Endless Thread - Defrauding Big Tech

Episode Date: September 27, 2024

They were scammers. But they weren't going to scam just anyone. They were going to scam Big Tech. And they almost got away with it. Earlier this month, federal prosecutors accused a North Carolina ma...n of stealing royalty payments from music streaming platforms for seven years. He allegedly used artificial intelligence to create songs by fake bands and then play those songs to get paid. The incident resembles a scheme between 2013 and 2015 when a Lithuanian man bilked Google and Facebook out of more than $100 million before getting caught. Endless Thread's Ben Brock Johnson and Dean Russell bring two stories of grifts gone wrong. ***** Credits: This episode was produced and co-hosted by Ben Brock Johnson and Dean Russell. Mix and sound design by Emily Jankowski. It was edited by our managing producer, Samata Joshi.

Transcript
Discussion (0)
Starting point is 00:00:00 Support for endless thread comes from MathWorks, creator of MATLAB and Simulink Software, to design and develop engineered systems, accelerating the pace of discovery in engineering and science. Learn more at Mathworks.com. Support for WBUR comes from Is Business Broken, a podcast from the Merotra Institute at Boston University that explores questions like, why is innovation in healthcare so hard? Is ESG just greenwashing? And, of course, is business broken? Listen, wherever you get your podcasts.
Starting point is 00:00:36 WBUR Podcasts, Boston. Dean, how do you get your music? Uh, oh, the normal way to answer this is to say that I, like, listen to Spotify for the most part, and then also, like, the radio, like, you know, main, main public classical. Yeah. So you do give yourself over to the algorithm, right? Like, do you put your hands in the air and wave them like you just don't care when a robot is requesting you to do it? I guess. I mean, I don't like dig the whole like creation of playlists for me.
Starting point is 00:01:23 But yeah, yeah, sure. Algorithms. So I think one of the interesting things about the glut of music now that is now available to us. the sort of prison of infinite choice that we reside in, is that as we sift through the heap of music that we can now access on digital platforms, there's this kind of like this other level of music that feels not just like indie artist or non-famous,
Starting point is 00:01:52 but specifically designed to be unremarkable. This is making me think of like, I mean, not to be insulting because I do listen to this too, but like lo-fi. Yes. Like, the vast majority of it feels. It's like by people of. Yeah, exactly, exactly. Yeah.
Starting point is 00:02:12 Yeah. So I have this friend from high school named Danny Shreg. Nice name. Danny was a musician, is a musician. You know, he's always kind of like tickling the ivories. He's a piano player. And he was always putting together these kind of like circular piano music songs, like composing these songs.
Starting point is 00:02:31 And as he became a full adult with a real job, so to speak, Danny started selling jet engines as his day job. Cool. And at night, he was recording these piano songs that he was composing. And what was interesting to me was that he was actually like putting out all of this different piano music under different album names on Spotify. Like he'd have like six songs. and instead of like putting them all out on the same sort of like when he was like uploading it as the artist, he wouldn't just put it out as like one six song EP.
Starting point is 00:03:10 He would actually split them up over three different quote unquote albums as the way he like uploaded them to the platform. Okay. So he started to get into this idea of like creating music libraries that the algorithm of Spotify would then pick up and put on its automated playlists for people. And for Danny, that was how he got more plays by kind of like spreading his stuff around. Yeah. But the story I want to talk to you about today hits different. I'm going to give you the quick headlines first. Let's hear them.
Starting point is 00:03:49 Man charged an AI-generated music fraud on Spotify and Apple Music. Cool. The bands and fans were fake. The 10 million was real. Ooh, like that headline. That's a New York Times headline. They know how to do it. Yeah.
Starting point is 00:04:07 Music producer accused of using AI songs to scam streaming platforms out of $10 million in royalties. That headline was written by AI itself, so, yeah. The call's coming from in the building. So we have two stories about how people are getting fed up with tech companies just enough to try to defraud them. Dean Russell. Ben Brack Johnson. Today's endless thread of chat about what happened
Starting point is 00:04:46 when two separate men came up with two separate schemes to get paid by gaming big tech. So, you know, I want us to talk about this music story that popped off this fall and included artificial intelligence, but maybe not in the way that we usually think about it. Michael Smith is 52 years old. He is a musician.
Starting point is 00:05:15 And, you know, this will be a surprise to no one. The old record business ain't what it used to be dean. Which means that like you have to basically be Taylor Swift to make good money in the streaming world. Right. Like you, it is really hard to make it. You're getting fractions of a fraction of a fraction of a fraction. of pennies on the dollar from Spotify. But Michael Smith has been doing this
Starting point is 00:05:45 trying to make money from music streaming, apparently since before 2017, at the very least. Do you want to know how the scheme worked? I absolutely want to know how the scheme works. Okay. And again, we've not talked to Mr. Smith. I'm just telling you this story from news reporting around the country that I've seen on this and a lot of Reddit threads
Starting point is 00:06:09 that I've been reading about this. Right. It seems as though Mr. Smith started to dabble in fan creation. Fan creation. He started trying to find ways to get his music streamed
Starting point is 00:06:26 more often by creating bot listener accounts on streaming platforms. Wow. So like, we all know what making it a music is. It's effectively a popularity contest, right? Right.
Starting point is 00:06:47 It seems that Mr. Smith has figured out that he could use bot accounts to generate, according to court documents here, 661,440 streams a day, yielding annual royalties of, $1.2 million. Brilliant. It's a brilliant idea. And by the way, he partnered with other people on this scheme. So, like, he worked with the founder of an AI company to, like, make this actually work because I think he needed partners in the scheme in order to make the money.
Starting point is 00:07:29 Yeah. But he created thousands of listener accounts because he realized, of course, that he would get caught very quickly if it was just, like, one listener, listening to a butt ton load of music. Right. Like that wasn't going to fly. He was going to get flagged for that by the platforms. So he started to use different thousands of listener accounts and then starts to use virtual
Starting point is 00:07:53 private networks or VPNs, which, you know, again, gives the impression that these users are connecting from different places. They're not all coming from the same IP address. And these automated accounts, you know, basically get him to generate. a huge number of listens on the streaming platforms that aren't going to get sort of like flagged by the platforms as fraud. Wow. And the other piece of this is that like very quickly Mr. Smith realized as he started to do this was
Starting point is 00:08:26 he might have been the, you know, he might have had the fastest fingers this side of the Mason Dixon line, but he's not going to be able to compose enough music to like build the other side of that, right? Like, he can make 20 little ditties a day and still struggle with, like, again, potentially getting flagged, right? What do you mean? Like, like, he couldn't, with the body of work that he had already put out there into the world, he couldn't, like, have all of these bots listening to that, that music over and over again because it would have been too suspicious? I think that's right. I think, again, that's, like, another way.
Starting point is 00:09:09 in which the platform moderators would probably say like, oh my God, this artist is doing really well. Like we should, you know, like it would start to raise suspicions because it would be like, oh, this artist is famous. Like, who is this artist? Right. Yeah. So on or around the day after Christmas of 2018,
Starting point is 00:09:31 according to the U.S. Attorney's Office, Smith emails his two co-conspirators about composing music using AI. And he says, these two conspirators have not been hauled in yet, by the way, at the time of recording this. Okay. So your friend could actually be involved in this,
Starting point is 00:09:53 Danny Shreg. Danny could be implicated. Danny, I hope you've been careful. So he emails his two co-conspirators. We need to get a ton of songs fast to make this work around the anti-fraud policies these guys are all using now. Hmm. And so he starts to create AI-generated music.
Starting point is 00:10:18 And that sort of helps him again kind of like spread things out so he doesn't get flagged for what he eventually gets flagged for, which he's charged with wire fraud and money laundering. Okay. To me, what is interesting about this is the argument that the U.S. attorney, which, you know, presumably is, you know, in touch with or working at the behest of the music platforms, what Mr. Smith did that was wrong as he's like generating song names like N lower dash 7A2BD2D. 74-1621-4-385, etc., etc. That was his problem, right? Like, it was like, Spotify was like, oh, there are so many people listening to N-3-4-5-6-7. Like, that doesn't make any sense at all.
Starting point is 00:11:20 But the argument that the U.S. attorney makes is, quote, Smith stole millions in royalties that should have been paid to musicians, songwriters, and other rights holders whose songs were, legitimately streamed. And I think that's like a really interesting argument. That is. Because like essentially what it's saying is, well, you're taking money out of the pockets of real musicians by creating fake musicians. It kind of is a precedent setting statement because I can very well imagine AI artists in the future and also artists using AI.
Starting point is 00:12:03 if they're not already, to create music. And so that becomes very complicated, very fast, if you're going to make a statement like, you know, this is illegitimate music because it's created by AI. And like, is it wrong? Was that wrong to create a set of robot musicians and be their manager? Or is he just using the system?
Starting point is 00:12:33 So I'm going to say something maybe a little bit controversial, but I kind of think it's the theme of this episode. I just, I don't know how much empathy we have for the tech companies these days. You know? Like, I just feel like, and I, you know, again, we don't know Mike Smith. We don't know what he's, we don't know the extent of what he's done. We don't know what he's going to say. we don't yet know as we are recording this, what his lawyer is going to say.
Starting point is 00:13:06 And look, you know, if you break the law, you break the law. But when I look at this story, I'm sort of like, I'm not sure who's getting screwed here other than the companies that refuse to pay musicians what I think musicians probably should be paid. Yeah, you could make an argument that the way in which the music business is conducted right now almost forces someone in a position where they really do devise such a scheme, right?
Starting point is 00:13:44 Like, this is an option that I have to actually make a decent amount of money making music. And of course it kind of turned into something else, right? because he starts making music using bots, and so, you know, it's conceivable that that sort of initial joy of creation of music was gone if it was ever there. But it's kind of you would expect it. You know, if you add in all the ingredients of this is how music has changed, this is how the music business has changed, this is the technology that we have, it actually kind of makes sense that this happened.
Starting point is 00:14:26 and probably, you know, will happen in the future in different variations. As per usual, you are wise and thoughtful on this topic, Dean. And now you got a story for me. You got another, was that wrong story for me? Which we'll get into. Yes. Right after the break. At Radio Lab, we love nothing more than nerding out about science.
Starting point is 00:15:15 neuroscience, chemistry. But we do also like to get into other kinds of stories. Stories about policing or politics. Country music. Hockey. Sex. Of bugs. Regardless of whether we're looking at science or not science, we bring a rigorous
Starting point is 00:15:31 curiosity to get you the answers. And hopefully make you see the world anew. Radio Lab. Adventures on the Edge of what we think we know. Wherever you get your podcast. There is something powerful about the sound of the human voice, beautifully produced audio has the unique power to connect and inspire. Tell your organization's story with a custom
Starting point is 00:15:52 podcast from City Space Productions, the Creative Studio from WBUR's business partnerships team. Become a thought leader. Recruit new talent, reach new audiences, whatever your goal, we can help. Discover how the magic is made at WBUR.org slash creative studio. All right, Ben. Have you ever gotten a fishing email? Yeah, absolutely. Often it's been like, hey, will you please look at this statement and then like a link to a PDF or like it's time for you to cancel your membership or be charged $420.69. Please click on this link. And the way, you know, kind of as you're describing, the way this typically works, you get an email, let's say from your utility provider. Like here's your bill. Click on the link. It's not your bill. But you. pay some rando, and now they have remote access to your computer and you're
Starting point is 00:16:57 out a thousand bucks, and yeah, that's how it goes. Mm-hmm. But one guy, he decided to go big. He thought, I'm not going to scam just anyone. I'm not going to scam Ben Brock
Starting point is 00:17:13 Johnson. I am going to scam Google. Okay. This starts around 2013. And I think it's best to kind of think of this as like a heist movie. So like if you can picture that, like, we'll start with like the ringleader, the like Danny Ocean character, the key perpetrator.
Starting point is 00:17:41 A Lithuanian man, his name is Avaldus Ramesowskis. Wow. That's a good name. I'm in already. I'm already in. To give you a visual on this guy, like one Reddeter put it, I think, really. well when they said he either looks like Benicio DiCaprio or Leonardo del Toro. He is like a mix of these two actors perfectly.
Starting point is 00:18:07 Okay, that's good. Yep, that's good. Ben, can you tell me, like, how does every heist movie begin? It's somebody picking somebody up as they get out of prison. Yep. There's like the assembly, I think broadly speaking, we could call it like the assembly of the crew. Exactly.
Starting point is 00:18:28 And that is what Rimasowskas first starts doing. Okay. Now, this crew that he puts together, they are, they're smart. This is not some guy in a basement sending emails, like fishing emails. They are all about research. You've got to know your mark. And the way that they do is the team starts calling Google's customer service. And not only Google, but they also decide, like, hey, we're also going to call Facebook.
Starting point is 00:19:00 Okay. The objective is to get as much information as they can about these companies. Key employees, contact information, organizational hierarchy. We're casing, in other words, we're casing the joint, I think, is the technical terminology here. Exactly, because none of this stuff is Googlable. So you got to call. You got to do your work. So Ramosauscus is 11.
Starting point is 00:19:28 You know, that's the crew. I'm going to call them that. I have no idea how many there are, but I'm just going to call them Rumsalskis's 11, which is really easy to say. While they are doing this research work, they learn about one company in Taiwan, named Quanta Computer.
Starting point is 00:19:48 So Quanta makes hardware, and two major clients, Quanta, Facebook, and Google. So the fraudsters think, like, hey, why don't we start a fake company? We'll set it up in Latvia and we'll call it Quanta Computer. If this is a movie, you know, it's a pretty long first act because, you know, all of this research, like I said, it takes years, it takes months, takes years. and after all of that legwork,
Starting point is 00:20:24 one of the crew makes two calls, simple calls, pretending to be the real Quanta computer. The caller tells each company to essentially update their bank information for an upcoming payment. And instead of sort of sending the money to Taiwan, Google and Facebook, they start sending their money to banks in Latvia and Cyprus. And then he also has to then launder all the funds to Slovakia and Lithuania and Hungary and Hong Kong.
Starting point is 00:21:01 And everything's going to plan. This is like the scene where they're like, yeah, it's all clicking. Everybody's doing their part. The TV screens have been switched to the static image instead of the live image so that no one will see them moving through the hallway. It's all coming up, Millhouse. They're sending emails with these fake invoices, fake contracts, fake signatures. These are all very well researched and in line with the real work that the real Quanta computer is doing. Except that the money is going to the fake Quanta to, you know, of Aldous and his 11.
Starting point is 00:21:52 I think what's interesting about this so far, and what you're shedding light on that's important is like most of the headlines about this story, I feel like as headlines often do, severely underrepresented the story. Like the headlines have basically been like, this guy said, I'm just going to send some bills to Google and Amazon and they paid them. Like that is like without reading the story, that is what you would assume would have happened. Right. And you are shedding light on the depth and complexity of this heist, which I think is very important. Because, again, most of the headlines are about this were like, they were like, they borderline on encouraging people to do this.
Starting point is 00:22:41 You know what I mean? I was just about to say, like, I wonder how many emails Google and Facebook got after this story came out. And it was just like, hi, Google. My name is Bob Smith. and you owe me $500. Thank you. Goodbye. Ted's balloon company is really swinging for the fences.
Starting point is 00:23:01 Oh, man. It is honestly, it is a brilliant plan. And it does. It pays big. By October of 2015. So this started in 2013. By October of 2015, the fraudsters have successfully grifted
Starting point is 00:23:22 $120 million. Wow. There is always, however, a third act and a twist. Yeah, that's true. Unfortunately for Rimasowskis and his 11, the twist is not very much not in their favor. Eventually. It's not a happy ending for the heisters.
Starting point is 00:23:46 Exactly. These giant tech corporations, one of which is one of the biggest email provider, in the world, eventually they catch on to these fishy emails. They alert the FBI, which, you know, swoops in. And they freeze Summer Ramosowskis' assets. In 2017, he's extradited to New York. Okay.
Starting point is 00:24:12 In March of 2019, the now 50-year-old Evaldus Rimmisalskis pleads guilty to one count of wire fraud, and he agreed to force. forfeit about $50 million. A few months later, he is sentenced to five years in prison. And that means if he's not out already, he's probably going to be out pretty soon, which, you know, we should talk to him. Actually, you know what we should do is we should start the next movie. I was...
Starting point is 00:24:46 By picking him up outside of prison. We need a sequel. We need a sequel. Which, you know, again, to be clear, Dean and I are not necessarily advocating. for people to do this. But it does to me, both of these stories to me represent a certain level of willingness on the part of people of the world
Starting point is 00:25:17 to not care too much about giant tech companies and their profits. Yeah, unless you're the FBI, then you care very much. Yeah. that's true. That is true. Or the tech companies, which presumably are, you know,
Starting point is 00:25:37 giving the tips to the FBI. But I think, like, in what heist movie, have you ever not rooted for the heasters? Right. And against the heistees. Yeah. And I just,
Starting point is 00:25:55 it's hard not to say, you know what? I don't blame them. I don't blame them. Yeah. But all you kids listening out there, do not do this. Please don't break the law. Definitely do not do this.
Starting point is 00:26:08 It is not an honest way to make money. Unrelatedly, Ben, I have a very good idea that we should talk about raising our listenership in Frenless Thread. Oh, perfect. I can't wait. I can't wait to talk about it offline. Let's meet in a coffee shop that's empty. Okay.
Starting point is 00:26:27 I'll send you some emails first. Okay. Perfect. Thanks, Dean. Today's episode was researched and co-hosted by myself, Ben Brock Johnson, and producer Dean Russell. It was sound designed without using AI by Emily Jenkowski. Sumitah Joshi edited us to continue our efforts for fan generation. The rest of our crew, who is waiting outside a prison with a bag full of clothes and a few bucks in their wallets is Paul Vicus, Grace Tatter, Amory Seabretson.
Starting point is 00:27:09 And if you want us to tell a story, hit us up at endless thread at wbUR.org. See you next week.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.