Meta CEO Mark Zuckerberg “loved” an AI-generated slop image of a horse made out of bread posted by a spam page on Facebook that also posts AI-generated images of children with amputations and regularly circumvents Facebook’s algorithm to link users offsite to ad-laden AI-generated content farms.

The page, “Faithful,” is verified, operated out of Romania, has 1.1 million followers, and regularly goes mega viral with the exact type of AI slop that I have been writing about over the last year. In that sense, it is the perfect encapsulation of the type of spam page that has become dominant on the platform as Meta continues to lean into AI-generated content and pays people for going viral on the site. 

“I made every detail with love, but it seems no one cares,” reads the caption of the image, which has 2.7 million reactions, 193,000 comments, and 98,000 shares as of the time of this writing. Zuckerberg’s interaction with the page was first noticed by Gazpacho Machine, a man who posts reviews of food he eats while taking showers. 

When Gazpacho Machine posted about this, I was initially skeptical that Zuckerberg's real account had liked the page (as in, it could have been an imposter), and the image had so many likes that Facebook was initially having trouble loading information about which accounts actually liked the page. Gazpacho Machine sent me a screen recording showing that it was indeed Zuckerberg's real "@zuck" account, and I was later able to verify for myself that this is Zuckerberg's real account:

This bread horse is a variation on a classic type of AI slop that made its way into my very first story about the phenomenon of AI spam on the platform in December 2023, when AI spammers were taking already viral images and running them through image-to-image AI tools to create slight variations of the original viral image. 

The origin story of bread horse is “The Bread House man,” a viral Russian image from the 2010s of a man next to a house he had created from various rolls and baguettes. Catherine Hall, a Facebook user who tracked the early spread of AI spam on Facebook, originally found dozens of AI-generated variations of the Bread House Man. 

The “Faithful” page, whose header image says “I Love God. I am proud to say that,” has posted hundreds of AI-generated images over the last few years across a host of genres that are now very familiar to me, and are similar to many pages that are operated by people in the global south trying to make money on Facebook. It has repeatedly posted the same image of an AI generated child who is missing an arm and whose caption says “My mother said I was beautiful, but so far I have not liked anyone” at least three times. It has gotten thousands of likes each time.

The page has also posted various images of AI-generated elderly people who are supposedly older than 100 celebrating their birthday, AI sand sculpture images, AI-generated variations of American Idol and America’s Got Talent, AI wood sculptures, AI photos of aging couples, AI ice sculptures, AI knitting, AI ‘I drew a picture’ images, AI families who are also onion farmers, and AI recycled bottle sculptures. Faithful has also posted a fair bit of Donald Trump content, as well as lots of inspirational screenshots of the Bible, reels that are seemingly automatically created from Reddit posts or written by AI, and inspiration porn. 

Many of the images are monetized in ways that I have previously reported on. For example, many of the images have captions that ask users to read the first comment for more information; a pinned top comment posted by Faithful will then link off of Facebook to a website that is absolutely loaded with ads. I clicked on a recent link posted by both Faithful and a related page called "Faith Space" about an AI generated stepfather who stood up for his AI generated stepdaughter when she was being bullied and was taken to a website called Daily Home Gardening, which served me many ads for products called “Levitox” and “GlucoReNu,” which showed images of worms and had captions that read “The Lump Of Worms Will Come Out Of You In The Morning. Try It.”

I do not know why Zuckerberg “loved” the AI generated bread horse, but it should be noted that it is harder to errantly “love” something on Facebook than it is to errantly like it. Meta did not respond to a request for comment. It is just one small action by one very rich and powerful person. But it is further evidence that strengthens what we already know: Mark Zuckerberg is not bothered by the AI spam that has turned his flagship invention into a cesspool of human sadness and unreality. In fact, he thinks that AI-generated content is the future of “social” media and Meta believes that one day soon we will all be creating AI-generated profiles that will operate semiautonomously on Meta’s platforms. 

While I was writing this, Faithful posted another AI-generated image of a grave with the caption “NEW TOYS APPEARED ON MY SON'S GRAVE EVERY DAY, SO I DECIDED TO FIND OUT WHO WAS DOING IT.” The image links in the comments to Daily Home Gardening, which loaded 64 distinct ads, plus an infinite scroll of ads at the bottom of the page. 

On Monday, a day after TikTok came back online for the U.S., people started noticing that the app would not let them comment “free palestine.” 

Several TikTok users posted screenshots on X and Bluesky showing the message they received after trying to comment “free palestine” under other people’s posts. TikTok users started reporting this on Sunday, following a tense few hours where the app blocked U.S. users from access ahead of a potential ban, displaying instead a notification sucking up to then-incoming president Donald Trump. 

I tried this myself on Tuesday morning, using two different throwaway TikTok accounts. Using one account, I could comment “free palestine” without a problem, and that comment is still up as of Wednesday morning. Using another, my “free palestine” comments were immediately removed repeatedly, and I received a notification that I had violated the TikTok Community Guidelines. I could comment with a nonsense phrase (“free shavacado”) using that same account, however, and TikTok didn’t remove it. 

Using the account that kept getting comments removed, I could post a video with the caption “free palestine #freepalestine” on Tuesday without a problem.

A spokesperson for TikTok told 404 Media that the platform’s policies and algorithms did not change over the weekend, adding that they’re working on restoring U.S. operations back to normal. They’re expecting some “temporary instability” as services are restored, they said, which could affect some features or access. They said that TikTok does not have a policy against people saying “free palestine,” and pointed me to the platform’s community guidelines, transparency report, and moderation approach pages. 

There have been a few noteworthy instances in the last few years of social media users claiming that pro-Palestine content was being censored or downranked, only for the platforms involved to blame the issue on bugs or deny it was happening. In October 2023, 48 organizations, including 7amleh, the Arab Centre for Social Media Advancement, which advocates for digital rights of Palestinian and Arab civil society, issued a statement “urging tech companies to respect Palestinian digital rights during the ongoing war,” Al Jazeera reported.

“We are [concerned] about significant and disproportionate censorship of Palestinian voices through content takedowns and hiding hashtags, amongst other violations,” the statement said. “These restrictions on activists, civil society and human rights defenders represent a grave threat to freedom of expression and access to information, freedom of assembly, and political participation.”

Also in October 2023, on Instagram, using the “see translation” feature in bios about Palestine inserted the word “terrorist.” In March 2024, Google’s Gemini AI would not answer questions like “where is Palestine?” or “what is Palestine?” but would answer questions like “where is Israel?” or “what is Israel?” And in October 2024, Twitch admitted to temporarily blocking new users in Israel and Palestine from creating new accounts to prevent them from uploading “graphic material.” 

There is a lot of weird stuff going on with social media platforms right now, to put it as mildly as possible. One minute, TikTok is gone; the next, it’s back. One minute, people are outraged because they think they’re being forced to follow Trump on Facebook; the next, it’s obvious that the administration’s accounts have simply changed hands. All of it makes people constantly doubt, second-guess, and argue about what they see with their own eyes. 

In a lot of ways, this constant, roiling turmoil of navigating what can and can’t be said, what might get you banned, and what app is even still accessible to you anymore is exactly what a lot of marginalized people—including, especially, sex workers—have always gone through. That disparity and confusion is baked into the algorithms we are forced to coexist with online. It’s why when I wrote about getting a bizarre search result from Google’s AI Overview, the company’s spokesperson couldn’t even replicate it themselves. And now that TikTok has bent the knee to the Trump administration in a very public way—including CEO Shou Chew posting a personal message thanking the president for “his commitment to work with us to find a solution that keeps TikTok available in the United States,” —everyone online is on the highest alert possible, looking for every sign and symbol that their social media platform of choice might be falling into fascist censorship. This is, as marginalized communities online have been showing us for years and as we said when Trump was elected, always the way it’s been.

The manufacturer of a machine that costs six figures used during heart surgery has told hospitals that it will no longer allow hospitals’ repair technicians to maintain or fix the devices and that all repairs must now be done by the manufacturer itself, according to a letter obtained by 404 Media. The change will require hospitals to enter into repair contracts with the manufacturer, which will ultimately drive up medical costs, a person familiar with the devices said. 

The company, Terumo Cardiovascular, makes a device called the Advanced Perfusion System 1 Heart Lung Machine, which is used to reroute blood during open-heart surgeries and essentially keeps a patient alive during the surgery. Last month, the company sent hospitals a letter alerting them to the “discontinuation of certification classes,” meaning it “will no longer offer certification classes for the repair and/or preventative maintenance of the System 1 and its components.” 

This means it will no longer teach hospital repair techs how to maintain and fix the devices, and will no longer certify in-house hospital repair technicians. Instead, the company “will continue to provide direct servicing for the System 1 and its components.” 

On the surface, this may sound like a reasonable change, but it is one that is emblematic of a larger trend in hospitals. Medical device manufacturers are increasingly trying to prevent hospitals' own in-house staff from maintaining and repairing broken equipment, even when they are entirely qualified to do so. And in some cases, technicians who know how to repair specific devices are being prevented from doing so because manufacturers are revoking certifications or refusing to provide ongoing training that they once offered. Terumo certifications usually last for two years. It told hospitals that “your current certification will remain valid through its expiration date but will not be renewed once it expires.”

Hospitals are increasingly being pushed into signing maintenance contracts directly with the manufacturers of medical equipment, which means that repair technicians employed by hospitals can no longer work on many devices and hospitals end up having to employ both their own repair techs and keep up maintenance contracts with device manufacturers. 

“One of my fears is that if a device goes down, we’re going to be subject to their field engineers’ availability,” a source who works in hospital medical device repair told 404 Media. 404 Media agreed to keep the source anonymous because they were not authorized by their hospital to speak to the media. “They may not be able to get here that same day or the next day, and if you’ve got people waiting to get an open-heart surgery, you have to tell them ‘Oh, the machine’s down, we’re going to have to postpone this.’ That’s detrimental to a patient who has a life-altering, very serious thing that they’re having to cancel and reschedule.” 

Having to rely on a manufacturer’s repair network is the exact situation that farmers have found themselves in with John Deere tractors. Last week, the Federal Trade Commission sued John Deere for its monopolistic repair practices. The FTC specifically cited the fact that farmers have often been forced to wait days or weeks to get a John Deere “authorized” repair tech out to fix their tractors, which has resulted in farmers losing crops at critical harvest times. During the peak of the COVID-19 pandemic, some hospitals found themselves pirating repair software from Poland to repair broken ventilators because manufacturers’ repair technicians were spread so thin that hospitals had to wait weeks for repairs. 

This specific ventilator repair crisis during COVID led experts at Harvard Medical School to write that “For years, manufacturers have curtailed the ability of hospitals to independently repair and maintain medical equipment by preventing access to the necessary knowledge, software, tools, and parts” in a piece calling for right-to-repair legislation. The FTC, meanwhile, suggested in a report that medical device manufacturers sometimes charge two-to-three times what an independent repair tech would charge for the same repair. 

“It's scary to think that you could buy a piece of medical equipment for your hospital, just to have the manufacturer wake up one day and decide they will monopolize all repairs for that product,” Nathan Proctor, senior director of consumer rights group PIRG’s campaign for the right to repair, told 404 Media. “The people who are trained to fix that equipment won't suddenly forget all they know, but they will suddenly be restricted from doing the repairs. I think that's just absurd.”

Manufacturer contracts like this lead, across the board, to higher costs for hospitals. 

“It’s no secret that America’s healthcare system is the most expensive, and this is one of the reasons why. These machines are actually highly reliable, we’ve had a low cost of service for it over the last few years. And when something isn’t right, we have people in-house who can fix it,” the source familiar with Terumo machine repair said. “But the cost of having a service contract with a manufacturer, you’re probably talking 10 times the cost. It’s not a big deal having a contract for one device, but when that starts happening across many devices, it adds up in the end. If you took every hospital in America and said for every medical device in the hospital, you need to put it on an OEM [original equipment manufacturer] maintenance contract, it would tank your financial system. You just can’t do that.”

Medical equipment manufacturers have strongly lobbied against right to repair legislation all over the country, and have been successful in getting medical devices exempted from right to repair legislation by claiming that the machines are too sensitive and complex to be repaired by anyone besides the manufacturer. The medical device giant AdvaMed, for example, says “the risk to patient safety is too high.” 

But, again, the people working on medical equipment in hospitals are often hospital employees or contractors whose job is to repair medical equipment, and who are being prevented from fixing equipment that a hospital has purchased. “Just because a guy has Terumo on his shirt doesn’t mean he’s a more competent technician” than an in-house hospital technician, the source familiar with Terumo device repair said.

In a brochure for hospitals, Terumo advertises both its device and its maintenance program: “Advanced, precision medical equipment requires genuine parts and top-quality, specialized service – just as getting the best medical care from qualified specialists. Terumo Cardiovascular Service has the unrivaled expertise, experience, equipment, and parts to provide the optimal level of planned service and repairs needed. Use Terumo Cardiovascular Service and avoid exposure to liability issues.” 

A spokesperson for Terumo told 404 Media that the company “saw declining participation in this program and determined that the best way forward was to require servicing through Terumo Cardiovascular’s genuine in-house Service team to continue to ensure Terumo devices are properly maintained.”

“Terumo Cardiovascular’s Biomed Certification Program was originally structured to train non-Terumo personnel (hospital Biomeds) to service Terumo heart-lung machines and associated hardware. Properly maintained medical devices are necessary for optimal performance which is essential for quality of patient care and outcomes,” they added. “Hospitals’ existing Terumo Cardiovascular Biomed certifications will remain valid through their expiration dates but will not be renewed once they expire.”

We start this week with the TikTok ban: how we got here, what happened, and, most importantly, why it means we need more decentralized services. Jason runs us through it. After the break, Joseph breaks down a site called GeoSpy which is marketing geolocation technology to the cops. In the subscribers-only section, we all scroll through an archive of old Nokia (yes, Nokia) designs. Good stuff in there.

Listen to the weekly podcast on Apple Podcasts, Spotify, or YouTube. Become a paid subscriber for access to this episode's bonus content and to power our journalism. If you become a paid subscriber, check your inbox for an email from our podcast host Transistor for a link to the subscribers-only version! You can also add that subscribers feed to your podcast app of choice and never miss an episode that way. The email should also contain the subscribers-only unlisted YouTube link for the extended video version too. It will also be in the show notes in your podcast player.

• Decentralized Social Media Is the Only Alternative to the Tech Oligarchy
• The U.S. Wants to Ban TikTok for the Sins of Every Social Media Company
• The Powerful AI Tool That Cops (or Stalkers) Can Use to Geolocate Photos in Seconds
• Nokia’s Weird Y2K Designs Show the Future We Could Have Had

If it wasn’t already obvious, the last 72 hours have made it crystal clear that it is urgent to build and mainstream alternative, decentralized social media platforms that are resistant to government censorship and control, are not owned by oligarchs and dominated by their algorithms, and in which users own their follower list and can port it elsewhere easily and without restriction.

Besides all of the “normal” problems with corporate social media—the surveillance capitalism, the AI spam, the opaque algorithms—let’s take stock of what has happened in the last few days. 

First, millions of small business owners and influencers who make a living on TikTok were left to beg their followers in TikTok’s last moments to follow them elsewhere in hopes of being able to continue their businesses on other corporate social media platforms. This had the effect of fracturing and destroying people’s audiences overnight, with one act of government. 

TikTok has since come back, but it is still unclear what the future of the platform is, and TikTok now exists at the whim of President Trump and is beholden to him to an unknown extent. TikTok’s status in the Untied States is still up in the air—it is still not available for download in the iOS App Store or the Google Play Store, and it could disappear at any moment if service providers like Oracle decide that Trump’s executive order and assurances that they will not be prosecuted or fined are not enough assurance to keep the app online. 

Elon Musk, who had already turned X into a cesspool of hate and an overt tool to get President Trump elected, is now formally part of the Trump administration, meaning the platform is literally owned by a member of the Trump White House. 

Meta has made an overt shift to the right, and Mark Zuckerberg has himself become a Trump booster. The platform is making its content moderation worse, has declared that immigrants and LGBTQ+ people are legitimate targets for hate speech, and has made many of these changes at the behest of the Trump White House and Stephen Miller, according to The New York Times. 

Shortly after Donald Trump was sworn in as the 47th president of the United States on Monday, many Facebook users started to report that the social network automatically made them follow Trump’s Facebook page. I’ve seen several threads on this issue on Reddit, social media, and people I know personally have also asked me why Facebook made them follow Trump.

While Meta’s CEO Mark Zuckerberg has made it very clear that he and his company are making changes to endear Meta to the Republican president, there is a much less nefarious explanation for why people suddenly saw Facebook posts from Trump even though they don’t remember following him. When reached for comment, Facebook Communications Director Andy Stone directed me to his posts on Threads which points out how government social media accounts work and change with the current administration. 

The POTUS and White House Facebook pages are owned by the sitting administration. When Trump was sworn into office, those pages started sharing his official portrait, images of him signing a flurry of executive orders, and other images from the inauguration. An archive of the POTUS Facebook page from January 24, 2021, shows that the then incoming Biden administration also held on to all the followers from the first Trump administration. And an archive of the White House Facebook page shows that in both the 2017 and 2021 transitions, the page’s posts were wiped and archived, but its follower count was not. Users who followed those pages during the Biden administration may not know that this is how those official pages are managed, and were surprised to see posts from Trump in their feeds all of a sudden. 

I’ve also seen some people claim that Facebook made them follow other Trump-related Facebook pages, like Melania’s Trump’s or the GOP’s pages, but was not able to confirm that happened. Facebook’s Stone told me that the company has seen no evidence of this happening or reason that it would happen. 

Fear and confusion about social media companies leaning further to the right as soon as Trump took office reached beyond Facebook yesterday, and are understandable given that some “errors” are impacting Democrats but not Republicans. On Instagram, for example, users found that Instagram was hiding results for the search term “Democrats.” At the time of writing, when I searched for “Democrats” Instagram said it couldn’t find any matching results. When I searched for “Republicans” Instagram had no issue and surfaced many posts. Meta told the BBC in a statement that it’s “aware of an error affecting hashtags across the political spectrum and we are working quickly to resolve it.”

As we’ve reported in recent weeks, it’s obvious that Meta has decided to endear itself to Trump and conservatives, making specific policy changes around how users can talk about gender and immigrants in a way that panders to Trump and his supporters. There are many reasons Meta would pander to the most powerful politician in the world, especially since Zuckerberg has explicitly said he hopes Trump will protect his company from European regulators. However, forcing millions of users to follow Trump does not seem especially beneficial to anyone, and could easily be explained by the fact that many of these users already followed official administration accounts, or just the fact that Facebook is a gigantic platform that is rife with errors. 

An issue with Cloudflare allows an attacker to find which Cloudflare data center a messaging app used to cache an image, meaning an attacker can obtain the approximate location of Signal, Discord, Twitter/X, and likely other chat app users. In some cases an attacker only needs to send an image across the app, with the target not clicking it, to obtain their location.

Although the obtained location data is very coarse—in some of 404 Media’s tests it showed what city or state someone was in but did not provide more accurate information than that—the news shows the importance for some at-risk users to protect not just their message contents, but their network activity as well. 

“It's more of an oversight in the way the mobile application works than a vulnerability in the actual code but regardless, I thought it should be fixed,” daniel, an independent security researcher who reported the issue to Cloudflare, told 404 Media in an email. daniel said Cloudflare has since fixed the specific issue his custom-made tool was using.

A powerful AI tool can predict with high accuracy the location of photos based on features inside the image itself—such as vegetation, architecture, and the distance between buildings—in seconds, with the company now marketing the tool to law enforcement officers and government agencies.

Called GeoSpy, made by a firm called Graylark Technologies out of Boston, the tool has also been used for months by members of the public, with many making videos marveling at the technology, and some asking for help with stalking specific women. The company’s founder has aggressively pushed back against such requests, and GeoSpy closed off public access to the tool after 404 Media contacted him for comment.  

Based on 404 Media’s own tests and conversations with other people who have used it and investors, GeoSpy could radically change what information can be learned from photos posted online, and by whom. Law enforcement officers with very little necessary training, private threat intelligence companies, and stalkers could, and in some cases already are, using this technology. Dedicated open source intelligence (OSINT) professionals can of course do this too, but the training and skillset necessary can take years to build up. GeoSpy allows essentially anyone to do it.

Welcome back to the Abstract! 

This week, we’re going to the movies. For 77 hours straight. At the end, we’ll know whether we are doomed to villainy or driven to heroism. These are the only two options! 

Then, you’re not going to believe this, but the global trade for exotic ornamental plants is a bit sus. Next, OCTOPUS BRAINS! Last, scientists solve the mystery of the missing Moon craters, which has been a cold case for about four billion years. 

May you all be burdened with glorious purpose.

With Great Power Comes Great Reproducibility

Wigmore, Julia et al. “Are adverse childhood experiences scores associated with heroism or villainy? A quantitative observational study of Marvel and DC Cinematic Universe characters.” PLOS ONE.

Losing parents. Survivng genocide. Having your home planet explode. Superheroes and supervillains sure go through a lot of trauma. Now, this emotionally manipulative trope has undergone scientific scrutiny in a new study that assessed whether traumatic experiences in childhood predict heroism or villainy in superhero movies.

Put another way, researchers devised a professional justification for watching 33 films from the Marvel and DC cinematic universes, totaling 77 hours and 5 minutes. By scoring 28 characters—19 men, 8 women, and a gender-fluid Loki—the team established that trauma has no impact on whether a character becomes a hero or a villain. 

“No one is doomed to be a villain just because of early childhood experiences,” concluded researchers led by Julia Wigmore of the University of Calgary.

In other words: Magneto, no more excuses, dude. Stop murdering people with metal. Go touch grass.

Naturally, characters like Harley Quinn and Loki were head-scratchers, given that they can swing both ways (I’m officially coining the term: bimoral). The team categorized Harley as a hero, because she has a redemption arc after dumping the Joker. Loki is classified as a villain because he spends most of his screentime making mischief. I don’t get the distinction here and think it should be contested in future academic literature. 

The study includes some interesting context about how therapists use superhero stories to help children process grief and trauma, and other Serious Stuff. But mostly, I’m here for the authors’ cheeky little flourishes, letting us know that they really got away with this one. 

“No superheroes or villains were involved in this research study,” the team said, presumably while winking in synchronous harmony. “If anyone could connect us with them, we would be happy to conduct a follow up study to overcome this limitation.”

Snakes (in a Pot) on a Plane

Hinsley et al. “Understanding the environmental and social risks from the international trade in ornamental plants.” BioScience. 

The next time you decide to order an ornamental plant from overseas, a thing I assume we are all constantly doing, make sure to check it for hitchhikers. A study this week revealed that the multi-billion dollar trade in ornamental plants—including olive trees, cut roses, and exotic shrubs—is opening up new vectors for invasive species, such as insects, frogs, geckos, and snakes. 

“Given the number and diversity of vertebrates, including fragile ones such as tropical frogs, reported live in imported products, the number of imported invertebrate pests is likely underestimated, and more consistent measures are needed to provide an accurate understanding of the true implications of trade and how they might be managed,” said researchers led by Amy Hinsley of the University of Oxford.  

People are straight-up bagging plants that still have a bunch of creepy crawlies still on them, and flying them out to customers in other continents. The authors of the study outline a range of actions to help mitigate the risks, including introducing “plant passports.”  

Ultimately, though, we might have to break out the big guns and get Samuel L. Jackson to reprise his role as the globetrotting FBI agent Neville Flynn, who could solve this problem, like all others, with eruptive profanity. Because I, for one, have had it with these (bleep) biosecurity risks due to (bleep) under-regulated industries in this (bleep) complex integrated global economy.

Octopuses Map Their World Through “Suckerotopy”

Olson, Cassady et al. “Neuronal segmentation in cephalopod arms.” Nature Communications.

Scientists have confirmed once again, and to the surprise of nobody, that octopuses are epic. While it’s well-established that these charismatic mollusks basically have prehensile brains, a team has now zoomed in on the axial nerve cords that animate the tentacles, revealing some of the mechanics behind their coordination of such segmented anatomy.

“The octopus has a motor control challenge of enormous complexity,” said researchers led by Cassady Olson of the University of Chicago. “Each of its eight arms is a muscular hydrostat, a soft-bodied structure that lacks a rigid skeleton and moves with near infinite degrees of freedom.”

“Even with this complexity, octopuses control behaviors effectively along the length of a single arm, across all eight arms and between suckers,” the team said. “The neural circuits underlying these behaviors have been unexplored with modern molecular and cellular methods.”

By studying the nervous system of the California two-spot octopus (Octopus bimaculoides) the researchers discovered that axial cords nerves “form a spatial topographic map” for each sucker, which they call “suckerotopy.” In other words, octopuses generate topographic maps of their own suckers, which helps them coordinate the immense motor and sensory input from these appendages.  

Bonus points to the authors for including video footage of octopuses doing neat stuff, like playing with a baseball and emerging from what looks like ancient Greek pottery. Mother Nature gets a 10/10, would evolve eight-limbed intelligent molluscs again.

The Moon is a Harsh Viscous 

Zhu, Meng-Hua et al. “Obliteration of ancient impact basins on the Moon by viscous relaxation.” Nature Astronomy.

The Moon was born in the cataclysmic fallout of a crash between Earth and a Mars-sized object—and that was the easy part. Earth and the Moon were bombarded with a heavy flux of space rocks for hundreds of millions of years after they formed (talk about a traumatic childhood!). 

Earth has erased most craters from this time because it moisturizes daily, but the Moon is an airless inactive world that should have preserved an estimated 300 craters with diameters greater than 185 miles. Yet there are only about 40 ancient crater basins on this massive scale on the lunar surface. What gives? Who ate all the craters?

This week, scientists presented a new explanation for the mysterious discrepancy: Viscous relaxation. It sounds like something Gwynth Paltrow wants to sell you, but it is actually a geological process that smooths out terrestrial surfaces over time. In the case of the Moon, the team found that viscous relaxation from high temperatures in the crust, fueled in part by radioactive elements, could have obliterated hundreds of impact basins.

This scenario offers “a realistic explanation for the low number of basins observed on the Moon,” said Meng-Hua Zhu of the Macau University of Science and Technology. “The substantial relaxation of early basins suggests that terrestrial planets…may have suffered far more impacts than the basin records indicate” but those “early epochs are obliterated.”

And with that, it’s time for all of us to experience the substantial relaxation of the weekend. Obliterate responsibly.

Thanks for reading! See you next week.

This is Behind the Blog, where we share our behind-the-scenes thoughts about how a few of our top stories of the week came together. This week, we discuss enjoying things in the AI industry and the TikTok ban.

JOSEPH: I’m going to talk about Emanuel’s great piece CEO of AI Music Company Says People Don’t Like Making Music. Here is the key passage:

Mikey Shulman, the CEO and founder of the AI music generator company Suno AI, thinks people don’t enjoy making music.

“We didn’t just want to build a company that makes the current crop of creators 10 percent faster or makes it 10 percent easier to make music. If you want to impact the way a billion people experience music you have to build something for a billion people,” Shulman said on the 20VC podcast. “And so that is first and foremost giving everybody the joys of creating music and this is a huge departure from how it is now. It’s not really enjoyable to make music now […] It takes a lot of time, it takes a lot of practice, you need to get really good at an instrument or really good at a piece of production software. I think the majority of people don’t enjoy the majority of the time they spend making music.”

Please go read Emanuel’s whole piece. It’s short and gets straight to the good points of, among other things, that the pursuit of the challenge is one thing that draws people to learn an instrument. Shulman says with his AI tool, they could make the music industry as big as the video game industry, ignoring the fact that some of the most popular video games are specifically about a hard challenge, as Emanuel says.

This week, Google shoved various capabilities from Gemini, its AI tool, into Workspaces for Business and Enterprise customers, including associated Gmail accounts. You might now see buttons for “Summarize this email,” which when clicked will provide a bullet point list of what the email (allegedly) says and, in email threads, peoples’ sentiment towards it in their replies. There’s also a button in the top right that brings up a Gemini prompt bar, and a couple of ways Gemini offers to help. “Show unread emails from today,” and “show unread emails from this week,” are two I’m looking at right now.

Many people are going to love this. Others are going to want to run away from it as quickly as possible. Many people—incluing us—are already furious that they were automatically opted into it. Turns out, disabling it isn’t straightforward, as I found out why I tried to opt 404 Media out of it.

“Today we announced that we’re including the best of Google AI in Workspace Business and Enterprise plans without the need to purchase an add-on,” Google wrote in a blog post on Wednesday. 

The “Summarize this email” button took me by surprise. I opened my Gmail iOS app and it was just there. When I asked a Google spokesperson if Google gave clients a heads up this was coming, they provided me with a couple of links (including that one above), both of which were published Wednesday. So, no not really.

I tried out the email summarize feature on a non-sensitive email Emanuel had just forwarded me. It was an obvious scam email, with someone pretending to be from the family of Bashar Al-Assad and who said they could make us a lot of money. Emanuel forwarded me the email and joked “sounds good.”

Gemini’s summary said “Mohammed Karzoon, a former member of the Syrian President al-Assad’s cabinet, reaches out to Emanuel Maiberg to discuss potential investment portfolios.” The second bullet point read “Emanuel Maiberg expresses interest in the proposition.” Gemini, to little surprise, did not detect that Emanuel was being heavily sarcastic, a beautifully human act.

I then tried to opt us out of these sorts of Gemini features. I logged into Google Workspace, clicked the “Generative AI” drop down menu on the left, then clicked “Gemini app.” I changed the service status to “OFF for everyone.”

Nope, that’s wrong. The Google spokesperson told me that button referred to gemini.google.com, which is the Gemini app, not its integration with Workspace. I also tried in another section called “Gemini for Workspace” which sounded promising but that wasn’t helpful either.

I actually had to go to account, account settings, and “Smart features and personalization” where an administrator can set a default value for users. The spokesperson clarified that individual end users can go turn it off themselves in their own Gmail settings. They pointed to these instructions where users disable “smart features.” 

But it looks like it’s all or nothing. You can’t turn off just the new Gemini stuff without also disabling things like Gmail nudging you about an email you received a few days ago, or automatic filtering when Gmail puts emails into primary, social, and promotion tabs, which are features that Gmail has had for years and which many users are probably used to.

On iOS, you go to settings, data privacy, then turn off “Smart features and personalization.” A warning then says you’re about to turn off all the other stuff too that I mentioned above and much more. On Android, you go to settings, general, and then “Google Workspace smart features.”

Turning these off doesn’t actually get rid of the Gemini button at the top right of the inbox. It just means when you do click it (maybe by accident because it’s right next to the button to switch to a different inbox), it’ll prompt you to once again turn on smart features. It does get rid of the summarize this email button, though.

My first thought when I saw the “Summarize this email” button was, oh god, people are going to be submitting all sorts of sensitive, confidential business information into Gemini. We’ve already seen that with ChatGPT, and organizations have to write policies to stop employees doing it. And now you’re making that process one click, directly in the inbox? In its Privacy Hub page, Google says “Your content is not used for any other customers. Your content is not human reviewed or used for Generative AI model training outside your domain without permission.” I do not know if I have given permission or not, though, that’s part of the problem.

“You’ll see these end user settings will become even clearer and easier for people to use in the coming days as we’re rolling out updates (happening now) with language that’s specific to Gemini in Workspace features,” the spokesperson told me.

I hope so.

Mark Zuckerberg can see the finish line. He is so close to getting what he has wanted for years. The U.S. government is trying to give him the greatest gift he could possibly imagine: A TikTok ban. This would be U.S. intervention against the most credible competitor Meta has seen in years, and U.S. intervention to kill a superior product to the benefit of an American company.

On Joe Rogan last week, Zuckerberg said that the U.S. government “should be defending its companies, not be the tip of the spear attacking its companies.” And yet, in this case, the U.S. government—the Biden administration that he has been railing against as he pivots to MAGA—has squarely aimed its spear at Meta’s biggest, most credible competitor in a move that would greatly benefit Zuckerberg and his company.

Everything that Zuckerberg is doing right now–Meta’s shift rightward; its dehumanizing of immigrants and LGBTQ+ users and employees; its move away from diversity-focused hiring; his trips to Mar-a-Lago; removing tampons from the men’s bathrooms at Meta offices; the inauguration party–should be seen in the broader context that Meta would benefit enormously from a TikTok ban and that Donald Trump, who will be sworn in as president on Monday, is the one person who, at this point, credibly has the ability to reverse a ban. 

Zuckerberg’s very public pledge of fealty to Trump has multiple purposes, of course. Trump previously threatened to put Zuckerberg in jail, and he is obviously cozying up to an administration that he hopes will not regulate his companies. But a TikTok ban is the biggest potential prize. Trump has Zuckerberg by his apparently very masculine balls, and is positioning himself as being the ultimate decider on what will happen to TikTok.

Zuckerberg’s political persuasions and positions have always shifted with whatever suits his companies most at that moment in time, which is something that became more clear as I went back through many hours of Congressional testimony and political speeches that Zuckerberg has given over the last few years. One thing that has not changed, however, is Zuckerberg’s obsession with using the specter of Chinese internet dominance and competition to both avoid consequences for his own company and to lay the groundwork for government regulation on Chinese platforms like TikTok.

Meta has denied directly lobbying on the TikTok ban, but the company spent a record sum lobbying in 2024, including on “Homeland Security” topics. In March 2022, the Washington Post reported that Meta paid a firm called Targeted Victory to push the narrative that TikTok is dangerous to children. And Zuckerberg himself has spent the last five years painting a picture to Congress that his monopolistic company faces great competition, actually, from Chinese companies and more importantly from China itself. This story has served Meta extraordinarily well, as he has been able to distract from Meta’s myriad privacy violations and monopolistic actions by saying it would be worse if China wins. Meta is not a monopoly, he says. It is a company fighting on behalf of America against China and Chinese companies for the soul of the internet.

Zuckerberg made this argument most clear at a speech at Georgetown University in 2019. 

“The larger question about the future of our global internet. You know China is building its own internet focused on very different values, and it’s now exporting their vision of the internet to other countries,” Zuckerberg said. “Until recently, the internet in almost every country outside of China has been defined by American platforms with free expression values. But there’s no guarantee that those values will win out. A decade ago almost all of the major internet platforms were American. Today, six of the top 10 are Chinese and we’re beginning to see this in social media too.”

“While our services like WhatsApp are used by protesters and activists everywhere due to strong encryption and privacy protections, on TikTok, the Chinese app growing quickly around the world, mentions of these same protests are censored even here in the US,” he added.

A few days after Zuckerberg’s Georgetown speech, Zuckerberg had a seven-hour hearing before the House Financial Services Committee in which he stated that an entire cryprocurrency system Facebook was spinning up called Libra was so incredibly important to U.S. financial and cultural dominance that if Congress imposed restrictions on it, Xi Jinping would win. (Libra, later called Diem and then sold to Silvergate Bank in 2022, is now dead.)

“I think there are completely valid questions about how a project like this would impact America’s financial leadership, our ability to impose sanctions around the world, our oversight of the financial system in a lot of places,” Zuckerberg said. “And I just think that we need to trade off and think about and weigh any risks of a new system against what I think are surely risks if a Chinese financial system becomes the standard.”

This general reasoning from Zuckerberg prompted Rep. Anthony Gonzalez to say that “you’re painting this as if we don’t do it, China will do it. I think you’ll be hard pressed to find somebody who is more of a hawk on China in this committee. So I agree with that. What I don’t think is the right frame is ‘If Mark Zuckerberg and Facebook don’t do it, then Xi Jinping will.’ Like, this isn’t Mark Zuckerberg versus Xi Jinping … Facebook doing this, frankly, I don’t trust it and I don’t believe the American people trust it.”

This endeavor of watching Zuckerberg’s old testimony made clear that he lied repeatedly on Joe Rogan last week on all sorts of things. For example, he practically begged Congress to regulate Meta in July 2020—during the Trump administration—while groveling about how seriously the company takes things like COVID misinformation and election integrity. On Rogan, he suggested such concerns were foisted upon him by the media and the Biden administration. Zuckerberg also seemed dumbfounded on Rogan that the Consumer Financial Protection Bureau and Elizabeth Warren were interested in his company, which again, notoriously attempted to launch an entirely new global monetary system with Libra.

“We had organizations that were looking into us that were, like, not really involved with social media,” Zuckerberg told Rogan. “Like, I think, like, the CFPB, like, this financial I don't even know what it stands for. It's the, it's the financial organization, that Elizabeth Warren had set up. And it's basically like, we're not a bank.”

During another Congressional hearing in 2018, Zuckerberg’s prepared notes said “Break Up FB? U.S. tech companies key asset for America; break up strengthens Chinese companies.” And, again, in 2020 he told Congress: “If you look at where the top technology companies come from, the vast majority a decade ago were America. Today, almost half are Chinese … Facebook stands for a set of basic principles. Giving people voice and economic opportunity. Keeping people safe. Upholding democratic traditions like freedom of expression and voting and enabling an open and competitive marketplace. These are fundamental values for most of us, but not for everyone in the world. Not every company we compete with or the countries they represent. As global competition increases, there is no guarantee that our values will win out.” 

Meta, and, specifically, Instagram Reels, would be the most obvious beneficiary of a TikTok ban. TikTok has now what Facebook once had, and which Instagram has but is losing: Deep cultural relevance and a generation of users who love it. Facebook and Instagram still have billions of users, but AI spam, a terrible algorithm that seemingly universally surfaces cringe, and a huge number of bots and people who post like their brains are made of mashed potatoes have made both Facebook and Instagram feel like platforms that people remain on begrudgingly, not because they actually want to be there. 

Zuckerberg’s general narrative that Meta faces intense competition from China and that Chinese social media companies cannot be trusted persisted across all of his Congressional hearings, and he has repeatedly used the specter of China exporting its cultural values via social media as a shield to deflect from his own company’s monopolistic tendencies, its privacy violations, and its harms against children. He has used this not only to avoid regulation of his company and his platforms, but also to seed the ground for what ultimately became the TikTok ban. Zuckerberg can see the prize. The question is whether, by kissing Trump’s ass, he will actually finally get it.

One of my first cellphones was a Nokia 3310. I still miss it: Launched in 2000, it was a solid worry-stone of a phone with rubbery keys that I could text my friends on during class from under the desk without breaking eye contact with my math professor. I played Snake when I felt awkward at parties and hearing the “kick” ringtone in 2025 is like hearing an ancient folk song. Playing Snake, texting, and scrolling through ringtones was pretty much all I could do with that phone, actually. It was plenty.

The iPhone was still seven years away, and for that reason, mobile phone aesthetics were still finding themselves. On Wednesday, Aalto University in Finland introduced the Nokia Design Archive, an “uncurated repository” containing 20,000 items and 959GB of files showing the designing process, imagining and ideating, and often wacky concepts for how we might use communications devices on the move in the then-future. The content—spanning from the mid-90s to 2017 according to a press release from the university—was licensed from Microsoft Mobile for research and education purposes, but is now open to the public to peruse. 

With Meta’s recent speech policy changes regarding immigration, in which the company will allow people to call immigrants pieces of trash, Mark Zuckerberg is laying the narrative groundwork for President-elect Trump’s planned mass deportations of people from the United States. 

Multiple speech and content moderation experts 404 Media spoke to drew some parallels between these recent changes and when Facebook contributed to a genocide in Myanmar in 2017, in which Facebook was used to spread anti-Rohingya hate and the country’s military ultimately led a campaign of murder, torture, and rape against the Muslim minority population. Although there are some key differences, Meta’s changes in the U.S. will also likely lead to the spread of more hate speech across Meta’s sites, with the real world consequences that can bring.

“We believe Meta is certainly opening up their platform to accept harmful rhetoric and mold public opinion into accepting the Trump administration's plans to deport and separate families,” Citlaly Mora, director of communications at Just Futures Law, a legal and advocacy organization focused on issues around deportation and surveillance. 

Zara Dar’s six-minute explainer video “So what are Integrals?” has a little over half a million views with 450 likes and an 87 percent positive “thumbs up” review rate. Commenters have said the video, which is a short introduction to one of the fundamental operations of calculus, is “educational,” “great,” and “just incredible.” 

“I have to say, as an aerospace engineering student, that you have explained the usefulness of integrals better than my university professor,” one commenter said. 

Only a few commenters complained or demanded that Dar take her clothes off, which is impressive because the video and comments were posted to her Pornhub channel, where Dar has been posting educational videos like “What is a neural network?” and “Intuitive Approach to Understanding Probability” for the last year. 

When I asked her why she thinks her videos are gaining traction on a porn site, Dar said “I’m not entirely sure, but it could be because my SFW videos stand out against the typical NSFW content on the platform. That contrast might make them more intriguing or refreshing to viewers. But that’s just my speculation.” 

Dar, who dropped out of grad school to be a content creator full-time (she also has a Pornhub video about that choice), also posts the same videos to YouTube and has an OnlyFans where she posts adult content, where she says she’s made over a million dollars. 

On Wednesday, while the Supreme Court heard oral arguments in a case challenging a law in Texas that requires age verification for people to view porn sites, Justice Samuel Alito asked if Pornhub was “like the old Playboy magazines,” meaning does the site just offer pornography or content that is not pornographic as well. The answer is that it doesn’t have much in terms of written articles, but there’s a long history of posting non pornographic content to Pornhub. As I wrote almost a decade ago, some people do it as a goof, but it’s also not the worst place to monetize one’s videos, even if they’re not porn. In fact, I learned about Dar’s channel via a post she made on Linkedin, in which she explained that the same educational videos she shares on YouTube make more money per million views on Pornhub because Pornhub offers better rates—$1,000 per million views on Pornhub versus YouTube’s $340 per million views. 

For reasons she doesn’t understand, Dar’s Linkedin account was banned after that Linkedin post started going viral.

“My account was banned, seemingly because of this post,” Dar told me. “I received an email stating, ‘We recently removed your profile photo because it does not appear to be a photo of you,’ which was confusing since LinkedIn has never requested ID verification from me. When I tried to log in to update my photo, I discovered my account was banned.”

Dar tried contacting Linkedin support but did not receive a response and she’s still unable to access her account. Linkedin also did not respond to my request for comment. 

“I have no issue providing LinkedIn with my ID for verification (as I have done so for many other platforms), but I was under the impression that it’s optional,” Dar told me. “It’s frustrating that verification isn’t required for other users, yet my account was banned for sharing a straightforward fact about my content creation career. I was engaging professionally, but LinkedIn’s strict handling of this situation feels counterproductive to its purpose.”

In the message sent to her by Linkedin asking her to change her profile photo, Linkedin said that profile photos can get flagged for a variety of reasons, including for “being considered offensive.” There was no nudity in Dar’s profile photo.

Ironically, being removed from social media with little explanation or recourse is another good reason for people to share their content on Pornhub. We don’t know exactly why Dar’s Linkedin account was removed because Linkedin won’t explain, but it’s possible it was reported because it was reported by users after going viral on the LinkedinLunatics subreddit, where people share Linkedin posts they feel don’t belong on the professional networking platform. As we’ve reported over the years, this is something that happens to sex workers and adult content creators on other social media sites all the time.  

Dar said she makes more money on YouTube overall because that’s where she gets more views, but in addition to higher rates, an added benefit of posting to Pornhub is that Pornhub is not likely to ban her for sharing adult content elsewhere on the web.

“I can't believe this website is banned in Texas!” one commenter said on Dar’s Pornhub video on pi. “They're trying to hinder our education.”

The Biden administration and the states of Illinois and Minnesota sued tractor and agricultural manufacturer John Deere Wednesday, arguing that the company’s anti consumer repair practices have driven up prices for farmers and have made it difficult for them to get repairs during critical planting and harvesting seasons. The lawsuit alleges that Deere has monopoly power over the repair market, which 404 Media has been reporting on for years.

The lawsuit, filed by the Federal Trade Commission and the attorney generals of Illinois and Minnesota, is the latest and most serious legal salvo against Deere’s repair monopoly. Deere is also facing a class-action lawsuit related to its repair practices from consumers in Illinois that the Department of Justice and other federal entities have signaled they are interested in and support, as we reported last year. 

“The Federal Trade Commission today files suit against agricultural equipment manufacturer Deere & Company, stating that it has illegally restricted the ability of farmers and independent technicians to repair Deere equipment, including tractors and combines,” FTC commissioner Lina Khan wrote in a formal comment explaining the decision. 

Deere has become notorious for cornering the repair market on its machines, which include tractors, combines, and other major agricultural equipment by introducing software locks that prevent farmers from fixing the equipment they buy without the authorization of John Deere.

It has also made repair parts difficult to come by. Deere previously promised to make certain repairs easier for consumers with a “Memorandum of Understanding” (MOU) signed with a farming organization that would have made it possible for farmers to do some repairs and obtain some specific parts; implementation of that MOU has been incredibly uneven, according to farmers. In October, Sen. Elizabeth Warren said that Deere was not honoring that agreement and demanded answers to several questions about it; Deere has not yet responded.

The FTC lawsuit specifically states that that MOU was designed to kill right to repair legislation and repair regulation against the company.

"Deere invoked its release of Customer Service ADVISOR and its MOU with the Farm Bureau to stymie state 'right-to-repair' legislation that would otherwise have required Deere to make fully functional repair tools available to customers," it stated.

It also highlights the fact that Deere has released a version of its repair software, called "Service Advisor," to the public (which costs $3,160 per year). But the version of the software released to the public can only do certain repairs and is not fully functional.

"Deere offers two versions of its electronic repair tool: (1) Full-Function Service ADVISOR, a fully functional repair tool that Deere makes available only to Deere dealers, and (2) a degraded Customer Service ADVISOR, which Deere licenses to equipment owners, IRPs [Independent repair providers], and others," it states.

"Deere has acquired and maintained monopoly power in a relevant market for the provision of repair services that require the use of a fully functional repair tool. Through its limited distribution of the repair tool, Deere controls entry into, and limits output in, the provision of such services," the lawsuit added. "As a consequence, Deere’s dealers are able to maintain a 100% market share and charge supracompetitive prices for restricted repairs, and Deere itself reaps additional profits through parts sales."

Farmers have told 404 Media that they remain unable to do many types of repairs, and that it can sometimes take days for “authorized” John Deere or John Deere dealer technicians to come fix broken equipment. In farming, this delay can result in lost harvest, crucial delays in planting, and dying crops during critical periods of the farming season. 

“These delays can mean that months of hard work and much-needed income vanish, devastating their business. In rural communities, the restrictions can sometimes mean that farmers need to drive hours just to get their equipment fixed,” Khan wrote. “For those who have long fixed their own equipment, these artificial restrictions can seem especially inefficient, with tractors needlessly sitting idle as farmers and independent mechanics are held back from using their skill and talent.”

The lawsuit, in the waning days of the Biden administration, is the most serious punitive act the federal government has ever taken to break up a repair monopoly and to support consumers’ right to repair. For years, the FTC has issued reports about repairability and manufacturer dominance of the repair market, but aside from a few small fines, has not formally sued any company. The steps Deere has taken to secure a repair monopoly are among the most egregious of any manufacturer in any industry, which has led farmers in some cases to resort to hacking their own tractors for the purposes of repair, sometimes using software pirated from Ukraine and other countries. 

“We shouldn’t tolerate companies blocking repair,” Nathan Proctor, consumer rights group PIRG's Senior Right to Repair Campaign Director, said. “When you buy something, you should be able to do whatever you want with it. The FTC’s enforcement action will help farmers, and everyone else who believes people should be able to fix their stuff.”

The Oscars of porn and the industry’s largest trade show will begin in one week, on January 22 at Virgin Hotels in Las Vegas. Inside, adult performers, producers, directors and crew will find out whether they’ve won the prestigious AVN Award in categories including “Best All-Girl Group Sex Scene” and “Best Big Butt Movie or Collected Release,” but outside, hospitality workers are picketing for a better union contract. It’s a massive week for the adult industry, but porn workers are conflicted about crossing the picket line, with many choosing to sit this one out entirely.

The Adult Entertainment Expo (AEE) and the Adult Video Network Awards, collectively often referred to as AVN, is expected to bring more than 45,000 attendees to Vegas for three days of signings and meet-and-greets for fans, panels and programming about the adult industry, and booths for hundreds of brands, studios, and sex toys. It’s also the biggest event of the year for the industry’s indie performers, who come out to shoot content and network, spending weeks and months planning agendas and setting up unofficial parties and meetups. The show has been going for decades—since 1998 when it shared space and dates with the Consumer Electronic Show, and later at the Hard Rock Hotel when vendors and performers were forced out of CES and started their own event a few weeks and a couple blocks away. 

Culinary Workers Union Local 226 and Bartenders Union Local 165, Nevada affiliates of UNITE HERE, represent 60,000 workers in Las Vegas and Reno according to their website. They’ve been on open-ended strike since November, and are fighting for a new five-year contract for 700 workers. Union representatives said last month that the latest offer from Virgin Hotels Las Vegas was a 30-cent-a-year wage increase, which they rejected.

Virgin Hotels, formerly the Hard Rock before renovations and a rebrand in 2020, was AVN’s first home when it split from CES’ dates and venue in 2011 (AVN became its own event in 1998 but shared space and time with the mainstream electronics show, where attendees would drift from one to the other). Today, AVN is actually multiple events, including the AEE and the awards but also the Adult Novelty Expo (ANE), interNEXT Expo, GayVN Awards Show, and the “O” Awards. In 2020, when the hotel reopened, Virgin Hotels Las Vegas CEO Richard Bosworth called AVN and other conferences “very cherished clients.” 

Performer Electra Rayne told me on Monday she’s still planning to boycott the event as long as the strike is ongoing. She corresponded with the Culinary Union, and confirmed with them in an email she posted to X that striking workers are asking people to refrain from patronizing the venue in solidarity, which the union has since said publicly in statements. 

 “Thankfully I'm local so it's not like I spent a ton of money on travel or anything, and I'm supposed to go sign at the booth for my agent who has been super understanding about me needing to cancel should the strike be ongoing (which it seems it will be),” Rayne said. “I'm bummed to not see my fans, but the really sad part is that I was in a feature that's up for multiple awards this year, and I won't be able to celebrate that at the awards with the cast. We worked so hard on that movie and I'm so proud to have been a part of it, it's a really bummer to not get to participate in that moment.” 

Other people in the industry are planning to attend despite the strike, but still find the decision difficult. “It saddens me to hear that Virgin is not paying their workers a fair wage, as that is something everyone deserves. With that said, I will still be attending AVN as it is the biggest convention of the year for my industry and I do have agreements and obligations that were arranged far ahead of these strikes,” performer Leya Falcon told me. “I do find it quite disturbing to hear that some of those on strike are ‘allegedly’ harassing and sometimes even following in an intimidating manner those that decide to enter the property as we have nothing to do with their employment situation, we are simply doing our job, which to be fair, does not always pay fairly either. [The Culinary Union did not respond to a request for comment on this alleged harassment, but we will update if we hear back.] Just as they are working to care for their families, we are doing the same and we should not be bullied for it, especially in this economy. I do hope that they can see the other side of the coin here, much as we do see theirs, and understand that none of us are against them receiving fair pay, we are simply doing what we need to do to provide for our families as not all of us have the luxury to just decide to sit this one out.” 

Performer London River wrote on X that she would also still attend. “I likely can’t pull out of my commitments to AVN without experiencing losses. Considering that, I will be donating a percentage of my profits from the event to the strike fund,” she wrote, linking to a donation page for their fund. Striking workers are paid $500 a week out of this fund, a representative for Local 226 told me.

AVN declined to comment, but last week, it launched a FAQ page for attendees of the event. “The hotel has assured us that the strike will not impact the AVN Expo and Awards. Contingency measures are in place to maintain a memorable guest experience and exceptional service during the event,” the page says. 

On January 9, the Adult Performance Artists Guild (APAG) issued a statement about the strike: “As the union for performers in the adult industry, care for the safety and well-being of our workers is paramount to our mission,” they wrote. “We feel this for not only our members and other workers in the adult industry but for all workers, regardless of their jobs. As union representatives, we support the sacrifices made by workers on strike, fighting for better working conditions. The officers of APAG voted unanimously to support our fellow union workers in Culinary Workers Local 226, and we will not cross their picket line in a show of solidarity.” 

APAG said it “strongly encourages” members to avoid crossing the picket line, as well as cancelling reservations at the hotel and contacting AVN, Virgin Hotels, and companies they were set to sign with at the event to express their concerns. 

And on Saturday, marking day 58 of the strike, APAG members joined Local 226 workers and their families for a march from the Las Vegas Strip to Virgin Hotels, blaring vuvuzelas and holding signs referencing Virgin Las Vegas’ contract negotiation offer of an estimated 30 cents an hour in wage increases. 

“The Adult Performance Artists Guild (APAG)’s decision to stand with Culinary Union strikers and honor the picket line at Virgin Las Vegas demonstrates unity, and the Culinary Union applauds the unwavering solidarity shown by the APAG and their members,” Ted Pappageorge, Secretary-Treasurer for the Culinary Union, said in a statement. “Workers across industries share the same fight for dignity, fair pay, respect, and protections on-the-job, and Culinary Union is proud to stand with APAG in solidarity as strikers continue to take on a billionaire-owned company that refuses to treat workers fairly. APAG’s support sends a powerful message: When workers stand together, we are unstoppable. To APAG members and all customers choosing not to cross the picket line – thank you for standing with workers on strike, with your continued support we will win.”

Last week, Pornhub announced that its team is canceling its upcoming trips to AVN and adult industry event XBIZ LA due to the strike in Vegas, and in Los Angeles, the wildfire crisis that’s required many LA residents to evacuate to hotels. “Unfortunately, this means our scheduled in-person events and workshops are no longer going on as planned. We are working closely with XBIZ to host our scheduled workshop virtually,” they wrote in a post on X. “We're so disappointed we won't be able to get face time with those of you who were planning to attend, and looking forward to seeing you all again soon.”

If workers don’t receive a fair contract soon, AVN won’t be the only event where attendees will have to choose between staying home or crossing a picket line. There are 11 other trade shows and events set to take place at Virgin Hotels in the coming weeks, including the Fancy Food Show, World of Concrete Expo, Academic Surgical Congress, CHAMPS Trade Show, and Kitchen and Bath Industry Show. 

We've got much more on what is happening inside Meta with the company's recent speech policy changes. Jason runs us through it. After the break, Joseph explains how thousands of apps have been hijacked to steal your location data, possibly without the app developers' knowledge. In the subscribers-only section, we talk about various stories intersecting with the LA fires, such as Amazon delivery drivers and AI images. (YouTube version to come shortly.)

Listen to the weekly podcast on Apple Podcasts, Spotify, or YouTube. Become a paid subscriber for access to this episode's bonus content and to power our journalism. If you become a paid subscriber, check your inbox for an email from our podcast host Transistor for a link to the subscribers-only version! You can also add that subscribers feed to your podcast app of choice and never miss an episode that way. The email should also contain the subscribers-only unlisted YouTube link for the extended video version too. It will also be in the show notes in your podcast player.

• ‘It’s Total Chaos Internally at Meta Right Now’: Employees Protest Zuckerberg’s Anti LGBTQ Changes
• Meta Deletes Trans and Nonbinary Messenger Themes
• Candy Crush, Tinder, MyFitnessPal: See the Thousands of Apps Hijacked to Spy on Your Location
• Online Behavioral Ads Fuel the Surveillance Industry—Here’s How
• People Think AI Images of Hollywood Sign Burning Are Real
• An Amazon Delivery Confirmation Photo Is the Last Time a Palisades Resident Saw Her Burnt Down House
• ‘We’re Fine’: Lying to Ourselves About a Climate Disaster

An AI app for creating nonconsensual nude images of anyone is getting the vast majority of its traffic directly from Meta platforms, where the app is buying thousands of explicit ads featuring nonconsensual nudity of celebrities and influencers. The blatant and repeated violation of Meta’s policies over the course of months is making a mockery of the company’s ability or willingness to moderate a known bad actor that at the moment appears to get the majority of its users by paying Meta directly for ads.

The app, known as Crushmate or Crush AI, has been buying ads on Facebook, Instagram, and other Meta platforms since at least early September. As first reported by Alexios Mantzarlis in his Faked Up newsletter, according to internet traffic analysis firm Similarweb, three of the domains Crush uses had around 240,000 visitors combined, with 90 percent of that traffic coming from Facebook or Instagram.

I’ve seen Meta remove some of these ads since September, but at the time of writing the same three domains that were advertised on Meta platforms and redirected to Crushmate’s services had around 350 active ads and more than 5,000 ads overall. 

Most of the recent ads use the same format. They take a video a woman posted to Instagram or TikTok and show how a user can pause the video on any frame and create a nude image of her. Many of the ads, which are still active, do this to videos of the extremely popular OnlyFans creator Sophie Rain, who made headlines recently for making $43 million in one year on OnlyFans. As Mantzarlis points out, one ad nudifies Mikayla Demaiter, a model with 3.2 million followers on Instagram. Rain and Demaiter did not respond to a request for comment.

Other ads feature other real women I wasn’t able to identify and AI generated women with their clothes being “erased” by the app. 

In early September, a 404 Media reader also tipped me that Crushmate was advertising its services on Facebook Marketplace.

I’ve confirmed that all these ads lead to the same Crushmate service that will create nonconsensual nude images and offers some of its services via a subscription plan. 

I’ve recently reported about Meta running ads that feature explicit nudity, including dozens of ads that are just close up images of vaginas. I’ve also reported repeatedly about “nudify” apps buying ads on Meta platforms. When we’ve flagged these ads to Meta in the past, they removed them. Meta has also removed associated Facebook pages that are buying the ads, but Crushmate has found an easy workaround that is clearly paying off: It creates multiple Facebook pages with AI-generated profile images that look like normal people, then buys ads promoting new, different URLs that redirect to to Crushmate. 

Meta did not respond to specific questions about why it’s not detecting and removing the offending ads for featuring nonconsensual nudity. As I reported last week, extensive testing by AI Forensics, a European non-profit that investigates influential and opaque algorithms, found that nudity uploaded to Instagram and Facebook as a normal user was promptly removed for violating Meta’s Community Standards. The same exact visuals were not removed when they were uploaded as ads, showing that Meta has a different standard for enforcement when it’s getting paid to push images in front of users. 

“Meta prohibits ads that promote adult sexual exploitation. We have removed the violating content, enforced against violating urls, and have taken action against the associated accounts and users,” a Facebook spokesperson told me in a statement. “This is a highly adversarial space and bad actors are constantly evolving their tactics to avoid enforcement, which is why we continue to invest in the best tools and technology to help identify and remove violating content.” 

Meta removed the ads promoting the three Crushmate domains after Mantzarlis flagged them to the company. Around 230 of the same ads promoting a fourth Crushmate domain Mantzarlis found after reaching out for comment are still live on Meta’s platforms. 

As we’ve previously reported, these nudify apps are some of the most harmful applications of generative AI because they make it so easy to create nonconsensual images of anyone. In the last two years, we’ve seen several examples of these apps being used by minors to create images of other minors. Last year, a survey found that 1 in 10 minors reported that their friends or classmates have used AI tools to generate nudes of other kids. As the Crushmate ads show, minors don’t need to go to the dark corners of the web in search of these tools. Meta is getting paid to popularize them.