Content warning: this article contains graphic descriptions of violence against people and animals.

An “error” in Instagram Reels caused its algorithm to show some users video after video of horrific violence, animal abuse, murders, dead bodies, and other gore, Meta told 404 Media. The company said “we apologize for the mistake.”

Sometime in the last few days, this error caused people’s Reels algorithms to suddenly change. A 404 Media reader who has a biking-related Instagram account reached out to me and said that his feed, which is “typically dogs and bikes,” had become videos of people getting killed: “I had never seen someone being eaten by a shark, followed by someone getting killed by a car crash, followed by someone getting shot,” he told 404 Media. 

To test this, the person let me login to his Instagram account, and I scrolled Reels for about 15 minutes. There were a couple videos about dogs and a couple videos about bikes, but the vast majority of videos were hidden behind a “sensitive content” warning. I will describe videos I saw when I clicked through the warnings, many of which had thousands of likes and hundreds of comments: 

• An elephant repeatedly stepping on and flattening a man
• A man attacking a pig with a wrench
• A close-up video of someone who had just been shot in the head
• A woman crying while laying on top of a loved one who had just been shot to death
• A man on a motorcycle stopping next to a pedestrian and shooting them in the head with a pistol
• A pile of dead bodies in what looked to be a war-type situation
• A small plane crash in front of a crowd of people
• A group of people beating a crocodile to death
• A few videos by an account called “PeopleDeadDaily” 
• A man being lit on fire
• A man shooting a cashier at point blank range

Most, but not all, of these videos were behind a “sensitive content” warning label, which won’t play the video unless you click a button that says “see reel”. In this person’s feed, I also saw video after video of people getting jumped, attacked, into fistfights, and being hit by cars, which were not behind sensitive content warning labels. A close-up video of a person falling out of a tower of terror-style amusement park ride, and patrons screaming, was also not behind a warning label. 

When this user reached out to me and told me he was seeing almost exclusively gore videos on his Reels algorithm, I wondered if it would be worth writing a story at all, because all of us at 404 Media regularly see incredibly disturbing content on Instagram. Was this any different than a normal day on Instagram, or had he just ended up on the “death algorithm” as Sam and Emanuel both called it and have both ended up on? Even the person who initially told me about this had this thought: “I’ve been telling people about this and hearing, ‘Oh it’s cuz u liked something, maybe someone u followed changed their profile,’” he said. “I feel like I’m in the Twilight Zone. I feel like nobody I’ve talked to today understands how disturbing what is being pushed is.”

Then I looked at the comments of many of these videos and it became clear that the issue was widespread. Here are comments I saw on the gore videos:

• “What happened today? My feed is full of shoot, killing, thriller clips, fighting, killing human/animals, murder, and torture. Feed got f**cked”
• “Today’s algorithm showed me around 70 murders, 100+ accidents, and around 115 violence videos, is anyone on Instagram noticing it?”
• “Bro today’s feed is not for beginners”
• “Por que me sale Gore en mi ap de racismo?” (Why am I getting gore on my racism app?)
• “Yo wtf is ig becoming”
• “What is Instagram on today?”
• “Algorithm is insane today” 
• “Why the fuck I have so many like this today???”
• “This isn’t a normal day, some DARKWEB shit going on here.”
• “Not Instagram more like instagore.”

Nearly every recent post on the Instagram subreddit, meanwhile, is about gore. Post titles include “Instagram is now 100% gore,” “Guys I can’t literally sleep,” “Meta can’t get away with this,” “This recent ‘situation’ has finally helped me make the decision to delete Instagram,” “What’s happening with Instagram? Violent reels everywhere for the last 24 hours!,” “can’t believe my feed full of cute cat videos got replaced with this shit,” a video of someone scrolling through a dozen straight “sensitive content” label videos, a post referring to children being traumatized while scrolling, various conspiracies about why this may have happened, and more. 

“I feel like I lost some humanity today when I exposed myself to seeing so many of those types of videos, Gore, deaths and what broke me the most was animal cruelty,” one post read. “Excuse me if I sound a bit exaggerated but death, pain and human suffering is something that makes me sick, I always imagine ‘What would happen if I or my family were there?’ i can't sleep when my mind replays those videos.”

In an email, a Meta spokesperson told 404 Media “We have fixed an error that caused some users to see content in their Instagram Reels feed that should not have been recommended. We apologize for the mistake.” They said that the problem does not have anything to do with Meta’s recent announcement that it would loosen some content moderation rules. 

To prove that this was actually happening, I sent Meta six links to graphic reels. These included two videos of people getting shot in the face, a video of a dead body with no context or obvious news value, a person getting lit on fire, the account called “PeopleDeadDaily,” and the tower of terror video. None of these videos have been deleted. 

One of the many problems preventing people from actually holding Meta to account for any of this is that everyone’s feed is so incredibly personalized. Like I said, because we report on the darker corners of the internet, my Instagram feed is full of horrific things on a daily basis, which is probably not everyone’s experience. 

When we talk about things like “content moderation,” the vast majority of the job is deleting videos of terrorism, murder, horrific violence, and things like this, not censoring specific viewpoints, which is part of why the job itself is so traumatizing. Meta has signaled that it intends to do less content moderation overall. And so the future of Instagram, for us all, may be one where you login to see what your friends are up to and instead have videos of people getting murdered shoved algorithmically into your feed.

A U.S. soldier who recently pleaded guilty to hacking AT&T and Verizon communicated with an email address that he believed belonged to a foreign country’s military intelligence service and attempted to sell the service stolen data, according to newly filed court records reviewed by 404 Media. The court document also says that the soldier searched for “U.S. military personnel defecting to Russia.”

The court filing in the case of Cameron John Wagenius, who used the handles kiberphant0m and cyb3rph4nt0m, discusses Wagenius’ unlawful posting and transferring of confidential phone records, including records belonging to high-ranking public officials. 404 Media previously revealed how hackers linked to the AT&T breach mined it for records associated with members of the Trump family, such as Melania and Ivanka Trump, Kamala Harris, and Marco Rubio’s wife. The court document does not say what specific data Wagenius tried to sell to the foreign intelligence service, or who that data belonged to.

The surveillance company Flock sent the creator of a website that maps its license plate-reading cameras a cease and desist letter demanding that he immediately stop using the name “DeFlock” on his website.

404 Media previously wrote about DeFlock, an open source mapping project created by Will Freeman that tracks the locations of automated license plate readers (ALPRs) from Flock and other companies. DeFlock currently maps more than 16,000 ALPRs around the world, which includes both Flock cameras as well as many created by Motorola. 

Late last month, Flock’s lawyers sent Freeman a letter demanding that he immediately “Cease and desist all use of the name ‘DEFLOCK’ or any variation thereof, remove all instances of ‘DEFLOCK’ from your Website, advertisements, promotional materials, and any other content, and Refrain from adopting any trademarks, trade names, or branding continue to or likely to cause dilution by blurring, dilution by tarnishment, and false advertising with respect to the Flock Marks in the future.” 404 Media has obtained a copy of the letter and uploaded it here.

“It has come to our attention that you are maintaining a website and promoting a project entitled DEFLOCK, which purports to track automated license plate readers across the country, and discusses the alleged dangers of [the surveillance]. While Flock believes in open debate, it takes misuses of its intellectual property seriously,” the letter, written by Sarah M. Katz of the law firm Adelman Matz, says. “While Flock does not object to the free dissemination of truthful information, your use of the Flock Marks as part of your brand DEFLOCK is being wrongfully used to make false statements about Flock and its products and is damaging both its reputation and the goodwill associated with the Flock Marks.”

It is not clear what “false statements” Freeman is making about Flock. The letter says that it should not be called DeFlock because not all of the cameras tracked by DeFlock are Flock cameras (some are Motorola), and says “the website also implies that various license plate readers are vulnerable to security hacks, which given that all of the readers are being imputed to Flock, provides a false impression about the security of Flock Products.” On the front page of DeFlock, there is a link to a 404 Media article about a security vulnerability in Motorola ALPRs. A security researcher on YouTube and Freeman previously showed that certain Motorola ALPRs are leaking data online, and 404 Media wrote about that research. The DeFlock site says “BREAKING: Anyone Can Access Motorola ALPR Data” and links to our article, but makes no claims about Flock ALPR security. 

Freeman is being represented by the Electronic Frontier Foundation and is not going to change his website, Cara Gagliano, a senior staff attorney at the EFF said in a response to Flock: “The claims alleged in your letter are groundless, and Mr. Freeman will not be complying with your demands,” the EFF’s letter says. “Because there is no legal basis for your demands, my client declines to comply with them.”

The cease-and-desist letter shows that Flock is both aware of the DeFlock website and is threatening Freeman with legal action. Flock’s letter argues that DeFlock is causing the “dilution” of Flock’s trademarks rather than “infringement” of them. This is a crucial distinction, Gagliano said. 

Companies can sue for trademark infringement when they believe that a consumer is likely to confuse the infringing product for the real one; dilution cases only apply to “famous” trademarks and can be pursued when a similar product would undermine or tarnish the brand of the original. Gagliano says in her letter that DeFlock is not diluting the Flock brand because it was specifically made for the noncommercial criticism of the surveillance company.

“Federal anti-dilution law includes express carve-outs for any noncommercial use of a mark and for any use in connection with criticizing or commenting on the mark owner or its products,” Gagliano wrote, adding that a false advertising claim made in Flock’s letter does not apply because DeFlock is a noncommercial website. 

“DeFlock is a grassroots project that aims to ‘shine a light on the widespread use of ALPR technology, raise awareness about the threats it poses to personal privacy and civil liberties, and empower the public to take action.’ It pursues that mission by providing information about ALPRs and maintaining an interactive, crowd-sourced map of ALPR installations,” she added. “The name ‘DeFlock’ references the project’s goal of ending ALPR usage and Flock’s status as one of the most widely used ALPR providers.”

Gagliano told 404 Media that Flock’s attempt to go after Freeman and DeFlock on a dilution claim raises serious free speech concerns. “Flock's choice to claim dilution rather than infringement is telling. Infringement requires showing that consumers are likely to be confused by the use; Flock clearly realizes how implausible that is here,” she said. “Dilution is a much more nebulous concept that we think raises serious constitutional questions. It's fortunate that dilution laws typically have enough explicit exceptions for claims to fail in their face in cases like this, but it's still much too broad a doctrine with little to justify it.”

Flock did not respond to a request for comment.

Update: After this article was published, Bluesky restored Kabas' post and told 404 Media the following: "This was a case of our moderators applying the policy for non-consensual AI content strictly. After re-evaluating the newsworthy context, the moderation team is reinstating those posts."

Bluesky deleted a viral, AI-generated protest video in which Donald Trump is sucking on Elon Musk’s toes because its moderators said it was “non-consensual explicit material.” The video was broadcast on televisions inside the office Housing and Urban Development earlier this week, and quickly went viral on Bluesky and Twitter. 

Independent journalist Marisa Kabas obtained a video from a government employee and posted it on Bluesky, where it went viral. Tuesday night, Bluesky moderators deleted the video because they said it was “non-consensual explicit material.” 

“A Bluesky account you control (@marisakabas.bsky.social) posted content or shared a link that contains non-consensual explicit material, which is in violation of our Community Guidelines. As a result of this violation, we have taken down your post,” an email Kabas received from Bluesky moderation reads. “We trust that you will understand the necessity of these measures and the gravity of the situation. Bluesky explicitly prohibits the sharing of non-consensual sexual media. You cannot use Bluesky to break the law or cause harm to others. All users must be treated with respect.” 

Kabas is challenging the deletion. 

“Hello—the post you have taken down was a video broadcast inside a government building to protest a fascist regime,” Kabas wrote in an email back to Bluesky seen by 404 Media. “It is in the public interest and it is legitimate news. Taking it down is an attempt to bury the story and an alarming form of censorship. I love this platform but I’m shocked by this decision. I ask you to reconsider it.” 

Other Bluesky users said that versions of the video they uploaded were also deleted, though it is still possible to find the video on the platform. 

Technically speaking, the AI video of Trump sucking Musk’s toes, which had the words “LONG LIVE THE REAL KING” shown on top of it, is a nonconsensual AI-generated video, because Trump and Musk did not agree to it. But social media platform content moderation policies have always had carve outs that allow for the criticism of powerful people, especially the world’s richest man and the literal president of the United States. 

For example, we once obtained Facebook’s internal rules about sexual content for content moderators, which included broad carveouts to allow for sexual content that criticized public figures and politicians. The First Amendment, which does not apply to social media companies but is relevant considering that Bluesky told Kabas she could not use the platform to “break the law,” has essentially unlimited protection for criticizing public figures in the way this video is doing. 

More importantly, the video Kabas posted was not a video Kabas made herself or that was totally devoid of context. As Kabas notes in her email back to Bluesky, the video was being played on television screens within a federal government office building, an obvious act of protest that she was reporting on, and an obviously newsworthy video when considering the context that the federal government is currently being gutted by these two men. (For what it's worth, Kabas has been doing some of the best reporting on Musk's dismantling of the federal government on her website The Handbasket.)

Content moderation has been one of Bluesky’s growing pains over the last few months. The platform has millions of users but only a few dozen employees, meaning that perfect content moderation is impossible, and a lot of it necessarily needs to be automated. This is going to lead to mistakes. But the video Kabas posted was one of the most popular posts on the platform earlier this week and resulted in a national conversation about the protest. Deleting it—whether accidentally or because its moderation rules are so strict as to not allow for this type of reporting on a protest against the President of the United States—is a problem.

Bluesky did not immediately respond to a request for comment.

We start this week's episode with Joseph finding out someone basically ripped off his book with a potentially AI-generated summary. Emanuel also updates us on some of the impact his reporting on AI in libraries has had. After the break, Sam tells us all about a Y Combinator supported startup that is straight-up dehumanizing factory workers. In the subscribers-only section, we talk about an apparent act of protest from inside the U.S. government involving an AI video of Musk and Trump.

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.

• SXSW event information
• A Slop Publisher Sold a Ripoff of My Book on Amazon
• Y Combinator Supports AI Startup Dehumanizing Factory Workers
• AI Video of Trump Sucking Musk's Toes Blasted on Government Office TVs

A venture capital-backed “AI performance monitoring system for factory workers” is proposing what appears to be dehumanizing surveillance of factories, where machine vision tracks workers’ hand movements and output so a boss can look at graphs and yell at them about efficiency.

In a launch video demoing the product, Baid and Mohta put on a skit showing how Optifye.ai would be used by factory bosses. 

Workers around the federal government are scrambling to figure out how and if they should respond to an all-government email sent Saturday at the behest of Elon Musk asking them to list five things they did at work within the last week. During the confusion caused by Musk’s email, workers at the Treasury Department received an email from a former Heritage Foundation staffer who is not the Treasury Secretary from an email address that billed itself as being from “Secretary of the Treasury.”

How and whether to respond to the “What did you do last week” email has itself resulted in much discussion and confusion, and efforts to clarify any confusion have resulted in additional confusion as well as worries about sharing classified or otherwise private information. FBI employees were told by new FBI director Kash Patel not to respond to the email, so were members of the military. Musk tweeted “Failure to respond will be taken as a resignation.” 

The Treasury Department email, seen by 404 Media and currently being discussed widely on Reddit, came from an email address with the name “*Secretary of the Treasury” but signed by John W. York, who is not the Secretary of the Treasury and who previously worked for the Heritage Foundation, the architects of Project 2025. The current Secretary of Treasury is Scott Bessent, not York. Treasury workers seem to not know who York is or why he is sending emails from an email address previously used by past Secretaries of Treasury. 

“It was used in the past rarely: wishing Treasury employees a Merry Christmas or noting there is a return to office mandate,” one source told 404 Media about the email address York’s email came from. “In the past, the emails included the title of the sender (Sec of Treasury, for example)."

In the email, York tells workers that they must respond to the “What did you do last week” email: “Given the voluminous and extremely important work that Treasury staff perform [sic] on a daily basis, we expect that compliance will not be difficult or time consuming.” 

“Your responses should be descriptive enough to show the significance of the work you performed; however, the descriptions should not reveal confidential, privileged, otherwise non-public, pre-decisional or deliberative aspects of that work, given that these responses will be sent outside Treasury,” he wrote. “If you have any questions about how to respond, please consult with your manager.”

Sources at the Treasury Department told 404 Media that they have not previously received any emails from John W. York, that they are not sure what his job is or whether he actually works for the Treasury Department, and that giving descriptive, substantial rundowns of their work tasks without giving “non-public” or sensitive information is not an easy task.

“John York had no title associated with his signature line (unusual as ALL Fed service employees are proud to put their title, Dept, etc in the sig line as a default),” one source told 404 Media. Employees at the Treasury Department have been doing research on York to attempt to figure out who he is. York worked for the Heritage Foundation before joining the Office of Personnel Management towards the end of Trump’s first term. His LinkedIn says he has worked as a “Strategic Human Capital Lead” at Accenture since March 2021. The Treasury Department did not immediately respond to a request for comment about whether he is now a Treasury Department employee.

Top comments on a Reddit post discussing this email are “Who the fuck is John W. York?” and “Schrodinger's phishing email. you're fired if you respond. you're fired if you don’t.” 

Other federal employees tell 404 Media that they have been receiving similar clarification emails from agency heads about how and whether to respond, and have been getting follow up emails from their supervisors about what to say if the things they work on are classified. The majority of these emails, which 404 Media is not sharing specifics on because they were in many cases sent to small teams of people, are begging employees to respond to the “What did you do last week emails” while threading the needle of sharing specifics but not sharing private or confidential information.

An AI-generated video of President Donald Trump sucking on Elon Musk’s feet, overlaid with the text “LONG LIVE THE REAL KING,” played on TV screens at the Department of Housing and Urban Development (HUD) headquarters in Washington, D.C., multiple journalists are reporting on social media.

Journalist Marisa Kabas posted the video on Bluesky, writing, “this video played on loop for ~5 mins on screens throughout the building, per agency source. Building staff couldn’t figure out how to turn it off so sent people to every floor to unplug TVs.” 

The Washington Post also obtained a recording of the televisions.

Last week, Trump called himself a king in a social media post. “CONGESTION PRICING IS DEAD. Manhattan, and all of New York, is SAVED,” Trump wrote on his platform, Truth Social. “LONG LIVE THE KING!”

Right to repair legislation has now been introduced in all 50 states, a milestone that, despite not all passing, shows the power of the grassroots political movement. Thursday, Wisconsin became the final state in the country to introduce a right to repair bill. 

So far, right to repair laws have been passed in Massachusetts, New York, Minnesota, Colorado, California, and Oregon. Another 20 states are formally considering right to repair bills during this current legislative session. The rest have previously introduced bills that have not passed; so far we have seen that many states take several years to move a given right to repair bill through the legislative process. 

Right to repair laws are designed to make it easier for consumers to fix their electronics, farmers to fix their agricultural equipment, for hospitals to fix their medical devices, and so on. Most right to repair legislation requires companies to sell repair parts to the general public, to make repair manuals available, and bans the use of technological protection measures (which are called “software locks”) that are designed to restrict repair only to authorized repair technicians. 

I have been following the right to repair movement for a decade, and, in the early days, a small group of consumer rights advocates worked in a couple states to get legislation introduced. 

Many of those bills were killed quickly by big tech lobbyists, who were successful at scaring lawmakers into believing that right to repair would make devices less safe or would be a boon for hackers. Over time, those same consumer rights advocates have been successful in convincing the general public that you should be able to fix the things you buy. The movement was endorsed by the Biden White House and Biden’s Federal Trade Commission, millions of consumers, and an increasing number of state legislators. The movement has gotten companies like Apple, Google, and John Deere to change their policies, inching toward a world where repair is more easily accessible.

"Now that Wisconsin filed their first Right to Repair legislation, we’ve completed the sweep of getting bills filed in all 50 states. Our legislative map no longer has any blanks,” said Gay Gordon-Byrne, Executive Director at Repair.org, which has been advocating for the legislation. “This proves that Right to Repair is needed everywhere—and we are well on our way towards making that happen."

“Americans are fed up with all the ways in which manufacturers of everything from toasters to tractors frustrate or block repairs, and lawmakers are hearing that frustration and taking action,” said Nathan Proctor, right to repair director for consumer rights group PIRG.

iFixit’s Kyle Wiens, meanwhile, said covering the entire map is a “tipping point” for the movement: “We’ve gone from a handful of passionate advocates to a nationwide call for repair autonomy. People are fed up with disposable products and locked-down devices. Repair is the future, and this moment proves it.”

Like some (many? most? all?) authors I sometimes check how my book is doing on Amazon and other booksellers. Recently while doing that, I came across another listing on the online retailer: “SUMMARY OF JOSEPH COX’S DARK WIRE,” referring to the book I spent years researching, investigating, and writing. It cost $4.99.

Curious whether this product was an AI-generated rip-off of my work, I bought a copy. Flicking through the digital pages, the summary, rather expectedly, condensed each of my chapters into a few page overview. Details I had gone to incredible lengths to get, including flying around the world to meet criminals face-to-face, or sneaking into a law enforcement conference, or slowly building trust with understandably scared sources was plopped into this new book with little context on how they got there or why they mattered.

For example, here is the original opening of my book, about a drug trafficker called Owen Hanson: 

Welcome back to the Abstract! 

It’s hard to keep up with all the news about all the giant gassy orbiters out there. I’m speaking, of course, about hot Jupiters, a class of planets that takes the concept of “inhospitable” to dazzling and creative new levels, and which had an epic news week.

Then, what did scientists find in cores taken from deep-sea trenches? The answer might surprise you. Next, mice administer “first aid.” Last, fish can see you for who you really are (though yummy treats will certainly not be refused). 

Hot Jupiters Are So Hot Right Now (and at All Other Times)

Seidel, Julia et al. “Vertical structure of an exoplanet’s atmospheric jet stream.” Nature.

Hot Jupiters are the low-hanging fruit of exoplanet discoveries. As the name implies, they are Jupiter-sized worlds that orbit extremely close to their stars, a proximity that makes them—you guessed it—hot. 

Given that they are both giant in scale and have short years lasting only hours or days, hot Jupiters are the easiest exoplanets to spot, which is why our catalog of distant worlds is packed with them. In fact, a study came out just this week that identified seven new ones.

But while it’s not all that novel to discover these worlds (which is kind of amazing in itself), scientists have now peered deep into the atmosphere of the hot Jupiter WASP-121, nicknamed Tylos, which is about 850 light years from Earth. It’s the first time several distinct atmospheric layers and processes have been observed on an exoplanet.   

“Ultra-hot Jupiters, an extreme class of planets not found in our solar system, provide a unique window into atmospheric processes,” said researchers led by Julia Seidel of the European Southern Observatory (ESO). “Here we show a dramatic shift in atmospheric circulation in an ultra-hot Jupiter” including “the first vertical characterization of a high-altitude, super-rotational atmospheric jet stream.”  

Tylos is slightly bigger than Jupiter, but it is so close to its star that its year lasts only 30 hours. As a consequence, it is tidally locked, meaning that one side is always facing the star, and the other always faces away. The star-lit side is about 2,300°C (4,200°F) which is, as advertised, quite hot. Using the ESO’s Very Large Telescope, the researchers spotted the aforementioned equatorial jet stream and saw flows of hot gas moving from the hot day side to the cooler night side—which is still pretty hot at around 700°C (1,340°F). 

The weather report on Tylos is permanently fatal with a chance of titanium rain, according to a third study that came out this week (that’s a hot Jupiter hat-trick). Taken together, the research represents a new emerging era of exoplanet observations in which astronomers can peek under the hood of these distant atmospheres and start to get a real vertical cross-section of otherworldly skies. 

Down the line, this will lead to better characterizations of the atmospheres of potentially habitable exoplanets, which could contain detectable signs of alien life. But for now, on this late winter weekend, let's be satisfied with warming ourselves  into certain oblivion in the bellies of hot Jupiters. 

From the Hadal to the Grave 

Hovikoski, Jussi et al. “Bioturbation in the hadal zone.” Nature Communications.

To cool off, we shall now dive straight into the deepest parts of the ocean, the hadal zone, where strange things are inherently afoot. Scientists took sediment cores from seafloors at depths of over 4.6 miles in the Japan Trench which is, in my opinion, asking for trouble. But in this case, the results revealed an activity that you might not expect to find in one of the most inhospitable places on Earth—farming.   

I should just say, the “farmers” are probably invertebrates, like sea cucumbers or bivalves, that cultivate microbes that help break down organic matter for them. Still, a basic form of “agrichnial” farming is preserved in trace fossils, like burrows, the team found in the cores. 

“The hadal zone, >6 km deep, remains one of the least understood ecosystems on Earth,” said researchers led by Jussi Hovikoski of the Geological Survey of Finland. The cores open a rare window into this otherworldly region and reveal “slender spiral, lobate and deeply penetrating straight and ramifying burrow systems…interpreted to include burrows of microbe farming and chemosymbiotic invertebrates.” 

The study also gets points for its title, “Bioturbation in the hadal zone,” which sounds like an early aughts prog rock album. \m/ 

Somebody Call an EMT! (Emergency Mouse Technician)

Sun, Wenjian et al. “Reviving-like prosocial behavior in response to unconscious or dead conspecifics in rodents.” Science.

Humans produce a lot of selfish psychos, if you hadn’t noticed, but one nice thing about our species is we generally share a prosocial instinct to help people during a medical crisis. As it turns out, we’re not alone in this behavior, according to a new study that monitored the reactions of mice to ailing, unconscious, or dead conspecifics. 

“Anecdotal observations across several species in the wild, including nonhuman primates, dolphins, and elephants have reported intriguing behaviors of animals toward unresponsive conspecifics that have collapsed because of sickness, injury, or death,” said researchers led by  Wenjian Sun of the University of Southern California. “These animals…display various behavioral responses, including touching, grooming, nudging, and sometimes even more intense physical actions, such as striking, toward the collapsed peers. Some of these actions toward incapacitated conspecifics are reminiscent of human emergency responses, especially those involving sensory stimulation.”

To bring these anecdotal reports in an experimental setting, the team videotaped mice responding to cagemates that had been anesthetized into unconsciousness, as well as their reactions  to dead mice. The r mice interacted with unconscious cage-mates  about ten times as much as with an active partner, and may have even performed basic versions of first aid.

“Our results suggest that the actions of mouth/ tongue biting and tongue pulling may have rescue-like effects, reminiscent of human first aid efforts in reviving unconscious individuals with physical stimulation and airway maintenance,” the researchers said.  

“The consequences of the behaviors, such as improved airway opening or clearance and expedited recovery, are clearly beneficial to the recipient,” they added, though they also cautioned that “it is challenging to determine the motivational needs behind these distinctive ‘reviving-like’ behaviors.”  

Familiarity played a strong role in the experiment's outcome; mice heaped much more attention on dead or unconscious cage-mates that they knew well compared to strangers. At the risk of anthropomorphizing, it’s kind of sad to think about these mice being confronted with their passed-out or dead friends, but the silver lining is an empirical validation of widespread prosocial behaviors. 

I’m also going to assume it means that the Disney franchise The Rescuers, starring mice humanitarians, is a documentary.

The Adventures of Left Hump and Friends

Tomasek, Maëlan and Soller, Katinka et al. “Wild fish use visual cues to recognize individual divers.” Biology Letters. 

The next time you go for an ocean swim, why not introduce yourself to some neighboring fish? They might learn to recognize you as an individual and start following you around, especially if you give them something nice to eat. That’s the conclusion of a new study that found fish can tell individual divers apart based on visual cues—and that they rapidly learn which divers are generous with treats (in this case: shrimp).

Researchers Maëlan Tomasek and Katinka Soller conducted several dives at the STARESO research station in Corsica, France. Soller was the designated shrimp dispenser, and the wild fish “volunteers” rapidly learned to distinguish her visually from Tomasek, the shrimp miser.

“Two species voluntarily took part in our experiments: saddled sea bream O. melanura and black sea bream S. cantharus,” said the researchers. “Of specific individuals, the saddled bream (Bernie) was first identified at dive 5 of the training, four black bream at dives 12 (Left Hump), 15 (Kasi), 19 (Alfi), 21 (Julius) and the last black bream (Geraldine) on the first session of experiment 1. Note that this marks the moment from which we were able to reliably identify them (i.e. identify with absolute certainty at each apparition from one dive to the next) but that they most likely appeared several days prior to this.”

First of all, fantastic names. I’m already shipping Julius and Geraldine as a celebrity fish couple called Juladine. Left Hump will officiate the wedding. But setting aside the fish fanfic, the team demonstrated that the fish learned to visually tell the researchers apart, leading to a clear preference for following Soller. 

“The fact that wild bream can discriminate between divers adds scientific evidence to the numerous accounts suggesting differentiated relationships between fish and specific humans,” the team said. “Our study thus encourages a reappraisal of the methodological avenues to study cognitive abilities of wild fish under natural conditions.” 

“It also demonstrates a potential difficulty when conducting such experiments that could be disturbed by fish following specific experimenters,” the researchers said, concluding with an implied wink: “Researchers might not always want to be followed all around by fish, but if they do, they will not be disappointed.”

Thanks for reading! See you next week.

The U.S. Department of Health and Human Services (HHS) is wasting workers’ time and taxpayer dollars on “a witch hunt to find any content deemed ‘bad,’” according to a source familiar with the work and internal communications viewed by 404 Media. Specifically, people who work on HHS websites are spending days scanning those sites and any documents they share in search of a list terms like “gay,” “sexuality,” “non-binary,” “inclusion,” “queer,” and “gender,” potentially so they could be later removed to comply with Trump’s executive orders attacking diversity, equity, and inclusion in the federal government.

“The most obvious issue to me about this list is that it’s being done in the name of ‘efficiency and saving money.’ It is not efficient to take engineers off their work to scan old content for any keywords this new administration hates. The bigotry is astounding,” the source who is familiar with the work and who asked to be anonymous because they were not permitted to speak to the press, told me. “If they were being true to the concept, sure, they could say that moving forward, we will no longer support creation of new data about these topics. But to go backward decades, scrubbing for stuff they hate, that’s not a savings of time and money, that’s a huge expenditure. It's hypocritical on top of it all.”

The source said that part of what makes the work so time consuming is that the current HHS administration doesn’t just want to know about every page on its sites that include these terms, but also pages that link out to .PDF files that include those terms. For example, last week we reported that the Trump administration added a note rejecting “gender ideology” on a Substance Abuse and Mental Health Services Administration’s website page that shared a .PDF of a study about substance abuse among gay, lesbian, bisexual, or other nonheterosexual adolescents. According to the source, HHS administrators want that page added to a spreadsheet of pages and documents that include the terms it's looking for because of the content of the study. 

Since HHS websites share thousands of .PDFs, the source said, “very expensive” engineers spent multiple days scanning the files for the list of terms instead of doing their regular tasks. 

Other terms on the list include “they/them” pronouns, “pregnant ‘people,’” “Biden,” and “intersex,” according to a copy of the list seen by 404 Media.

The fact that the government is wasting resources finding every instance of a term it finds objectionable directly contradicts Trump’s and Musk’s stated goal of “government efficiency.” Finding these terms in thousands of studies and papers and potentially removing them is not saving any taxpayer dollars, but just purging government sites with a perspective it disagrees with. Other agencies have also scrambled to find similar terms. Axios reported that DOGE representatives at the National Oceanic and Atmospheric Administration are searching for “DEI content” and Stat News has reported that a number of federal health agencies are searching grants for “taboo words” like “trans” and “diversity.” 

“The spitefulness is such a waste of time and money. It's infuriating,” the source said. “Sure, they might argue not to do anything inclusive or helpful in the future, but to burn so much time and money trying to scrub out any content he [Trump] hates from past decades is ... I'm kinda at a loss for words.”

At the moment, it appears that HHS is not removing pages that contain the terms it’s looking for because a federal judge ordered it and other agencies to restore several webpages they removed as a result of Trump’s executive order. The court ordered the administration to restore the webpages to their versions as of January 30, 2025, meaning they were supposed to revert the webpages to what they looked like on January 30 with no changes. The versions that have been restored now have this additional disclaimer about “gender ideology” we reported on last 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 the new Murderbot show, ChatGPT for journalism, and birdwatching from afar.

EMANUEL: Yesterday Apple released the first two images from its upcoming sci-fi show, Murderbot, and announced that it will debut on May 16 this year. I, like many fans of the The Murderbot Diaries books the series is based on, am very excited about the show, but also already disappointed with one major deviation from the source material that’s obvious just from these two still images. The gist is in the show Murderbot is played by Alexander Skarsgård, who in the images looks like a guy, while in the books Murderbot is neither a he or a she, but an “it,” and while it’s not at all the focus of the story, the fact that the main character is androgynous make it much more interesting. 

To back up, The Murderbot Diaries are set in the far, far future and follow a “SecUnit,” a super advanced, super lethal cyborg who does private security for scientists and corporations exploring deep space and dangerous planets. Eventually the SecUnit, who we come to know as Murderbot hacks the governing module that keeps it enslaved and has to choose what to do with its independence as it goes off on a series of pulpy space adventures. 

Ziff Davis, the $2 billion media conglomerate that owns dozens of sites including PCMag, Lifehacker, IGN and CNET, is quietly taking diversity, equity, and inclusion information off of its website, 404 Media has learned. 

In the past month, the company removed information about diversity-focused employee resource groups, inclusion-based hiring goals, and diversity training for its workers and managers from its corporate website. 

The changes were first spotted by a Ziff Davis employee. 404 Media granted the employee anonymity to speak candidly. 

An archived version of Ziff Davis’s DEI webpage saved on January 19 states, “Ziff Davis is proud to offer Employee Resource Groups (ERGs), voluntary employee-led groups mentored by executive sponsors and overseen by our Global DEI and HR Programs team. They represent s​even identity groups​: Asian, Black, 2SLGBTQIA+, Latinx/Hispanic, family of all kinds, women and gender minorities​, and interfaith.”

This article was produced in collaboration with Court Watch, an independent outlet that unearths overlooked court records. To subscribe to Court Watch, click here.

A Beverly Hills plastic surgeon’s patients filed a class action lawsuit against him earlier this month after they say he didn’t tell them that his patient information database had been hacked twice, and that their personal information and nude photos of themselves undergoing surgery had been posted online. 

The lawsuit alleges that the surgeon, Dr. Jaime Schwartz, did not secure his patients’ information with industry-standard safety protocols, and that he lied about the scope of the first hack when patients asked him about it. Schwartz may also be familiar to some reality TV viewers, having appeared on shows such as “Botched,” according to his website.

“Despite charging clients thousands of dollars and having access to their deeply private medical information, Dr. Schwartz disregarded basic security measures necessary to protect that information from malicious cyberattacks,” the lawsuit states. “As a result of his negligence, he allowed his network to be compromised twice in less than a year [emphasis in original].”

In both cases, Dr. Schwartz did not notify his patients of a hack until some of them found their information—including nude photos of themselves with their faces visible—online, according to the lawsuit. 

The lawsuit alleges that Dr. Schwartz was first notified of a hack of his patient database in October of 2023, when the hacking group Hunter International posted that it had access to his data. 

“The hackers had exfiltrated 1.1 terabytes of data from Dr. Schwartz, consisting of 248,245 files,” the lawsuit states. “The dark web posting included four patient photos, including one nude photo with the patient’s face visible.” 

Schwartz refused to pay the ransom, according to the lawsuit. One month later, the hackers updated the post with a note to him.

“Seems like you don’t want to protect your data at all,” the lawsuit quotes the note as reading. “More than 30 days had passed already since your network has been breached. You have been provided with everything you have asked about…But you keep begging for proofs [sic]. This is not the way we going to make business with you. Maybe you will do us a favor and transfer half of the money to prove that you can pay for your data?” 

The lawsuit does not specify how much money the hackers had asked for as part of the extortion. About two weeks after the note was posted, the lawsuit states, the hackers put up another update including nude photos of patients. “If you find your private data here just email us and we will let you know how to proceed further with actions against this DOCTOR!” the last update read, according to the lawsuit. 

The lawsuit alleges that Schwartz did not notify his patients of the hack until some of them found information about it online. One plaintiff reached out to him to ask whether her data was compromised as part of the breach. 

“Thereafter, a person claiming to be in charge of cybersecurity for Dr. Schwartz called [the plaintiff],” the lawsuit states. “[She] is informed and believes that the person was Dr. Schwartz’s brother.”

According to the lawsuit, the head of cybersecurity told the plaintiff that the breach had only affected six people, that her data was not included, and that Schwartz was “working with the FBI and had completely overhauled the computer system to prevent future cyberattacks.” 

The medical world is not new to this kind of extortion. Both major hospitals and private clinics have suffered data breaches in recent years, and over 500 breaches of varying degree were reported to the U.S. Department of Health and Human Services in 2024. The American Medical Association found in 2019 that 83 percent of doctors in the U.S. had experienced some kind of cyber attack. 

Plastic surgeons, however, have recently become a popular target because of the kind of data they retain. A patient’s file includes not only their medical and financial information, but also photographs taken as part of the treatment process. Depending on the surgery, many of those photographs are taken nude. Even as far back as 2017, hackers targeted a plastic surgeon whose clients allegedly included royal families and stole a wealth of highly personal photos. 

“Seems like you don’t want to protect your data at all.”

“This information is particularly valuable for purposes of sale on the dark web to facilitate identity theft and for purposes of ransom/extortion against physicians and patients,” the lawsuit states.

In October of 2023, the FBI released a public service announcement that hackers were targeting plastic surgeons. The announcement said that hackers would phish plastic surgeons’ offices to get access to their patient information databases, then use “open-source information” like patients’ social media profiles as leverage.

“Once successful, cybercriminals use social engineering techniques to enhance the harvested data and extort individuals for cryptocurrency,” the announcement stated. 

Yet the lawsuit alleges that Schwartz did not take any extra precautions after his October hack. And, in March of 2024, it claims, he was hacked a second time. In this breach, the lawsuit alleges, “The entirety of Dr. Schwartz’s patient data was compromised.”

“[Schwartz] failed to notify his patients as required by federal and state law,” the lawsuit states. “He waited to do so until after the hackers posted a public website announcing the hack and leaking patients’ names, contact information, and nude photographs, and began contacting his patients directly. Despite knowing that his patients’ most private medical data was in the hands of malicious actors, Dr. Schwartz waited almost 10 months to notify them [emphasis in original].” 

“Maybe you will do us a favor and transfer half of the money to prove that you can pay for your data?”

Schwartz sent his patients a generic message about the second hack in January of 2025. He wrote that, “An unauthorized third party utilized a third-party vendor’s credentials to access the practice’s medical billing and practice management system…It was determined that some of your personal information was present in the impacted data set. We then took steps to notify you of the incident as quickly as possible.”

Despite the head of cybersecurity’s promise of a full system overhaul, the lawsuit alleges that Schwartz’s team did not sufficiently secure its network-connected devices, did not train its staff to avoid phishing emails, and did not properly vet or secure its third-party vendors with access to sensitive patient data. It also claims Schwartz did not adequately monitor its network activity or implement “appropriate network ‘traffic’ controls to prevent the exfiltration of large amounts of data.” The lawsuit additionally claims Schwartz did not have appropriate anti-malware software or firewalls in its system. 

The lawsuit also alleges that, when it was filed, Schwartz had not yet contacted the California attorney general or the U.S. Department of Health and Human Service about either hack. 

“To date, the hackers have posted approximately 30 patient files,” the lawsuit states. “They have warned that they will continue releasing patient files, in alphabetical order, until Dr. Schwartz contacts them to address the matter.”

The plaintiffs are demanding damages of up to $3,000 per violation per person, amounting to more than $5 million, as well as a potential jury trial. 

Schwartz’s office did not respond to a request for comment.

Hoopla, a service that provides public libraries around the country with ebooks, announced that it will do more to prevent the spread of low quality AI-generated books after a 404 Media investigation showed that they were common on its platform.

“At hoopla, customer satisfaction is at the core of everything we do, and we deeply appreciate the feedback we’ve received regarding our content, including AI-related titles,” Ann Ford, VP of Sales & Customer Support at Hoopla, said in an email sent to librarians on February 10, which 404 Media then obtained. “We want to assure you that we take your concerns very seriously. Your input is invaluable in helping us learn, grow, and continuously improve. In response, our senior management has come together to develop a thoughtful and comprehensive plan of action.”  

While the exact details of the plan Hoopla is putting together to prevent low quality AI-generated books from flooding its platform are still not clear, Hoopla emailed librarians again on February 14 to share more information on actions it has already implemented. This includes revising its “collection development policy to ensure we adhere to and evolve with industry best practices,” offering librarians better ways to manage the Hoopla catalog by contacting Hoopla directly, and the removal of all “summary titles from all vendors, with some exceptions,” such as HMH Books, the publisher of the popular CliffNotes series. 404 Media also obtained a copy of this second email.

As 404 Media’s investigation into Hoopla showed, books that seemingly use AI to summarize existing, human-written books, are some of the most common low quality content on Hoopla as well as other ebooks providers, including Amazon. For example, one publisher called IRB Media had hundreds of summaries available to lend via Hoopla when I published my story on February 4. At the time of writing IRB Media still has about a dozen summaries I could find on Hoopla, but most of its books had been removed. 

Hoopla’s second email to librarians also announced that the company has removed some publishers and authors it identified as providing “poor-quality and/or poor-quality AI-generated content” using “industry metadata standards to identify AI-generated content.” Some of the low quality AI-generated books I highlighted in my story, like a fatty liver diet cookbook by an author that doesn’t appear to exist and has an AI-generated headshot, were removed. Other books, like an AI-generated book about Elon Musk, are still on Hoopla but can’t be borrowed. Other books by the same author of the Elon Musk book, Bill Tarino, have been removed as well. 

“It is important to note that libraries may still choose to opt out of all publisher-tagged AI-generated content by contacting their sales representative,” Ford said in one of the emails to librarians. “It is our hope that you are already noticing the positive impact of these actions.”

While it’s notable that Hoopla is actively removing AI-generated books on its platform that it previously ignored, librarians think the company still has a lot of work to do.

“Librarians select, purchase, and lend materials in service to the public, and they put their trust in hoopla to provide a curated and high-quality catalog of materials,” Jennie Rose Halperin, executive director at Library Futures, an organization of librarians, told me in an email. “Hoopla has broken this trust in favor of a profit-motivated, exploitative model that flies in the face of professional values. This statement, which is very light on details, continues to avoid accountability for the expensive and shoddy product they are vending. Around the country, libraries are under attack by censors and book banners for simply providing access to quality resources that serve the needs of their communities, and hoopla’s model puts them further at risk. The misalignment of values between big vendors and libraries has never been clearer.”

This article was produced in collaboration with Court Watch, an independent outlet that unearths overlooked court records. To subscribe to Court Watch, click here.

Meta is suing a scammer who allegedly and prolifically extorted people by banning and unbanning their Instagram accounts.

The lawsuit, filed on Tuesday, is against Idriss Qibaa, who ran the “Unlocked 4 Life” extortion scheme, according to an earlier criminal complaint filed by prosecutors. Qibaa, a self-described “professional when it comes to the banning and unbanning of Instagram accounts,” admitted on Adam 22’s No Jumper podcast in January 2024 that he had over 200 people who pay him monthly to maintain access to their accounts, and claimed he made more than $600,000 a month with this scheme. On the podcast episode, Adam mentioned that celebrities have fallen victim to similar extortion crimes, and Qibaa responded that they’re “getting extorted.”

Scammer Allegedly Makes $600,000 a Month Holding Instagram Accounts Hostage

The case of Unlocked4Life, who outed himself on Adam-22’s No Jumper podcast, shows how Instagram account scammers have escalated to violence and intimidation too.

404 MediaSamantha Cole

Part of the “Unlocked 4 Life” extortion scheme included threatening to murder victims if they didn’t cooperate, according to the criminal complaint. A federal grand jury in the District of Nevada indicted Qibaa in August 2024 in a case that’s ongoing, charging him with two counts of violating interstate communications law for sending messages threatening to injure or kill two victims. The indictment goes into detail about the harassment Qibaa allegedly doled out against people who didn’t comply with his scheme, including sending hundreds or thousands of text messages, racial slurs, messages threatening to kill them, and photos of a man with a bloodied face. “Here’s the last guy who came to take photos/came near my home,” that text said. In one case, he threatened to “blast out” a victim’s social security number and demanded she pay him $20,000 to stop harassing her, according to court documents.

“We will consider all enforcement and legal options to protect people on our platforms. These particular abuses target users and violate our policies, and we are committed to countering these malicious activities," a Meta spokesperson told 404 Media in a statement.

Meta’s new complaint accuses Qibaa of selling “unauthorized Instagram services including (a) the ability to disable user accounts; (b) user account reinstatement services intended to circumvent enforcement actions taken by Meta in response to users who violated the Instagram Terms of Use (‘Terms’) and other rules that govern access to and use of Instagram, including Instagram’s Community Standards (collectively, ‘Instagram Terms and Policies’); and (c) fake engagement services intended to artificially inflate followers on Instagram user accounts, among other things.” The complaint also claims that Qibaa was running the same grift on X, YouTube, TikTok, Snapchat, and Telegram.

In February 2024, Meta sent Qibaa a cease-and-desist letter, revoked his licenses to access Facebook and Instagram, and disabled his accounts, according to the complaint. But Qibaa made new Instagram accounts to get around the bans, Meta alleges.

Meta’s complaint is a look into how easy it is to manipulate its own reporting and moderation features. The company says Qibaa got people’s Instagram accounts banned by simply submitting fake reports claiming they were violating the platform’s terms. When Qibaa submitted the misleading reports, Meta alleges, Instagram disabled the account on the same day, and in some cases, reinstated it on the same day, too. 

Updated 2/19 12:35 p.m. EST with comment from Meta.

Roughly 10 months after it was released, the Humane AI Pin, a terribly executed and terrible product is now officially a $700 brick of e-waste after the company sold its software to HP and told its dozens of customers that they are now out of luck. 

Every Humane AI Pin ever created will stop functioning at the end of the month. Well, that is not exactly correct. As Engadget has pointed out, Humane told customers that nearly every function of the AI pin will stop working on February 28, but that true diehards can continue to access “offline” features, which primarily seems to be checking whether the battery is charged or not: “After February 28, 2025, AI Pin will still allow for offline features like battery level, etc., but will not include any function that requires cloud connectivity like voice interactions, AI responses, and Center access.” Humane went on to say that “We encourage you to recycle your AI Pin through an e-waste recycling program.”

We Can, and We Must, Clown on the Humane AI Pin Forever

The Humane Ai Pin joins a rich tradition of terrible tech products that includes the Juicero, Coolest Cooler, and Magic Leap that we must remember forever.

404 MediaJason Koebler

If you are not familiar, the AI Pin is a $700 piece of junk that was supposed to be an “AI assistant” but instead barely worked, was perhaps a fire hazard, and whose main functionality was triggering fragile venture capitalists on Twitter who self-immolated when the reviews were understandably very bad.

There is very little to say about the Humane AI pin right now other than they are very lucky that the vast majority of tech journalists in the United States are too busy writing about the Elon Musk-led ransacking of the federal government to dunk on this company in the way it truly deserves (we are also doing this but need a break for five minutes). 

The company and the tech was wildly hyped, wasted gazillions of dollars (it raised $240 million in funding), made something terrible, existed for less than a year, and are now hazardous e-waste that is a huge pain in the ass to safely dispose of. The saving grace of all of this is that Humane sold so few devices (roughly 10,000) that the number of consumers who are affected is relatively low as these things go and therefore, there are fewer of them that need to be recycled. 

The Humane AI Pin is the latest in a long line of internet of things devices that cost a lot and then became e-waste when the company decided to stop supporting it or went out of business.

On recycling: I have been to electronics recycling centers, and small wearables like this are labor intensive to recycle because they have small, difficult-to-remove batteries. An iFixit teardown wondered whether Humane pin was one of the “worst devices ever,” and stated that both the Humane AI pin and the Rabbit R1, another AI wearable, “have batteries that are a pain to remove, hidden behind thoroughly glued-down panels,” and that “making the battery so difficult to reach is perplexing at best.”

Anyways, we must never forget the Humane AI Pin. Good job everyone.

This week we start with Jason's story about anyone being able to push updates to DOGE.gov website. Then we talk about other stories with the DEI.gov and Waste.gov sites. After the break, Sam tells us all about some lawyers who get caught using AI in a case. In the subscribers-only section, we chat about a true crime documentary YouTube channel where the murders were all AI-generated.

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.

• Anyone Can Push Updates to the DOGE.gov Website
• Elon Musk's Waste.gov Is Just a WordPress Theme Placeholder Page
• Lawyers Caught Citing AI-Hallucinated Cases Call It a 'Cautionary Tale'
• A ‘True Crime’ Documentary Series Has Millions of Views. The Murders Are All AI-Generated

A worker at the General Services Administration told colleagues in a Slack message Tuesday that they have resigned in protest after Elon Musk ally Thomas Shedd requested “admin/root access to all components of the Notify.gov system,” which is a government system used to send mass text messages to the public that contains information the worker said is highly sensitive and would give Shedd unilateral, private access to the personal data of members of the public.

Shedd is a former Tesla engineer who now runs Technology Transformation Services (TTS), a group of coders and software engineers within the GSA, who is closely allied with Elon Musk and DOGE. Notify.gov contains not just the phone numbers of everyday people but also information about whether they participate in government programs such as Medicaid, which is based on a person's financial situation. In recent days, Musk has become obsessed with the idea of "fraud" in Medicaid, Medicare, and Social Security, and in identifying those he suspects are committing fraud.

“The TTS commissioner, Thomas Shedd, has required us to provide admin/root access to all components of the Notify.gov system,” the Slack message, seen by 404 Media starts. It then says this would allow Shedd to “view all personally identifiable information (PII) moving through the Notify system, including phone numbers and variable data for members of the public.” It says Shedd “would be able to download and store this data without anybody else receiving a notification.”