B (4:01)

Yeah, just another way in which some large economic and geopolitical decisions can impact four journalists trying to sell T shirts with the doom code on them basically. I mean it's impacted untold number of businesses in all these different ways and it's impacted us in a fairly straight forward way probably like hey, price went up, you know, but hopefully people, you know, go check out the merch refresh. I'm definitely glad that the doom one is now on a long sleeve shirt. There was no way I was ever going to wear the tank top. Although I know Jason rocks it.

A (4:44)

Don't think Sam. Sam's a big fan.

C (4:46)

We're brains Suns out, guns out.

B (4:49)

Not for me. Uh, so yeah, I'm glad the new stuff is in. And with that let's talk about this week's stories. The first section, lob out Zora 2. I don't even know if I did that.

A (5:03)

Correct. That sounds like a Z Sora, like a Kingdom Hearts character.

B (5:09)

I think that's why my brain refuses to pronounce it because I'm not, I'm not entertaining the idea of Kingdom Hearts. I find it interesting from ip like I'm not, I don't want anything to do with that.

A (5:22)

Okay.

B (5:22)

It's about Sora OpenAI's new AI slop machine app video model. And the first one is OpenAI's Sora II copyright infringement machine features Nazi Spongebobs and criminal Pikachus. First of all, Jason, what is the new app exactly? Because it's not just a new video model, which we see every six months or whatever, right?

A (5:51)

Yeah, I mean, it's OpenAI's attempt to make a TikTok competitor. So it's a social media platform, or at least it's trying to be. And that's kind of the new thing that AI companies have been trying to crack. The new problem they've been trying to solve is how to make this stuff social. We've seen META try to put it into its products, but. But Mark Zuckerberg has talked about trying to, like, make its own AI focused social media app. And so this is OpenAI's attempt at that. So basically, you download the Sora app, you scan your face. Like, that's. You can't really use it unless you scan your face, and that's like the first thing that you do.

B (6:36)

So do you literally have to do it or you can just scroll if you don't?

A (6:41)

Uh, I'm not sure, actually, but it's the first thing that you're prompted to do. And you can't, like, do much of anything on it if you don't scan your face. It. I should have. I should have checked. I just did it because I was trying to figure out, like, how this worked and so.

B (6:57)

But it's shoved in front of you.

A (6:58)

Yeah, yeah. So you open it up and like, the camera turns on, and then you do a selfie and you, like, take a video and you hold it to your face for a few seconds and then it asks you to look left, look right, look, look up. This takes like maybe 10 seconds. And then it asks you to say three numbers. So it'll be like 42, 87, 91. And that's it. Like, that's. You give it your voice by just saying that. And I bring that up because maybe like four or five years ago, on an episode of Cyber, which is the old podcast we did at Vice, we did a deep fakes audio episode where we cloned the voice of Ben McCoo, who was our host at the time. And it. We had to, like, find bespoke software engineers to figure out how to do this. Like, they. They had created a model that would, like, clone his voice. It was pretty sophisticated. It was running on their, like, own server. And Ben had to read maybe like 20 minutes worth of audio, which is a lot of audio.

B (8:05)

It was a whole thing.

A (8:06)

And at the time they're like, we've gotten this down from, like, this is state of the art, like, you only have to read 20 minutes worth of audio versus like hours and hours and hours worth of audio. And it didn't sound even that good. Like it, it, it did sound like Ben, but it was like a days long process. He had to do all this stuff and now to make pretty convincing video with synced audio that sounds like your voice. It's like a 17 second process. And Sam, I feel like you have done like quite a lot of reporting on like how that tech has changed. Like, I don't know, I feel like you've like cloned yourself various times.

C (8:46)

I mean, when Deepfakes first came out in I guess late 2017, I tried to make one of myself. And at the time it was like, I don't know, you needed to, you needed to have like pretty extensive programming knowledge to even just set up the adversarial networks that you needed to make a deep fake visually. And I gave up. I mean I was like, I don't have the, I don't have the patience or even the knowledge to like follow a tutorial at the time. And now it's like, you know, we can do it on an app and it's literally one photo or in some cases not even a photo if the person is like well known or a public figure. You just use text. But yeah, that's the case with all of every format of this technology is so much better than it used to be to the point where you just need like you said, a couple seconds worth of audio. It's also interesting to me because I had to go through this recently with my bank with doing the voice verification stuff to make a voice print. And it's basically the same exact thing. It's like you read like eight words or something or numbers and you instantly are like, I know your voice now. It's like, this doesn't seem right. It seems like you should need more than that.

A (10:00)

It seems like you should need more.

B (10:02)

Yeah, well. And with bank accounts specifically, I used the service 11 labs a few years ago now and to clone my voice with that, I had to read out, I think, a few sentences and it was weird. It was like some romance novel if I'm remembering correctly. And it got like a bit steamy. I'm like, why the fuck am I reading this? And then I did that, cloned my voice. And then I used that cloned version of my voice to break into my bank account which was protected with voice id. So it's a big mess. And as you say, Jason, it's crazy easy now to do It, Yeah, and.

A (10:35)

I bring this up and we dwell on this because like, that is the big advancement from Sora, in my opinion, is like there have been other like new defy apps and AI apps where it's like, oh, you put in a picture and then it makes like a crude version of it. And those have been very damaging and we've like written about them, but there hasn't really been a video model, to my knowledge, where you can deep fake yourself or, you know, for lack of a better term, you can like clone yourself in like three seconds. And that is like how the app works is basically like you scan yourself, it's connected to your profile, and then you can use text like in any other AI app to say, like, Jason writing an article in a cyberpunk future. And like, it will make a video of that and it will have my voice and it will. You can like write the script like, blah, blah, blah. And to be like, very honest, it is pretty damn good. Like, it is shockingly good. I think that there's still in many of the videos, like quite a tell where it's like, oh, this is AI generated but with things like animation, which we'll get into in a minute. Like you can tell it to make an episode of spongebob and it will look like the actual animation and the videos like look pretty realistic. The audio sync pretty well. The voice sounds like my voice. It's like quite concerning. And then you can, it's called Cameo. You can like put anyone else who has the app into your videos and by just like tagging them and like generating a video with them. So basically, like people have been making videos of Sam Altman, the CEO of OpenAI, by tagging him and then making him say and do like crazy shit. And notably, and I think to like OpenAI's credit, this is like an opt in system. And so you can be on the app and say, don't let anyone make videos of me. You can also say, only let my friends make videos of me. And you can also say you have to like manually approve any video that has you in it before it is published publicly.

B (12:52)

I wonder if does Sam Altman, is he doing that to all of the ones in him in Nazi uniforms and.

A (12:57)

Shit, just taking his notifications? No, no, I think you can, I think you can make it just like an anything goes situation. But like, there is a setting where it's like I have to manually prove each and every video that's made of me. Although, like, I actually haven't tested it enough to Know, because you can generate drafts that live on, like, the person's phone, and then that person can download the video, which I think is actually one of the, you know, concerning things about Sora. And what we're going to see and already are seeing is people using Sora to generate video, downloading it and then posting it elsewhere. Because, like, if you stay in the Sora app, which, again, it's just like a TikTok clip, so everything on the app is, like vertical videos that you can scroll endlessly, and they're all AI generated. It's like, if you're in that app, you know it's fake because it's like an AI app. So, like, on that front, it's actually slightly better than, like, my Instagram feed, where my Instagram feed is full of AI shit.

B (13:59)

But, like, which isn't disclaiming either.

A (14:01)

Yeah. Whereas, like, everything on Sora is AI generated. But, like, Sora is so good at generating AI video that you can just, like, generate the video, download it, and post it elsewhere. And, like, that's probably advantageous for the people doing this because the other platforms are monetized, as we've, like, talked about endlessly.

B (14:25)

Yeah. And we'll get to more about spreading these videos and sort of misleading people. But as well as uploading your own face, you mentioned there's spongebob, and of course, the headline mentions Criminal Pikachus as well. Well, when you first open the app and you start scrolling, what are some of these examples of stuff you're seeing? Because the article starts with basically a very long list to give readers an idea of what people are seeing.

A (14:50)

Yeah. So I got this app the day that it came out. And the way that I got it is I bought an invite code on ebay because I was not invited by OpenAI, which we also covered, to be.

B (15:04)

Clear, because that's interesting in its own right.

A (15:06)

Yeah, yeah. And it was full of just, like, copyrighted stuff. And so, like, I think the first video I saw was spongebob, you know, saying, like, I'm on the Sora app. Oh, my God. And, like, talking to Patrick, and it sounds like spongebob. It sounds like Patrick. It sounds like Squidward. Whatever. Then I saw there's so much Pikachu. There's like, tons and tons and tons of Pikachu. I actually saw a video of Sam Altman grilling a dead Pikachu on a charcoal grill and him going like. And, like, cutting into Pikachu and being like, I'm gonna get sued for this. Which I thought was actually pretty funny. There's like tons and tons and tons, like Mickey Mouse, you know, Bart Simpson, Peter Griffin from Family Guy. Like any copyrighted character you could possibly imagine was on there and people were doing all sorts of shit with it. Like, you know, there was crossover stuff. There was a lot of people putting them into like Minecraft. There was like Fortnite, but it has, you know, Peter Griffin shooting Stewie or whatever. And that was like the vast majority of the videos that I saw. It was like a mix of that. A mix of like Sam Altman videos and then like random people, you know, posting themselves, just trying the app out. But like, the stuff that was going viral that was being delivered to me by the algorithm was primarily copyrighted content.

B (16:36)

Yeah. And you write in the piece, quote, with the release of Sora 2, it is mad. It is maddening to remember all of the completely insane copyright lawsuits I've written about over the years. Some successful, some thrown out, some settled. In which powerful companies like Nintendo, Disney and Viacom sued powerless people who were often their own fans for minor infractions or for use of copyrighted characters, there would almost certainly be fair use. That was a long quote. Sorry, but what do you mean by that? Because you're seeing all of this wave, this flood of copyrighted material in this app. Clearly OpenAI has scraped these characters and this material from the web in some fashion. And then you have companies like Nintendo who are usually crushing random ass fans for making stuff about their games. What do you think about that disconnect?

A (17:26)

Yeah, I mean, I think that there's a huge disconnect and with a caveat that we'll talk about in a second of what has happened over the last week. But basically, Nintendo is extremely litigious. They go after people who pirate their games, which makes sense, but then they also go after people who mod their games or emulate their games. I wrote about one time Pokemon company and Nintendo suing this fan who threw an unofficial Pokemon party at a game dev conference. Like not even affiliated with the game dev conference. It wasn't like it was at this big event. It was just like at some satellite bar. And they sued him because they put a Pokemon on the poster of that tiny, tiny event. And so then to see, just like this, anything goes, Extremely blatant copyright infringement by a company that's worth billions and billions of dollars and has all this funding is kind of insane to see, like how far we've come. We've talked about the copyright issues with AI before, so we won't get into it too much here because a lot of it is like unsettled law as to whether this is fair use. But like it was only a year and a half ago or so that most of these big AI companies, like, wouldn't even say whether they were training on copyrighted content, even though it was obvious that they were. And now it's like you can have a picture perfect episode of the Simpsons with all of the voices synced up. It's like, obviously they were training on all of this. And so it's kind of insane. And to be clear, there have been a couple high profile lawsuits against AI companies. Like Disney is suing Mid Journey. But you would think like when you have like Mario, like when you have Sam Altman, like killing Pikachu, you'd think like, hey, maybe they would be mad about this.

B (19:22)

Yeah. And there's still time obviously for like lawsuits to happen and stuff, but companies can opt out of their IP being used in the app. And as far as I know, companies have to do that on like a per character basis. Like Disney or Nintendo can't go and say, take all of our stuff out, please. They have to go, could you take out Pikachu, please? This is what Pikachu looks like. And I mean, I remember way back when Pokemon was what, 151Pokemon? How many is it now?

A (19:55)

I mean, I think it's up near a thousand, if not over a thousand. Yeah.

B (19:59)

So Nintendo's gonna have to send a thousand letters and open a thousand different cases about Pokemon that I know, I don't know how many people care about the latter. A series of Pokemon. That's my bias showing. But Disney's gonna have to do the same for that. And maybe that will please these companies. But yeah, sure, a lawsuit could happen. And with that said, because now the app's out and companies have had a minute, a few days to opt out certain characters. Have you seen any change on the app? Like, oh, actually I can't make spongebob Nazi anymore. Have you noticed anything?

A (20:37)

Yeah, yeah. So actually, yeah, I didn't mention, but like there's, there's, there are videos of spongebob being a Nazi that I saw like right after I opened the app. And then like one thing you can do in Sora is you can then tweak things that have already been published. And so people were like, now make it Pikachu. Now make it Goku. Now make it Rick and Morty now. And so there was just like an endless scroll of like Nazi ip, like with swastikas in the background. Like it was crazy. It was crazy. But anyways, over the last week a lot of the characters have been opted out by the companies that own them. And that is now one of the main things that people on Sora are making videos about and talking about where basically like Sora is no fun anymore because everything I try to do is a content violation. And so a lot of the videos are about content violations and getting like Sam Altman to be like, say things that are like, you can't have fun on my app anymore. Which is like, the app only came out six days ago. And so I mean that, that raises questions about whether this app will have like any sort of staying power whatsoever. We've seen it time and time again. Like big, fancy new social media app is released. There's like an invite system. So it's hard to get on.

B (22:02)

So it seems really that sort of thing.

A (22:04)

Yeah, it like seems really exclusive and then it's like, oh, we got an invite, like, got to get on this app. And then it's popular for a few days or a few weeks and then it kind of peters out and it's like, I don't know if this will have any sort of staying power. Like obviously ChatGPT has, but not everything that OpenAI has made has been like some massive hit. And so yeah, it's like a lot of the characters have been opted out and people are now like complaining about it. And let me just say that the opt out system is like a system that OpenAI devised. It's like that's not like a legal framework. I mean very, very easily. Like a company could be like, no, take down everything or we're going to sue you. And I don't know how that would go, but it's worth saying that this is like in interesting new like paradigm that they're trying where it's like, we're going to put you in the app unless you specifically tell us you're not going to be in there.

B (22:59)

That's not how copyright works. It's literally the opposite of how it's supposed to work. Yeah, and I mean, I guess maybe to wrap it up before we go to the watermarking thing, just with all of that IP now being removed by the copyright holders, it almost just brings up the question of, well, what was the fucking point of this app? Like, literally what is the point of this app? Like, what do you think?

A (23:26)

Well, I mean, I think it's like AI generated video is very popular on the Internet just because of the AI Slop phenomenon and all that. And it's like this is the best, easiest slop maker. I mean, I think that people will find ways to jailbreak it, for lack of a better term, like to get around the guardrail and things like that. And then also it's like most AI slop that we see on the Internet doesn't have characters in it and stuff like that. And so they've made like the best, easiest AI slop generator that there is. And I think that to the extent that there will be staying power for this, I think it will probably be from the slop manufacturers who are using the app to make stuff, but then are taking it off the app and putting it elsewhere to go viral in the ways that we've talked about before.

B (24:20)

More like generic Monster Does XYZ or something. It doesn't have to be some IP from Stranger Things or something. It's just going to be more generic creations which don't violate copyright, probably. Maybe. I'm really sorry if this is so dumb and this was like the whole in joke the entire time, but I'm just sort of realizing that Sora from Kingdom Hearts, which is a video game which has all of this IP crossover from Disney with Goofy and all of that, and then OpenAI uses that name for its massive copyright infringement app. I'm sorry, I'm only just putting two.

A (24:56)

And two together, but I don't think that's why it's called Sora.

B (24:59)

But that's right, exactly.

C (25:01)

There's also Cameo exists too. They're using the word cameo for literally what cameo is, which is put a celebrity in something to make them say what you want them to say.

A (25:10)

App.

C (25:11)

Does Cameo not own that trademark?

B (25:14)

Surely they have a trademark on that. I'm actually going to look that up. While Sam, I'm going to ask you about this next story. And the headline is. And Matthew, our regular contributor, wrote this. Sora 2 watermark removers flood the web. Basically every Sora 2 video comes with a watermark. It's sort of like a little spinning icon that, as Jason says, you can be like, oh, that comes from this app. So it's going to be AI. But as Matthew's reported, there's now this wave of watermark removers where you just upload the video and it takes it out automatically. I'm just wondering what you think of that because of course you've covered a ton of abusive deepfake and AI videos as well, which don't present themselves as AI generated. They don't have a watermark very much deliberately. What do you make of Sora trying to have a watermark and then it being defeated in 20 minutes or whatever it was.

C (26:12)

I mean, it's like very predictable. Like, you can take watermarks out of pretty much any. Like, TikTok has watermarks where it's like little TikTok logo bounces around. It's just not like watermarks in general to be a solution for AI generated content. It's like a good start, but it's relying on people to keep it in the video to like not use something like this exactly like this to take it out. It's. Watermarks are just such a flimsy indicator of AI content in general. And it's also. You can put it. You can put a watermark on a real video. Like, you could edit a video to put a sore watermark on it and say, oh, that's not real. There are water. There are types of watermarking that aren't. We're talking about like the little logo, right? Like the thing that is on the video that you can see with your eyes. Um, there's types of watermarking that's like. It's more like thumbprinting or hashing where it's not visible to the naked eye and it's much harder to remove. You have to like, edit the video to actually be able to see it. Google has something like this for vo. But this particular type of watermark is like, very predictable that people would try to take it off of Sora videos to make them seem like they're real videos and then they. They repost them to like reels and other platforms where, you know, it's like they're mixed in with real stuff. Yeah, this whole story was very. It's like, yeah, that's what people are going to do with this.

A (27:42)

It's.

C (27:43)

It sucks that this is like the, the cottage industry that pops up every time. So it's definitely worth reporting on and noting that here it is again, like always.

B (27:56)

Yeah, and it was just done so quickly and it can be automated by tools that anybody can then use. I'm looking at the Cameo trademark, and I'm not a copyright or a trademark lawyer. I think Jason would probably know more about this. But, you know, it says in its Goods and Services section, downloadable software used to create personalized video messages for entertainment and structural, inspirational or greeting purposes featuring athletes, actors, entertainers, influencers, etc. Etc. I don't know, it sounds like you could do that on Sora, so I guess we'll see. I know, but then there's the specific style of the cameo logo. Whatever. Yeah, I'm pretty sure there's going to be at least one lawsuit after this app in some sort of direction. All right, let's leave that there. When we come back after the break, we're going to talk about the app Iceblock and Red Dot, which you know are both for reporting sightings for ICE officials and how Apple and Google has removed those. We'll be right back after this.

A (29:12)

You ever have one of those moments where you realize you started something big and suddenly you're the CEO, the designer, the marketing team, and the IT department all at once? When we launched 404 Media, we had that moment. We're not just the journalists, we're the business people and everything else. Thankfully, when it came time to set up a merch shop, we had Shopify in our corner. Shopify is the commerce platform behind millions of businesses around the world, powering 10% of all E commerce in the US from household names like Mattel and Gymshark to, well, 404 Media. Shopify gives you the tools to build, sell and grow anywhere. Shopify makes it easy to create a store and to refresh your items when the time comes, just like we just did with a new line of merch. With hundreds of ready to use templates, you can create a stunning online store that perfectly matches your brand style, even if you're not a designer. And when it's time to grow, Shopify gives you the power to market like a pro, launching email campaigns, running social ads, and connecting with your audience wherever they scroll or stroll. Shopify is your commerce expert, helping you with everything from inventory management to international shipping returns and more. Because when you're ready to sell, you're ready for Shopify. Turn your big business idea into With Shopify by your side, sign up for your $1 per month trial today and start selling@shopify.com media. That's shopify.com media.

D (30:41)

This is a paid advertisement from BetterHelp. October 10th is World Mental Health Day, and this year we're saying thank you therapists. Thankfully, therapy has been destigmatized over the last few years, and a lot of that is thanks to easier access to qualified therapy thoughtful therapists through services like BetterHelp. If you've done therapy, you know that the right therapist can change everything, helping you cope with a world that, let's face it, has been quite a mess lately. Finding a therapist you vibe with can be hard, which is why BetterHelp does the initial matching work for you so you can focus on your therapy goals. Just fill out a short questionnaire which helps identify your needs and preferences. And BetterHelp's more than 12 years of experience means you typically find someone you you like for the first time. But if you're not satisfied with your match, you can switch to another therapist at any time. BetterHelp has more than 30,000 therapists and has served more than 5 million people globally, meaning that you're sure to find a therapist that works for you. This World Mental Health Day, we're celebrating the therapists who have helped millions of people take a step forward. If you're ready to find the right therapist for you, BetterHelp can help you start that journey. Our listeners get 10% off their first month at betterhelp.com 404 Media that's BetterHelp H E L P.com 404 Media Fall is planting season.

A (32:10)

Did you know a lot of plants actually do better when they're planted this time of year? That's why I love fastgrowingtrees.com doesn't matter if you're in the sunny south or where the air is getting chilly, their plant experts can help you find the perfect fit for your space. I've loved my experience with Fast growing trees. Picking out my plants and trees was a breeze. There was no shoving the plant into my car, getting dirt and leaves all over. My beautiful arbequina olive tree, dragon fruit plant and birds of paradise all showed up to my doorstep happy and healthy. In the few weeks since I planted them, I can already tell they're going to be a great fit for my backyard. And I've really enjoyed reading the specialized care instructions and tips I found on Fast Growing Trees website. Fast Growing trees has over 6,000 plants for any space. Fruit trees, flowering trees, shrubs, even indoor plants. Every single one goes through a 14 point quality checklist and comes with their alive and Thrive guarantee so you know they'll last this fall. They have the best deals for your yard, up to half off on select plants and other deals and listeners to our show get 15% off their first purchase when using the code first 404 Media at checkout that's 15% off at fastgrowingtrees.com 404 Media using the code 404 Media at checkout now's the perfect time to plan use 404 Media to save today. That's fastgrowingtrees.com 404 media use code 404 Media at checkout offer is valid for a limited time. Terms and conditions may apply. Check out the link below or in the show Notes and support the show.

B (33:54)

All right, and we are back. Jason, do you want to take the lead on the story?

A (34:01)

So this is a story that you wrote. You've actually written a handful of articles about this and this phenomenon. The title is Ice Block Owner. After Apple removes app, quote, we are determined to fight this, I guess. Let's just start. What is the iceblock app?

B (34:20)

Yeah, so iceblock is an app that was only available on Apple devices and you downloaded it and you could anonymously report sightings of ICE officials in your local proximity so it would have access to your location data. They say they treat this, you know, they're very privacy centric way. And a security researcher who looked into the app largely corroborated that as well. Didn't find anything sketchy going on with the location data. But the idea is that, oh, I can turn that on and then if I see an ICE official in this location or a raid going on or somebody not having their due process rights or whatever, I can report it on the app and that will then alert people, say in that neighbourhood or maybe the next neighbourhood over, depending on how close obviously the activity is. And this is one of several apps like this and we'll talk about another one in a minute. But it really rose to prominence in June when CNN reported on the iceblock app and the Trump administration was very, very mad about it to the point where they even floated these ideas that we're going to try to find a way to charge CNN for covering and allegedly promoting this app. Obviously that's ridiculous and I don't think would have any legal basis, but you know, sort of, who knows in this day and age. But that's how it grew and grew and grew to the point where it became pretty popular. And when people think of ice spotting apps, they typically think of Ice Block just because it's sort of the most well known at this point.

A (36:09)

Yeah, but this app has been pretty widely used by activists, by people in their neighborhoods, et cetera, to essentially tell people where raids are happening. I again live in la and during the pandemic, I mean there's still, it's still happening. It's still happening like everywhere, especially in Chicago, especially in la, happening in Portland, but it's happening all over the country. It's like my neighbors were like, download the Ice Block app. Like you'll walking around, you'll see flyers that a lot of like coffee shops will put on their doors. And things like that, which is like what to do in case there's an ICE raid, like how to support the people who are being, you know, accosted and detained and things like that grabbed off the street. And a lot of them are like, download the ICE Block app to see to like report this. So that that's like very notable, I would say. So anyways, it gets the attention of the Trump administration because CNN reports on it and then Apple takes it down last week. What is like Apple's reasoning for taking it down?

B (37:18)

Yeah, so I spoke to the developer of iceblock and they shared the email they got from Apple after the app was removed. And they point to a couple of parts of their terms of use of their app Store policies or whatever. The first one is 1.1 objectionable content and in action include what they call as offensive, insensitive, upsetting, intent to disgust, in exceptionally poor taste or just plain creepy. And then they include some examples and they point to defamatory, discriminatory, mean spirited commentary about religion, race, sexual orientation, gender, ethnic origin or other targeted groups. I emphasize that because that's the only one I could possibly in a million years see this app kind of at a stretch, touching. But that is what Apple says. And crucially, this came after there was a shooting at an ICE facility, I think, a few weeks ago at this point, where a shooter shot into a van near the facility and authorities claim that the target was ice. Officials, probably based on messages written, are allegedly written on the bullet casings. Unfortunately, someone died in that event. It was a detainee in the van and others were seriously injured. But that sort of brings up ICE Block again in front of the Trump administration. And it seemed. Well, I mean, it is clear that pressure from the Trump DOJ and Pam Bondi's DOJ directly led to Apple removing this. I mean, Pam Bondi went to Fox herself, again, that's the Attorney General, and said, you know, we don't want this to happen, basically. I'm paraphrasing slightly. You can see the full quote in the article. But it was direct pressure from the DOJ leading to this. And I think you found that particularly wild, Jason.

A (39:29)

Right. Well, I just saw a couple of things. One, like during COVID the Biden administration sent some like, relatively innocuous like emails to Facebook and Twitter being like, hey, can you take down this disinformation about vaccines and things like that? And they did. And this became the basis of Elon Musk's the Twitter Files, which after he bought the the app, like released all of this internal Twitter stuff that was like evidence of a massive government attack on free speech because of pressure from the Biden administration on these social media companies to take down content it didn't like, like that. And so basically like this has underpinned like a gigantic right wing grievance for years at this point. Like this, these couple of emails have sustained like frankly, a year's long like.

B (40:36)

Outrage cycle and careers are built on it, basically. Grifting careers.

A (40:40)

Yeah, grifting careers are built on it. Like lots of people been fired. Like, you know, like Elon Musk came in and fired everyone at Twitter and like anyone who had anything to do with anything involving that. I don't know if there's been like threats to prosecute like the Biden officials who sent those emails. But like it's a big fucking deal like in right wing circles for a very long time now. This is like literally the exact same thing, arguably even worse because like for a variety of reasons, because not, they're not saying like, hey, take down a couple posts, they're saying, hey, delete this app or else essentially. I mean, I think that if the DOJ comes to you and says like, delete this app, the or else is implied, I would imagine, especially like with this administration and especially when you have like Tim Cook going to the White House giving like a glass ornament to the president, like things like this. And so you have this happening and it's like, I don't know how many people use this app, like millions, hundreds of thousands. I don't know. It's a pretty popular app. It was top of the App Store for a little while or near the top like that. That's, that's not good. And it's definitely, I don't know, it's like First Amendment violation adjacent at, at the very least. And so, you know, the creator of this app says that he's going to push back against it, I think. Let's talk about the Google situation and then we can talk about some of the like reasoning for this and the, you know, terms of service and things like that. So this, the second story is called Google calls ICE agents. A vulnerable group removes ICE spotting app Red Dot. So basically ICE Block was never on Android, right?

B (42:35)

It was never on Android because the developer says they couldn't do the privacy protecting things they like to do on Android. Some people disagree with that, but that's the state of reason.

A (42:46)

Yeah. Okay, so there's another app called Red Dot and it also gets deleted from the Google Play Store what happened in this case. And this happened immediately after the Ice Block situation?

B (43:00)

Yeah. As far as I can tell, this wasn't simultaneous to iceblock. I think it happened afterwards. Of course, if anybody has more information on that, you know, especially from Google's decision, let me know. But that's sort of my understanding at the moment. Yeah, Red Dot was a very, very similar app. I mean, at the end of the day, all of these apps are going to be pretty similar. Right. All you're doing is reporting a location saying there are ICE people there. Red Dot did much the same thing. Both Apple and Google removed this app. You can't get this app on the Apple App Store. Apple did tell me it removed multiple apps when it removed Ice Block, but the company stopped responding when I asked, well, was it Red Dot? Because they found that the day after I then pinged Google because that's even more interesting, because it signals that, oh, there's like a broader crackdown of these ICE spotting apps and it isn't just limited to Apple. Right. So I reach out to Google and I'm going to try and summarize. I'm going to read basically what we put in the article because I'm sure, for reasons that Jason will bring up, it was a bit of a weird response. First of all, Google said it didn't get any outreach from the doj. So I asked that explicitly, did the DOJ tell you to take this down? They said no. They didn't talk about the ICE facility shooting specifically. But in this very roundabout way, Google said it removed apps that share the location of what Google described as a vulnerable group after a recent violent act against them connected to this sort of app, which is like, wow, that's a lot of words to say that you removed ICE Spotting apps after the shooting at an ICE facility and most importantly, that you consider ICE officials a vulnerable group. Jason, when I initially filed it, I don't think we changed the copy. Well, no, we did later on a little bit. But the headline, crucially was, was not stressing the vulnerable groups part. You pulled that in. You pulled that out quite rightly. Why did that stand out to you? Calling ICE officials a vulnerable group is an obvious question.

A (45:16)

Yeah, so we used to report a lot on terms of service and content moderation on social media platforms. Specifically, like after the 2016 election where there was like. And during, like Black Lives Matter and like the MeToo movement and things like this, there was like a real effort from social media companies to devise rules that would protect, you know, quote, unquote Vulnerable groups. And vulnerable groups is defined differently by every social media company. But like, broadly, it means ethnic minorities, gender minorities, religious minorities, people who are traditionally discriminated against or the subject of hate speech. So trans people, black people, you know, in, in specific countries, it's like religious minorities that are subject to like, hate speech and genocide and things like that. Like, there are so many examples of just like really awful things happening on social media, you know, like Facebook being credibly accused of facilitating a genocide in Myanmar, things like this. And so they write rules in response to these like, really awful real world situations. And the catch all term that social media companies use is vulnerable groups. And like, in no world was like police officers, law enforcement, like a vulnerable group when these rules were written. And I think to call them a vulnerable group now is like, insane. And it's also, but it also mirrors like what the Trump administration has been saying where they, they say over and over and over again that these ICE agents are risking their lives, that they're trying to keep America safe, that they're being targeted, that they're being doxed, that they're being harassed. Which has like, been the underpinning for why they're not including ICE agents. Names on like, indictments in court and things, things like that. Affidavits, which is very rare.

B (47:30)

And masking, of course.

A (47:32)

Yeah, yeah, masking. Like refusing to say their badge number, refusing to identify themselves. Which is why you have like, sometimes what seem to be like plain clothes officers, like abducting people off the streets because they're like, oh, well, we're being targeted so we can't identify ourselves. And so this is now Google, like, parodying, parroting that language and also using these guard, these like, guidelines that they wrote years ago and retrofitting them to like, come up with a pretext to ban this app. And Apple didn't say the exact same thing, but it was somewhat similar where it was like, you know, you already said what the reasoning was.

B (48:14)

Targeted groups.

A (48:15)

Yeah, yeah. And it's like, okay, I don't know, like, not good. Not good. I don't know what to say other than it's like we're now in a world where like, trans people are not protected on the Internet. Like, you know, actually targeted groups are not protected on the Internet, but police officers are.

B (48:35)

I wonder if Google considers people like the woman who was shot seven times by an ICE official a few days ago in a hotly disputed event in Chicago. Right. I wonder if they consider that person part of a vulnerable group as well. Yeah, it is very much co opting the language and the methodology and the thinking from a content moderation approach from a different time. This doesn't. Well, they're trying to apply it here in a time that is radically different to, you know, 2016 or even 2020 or anything like that, really. I did get a statement from Fire App, which is another similar app and I should say another one was removed. I actually can't remember the name of it right now because there's so many of these, weirdly. But another one got removed. Fire App reached out to me and they're much more. I mean they're closer to like Citizen, they're sort of broader. But when I checked the Fire App website, and interestingly, it's a website as well, you can view it, you can't report on there yet. But of course a website's harder to block than the app. Right. I went on there and they sent me an email and they said that it's concerning to see platforms similar to ours, a Citizen style app, removed despite their clear compliance with legal standards. This raises serious questions about fairness and transparency. This action seems to be based more so on fear of retaliation and retribution from the Trump administration. And in line of kissing the metaphorical ring, something we unfortunately have seen many top executives do in order to placate Trump on their side. I mean, I think that puts it pretty well. How about we leave that there? I'm definitely going to keep an eye on all of these ice spotting apps being removed. If you're listening to the free version of the podcast, I'll now play us out. But if you are a paying 404 media subscriber, we're going to talk about a very, very significant update to a story we broke back in May. This radically changes and adds to the narrative the police and flocks the surveillance company gave at the time. You can subscribe and gain access to that content at 404 Media co. As a reminder, 404 Media is journalist founded and supported by subscribers. If you do wish to subscribe to 404 Media and directly support our work, please go to 404 Media co. You'll get unlimited access to our articles and an ad free version of this podcast. You'll also get to listen to the subscribers only song section where we talk about a bonus story each week. This podcast is made in partnership with Kaleidoscope. Another way to support us is by leaving a five star rating and review for the podcast. That stuff really helps us out. Soon I'll start reading through the Spotify ones. I've just been reading the Apple ones out. I'll go through those as well. Mr. Bean. 404 Media will see you again next week.