Loading summary
Transcript71 lines
- [00:01]
Rachel Toback
In today's world, data breaches happen all the time. And even the most secure companies can't always protect their employees personal information from ending up in the wrong hands. That's where DeleteMe comes in. DeleteMe is a service that removes your employees sensitive information from hundreds of data broker websites, sites where hackers can find phone numbers and emails within seconds. Rachel Toback, CEO of Social Proof Security.
- [00:27]
Jason Kebler
Says attackers use this data to target.
- [00:29]
Rachel Toback
Employees with phishing messages and AI powered phone scams. But Delete Me makes it harder for these bad actors by scrubbing your employees details regularly. It's simple. Attackers are lazy. If it's too hard to find contact info, they'll move on to easier targets. Deleteme takes care of this for you.
- [00:49]
Jason Kebler
Doing the heavy lifting so you don't have to.
- [00:51]
Rachel Toback
And over time, they keep removing the information so it stays down, protecting your team from constant exposure. If your business has a social presence or deals with clients, you need DeleteMe. Visit DeleteMe.com 404Media and start safeguarding your team's information today. That's DeleteMe.com 404Media.
- [01:18]
Jason Kebler
Hello and welcome to the 404Media podcast. Joseph is out this week, so this week I'm your host, Jason Kebler. With me are Sam Cole and Emmanuel Mayberg.
- [01:29]
Sam Cole
No room to say hi. Hi.
- [01:30]
Jason Kebler
No room to say hi. So a few things. Few things. Well, Emmanuel, do you want to say hi?
- [01:35]
Emmanuel Mayberg
Hello.
- [01:36]
Jason Kebler
Few things. I bought a teleprompter, so the eye contact during this is going to be insane. If you're watching on YouTube, I'm going to be on point. The thing is that I haven't written a lot of dialogue for me to say, but this thing is very cool. I really feel like I'm on MSNBC right now. We just need to get the scroll going. All right, let's hop into it. Let's just hop into it. So we are going to talk about OpenAI's new anime generator. That's not what it is, but we're going to talk about the Miyazaki AI. And the title of the story is Hayao Miyazaki, who said AI is an insult to life itself reduced to AI generated meme by OpenAI. Emmanuel, you wrote this story with Matthew Gault, who is going to be writing a little bit more for us coming up. Should we start with Miyazaki first?
- [02:33]
Emmanuel Mayberg
I was going to suggest if we could each establish our relationship with anime and Miyazaki individually. I feel like it's not necessary, but it's useful Information for our readers.
- [02:45]
Sam Cole
It feels like a trap. Establish your relationship with anime. Feels like a trap.
- [02:49]
Emmanuel Mayberg
Yeah.
- [02:49]
Jason Kebler
Total transparency. Yes.
- [02:52]
Emmanuel Mayberg
Personally, I am. I've had. I had an anime phase as a teen which included Princess Mononoke, which is my favorite. Miyazaki in general. I'm not a huge fan of anime in general, but his movies, Studio Ghibli's movies in particular. But I like Mononoke. That's my. That's my favorite one. Usually I just find his work to be like. Too, I don't know, magical realism. It's like very. It's very loosey goosey. Like there's not a lot of plot to follow. It's like vibes and amazing looking visuals, but not a lot of story I can latch onto. Sam, I feel like you're. You're more of an anime fan than I am.
- [03:33]
Sam Cole
I don't know. I mean, I'm. I had a phase also as a teen. I was really into Fruits Basket, which is a. An anime about boys who turn into cats. But yeah, I loved. I love Totoro, obviously. I was introduced to Miyazaki kind of late, like in college, and a boyfriend was like, you should watch Totoro. And I was like, we watched it together, so it has like a nostalgic vibe. I haven't watched a lot. I mean, I haven't watched a ton of Ghibli, but yeah, I'm a fan. I'm, you know, I would consider myself a fan for sure. I saw Mononoke in theaters and it was incredible.
- [04:11]
Emmanuel Mayberg
So, Jason, where do you stand?
- [04:14]
Jason Kebler
So we've established that you two are both anime fanatics. I have not seen anything by Miyazaki or Studio Ghibli. Nothing at all. I've not. I've not.
- [04:26]
Sam Cole
We should end the pod before we get canceled.
- [04:30]
Jason Kebler
I'm going to be the AI Copyright discussor on this podcast because. And. And also the stand in for people who have no idea what we're talking about, because I've not seen any anime other than like Pokemon.
- [04:47]
Joseph
Okay.
- [04:47]
Emmanuel Mayberg
That's anime. That's.
- [04:49]
Jason Kebler
Well, I know I like. Big fan of Pokemon. I watched like a very small amount of Yu Gi. Oh, I would say, but. And I saw a little bit of like Ranma 1/2, I believe it's called. No, not familiar.
- [05:06]
Sam Cole
I haven't heard of it.
- [05:07]
Jason Kebler
Wow, it's a deep cut. It's really not. It's very, very, very famous, considering that I've seen it, but I forget what it was about. In any case, let's talk about so what is, what is Studio Ghibli and who is Miyazaki like, for people like me who do not know, I know Spirit. I know that the, the name. I know that he made beloved movies. That's all I know.
- [05:34]
Emmanuel Mayberg
Yeah, it is the most celebrated and I think award winning anime studio and it is, I don't know if it's the most mainstream because you have things like Pokemon and Dragon Ball Z. But Studio Ghibli, which is led by Miyazaki, he's the co founder of the studio, has been around forever, has a very distinct visual style. And everybody knows about it because the movies have wide appeal. They've been nominated for Oscars when they're dubbed to English. Like big movie stars. Dub the. Dub the movies. And the reason we're talking about it is that last week OpenAI updated ChatGPT with a new AI image generator which immediately is, you know, it's a race to like, who has the best image generator. It feels like every week there's a new one that can do something really impressive. One thing that people figured out very quickly with this one and that made it go viral is that it is very good at copying the Studio Ghibli style, which is instantly recognizable if you've ever seen it. It's not quite like other anime that you've seen. It has like a warmer palette. It's a bit like softer angles when you think about, I don't know, Goku's spiky hair and things like that. And a style that people really love. And all you have to do is feed it an image and ask the model to recreate that image in the style of Studio Ghibli. And. And it does that very, very well. And that caught on a. Because understandably, it's like pretty novel and fun to feed it an image of your family or your friends and see what you would look like if you were part of a Studio Ghibli movie. But then also, I think people were having fun with it and testing the limits of what the model would do. And they found that it is way more permissive compared to previous OpenAI tools and OpenAI image generators. So they fed it like images of historical figures. Stalin. The top image in our article is an image that Matthew Gall generated, which is the moment that Saddam Hussein was captured in Iraq by the US and it will produce similarly kind of graphic or suggestive images. Napalm Girl, which is a famous image from the Vietnam War, is something that would recreate it in the Studio Ghibli. Style the Saigon execution. Another like very famous horrible image of a person being executed in Saigon during the war. It created a Studio Ghibli version of that as well. Not going to blew up all over the Internet. If you were on social media at all, I'm sure you've seen it.
- [08:46]
Jason Kebler
Yeah, I think that I guess some additional context. Studio Ghibli Animation. I learned this like reading articles about, about this. But like it's notoriously like painstaking work. It's like the artists who do it sometimes spend like a year on a single scene. Is that correct?
- [09:07]
Emmanuel Mayberg
Yeah, it's over the years. This is true in like Western animation studios, but also in Japanese animation studios, a lot of work has been outsourced or I mean, if you know Disney movies at all, it was like a big deal when Disney started doing 3D for some of like the background stuff, which saved them some of this labor. But yeah, animation is, you're doing I think what, 24 frames per second and you're hand drawing all of that if it's a Studio Ghibli movie. So they're very, very labor intensive works of art that many people work on. And the quote in the article comes from a documentary that came out in 2016 about Miyazaki and Studio Ghibli. And what happens there is that I think it's a few students that come in and they're demoing a 3D model that is animated with AI. So if you're making like a Toy Story, a Pixar movie, and you're moving a 3D model, you're kind of puppeteering that 3D model frame by frame to make it walk or do anything. And they created this AI tool that animates the character, which in this case I think was kind of a zombie that was crawling, laboring across the screen. And Miyazaki sees that to this group of excited students and he just shuts them down and says that it's an insult to life itself.
- [10:52]
Jason Kebler
And yeah, so. So here's the full quote. He says, quote, whoever creates this stuff has no idea what pain is whatsoever. I am utterly disgusted. If you really want to make creepy stuff, you can go ahead and do it. I would never wish to incorporate this technology into my work at all. I strongly feel that this is an insult to life itself.
- [11:12]
Emmanuel Mayberg
Yeah, Pretty strong condemnation of this demo that, that he got, which of course kind of circles back to the fact that almost 10 years later his signature style is reduced to an AI generated AI powered meme.
- [11:32]
Joseph
Yeah.
- [11:33]
Jason Kebler
And I think also notable, like you know, Sam Altman has shared images in this style. Sam Altman, you know, the CEO of OpenAI also changed his profile picture on X to a G version of himself. And I think that this has started. I mean, the conversation has been going on for a very long time, but it started yet again, the conversation about one AI generated art. But then also, you know, the theft of this art and the fact that this is clearly trained on, you know, tons and tons and tons of copyrighted work. And then, as you mentioned, it feels like the guardrails on this new tool are worse than ever. I mean, I've seen not just, you know, Miyazaki Studio, Ghibli AI versions of, you know, horrifying things, but I've also seen people just say, like, hey, generate me an image of the Pope. And it does. Generate me an image of LeBron James. And it does. That's not. That's something that in the past, OpenAI has attempted to prevent. And it suggests that the company is, like, really flouting this idea that it's going to win these lawsuits that are against it, that it is trying to take off as many guardrails as possible, I guess.
- [13:00]
Emmanuel Mayberg
Yeah. So a few things about that. One is that since we published this article, Sam Altman has said a bit more about what a big moment this was for the company. It wasn't just like a viral thing that was happening on social media. He said that the biggest week the company had was when they first launched ChatGPT3, and that was like 500,000 users joined in a day. And he said that a million people joined in the two days or a day after this update. So it's like a big.
- [13:35]
Jason Kebler
Well, like last night when we're recording this. We're recording this Tuesday morning. On Monday night, he tweeted that a million people joined in an hour.
- [13:44]
Emmanuel Mayberg
Yeah, right. Okay. Yeah. So that's even more impressive than I thought. It's like. It's like legitimately a big deal for their business. So in terms of the guardrails, I think there's two things that are happening. One is that OpenAI has taken a lot of flack and is also facing real lawsuits from artists who are mad that previously you could go to Dall E, which one of its AI image generators, and say, generate me an image in the style of. Fill in your specific artist. And they're getting sued over that. And what they said in response to our request for comment is that they now will prevent generation of images in the style of a specific artist, but they're okay with generating images in the style of a specific studio. So you could say, generate an image in the style of Pixar, generate an image in the style of Studio Ghibli, but not generate an image in the style of Samantha Cole's watercolors, you know, which is. I don't know. I don't know. Clearly, this is in response to the, to the lawsuits, but I don't know what the legal thinking is behind the scenes that they think this is somehow that's a lawsuit that they're less likely to lose. The other thing is that I think it was in 2023 that Sam wrote about how Dali refused. No, it was actually, yes, it was Dolly that refused to generate images of Julius Caesar. That is how strict the guardrails were back then. And those are just not there anymore. And Sam Altman has talked about this a little bit, that they want to have more open models, they want less restrictions. And I think, honestly, it's like, I don't know the thinking behind that either, but that seems to me a reaction to the competition and other models being more permissive. And then also just the fact that Trump is in the White House and they think that restrictions are like woke and left coded and they feel like they're not as likely to get in trouble for doing that kind of thing.
- [16:06]
Jason Kebler
Yeah, I mean, it feels to me like, you know, if you can't generate, you know, a fucked up image on ChatGPT, you'll just go somewhere else and do it. And Sam Altman might see this as like, oh, we're like losing users to Chinese image generators. We're losing users to like these startup image generators, so on and so forth, and they're just making a bet that they're not going to face some sort of enforcement here. And it's interesting we're going to talk about this in the subscriber section, but I did an article about brain rotation AI and how it's taking over Instagram, and a lot of these are like really graphic depictions of like, sexual situations between LeBron James and Steph Curry and Diddy and things like that. And it's like, I don't know, you can imagine what these things might be. And the way that those are being created is they're not being created on Chat GPT, but the people who are making it are having Chat GPT write the prompts that they can then go put into another image generator. And so it's interesting that the people who are like, quote unquote, doing this for a Living or doing this to make money, are using. They're like, they're using an interplay between a bunch of different image generators and video generators and, like, finding out where one guardrail ends and where one doesn't. And maybe Sam Altman is making the calculation that, like, fuck it, let's just let people do whatever they want on our platform. Trump is in office. FTC is not going to come after us. Who knows?
- [17:41]
Emmanuel Mayberg
Yeah. And I think we're also going to see there was some confusion about when OpenAI was allowing people to do this and when it wasn't. And the short answer to that is that some people were testing an older model of the image generator, which had more guardrails, and the newer one, the latest one, doesn't. And it makes me wonder if, because Open, Sorry, Sora is very guarded right now, but when they release the next Sora model, is that going to be more laxed? And then are you going to see the type of thing that we're going to talk about later? Just coming directly out of OpenAI tools without all these workarounds? Just like. Another thing that I really wanted to mention is that, as you said, animation is very labor intensive and Studio Ghibli's stuff is very labor intensive. And I suspect that they are not going to change that because that's part of the studio's legacy and they're very proud of how they do this work. But there's been a long history in animation of outsourcing a lot of the work. Do you know about, like, keyframe? Have you ever heard the term in betweener? Do you know about this?
- [18:57]
Jason Kebler
I know that technology has, like, come for animation over the years, and that, like, not every single cell is animated by the same person necessarily, or not every single cell is even animated by hand anymore.
- [19:16]
Emmanuel Mayberg
So it used to be like in the 90s that American studios and I think some Japanese studios, they would draw the key frames, as it's called. So it's like one frame every second or a couple of frames every second that are the key positions of the character, like the primary action. And you kind of send all these key frames to another studio. I believe Korea was a big country that did this at the time. And they, like a totally different studio would draw all the in between frames in the style that you gave them to kind of like tie all those key frames together. So that has happened for decades. And now some Japanese anime studios are outsourcing that exact same labor to AI tools, which to me seems like one of those applications of Generative AI, that makes sense. It's like very rote, difficult work that is already outsourced and you're just outsourcing it somewhere else now. So I can imagine very easily that Generative AI has a role in animation in the future. But the thing that I think people found very annoying, and I found very annoying, is that, like, the AI booster crowd conflates the ability to replicate the Studio Ghibli style with the people who are using this tool somehow being the equivalent of a Miyazaki, which is obviously not the case and obviously missing the point. And I don't know, this is a thing that has happened with Generative AI the entire time we've been covering it. But, yeah, it's like, there's a tool, there's probably a practical application of it in the future. But the AI booster crowd has so much glee about the possibility of putting artists out of work or somehow devaluing their work. Like, there was this meme that was going around like, oh, it's over for illustrators, it's game over for animators, which I'm sure there will be an impact. But that is not what the tool is. The tool is a proof of concept and a fun fad. But replicating the style of Studio Ghibli does not make you a master artist like a Miyazaki.
- [21:52]
Jason Kebler
Yeah, I want to hear from Sam in a minute because she's done the most sort of reporting on the fact that all of these things are trained on, you know, tons and tons and tons of, like, you know, cartoons, videos, YouTube videos, real photos, et cetera. Like, this is all trained on the work of real artists. But one thing that I saw going repeatedly viral would be like, VC or Silicon Valley worker MIA's occupying an image that they took of, like, themselves with them, with their bros and being like, yo, look, I turned. I turned this into art. And I thought that was really interesting and smart point you just made because it's like taking a picture of five of your co founders on a couch and turning it into a Studio Ghibli anime is not turning it into art. Like, that's not art. Uh, it's not animation. It's. There's no story there. There's no, like, taking an. An already existing image and just like, running it through a filter is something that we have been able to do for an extremely long time. And that is not creating stuff that people are, like, willing to pay for or consume. And I think that that's very stupid. Um, that said, if I were an Illustrator, I would probably be pretty concerned because there, there clearly is going to be some impact, and it already has been some impact on like graphic designers and artists. You know, we've seen a lot of like corporate clients, like people who used to do a lot of high paying design work for I don't like PowerPoints and stuff like that, like, stop doing that and just replace it with AI because they're doing things that don't matter that much or maybe aren't public facing or maybe are intended for clients who don't have like a particularly discerning eye. And so I know that a lot of graphic designers have seen their, you know, illustration, like clients kind of dry up, which is, is really concerning. Sam, we don't know exactly what this was trained on, of course, but you've done a lot of reporting on other companies, you know, training on just like tons of YouTube channels, things like that. I mean, what do you. I think it's pretty obvious what is going on here, but is there anything new for us to learn or is it same old, same old?
- [24:23]
Sam Cole
I mean, yeah, it's pretty obvious what's happening here. And I think just like what I keep seeing and what people are as a direction I think is new compared to like what you said with the filters that are art that have been happening for like five years almost, where it's like, oh, I turned my face into stained glass before we even called it like generative AI or whatever. It was just like filters people have. People who are into the Ghibli stuff as like a personality almost seems to be a thing. And it's not that they're like fans of the studio even. It's that they really gleefully hate artists. And I think that there's something going on there that maybe a lot of people wouldn't even admit. It's like, oh, I don't hate artists. I love art. Look at how much I love art. I'm imitating it. But there is just kind of like this real disdain for the process of making art itself that seems to be happening. And there's like people who super hate generative AI and there's people who super hate the concept of artists continuing to make money. And that there are people that's. Those are extremes. And there, there's few people on those sides and there's more people in the middle who are kind of like, oh, this is fun, whatever, I'm gonna put it on my Instagram. But that rift is getting wider and like people are more and more fighting over those concepts. As people actually lose their jobs over it. You know, it's like people are losing their jobs left and right in every industry. So you know, know, of course AI is impacting graphic design as well as like programming and coding. I don't know, it's just like it's. I got, I was really pissed off at the Studio Ghibli stuff when it was happening like Thursday, Friday, because of that attitude that I kept seeing where people were like, haha, like they were generating like that scene from that documentary of Miyazaki being like, this is an affront to like life itself or where whatever he says. And just like, I was like, okay, Sam Altman is spitting in the face of like artists clearly and blatantly, and people think it's funny. And then I read Emmanuel's behind the Blog and he was like, this is a trend, this is like a stupid trend that people do all the time with technology and with these kind of consumer tools. And that calmed me down a lot. I was just like, okay, you know, people are, people have been doing this forever. But that cultural rift I think is real and like worrisome. And I try not to contribute to it by getting super worked up, but like it does piss me off to see people with that kind of attitude. I don't know.
- [27:15]
Jason Kebler
Yeah, I mean, I guess the last thing I'll say is I'm very interested in the fact that they said that they gained a million users in an hour Yesterday because when ChatGPT launched, they did gain a lot of users very quickly, but a lot of those users fell off very fast. And of course, like ChatGPT, I know people who use it and rely on it like for all sorts of stuff. Not saying it's good, I just know that it's become a daily habit of people when they want to look up something or, I don't know, rewrite something. And I am curious if this is a million users who are like, haha, I'm making one anime image and then never use it again, or if these people are then going to use it to write emails and become subscribers and so on and so forth, surely some amount of people will stick around. But I don't know, it still feels very flash in the pan to me. If you've come there to make a specific meme or to run an image from your vacation through this and send it to the group chat, like, I don't know. I, I'm very curious. I, I know that OpenAI just raised like 40 billion more dollars yesterday, but the staying power of something like this. Like, I. I really don't. I really don't know. All right, let's leave that one there. When we come back, we are going to talk about only fans.
- [28:59]
Joseph
Big Tech Social media algorithms have made finding and consuming the news exhausting. It's tough scrolling through timelines full of content that have been reverse engineered to perform well in an algorithm have risen to the top of your feed with the help of targeted advertising and content that's just generally designed to keep us in the dark. I recently discovered Ground News, a news platform that doesn't use manipulative algorithms and which lets me quickly see how news outlets around the world are covering a story. For example, I've been really interested in the impact tariffs might have on the economy and my day to day existence. A recent page on ground news aggregated 100 different news sources about tariffs on Canada and allowed me to easily filter through American and Canadian perspectives on the tariffs. I could also find out what people on the left, right and center were saying about it. From the Associated Press A Senate vote to reverse Trump's tariffs on Canada is testing Republican support from Canada's the Globe and Mail. Trump's proposed tariffs likely to cause shortages of Canadian made drugs in the U.S. no matter your political leanings, I always feel like I want to know what the discourse is about any topic I care about. Ground News lets me do that quickly without the algorithms. Go to groundnews.commedia to get 40% off the ground News Vantage Plan and get access to all of their news analysis features. That's groundnews.com media for 40% off the ground News Vantage Plan for a limited time only. GroundNews.com Media One of the smartest apps you can have on your phone in 2025 is upside. Upside gives you the power to earn more on the stuff you buy all the time. Whether you're filling up your tank, going out to eat, or shopping for groceries, the Upside app lets you earn cash back on every purchase. Upside users are making as much as $280 a year. I used the Upside app the other day to go to the gas station I always go to and I got 25 cents back on my first tank of gas.
- [31:06]
Jason Kebler
That's what I like best about Upside.
- [31:08]
Joseph
I'm able to get cash back for shopping at the places I already go to. To get started, download the free Upside app, use my promo code 404 and get an extra 25 cents back for every gallon on your first tank of gas. After that, use upside at any one of more than 100,000 gas stations, grocery stores and restaurants to earn an average of 8% back on restaurant and grocery purchases.
- [31:33]
Jason Kebler
The best part?
- [31:34]
Joseph
These stack on top of whatever credit or debit card rewards you might already be earning. Start getting hundreds of dollars back per year. Download the free Upside app and use promo code 404 to get an extra $0.25 back for every gallon on your.
- [31:48]
Jason Kebler
First tank of gas.
- [31:50]
Joseph
That's an extra $0.25 BACK for every gallon on your first tank OF gas. By downloading the Upside app and using code 404.
- [31:59]
Unknown
Hey, it's Joseph again. If you're a new listener to the 404 Media podcast or even a long time one, you might not be aware of all of the impact our journalism has had recently or how we even got here in the first place. In 2023, the four of us quit corporate media to go independent. We were sick of working for a VC backed company that put profits before journalism that gave birth to 404 Media. Since then, we've stopped the spread of AI books and public libraries, triggered class action lawsuits against AI companies, got Congress to pressure big tech in various ways, and we've even shut down surveillance companies. This real world impact is only possible because of our paying subscribers. As a journalist owned business, they are the engine that powers our journalism and where the vast, vast majority of our revenue comes from. So please consider signing up today for $10 a month or $100 a year at 404 Media Co membership and get bonus content every week and access to all of our articles. Thank you and enjoy the rest of the podcast.
- [33:22]
Jason Kebler
Okay, we are back. Sam, this is yours and it was published today. This morning. OnlyFans sued after two guys realized they might not actually be talking to models. First of all, ha ha ha ha ha. Lol. What's. What's going on here?
- [33:42]
Sam Cole
Yeah, I'm gonna try to talk about the story without like litigating the case myself because it is pretty absurd. But it also touches on something kind of serious. So these two men in Illinois, both of them say that they started suspecting that the onlyfans models that they were subscribed to who have like hundreds of thousands of fans in some cases subscribers, and they were one of you know, 700,000 fans of one particular model. They say that they started getting suspicious that they weren't actually speaking directly to her personally, that she wasn't texting back to them herself. And it's a couple of models are mentioned in the lawsuit but they the basis of the lawsuit is that OnlyFans is allowing models to use chatter agencies. So the agencies pay a group of people to manage someone's OnlyFans accounts and reply to the probably a million messages they get every day directly to fans. So you're not talking directly to the model. You're talking to someone that she employs. And this is a huge industry. There are so many. There are probably hundreds of agencies at this point. But it's a really common practice. It's been talked about for years. I think some of the really big models use it and use different agencies to represent them, to help them kind of schedule out their content, to do lots of different tasks other than just dming back their fans. But that is part of what some of the agencies offer is we'll talk to your fans for you so that you can kind of focus on making the content or whatever it is. But it was, you know, it kind of blew open a couple years ago with the New York Times story, and then like Cosmo wrote a story and Viceroy a story and Washington Post wrote a story. It was like everyone wrote a story about agencies around 20, 22, 2021, 2223, where it was like, I went undercover as a chatter for an OnlyFans bottle. And this is what I saw. That kind of thing.
- [36:04]
Jason Kebler
Yeah. So, like, broadly speaking, like, how it works, as I understand this, is based on reading some of those articles. And then also I have, like, gotten some employment handbooks from some of these agencies just while clicking around on discord and things like that. Because it's a. It's a job. It's like a side hustle that some people do, but basically, like, a model will take photos and videos, et cetera, and post them herself. And sometimes they're. They're managed by an agency. Meaning, like, they sort of have a boss. Is this. Is this correct? Before I keep going. Like, sometimes they do it on their own, and sometimes they have like a manager, a boss, something like that. And then they post their images to OnlyFans. And then, you know, the people who subscribe to them will start messaging them. And then, you know, like essentially an army of chatters of people who just like, their only job is to talk to people on only fans pretending to be these models will have either scripts or guidelines for what they are supposed to say. And if. If they're like, sexting with a client, they are given, like, in some cases, images or videos that they can then share, like a separate cache of photos that they can use to keep the conversation going. And then sometimes AI is Involved to some extent. I don't know if on OnlyFans that is the case, but I know that on some other only fans, competitors, like you can outsource all of this to AI. Like, I know a lot of models have created like telegram bots where you can pay to talk to someone on Telegram, and that is almost fully hands off just for people who are not familiar with this world. That's vaguely how it works, correct?
- [38:03]
Sam Cole
Yeah, pretty much, yeah. And this, obviously, if you're subscribing and you're thinking, oh, I'm subscribing to this woman who wants to talk to her fans. Well, I'm a fan, she must want to talk to me. And then, you know, you start to kind of see inconsistencies, which is what the lawsuit claims, inconsistencies in messages. Or you're like, oh, I wonder, like, how is she making so much revenue? She has so many, like hundreds of fans, but she's talking to each one individually. That can't be possible. You know, putting two and two together, I guess it kind of comes to light. It's like, oh, maybe this is someone who's using one of these agencies that is very widely publicized and documented. And maybe I am talking to some rando in their own house, like one of hundreds of chatters for this one model. So that's what the lawsuit is bringing OnlyFans task for. And there a lot of the lawsuit, it's like, I don't know, it's like 60 something pages, it's really long. But a lot of it is kind of building on this idea that OnlyFans is marketing authenticity. Authenticity and direct connection and safety and moderation and all this stuff. And it's basically saying OnlyFans is defrauding users by breaking that promise of authenticity and not being upfront about not being upfront and also just letting people use agencies to begin with through the platform. So they're not even suing individual models, they're suing OnlyFans as a platform itself.
- [39:40]
Jason Kebler
Yeah. Two extremely funny lines from this lawsuit. Plaintiff created an account primarily in order to engage in friendly conversations with models and share photographs of his cooking creations.
- [39:54]
Sam Cole
Yeah, for sure, bro.
- [39:56]
Jason Kebler
Very, very funny. And then the other point is that they basically became suspicious because they saw one of the models had 700,000 fans and saying, like, how could they talk to all these people? And something that like, really stood out to me is that there's no, like, smoking gun in this lawsuit. Like, they don't say, we are sure that they were using chatters because, you know, I spoke to one of the chatters and they said that they were not actually the, the model. It's just that they saw that they had a lot of followers, a lot of fans, and they're like, well, it would be impossible. And I, I don't know, I just don't think that that is probably a. The basis of a select of a successful lawsuit. Like, generally I would think you would need more evidence of this. And it's also like, not clear that again, this is like a really well known industry. This is not something that it's like, oh yeah.
- [40:58]
Sam Cole
Which they say in the lawsuit. They're like, onlyfans fully knows that this is occurring. And what they use as examples is like years of reporting about the agencies in these major publications. It's like, onlyfans knows. Onlyfans knows. But why did the guys not know? Yeah, you know, it's like, does it not go both ways? I don't know. I mean, that's obviously, that's for the court to figure out. And I think they're probably banking on this going to trial as a class action because they're seeking national level and also people specifically situated in Illinois. But they're hoping that it shakes out in court, I'm sure, where it's like there's some kind of discovery process where it's shown that they have to then go through every claim that every person in the class action is making against a specific model, which would be a fucking nightmare. Just like logistically. But you know, it's. There was a trial, just went to a trial, just went to cases, went to trial, or it was set to go to trial in two years from now. It's like they decided that it would go to trial eventually.
- [42:10]
Jason Kebler
Last month I saw that in your article. And that's really crazy. It's crazy that they're like, this is going to this lawsuit from 2024, will go to trial in 2027. That's a really long time just to get this thing started.
- [42:24]
Sam Cole
Yeah.
- [42:25]
Jason Kebler
Sam, just to be clear, like your $12 a month does not get you unlimited attention from models. Personal, like reviews of your cooking creations and things like this. Like, this is not part of the, the, the deal.
- [42:41]
Sam Cole
To be fair to our cooking enthusiasts in this particular story, a lot of the bios and like the advertising that the models use is like, talk to me. You know, it's like talk directly to me about your day. It's like that is kind of the pitch in a lot of ways is like you get to Talk to me. And I think a lot of models do it that way actually. Like, they do actually put in the time and the work to talk directly to their fans. They do use automated systems a lot where it's like they'll send. If you subscribe to some, some particular tier, you'll get an automatic like short video clip about her day or something. But like, a lot of them do have levels where it's like you can talk directly to them, almost like a sexting kind of situation. But there are platforms for that also. It's like you can go use like Sex Panther or something and like talk directly to a model. You know, it's like, I mean, maybe those platforms are the same problem, but that is the pitch in a lot of places. And a lot of models do use that kind of language. But you know, it's, I don't know, like, like you said, it's like it's not built out in the lawsuit or in the complaint whether or not they actually were found to be, you know, caught red handed using these agencies. And a lot of people do use them, so they're not illegal to use.
- [44:04]
Emmanuel Mayberg
Can you imagine if, regardless of how the lawsuit shakes out, that only fans starts requiring creators on the platform to disclose whether they use a service like this or not? I can imagine them doing that just to like, not, not have this be an issue. And I don't necessarily have a problem with it. I think it's pretty crazy to bring a lawsuit forward about this and to not understand that a model with 700,000 followers would obviously not be texting you. But I also don't see a problem with just like resolving this. You know, there, there are, there are dumbo warning labels and disclosures on the Internet and other products, you know.
- [44:53]
Sam Cole
Yeah, I could see that for sure. I could also see them limiting logins from various locations, which is something that the complaint mentions that OnlyFans does send you like a, an email that says new login detected. Like every other platform, it's like new login detected at this location. So OnlyFans knows where the logins are coming from or how many people are logged into an account at one time. So, you know, do they know that people are logged in to one specific models account at the same time? If it's like 50 people logged in at the same time, maybe that's something that triggers some kind of, I don't know, like warning for the model or something. Or maybe they're using platforms that don't even, you know, like schedulers that you use for social media. It's like maybe they're not even required to be logged in directly. So.
- [45:43]
Jason Kebler
Yeah, I mean I was going to say I, I could see a world where that happens. I don't, I don't necessarily think that is fair to the models depending because it's like, I don't know, we schedule some of our social media posts to go out. It's like do you need to know if I like click the button at that, at that moment or if someone else is clicking the button or what? And I know that there is a difference between like we scheduled a Blue sky post to go out in 15 minutes and it went out and there wasn't a human being clicking that button at that time versus like a back and forth text exchange or chat exchange. But you know, I could see that having like quite a, quite an impact on models finances of course. And I bet that there's like, there's like layers of gray to it I guess where there, there's like fully automated, semi automated. Like where, where is the line? I guess would probably be somewhat tricky.
- [46:53]
Sam Cole
Yeah, I would think some kind of like disclaimer like Emmanuel said would be a middle ground. You know, it's like if we were, if our Blue sky bio said we're skating directly at you personally, every 15 minutes we push the button. I don't know, it's like maybe we should be pushing the buttons, but it doesn't say that. So maybe some kind of like I like OnlyFans is so fucking full of like restrictions and like rules and stupid arbitrary shit because of the credit card companies at this point. It's like you hate to see another goddamn lawsuit that could end up more rules and more regulations and more things that would push models off the platform and onto like different platforms and those platforms get attacked. Do you hate to say there would.
- [47:42]
Emmanuel Mayberg
Be false positives and stuff? It would just be another thing that goes wrong? Yeah, yeah.
- [47:46]
Sam Cole
And once that precedent is set, you know, it's like then the floodgates are open. People can be, you know, suing for this all over the place. I don't know, it's just like it's such a. I don't know. Makes me nervous to see this kind of thing go to the courts.
- [48:02]
Jason Kebler
But yeah, yeah, there it is. Well, I mean this is definitely like local, local men upset they have been swindled lawsuit which is a, it's a good topic for us. I, I like this sort of thing. I like when this happens, reporting wise. But I do think that it's a good chance to talk about some of the economics behind only fans and some of the technologies and platforms that are used here. Anything else on this one I'm going.
- [48:35]
Emmanuel Mayberg
To sue about there not being actual hot singles in my area.
- [48:40]
Jason Kebler
Dude, that's really good. That's really good. Okay, we'll leave it there for now. If you are a paying subscriber to 404 Media, in the bonus section we're going to talk about where Brainrot AI comes from, which is super viral AI on Instagram. You can listen to that segment if you subscribe to us by going to 404 Media Catch and clicking subscribe, you'll get a super personalized sent directly by us RSS feed where you can get that content. Otherwise, we will see you next week.