecting their craft. Yes, that's exactly how these criminals work. And artificial intelligence is the perfect co-worker, co-conspirator. Of course, in some cases these kinds of crimes just seem like AI but might be old-fashioned photoshop, or a fake voice instead of voice cloning. But just how common are these dark uses of this powerful new technology? To get a sense of that, we asked Sean McGregor here today. He's a fellow at Harvard University. He works on AI safety, and he helps manage something called the AI Incidents Database.

[00:27:53] Sean McGregor: My background is in machine learning. I do a lot of AI safety work, and that cuts from everything from the machine do the right thing and safely, all the way to people misusing those things and doing bad things.

[00:28:09] Bob: And yes, when Sean heard Bill's story, he was shocked but not surprised.

[00:28:16] Sean McGregor: My general reaction to the scam is while people are inventive and sometimes horrible, and as soon as you see someone, someone do the thing, the bad thing, you're like yeah, that's something we have to deal with in the world now, that this is something that technology has made possible. It can do many amazing, wonderful things, but it also empowers the bad actors.

[00:28:40] Bob: And, and taking a tool and abusing it or using it for abuse, that's, that happens at every stage of technology, right?

[00:28:47] Sean McGregor: That's right. It's a tale as old as time.

[00:28:50] Bob: How hard is it to generate a photo using say you know I go onto someone's Facebook page, I find a picture of their pet, and then I tell a computer, make a photo of this pet looking like it's in dire need of surgery. How hard is it to do that?

[00:29:01] Sean McGregor: It's very easy. You don't need to be a computer scientist, you don't need to have any great degree of technical sophistication, you just have to search around a little bit, find where the tool is, plug the photo in and you're good to go.

[00:29:14] Bob: In fact, Sean helps track stories like this one in that AI Incident Database, and yes, Archer's tale is in there. It's Incident 1478. But back up a second. What is this database?

[00:29:28] Sean McGregor: I started a project and an organization behind it about 8 years ago that's dedicated to indexing what are called AI incidents. And AI incidents can be a misuse, they can be scams, they can also be malfunctions. It's basically any time someone's harmed out in the real world we want to collect that and put it in the database because this is how you make safety. This is how you make AI safer. It's inspired by some other databases in aviation where a plane crashes, you record the circumstances of that crash. And that collectively gives the aviation industry the charge of making sure it doesn't happen again in that way. And so as a result of that, I'm one of the most well-read people in the world about bad things that happen with AI, and it's a funny point to me because I more or less got my start in doing good things with AI, figuring out how to solve problems in ecology and wildfire and all environmental management problems, and here I am reading about all the bad things.

[00:30:35] Bob: Uh this sounds like the police blotter to me. I don't know if you remember police blotter in newspapers, but it always on like on page 2 or 3, and it was just all the small bad things that happen in your town, but yet it was always the most read thing in the newspaper. This is like the police blotter for artificial intelligence it sounds like to me.

[00:30:51] Sean McGregor: Yeah, you could call us that, police blotter for AI. That fits.

[00:30:55] Bob: Why do we need a police blotter for AI?

[00:30:59] Sean McGregor: AI is increasingly everywhere in the world, and the problem is AI isn't always all that robust to all the ways in which the world varies. It often fails and is unsafe as a result of that, and I ask the question of what is safety, how do we build effective safety culture and processes to make these systems something that we actually want to have in the world.

[00:31:22] Bob: So give me an example of like an exception condition, or something weird that happens that AI would handle badly.

[00:31:27] Sean McGregor: Sure. When I think about the physical system side of things, is I once saw a dog that was in someone's car on the freeway and we were in stop and go traffic, and the dog jumped out of the car onto the middle of the freeway. An autonomous car hasn't seen a lot of instances of a dog in the middle of the freeway, and people know, let's stop driving right now. The whole freeway's stopped to let the person get out of the car, collect the dog that looked none too happy about this situation it got itself into, and got back in the car and then freeway kept on going. But we don't actually know how an autonomous car would react in this scenario. It's built to try and proceed down the freeway. And this is an example of the, there's a big, wide-open world and we need to build systems that are able to handle that.

[00:32:14] Bob: So initially, the harms that Sean logged were failures of the technology to work, to detect things like runaway dogs on highways. But that's changing.

[00:32:25] Sean McGregor: I think that the story of AI safety is one where new technology is introduced and initially it's unsafe because it just hasn't gotten operational history in the world. We haven't learned from the collective experience of the technology in society. Then through time, particularly with these systems that are based on a technique called machine learning, which is basically the system is produced by data, we accumulate data and the system gets safer to operate in the world. Then as that continues and as it moves from a place of it's unreliable to reliable, then it gets more reliable for the bad actors and the bad people, and then you have to solve that, what's in our community called the dual use problem of it could be used for good, and it could be used for bad. And so there's this kind of transition from we need to care about, it's just not working to we need to care about it's working for the wrong people.

[00:33:25] Bob: Working too well in some cases. That's really interesting. It's a dual use problem, huh.

[00:33:29] Sean McGregor: Right.

[00:33:30] Bob: AI is now working for the wrong people. Sometimes anyway.

[00:33:36] Sean McGregor: North Korea, for instance, a country that really needs hard currency. This is something that will keep them, keep the government afloat and it actually has a large number of people that are trained to go out there and figure out how to do cryptocurrency scams. There was one incident where, actually we've seen several incidents wherein important business people were impersonated and used to instruct others to transfer funds. That the capacity for impersonation has gotten much greater than before where you might have a scam previously of someone's grandson calling supposedly from Mexico and saying, I'm in jail and so forth, and it doesn't sound like them but they make the line noisier so that it's harder to tell. Now we can have people sound like the grandson, sound like the CEO, and extort funds or not extort, just command funds to be transferred. And so we've seen a quite a few of those as well.

[00:34:39] Bob: And since we're talking about executives, that, that can be hundreds of thousands of dollars, right?

[00:34:44] Sean McGregor: Yeah, they, they can move around quite a lot, the funds, without it, so it's not an unusual thing to do.

[00:34:52] Bob: Until very recently, we conclude many stories like this by saying something like, well watch for bad grammar, or stilted weird speech patterns. That's not really useful advice anymore.

[00:35:05] Bob: If I saw a picture of my dog in pain, I, I'd do anything, right, and I, and I kind of want to make that point, like really drive it home for people that, that you can't believe anything you see anymore, even in these emotional situations, right?

[00:35:18] Sean McGregor: Yeah, you need to approach digital information with skepticism because it's much easier to create very convincing contact than it was previously. And there is this distinction in the community of cheap fakes vs. deep fakes that I think is good to tease out. That a cheap fake has been possible for a while. It's using photoshop, it's using photo editing tools. You can cre--, create those very convincingly without necessarily using a advanced AI model to, to produce it. The thing that's different is that took some level of skill and time to produce and most scams fail. It's a volume and a numbers game, and so what we have here is it got a lot cheaper, a lot faster to generate these kinds of highly personalized attacks, and so we see a lot more of that now and we're going to see a lot more in the future. Another way of, of thinking of this is if you, if you plug a computer into the internet that hasn't touched the internet for a few years, it will pretty instantly get a virus. Like it, there's this kind of background radiation that is permeating computers that the computers don't get into trouble just because they've been hardened through time; they're robust to that background radiation, and we are moving that kind of radiated state, that scale of just there's always something roaming around trying to attack systems. We're moving that into the human space because we're making the human space available to the machine and making it possible for AI to generate and interact with people more directly.

[00:37:04] Bob: And, and so that's, that's a fascinating comparison actually. None of us have this sort of built up immunity that our computers do to computer-related crimes.

[00:37:13] Sean McGregor: No.

[00:37:14] Bob: Has all this work on the police blotter changed any of Sean's views on AI?

[00:37:19] Sean McGregor: I don't think so. I think I've always been somewhat measured and just in awe of what we can do and how we can work with these mag--, magical chips that allow for doing such useful things at fantastical speed. It's just a, a sense of we need a corresponding response to that power for making and preventing things like scams, and I find that we haven't nearly adequately invested in that. And even looking at the AI Incident Database, we're leading in the world in that particular kind of indexing of very critical information for making the technology safer. But we're still very much subscale and if anyone wants to support the program, you can go to, it's DatabaseAI and donate to it. We, every dollar goes to covering these things and making it so people can be better protected and that the safe-, the technology would be safer.

[00:38:14] Bob: So on one side of this ledger we have the, the world's largest companies making the world's largest investment ever in data centers, and on this side of the world, we have what sounds like a bake sale to me.

[00:38:26] Sean McGregor: That's probably a fair description. We're making some tasty Rice Krispie treats over here, but uh...

[00:38:32] Bob: (chuckles)

[00:38:34] Bob: So what can we do about this future?

[00:38:36] Sean McGregor: One of my favorite sayings often attributed to sci fi author William Gibson, is "The future is already here. It's just not evenly distributed yet." There is a great many ills that have been produced as a proof of concept. You can find a, a lot of those in the AI Incident Database, and you can see then what's coming around the corner, what things that you need to start worrying about. A lot of these stories about deep, deep fake dog, they're just moving from the perfect concept phase to this is a scam deployed at great scale. And so you can watch what has been proven and what happens, and you can use that to protect yourself, but if you want to be ready for that future that, that's going to be increasingly distributed, the best way to operate is to just try and figure out what is the unfakeable ground truth. How do you operate in a manner that doesn't trust the pixels on the screen in front of you or the, increasingly the voice coming out of the speakers. And that's run things to ground, have a, a physical place, that's see the person in, person in-person because a lot of these scams are not operating from near you physically because that would make them vulnerable to law enforcement in a way that uh, they're not then operating internationally.

[00:40:04] Bob: Unfakeable ground truth. What does that expression mean?

[00:40:08] Sean McGregor: That is looking at what can't be produced at the moment by computers, and that's often not being mediated by a computer in some form. It, it is going physical to an event that can be safe, that can choose its own hazards. But uh, going physical, looking at authoritative, trusted resources, and making sure that you're communicating via those validated channels, verifying the domain of the email address that you're receiving the emails from, being skeptical of anything that is not associated with a real identity, with the real world. Ask yourself, could I find this person given a, a few hours and determination? Could I knock on their door? And if you can't answer yes to that, then be very skeptical, even if you can't answer yes to that, you've got to be careful.

[00:41:00] Bob: Sean had a few other suggestions about how to be careful in the digital age.

[00:41:05] Sean McGregor: So for people to protect themselves from this particular threat, the best starting point is likely finding ways to not give the photograph to people that would do this. This might conflict a little bit with you want your dog's picture everywhere when you want people to be on the lookout for your dog. But uh you can also remember that people generally do bring dogs to shelters and you can contact shelters and express the picture to them more specifically. That would prevent someone from trying to do that to you, but if you are so cognizant of the potential of someone generating a photo and sending it to you, then the best way to proceed is just to be skeptical of it. To never issue payment in a form that can't be rolled back. To never feel that sense of urgency if you have to respond to this person now or something bad will happen to your dog. They, your dog is in good shape if it's with a person that's contacting you. You can take a moment. If they're rushing you, if they're not willing to give you reliable information, if they're not allowing you to see your dog, then you need to be very skeptical.

[00:42:14] Bob: As, as being surrounded by all this, has it changed anything in the way that, that you personally behave with, or with your family?

[00:42:23] Sean McGregor: So I, I've never been tested, so to speak. I've never had that phone call that sounds like my sister or mother or grandparent urging immediate action to prevent something bad from happening to them. Were it to happen, I think that I would probably do something like ask a question that I know to be false or have something that is just a useful test of the circumstances. We are approaching this world where there's this constant low-level, personalized attacks on, on people. I don't think we're there quite yet. People need to be vigilant, but it's coming and I have told proactively people in my life because I am active in this space of practice, that I don't need their money. I don't need them to do any--, anything urgently. If I'm asking for that, it's not me, and they should hang up the phone and give me a call.

[00:43:21] Bob: You know that's a really good piece of advice. We usually talk about it the other direction, but tell all your friends and loved ones, I'll never ask you for money, period.

[00:43:29] Sean McGregor: Yep, yep.

[00:43:30] Bob: I'll never ask you for money, period. I like that a lot. You could consider going ahead and telling your loved ones something like that, and sadly, just don't believe everything you see for the foreseeable future. For The Perfect Scam, I'm Bob Sullivan.

(MUSIC SEGUE)

[00:44:09] Bob: If you have been targeted by a scam or fraud, you're not alone. Call the AARP Fraud Watch Network Helpline at 877-908-3360. Their trained fraud specialists can provide you with free support and guidance on what to do next. Our email address at The Perfect Scam is: theperfectscampodcast@aarp.org, and we want to hear from you. If you've been the victim of a scam or you know someone who has, and you'd like us to tell their story, write to us. That address again is: theperfectscampodcast@aarp.org. Thank you to our team of scambusters; Associate Producer, Annalea Embree; Researcher, Becky Dodson; Executive Producer, Julie Getz; and our Audio Engineer and Sound Designer, Julio Gonzalez. Be sure to find us on Apple Podcasts, Spotify, or wherever you listen to podcasts. For AARP's The Perfect Scam, I'm Bob Sullivan.

(MUSIC OUTRO)

END OF TRANSCRIPT