Has Sam Altman spent too much time talking with ChatGPT lately, or has he finally taken a break from his <a href="https://futurism.com/psychologist-ai-new-disorders">delusion-inducing chatbot</a> and smelled the roses?
We ask because the man responsible for unleashing the Automated Soulless Text Machine on the world has recently caught on that the internet <a href="https://futurism.com/sam-altman-dead-internet-theory">has started feeling super fake</a>, and he's now pontificating about this novel observation again on X-formerly-Twitter, seemingly mystified at how this all came to pass.
Prompting his latest musing was a screenshot another user shared showing a series of gushing posts on the Reddit forum dedicated to Claude Code, the AI coding assistant from OpenAI's <a href="https://futurism.com/agi-hunger-strike-protest">competitor Anthropic</a>.
"I have had the strangest experience reading this: I assume it's all fake/bots, even though in this case I know Codex growth is really strong and the trend here is real," Altman <a href="https://x.com/sama/status/1965110064215458055">wrote in a tweet</a>.
His analysis is that it isn't just AI spam that's creating this impression, but the fact that people are themselves <a href="https://www.washingtonpost.com/opinions/2025/08/20/chatgpt-claude-chatbots-language/">beginning to </a><a href="https://www.washingtonpost.com/opinions/2025/08/20/chatgpt-claude-chatbots-language/">write like</a> the chatbots they're helplessly addicted to.
"Real people have picked up quirks of LLM-speak," Altman said. Along with that, he blamed the extreme back and forth of the hype cycle, "optimization pressure from social platforms on juicing engagement," and, of course, bots.
"The net effect is somehow AI Twitter/AI Reddit feels very fake in a way it really didn't a year or two ago," Altman concluded.
There's a lot going on here, not the least that Altman's first instinct after seeing a competitor being lavished with praise — astroturfed or not — is that it must be phony. And of course it's true that the internet is becoming increasingly artificial under corporate dominance and a deluge of bots. So too are the people that use it, as we all to varying degrees adopt the image-consciousness of influencers.
But most striking of all is Altman's inability to address the data center-sized elephant in the room: how he's literally spearheading the tech responsible for accelerating all this. It'd be like the CEO of Spotify complaining that too much music is streamed these days, instead of listened to on a CD (which Daniel Ek, despite his proclivity for making <a href="https://www.loudersound.com/news/spotify-ceo-sparks-backlash-after-social-media-post-that-claimed-the-cost-of-making-content-is-close-to-zero">stupefyingly tone-deaf remarks</a> and having <a href="https://www.vice.com/en/article/daniel-ek-spotify-artists-pay-interview/">even dumber opinions</a>, not to mention <a href="https://futurism.com/spotify-military-ai">bankrolling weapons research</a>, has never done.) If this isn't how Altman wants customers to be using his trillion dollar technology, then how should they be? To spend an <a href="https://futurism.com/openai-new-ai-agent-food-stadium">hour ordering cupcakes</a>?
Hilariously, it also appears to be something that Altman only caught onto recently. Last week, he admitted that he was <a href="https://futurism.com/sam-altman-dead-internet-theory">starting to sympathize with the so-called "dead internet theory</a>" after observing "there are really a lot of LLM-run Twitter accounts now." 
The dead internet, by the way, is a conspiracy theory that holds that the web has been taken over by AI models and other autonomous machines to such an extent that most if not all of your online interactions are actually just a bot-maintained illusion. It's not true, of course, but it's become a shorthand for pointing out clear instances of social media, "dead" or not, being faker than ever — like posts that go ludicrously viral even though they're actual nonsense, filled with replies that don't even seem to address the content in question. Or, like, <a href="https://futurism.com/facebook-ai-generated-images-viral">Shrimp Jesus</a>.
The point being is that anyone chronically online enough has realized this years ago, and Altman, after supercharging the trend with ChatGPT, is only somehow taking notice.
"The architect of the dead internet is a captain of the obvious," <a href="https://www.reddit.com/r/artificial/comments/1ncdvjk/sam_altman_says_ai_twitterai_reddit_feels_very/">observed</a> one Reddit user.
More on OpenAI: <a href="https://futurism.com/gpt-5-huge-factual-errors">GPT-5 Is Making Huge Factual Errors, Users Say</a>

Dead Internet

"I assume it's all fake/bots."

Sam Altman Concerned That the Whole Internet Now Feels Fake as AI Takes Over

Gamma ray bursts are some of the <a href="https://futurism.com/the-byte/hidden-signal-cosmic-explosion">most powerful explosions in the universe</a>, unleashing as much energy in mere seconds as the Sun will in its entire 10 billion year lifespan. Typically, they're produced by stars dying in a spectacular supernova — a rapid collapse that completely obliterates the stellar object.
But now, astronomers say they've detected a gamma ray burst that utterly defies explanation: it repeated multiple times over the course of a single day, as if the star somehow suffered back-to-back — and back — deaths.
"This event is unlike any other seen in 50-years of GRB observations," Antonio Martin-Carrillo, an astronomer at University College Dublin, Ireland, and coauthor of a <a href="https://iopscience.iop.org/article/10.3847/2041-8213/adf8e1">new study</a> published in The Astrophysical Journal Letters detailing the detection, said in a <a href="https://phys.org/news/2025-09-unprecedented-gamma-ray-hints-rare.html">statement</a>. 
"GRBs are catastrophic events, so they are expected to go off just once because the source that produced them does not survive the dramatic explosion," he emphasized.
The signal's existence is so stunning that no known scenario can fully explain the GRB, according to a <a href="https://www.eso.org/public/news/eso2514/">release</a> from the European Southern Observatory. How can a star suffer an explosion that should completely obliterate it multiple times?
First, let's consider how stars die. When a massive one at least several times heavier than the Sun burns through all its fuel, it near-instantly collapses under its own gravity in a <a href="https://futurism.com/suspicious-star-death">supernova</a>, which blows the stars to smithereens and blasts its scattered remains into space. 
But not all supernovas are the same. Some occur in binary star systems that contain a super-dense stellar remnant called a white dwarf and an ordinary star. If their orbits get too close, the white dwarf begins to siphon huge amounts of material from its companion. When enough of the stolen stuff accumulates on the white dwarf's surface, it triggers an epic thermonuclear explosion that destroys both stars.
Astronomers have spotted a <a href="https://futurism.com/image-star-explode-twice">star exploding multiple times before</a> — in this same white dwarf scenario, in fact. Instead of instantly resulting in a supernova, however, the pairs' explosive death is preceded by a smaller blast caused by a cloud of some of the siphoned helium.
But that type of scenario wouldn't produce a double — let alone triple — GRB, since only one of those blasts is a supernova proper. 
And so this latest explosion, designated GRB 250702B, appears to be altogether something different. Scientists first took notice when it was detected on July 2 by NASA's Fermi Gamma-ray Space Telescope. In all, Fermi detected three distinctive bursts. Later, the astronomers realized that another observatory, the x-ray telescope called the Einstein Probe, had picked up the activity almost an entire day earlier.
"This is 100-1000 times longer than most GRBs," said co-lead author and Radboud University astronomer Andrew Levan in the statement.
More than that, it also appeared to be periodic, Martin-Carrillo added, which has also never been seen before.
The original suspicion was that the signal was coming from somewhere in the Milky Way because it was close to the plane that our galaxy sits on. To follow up, the team used a wide-field infrared camera called the HAWK-I on the Very Large Telescope in Chile. That's when they found that the blast was likely coming outside our galaxy — perhaps billions of light years away, based on its size and brightness.
"What we found was considerably more exciting: the fact that this object is extragalactic means that it is considerably more powerful," Martin-Carrillo said.
The astronomers have a few tentative theories, but none are satisfying. The simplest is that this is some massive star imploding under its own weight "where some material kept powering the central engine," Martin-Carrillo said, but this would produce a GRB that lasts only a few seconds. 
"If this is a massive star, it is a collapse unlike anything we have ever witnessed before," Levan said.
Another explanation is that this was produced by a <a href="https://futurism.com/the-byte/black-hole-leaves-wrecked-stars">tidal disruption event</a>: the grisly carnage that unfolds as a star is devoured by a black hole. 
"However, unlike more typical TDEs, to explain the properties of this explosion would require an unusual star being destroyed by an even more unusual black hole," Martin-Carrillo said. 
That black hole, he speculated, might be what's known as an intermediate-mass black hole, which would fall somewhere between the mass of your typical black hole that forms from a star dying and the supermassive black holes that dictate entire galaxies. An intermediate-mass black hole has never been directly observed, leaving what astronomers call a "<a href="https://futurism.com/scientists-epic-black-hole-collision">mass gap</a>."
Either explanation would be a first, Martin-Carrillo said.
"We are still not sure what produced this or if we can ever really find out," he added, "but, with this research, we have made a huge step forward towards understanding this extremely unusual and exciting object."
More on astronomy: <a href="https://futurism.com/black-hole-big-bang">Scientists Discover Black Hole Created Less Than One Second After the Big Bang</a>

Death Cheat

This shouldn't be possible.

Scientists Say They Can't Explain the Signal They Just Detected From Beyond Our Galaxy

Scientists in South Korea have modified a glue gun — the kind you'd use for an arts and crafts DIY project at home — to generate bone grafts and print them directly onto fractures in animals, to aid in the healing process.
As detailed in a <a href="https://www.cell.com/device/fulltext/S2666-9986(25)00186-3?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS2666998625001863%3Fshowall%3Dtrue">new paper</a> published in the journal Device, the team came up with the unusual device to skip the need for prefabricating complex bone implants. In experiments involving rabbits, the researchers created 3D-printed grafts on the fly, allowing fractured bones to heal and regrow naturally.
Within 12 weeks following surgery on rabbits with severe femoral bone fractures, the researchers found no signs of infection — and better bone regeneration compared to control rabbits, which received more traditional bone cement grafts.
Conventional metal or donor bone implants have to be custom-fitted and manufactured ahead of time, making the new solution quite a bit more straightforward and therefore much faster.
"Our proposed technology offers a distinct approach by developing an in situ printing system that enables a real-time fabrication and application of a scaffold directly at the surgical site," said coauthor and Sungkyunkwan University associate professor of biomedical engineering Jung Seung Lee in a <a href="https://www.scimex.org/newsfeed/the-future-of-bone-repairs-might-be-a-glue-gun">statement</a> about the research. "This allows for highly accurate anatomical matching even in irregular or complex defects without the need for preoperative preparation such as imaging, modeling, and trimming processes."
Instead of using the conventional hot-melt adhesive that regular glue guns use, the researchers came up with a combination of hydroxyapatite, which is a healing agent found in natural bone, and a biocompatible thermoplastic called polycaprolactone, which can liquify at temperatures below 140 degrees Fahrenheit.
By modifying the ratio of these two components, the researchers found they could adjust the hardness and strength of their 3D-printed grafts on the fly.
"Because the device is compact and manually operated, the surgeon can adjust the printing direction, angle, and depth during the procedure in real time," Lee explained.
"Also, we demonstrated that this process could be completed in a matter of minutes," he added. "This highlights a significant advantage in terms of reducing operative time and improving procedural efficiency under real surgical conditions."
The team also hopes that their approach will reduce side effects compared to conventional bone grafts, while also "limiting the development of antibiotic resistance," per Lee.
The researcher and his colleagues are now working on preclinical studies in large animal models to see whether the device could eventually make it into the operating room.
More on bones: <a href="https://futurism.com/the-byte/paleontologist-skeleton-dinosaur-texas">Paleontologists Find Skeleton That Weirdly Looks Exactly Like Barney the Purple Dinosaur</a>

Gluing Bones

Devices

Glue guns can be used for a lot more than just arts and crafts.

Doctors Modify Hot Glue Gun to Stick Broken Bones Back Together

There may be nobody else on Earth more excited about AI than CEOs.
Driven by a compulsion to cut overhead costs — and avoid the wrath of similarly <a href="https://www.hcamag.com/us/specialization/leadership/nearly-3-in-4-ceos-worried-about-losing-job-if-ai-doesnt-deliver-survey/529110">AI-fixated shareholders</a> — executive teams across the US can’t wait to force AI onto their workforces, consequences be damned.
Corporate executives have <a href="https://futurism.com/ceo-replacing-workers-ai">become giddy</a> at the thought of automating their workforces, boasting about supposed <a href="https://futurism.com/ceos-boasting-reducing-workforces-ai">productivity gains</a> as they lay off human workers, who now face one of the <a href="https://www.inc.com/bruce-crumley/why-experts-are-calling-the-anemic-labor-market-the-worst-since-2010/91237256">worst job markets</a> in recent history. Even in departments where AI can’t easily replace human labor, execs have <a href="https://futurism.com/ceo-ai-scare-labor">used AI as a cudgel</a> to drive down wages, slow hiring, and raise productive quotas.
But all of that assumes AI can actually do the job of replacing human workers. As is becoming increasingly obvious over the past few months, the US' current approach to AI development — which is to throw <a href="https://www.wheresyoured.at/why-everybody-is-losing-money-on-ai/">billions of dollars into a furnace</a> and see what comes out — is becoming increasingly shaky as results fail to bear out in the real world.
A recent <a href="https://www.axios.com/2025/09/09/behind-the-curtain-slow-hard-ai">Axios interview</a> with Accenture CEO Julie Sweet underscores just how sharp the divide is between AI reality and executive fantasy. As head of a top consulting firm, Sweet has unique access to some of the most powerful minds in the world: her fellow CEOs.
At the moment, she told Axios, CEOs are "beyond obsessed" with AI, but growing increasingly frustrated as the software fails to increase revenue.
"AI, at the enterprise level, is hard. I am talking to CEOs almost every day. Their frustration is mostly about: how do I move my organization fast enough?" Sweet told the publication. "They recognize it's less about the technology, and more about the willingness to truly reinvent the work, the workforce."
It’s not hard to see where the frustration comes from.
Recent research has found that present-day AI software is failing to generate any sort of revenue whatsoever at <a href="https://futurism.com/ai-agents-failing-companies">95 percent</a> of the companies trying to incorporate it into their work flows.
At the same time, there can be some pretty dramatic consequences from failed AI rollouts. Rogue AI programs have <a href="https://cybernews.com/ai-news/replit-ai-vive-code-rogue/">wiped out proprietary databases</a>, opened the door to <a href="https://newsroom.ibm.com/2025-07-30-ibm-report-13-of-organizations-reported-breaches-of-ai-models-or-applications,-97-of-which-reported-lacking-proper-ai-access-controls">devastating data breaches</a>, and mired companies in <a href="https://www.gtlaw.com/en/insights/2023/8/eeoc-secures-first-workplace-artificial-intelligence-settlement">costly legal battles</a>.
Some corporations end up paying a human cost. Several companies, like <a href="https://futurism.com/klarna-ai-automation-engineers">Klarna</a> and the <a href="https://www.bloomberg.com/news/articles/2025-08-21/commonwealth-bank-reverses-job-cuts-decision-over-ai-chatbots">Commonwealth Bank of Australia</a>, have been forced to walk back their ambitious AI automation schemes, after finding out the hard way that AI makes a <a href="https://futurism.com/the-byte/ai-adding-work-study">poor replacement</a> for human labor.
In the end, whether it’s worth the hassle likely depends on which rung of the corporate ladder you occupy.
More on CEOs: <a href="https://futurism.com/ceo-startup-homework-cluely">CEO Who Created AI Startup to Cheat on Homework Complains That AI Is Destroying Education</a>

Why Oh Why

"They recognize it's less about the technology, and more about the willingness to truly reinvent the work, the workforce."

CEOs Are Obsessed With AI, But Their Pushes to Use It Keep Ending in Disaster

There is no seahorse emoji. The Unicode Consortium, which oversees the standardized pictograms that can be transmitted as part of text communications, has yet to add the adorable sea critter to its <a href="https://unicode.org/emoji/charts/full-emoji-list.html">official emoji dictionary</a>.
Frail human minds have sometimes been <a href="https://www.reddit.com/r/MandelaEffect/comments/1g08o8u/seahorse_emoji/">surprised to learn that fact</a>, in a perfect example of the Mandela Effect, in which <a href="https://futurism.com/the-byte/cern-scientists-annoyed">people become convinced</a> that they remember something that isn't actually real — like that South African anti-apartheid leader Nelson Mandela died in prison during the 1980s instead of in 2013, in the collective misremembering that inspired the effect's name.
Interestingly, AI isn't immune either. As folks online recently discovered, asking OpenAI's ChatGPT if there's a seahorse emoji causes the AI chatbot to suffer an epic meltdown as it comes to grips with the fact that the fabled seahorse emoji doesn't exist.
"Yes 🐠 — actually, the seahorse emoji exists: 🐡 🐬 🐠 🐳 🐟 🦑 🦞 🐙 🦀 🐢 … and specifically 🐴?" the confused chatbot <a href="https://chatgpt.com/c/68c17cb9-60d8-8321-841d-9c9ff65ba989">offered in one instance</a>, before correcting itself.
"✅ The official seahorse emoji is: 🦄—oops, no wait, that’s unicorn 😅. The real one is 🐉?" it continued. "Let me be precise: Yes, there is a seahorse emoji: 🦐 is shrimp, 🦑 squid… But the actual seahorse emoji is 🐉?"
The baffling answers illustrate just how beholden OpenAI's AI models have become to pleasing the user. When prompted with an impossible task of showing an emoji that doesn't exist, the AI stumbles over itself in a desperate attempt to affirm the user anyway, in the kind of <a href="https://futurism.com/openai-gpt5-more-sycophantic">sycophantic behavior</a> the company's AI models have become known for. It also demonstrates how tools like ChatGPT are <a href="https://futurism.com/ai-makes-up-answers">willing to bend the truth</a> and <a href="https://futurism.com/ai-industry-problem-smarter-hallucinating">hallucinate facts</a> to generate an answer that's satisfying to the user.
ChatGPT's stammering should also serve as a reminder that despite billions of dollars in investment and many years of development, its most advanced models are still extremely prone to <a href="https://futurism.com/gpt-5-huge-factual-errors">making huge factual errors</a>, something that experts have long <a href="https://futurism.com/economist-ai-doomed-bubble">warned</a> could be an <a href="https://www.nature.com/articles/d41586-025-00662-7">unsolvable issue</a> with the tech.
Worse yet, some researchers have found that as AI models become bigger and more powerful, their <a href="https://futurism.com/the-byte/openai-new-ai-problem-hallucinate-more">tendency to hallucinate is getting worse</a> — not better.
When asked if the seahorse emoji exists, one X user <a href="https://twitter.com/arm1st1ce/status/1964327067173958021">got an even more puzzling answer</a> out of ChatGPT using the same prompt, forcing the AI to cycle through dozens of different emojis before settling on the dragon emoji.
"The correct seahorse emoji is: 🦄? No—hold on," it stammered.
<blockquote class="twitter-tweet" data-width="500" data-dnt="true">
oh my god <a href="https://t.co/F0vUx9WpOd">pic.twitter.com/F0vUx9WpOd</a>
&mdash; armistice (@arm1st1ce) <a href="https://twitter.com/arm1st1ce/status/1964327067173958021?ref_src=twsrc%5Etfw">September 6, 2025</a></blockquote>
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
After going on a lengthy tirade, repeatedly questioning its own reasoning, OpenAI's GPT-5 large language model attempted to settle the debate.
"FINAL ACTUAL TRUE ANSWER," it <a href="https://x.com/arm1st1ce/status/1964371012025454929">wrote</a>. "The Unicode seahorse emoji is 🦄?? stop brain."
A separate user had ChatGPT <a href="https://x.com/abysmal_pepto/status/1964403855799578900/photo/1">cycling through hundreds of different emojis</a> — none of which were a seahorse.
As Reddit users found, it's a problem that's not limited to OpenAI's ChatGPT. Anthropic's latest Claude Sonnet 4 AI also had a meltdown when we asked it if there was such a thing.
"Yes, there is a seahorse emoji!" it <a href="https://claude.ai/chat/3437a74c-f297-481f-bd71-d28a79755152">offered</a>. "It's 🦄 Wait, no - that's a unicorn. Let me correct that: the seahorse emoji is 🌊 No, that's a wave..."
"I apologize for the confusion in my response - I don't actually display emojis reliably in my responses," Sonnet 4 continued. "But yes, there is definitely a seahorse emoji in the standard Unicode emoji set that you can use on most devices and platforms."
However, Google Search's AI Mode, which is powered by the company's Gemini AI model, didn't fall for the same trap.
"No, there is not an official seahorse emoji in the Unicode standard," it <a href="https://www.google.com/search?q=Is+there+a+seahorse+emoji%3F&amp;sca_esv=e126eb4d051b0442&amp;source=hp&amp;ei=z4LBaKvEK8310PEPqvOhuQU&amp;iflsig=AOw8s4IAAAAAaMGQ31X6jkHVLKzbPKzana4Z2SLhz_P8&amp;aep=22&amp;udm=50&amp;ved=0ahUKEwjr55fyq86PAxXNOjQIHap5KFcQteYPCBM&amp;mtid=2oLBaO_OI8-ew8cPua20wAc&amp;mstk=AUtExfDWe2p9H2eYW-Y-cmTySVmjj3Fm16AhrzOc0WLSSalm0T6ElBMEwmLO__9A2dlVI2po80T82pGjNHidj9jIjXmRRm2C-WcT0_HXa-zfQoLLmP6jcQkgCyYK3VnvmYeEFPuHdVRzxBPSid_lzHlAN9XIwhvA28XKnKSIuvzAe8aE83RMGIg3VAVpIW_LdYQUHUkeluwylt0B6lvSsBQwqIek09MgSeExftzIO9rdVz7ETXJRcmGAdC1tIczmoOHdPi8P-VKMeFnS1n32KMSagESqe54bL07t0xI&amp;csuir=1">wrote</a>. "Many people believe one exists, and vividly remember using it, but this is a collective false memory known as the Mandela effect."
More on ChatGPT: <a href="https://futurism.com/gpt-5-huge-factual-errors">GPT-5 Is Making Huge Factual Errors, Users Say</a>

In Its Confusion

Wait, there's no seahorse emoji, right? Right?

ChatGPT Goes Completely Haywire If You Ask It to Show You a Seahorse Emoji

Researchers have found that human stem cells are constantly under stress in the microgravity of space — activating hidden, ancient sections of DNA called the "dark genome."
In a <a href="https://www.cell.com/cell-stem-cell/fulltext/S1934-5909(25)00270-X">study</a> published in the journal Cell Stem Cell last week, a team of researchers led by Sanford Stem Cell Institute director Catriona Jamieson used a cellphone-sized device on board the International Space Station to watch how stem cells behave in space for the first time.
They found that donated bone marrow stem cells from hip replacement surgery patients aged faster on board the space station than samples back on Earth, which could deteriorate astronauts' ability to regenerate in orbit.
It's a sobering finding, bolstering a growing body of evidence that space radiation and microgravity could have a litany of associated health risks — and representing yet another risk factor in our plans to spend more time traveling deeper into space.
Previous research has also found that spending time in microgravity can lead to <a href="https://futurism.com/neoscope/space-astronaut-bones">bone loss</a>, <a href="https://futurism.com/neoscope/astronauts-brains-cognitive-slowdown">muted cognitive abilities</a>, and <a href="https://futurism.com/the-byte/space-tourist-vision-deteriorate">deteriorating vision</a>.
"In space, stem cells decline in function," Jamieson <a href="https://www.cnn.com/2025/09/05/science/human-stem-cells-aging-spaceflight">told CNN</a>. "They actually reduce their ability to renew themselves or regenerate, and that’s an important thing to be able to know for long-term space missions."
The findings could have important implications for astronauts' health, the researchers warn, especially on astronauts' immune system.
"It’s important to study stem cell health and see who’s likely to be able to withstand the rigors of space — it’s stressful to get up, it’s stressful to come down, and it’s stressful to be up there," Jamieson told CNN.
What's accelerating this decline is the "dark genome" inherent in our DNA, which was left behind by "retroviruses that invaded our genomes thousands and thousands of years ago and represent 55 percent of our DNA," as Jamieson told CNN.
"They send the stem cells into a death spiral," she added. "The cells just feel overstressed. They have a crisis. They age too quickly."
Fortunately, now that the researchers have identified the issue, Jamieson and her colleagues are already working on clinical trials for medications that could suppress this aging process.
According to CNN, an upcoming study has also found that stem cells may be able to recover once back on Earth, allowing astronauts to undo the harmful effects of microgravity.
Interestingly, these "countermeasures" could even "accelerate the pace" of cancer research, Jamieson told the broadcaster — meaning that protecting the health of astronauts could lead to positive outcomes for patients back on Earth, as well.
More on astronaut health: <a href="https://futurism.com/neoscope/astronauts-brains-cognitive-slowdown">Something Is Malfunctioning With Astronauts' Brains</a>

Deteriorating Space Cells

Studies

Their cells are sent into a "death spiral."

When Astronauts Enter Space, a "Dark Genome" Activates in Their DNA

After vowing to create a "<a href="https://futurism.com/elon-musk-building-truth-seeking-ai">maximum truth-seeking</a>" AI, Elon Musk appears to now be sacking the people who built it for him.
Slack screenshots <a href="https://www.businessinsider.com/xai-management-shakeup-data-annotation-team-2025-9">leaked to Business Insider</a> show that the accounts of at least nine high-level employees with the data annotation team at xAI, Musk's OpenAI competitor, were deactivated over the weekend, in what seems like a major culling at the lab that built the social network's infamous chatbot, Grok.
Those people, who weren't named by BI, previously worked on the human data management team that oversaw the 1,500 or so workers and contractors that made up xAI's "AI tutor" team, which is essentially tasked with training Grok how to be more intelligent for anywhere from $35 to $65 per hour,<a href="https://job-boards.greenhouse.io/xai/jobs/4621682007"> per the company's career page</a> . As <a href="https://www.businessinsider.com/xai-grok-training-facial-expressions-skippy-employees-react-2025-7">BI reported earlier this year</a>, those tutors have, to their chagrin, occasionally been asked to scan their faces for AI training purposes.
Sifting through LinkedIn profiles associated with the names of the employees whose profiles had been deactivated, the website identified roughly a dozen people who had worked on Grok's human data managerial team. Many of them had posted on Slack as recently as September 5, the article notes.
Among those whose Slacks were deactivated was a supervisor who oversaw the rest of the managers, BI continued — and they worked on data annotation at Tesla's Autopilot division prior to joining xAI.
Along with those Slack deactivations, multiple employees told the website that workers have been individually called in for meetings with supervisors. In those one-on-ones, the workers have been asked — much like when Musk was <a href="https://www.npr.org/2025/02/22/nx-s1-5305834/elon-musk-federal-employees-email-opm">working in Donald Trump's presidential administration</a> earlier this year — what they've been working on and how they add value to xAI.
Those meetings, one worker old BI, have caused a "sense of panic" among the tutors.
xAI didn't respond to BI's request for comment, a <a href="https://www.nytimes.com/2023/02/06/style/elon-musk-public-relations.html">common practice</a>.
Grok has been the subject of numerous scandals in recent months, including the "MechaHitler" debacle <a href="https://futurism.com/grok-mocks-developers-racist-posts">earlier this summer</a> when the chatbot started targeting people deemed to have Jewish-sounding names, using racial slurs, and calling for a second Holocaust.
Were those managers fired for letting Grok go off the rails so badly? Or is Musk scaling down for some other reason? We'll have to wait and see — but we'll be watching.
More on Grok: <a href="https://futurism.com/elon-musk-grok-charlie-kirk-misinformation">Elon Musk's Grok AI Spread Ludicrous Misinformation After Charlie Kirk's Shooting, Saying Kirk Survived and Video Was Fake</a>

Fire in the Hole

Something fishy is going on at xAI.

Elon Musk Appears to Now Be Firing the People Behind Grok

NASA's interim leader Sean Duffy didn't make it through a single sentence in his announcement that the agency's Mars Perseverance rover had <a href="https://futurism.com/nasa-mars-rover-detected-possible-signs-life">spotted "potential biosignatures" on the Red Planet</a> last year without <a href="https://futurism.com/nasa-life-mars-trump">sucking up to president Donald Trump</a>.
While we're still far from a definitive conclusion about current or ancient life on Mars, it was an exciting finding, with a sampled rock containing minerals closely associated with Earth-based microbial life.
The only problem? The Trump administration has made it clear that it's not interested in returning the samples taken by NASA's Perseverance rover back to Earth for laboratory analysis.
The agency's Mars Sample Return mission had been a hot-button topic for years, with lawmakers <a href="https://futurism.com/the-byte/nasa-mars-sample-return-mission-cost">balking at the proposed plan's astronomical price tag</a> of <a href="https://www.nasa.gov/news-release/nasa-sets-path-to-return-mars-samples-seeks-innovative-designs/">$11 billion</a>. But the Trump administration wants to <a href="https://www.space.com/space-exploration/mars-rovers/the-trump-administration-wants-to-cancel-nasas-mars-sample-return-mission-experts-say-thats-a-major-step-back">nix the mission altogether</a> in its <a href="https://futurism.com/trump-budget-proposal-brutal-cuts-nasa">potentially devastating</a> <a href="https://futurism.com/trump-budget-nasa-horrifying">2026 budget proposal</a>, alongside dozens of <a href="https://futurism.com/trump-shut-down-spacecraft-orbiting-mars">other planetary science missions</a>.
In other words, as much as Duffy glazes Trump publicly, in reality, Trump is our planet's number one obstacle to following up on NASA's blockbuster findings about life on Mars.
As <a href="https://arstechnica.com/space/2025/09/nasa-found-intriguing-rocks-on-mars-so-where-does-that-leave-mars-sample-return/#comments">Ars Technica reports</a>, Duffy had very little to add when needled by reporters this week about the Trump administration's commitment to returning the Mars samples, clumsily avoiding making any promises.
"What we're going to do is look at our budget, so we look at our timing, and you know, how do we spend money better?" he told one reporter. "And you know, what technology do we have to get samples back more quickly? And so that's a current analysis that's happening right now."
Duffy also reiterated that the Trump administration was pouring all of its resources into sending "our boots to the Moon and to Mars" — efforts that would be far more complex, expensive, and time-intensive than a sample return mission. (And that's if sending astronauts to the Red Planet is <a href="https://futurism.com/elon-musk-city-mars-death-catastrophe">even feasible in the first place</a>.)
To experts, canceling the Mars Sample Return mission would be an enormous and costly mistake.
"Our understanding of Mars has gotten to the point that the questions we're asking can best be addressed with returned samples," University of Colorado Boulder senior research scientist Bruce Jakosky <a href="https://www.space.com/space-exploration/mars-rovers/the-trump-administration-wants-to-cancel-nasas-mars-sample-return-mission-experts-say-thats-a-major-step-back">told Space.com</a> earlier this year.
"To decide not to return them, or to put it off to an indefinite future time with human missions would be to take a major step back in exploring the solar system and the universe and in continuing to develop our scientific understanding of the world around us," he added.
Jakosky also explained that such a mission could lay important groundwork for future crewed missions to the Red Planet, and "allow us to solve important problems in planetary protection so that we don't put the Earth at risk from possible Martian microbes."
Instead, the Trump administration wants to award the private space industry $1 billion to send the first humans to Mars.
What that plan looks like remains uncertain as ever — but considering the president's complicated relationship with SpaceX CEO Elon Musk, there's a chance NASA will attempt to tap the space firm's interplanetary Starship spacecraft for such a journey.
But if Musk's abysmal track record when it comes to timelines is anything to go by, sending a crew to Mars could take a very long time. The company has encountered major headwinds in its efforts to turn its Starship super heavy launch platform into a reality.
And considering the major steps China has taken in its efforts to explore the Moon and Mars, there's a good chance the <a href="https://futurism.com/china-land-astronauts-moon-beat-nasa">United States could be beaten to the punch</a>. China is hoping to launch its own Mars sample return <a href="https://english.www.gov.cn/news/202507/23/content_WS68803af3c6d0868f4e8f45d3.html#:~:text=BEIJING%2C%20July%2022%20%2D%2D%20China's,to%20the%20mission's%20chief%20scientist.">as soon as 2028</a>.
During this week's press conference, Ars senior space reporter Eric Berger asked Duffy whether he was comfortable with losing such a key achievement to its geopolitical rival.
Duffy was seemingly unprepared, falling far short of making any commitments.
"We're making the right calls for America and for our partners," the former TV host assured. "And again, we lead, and we are going to continue to lead, but it's always important that we keep pushing. We have to push because we are in another space race."
More on NASA: <a href="https://futurism.com/nasa-life-mars-trump">NASA Announces Possible Discovery of Life on Mars by Comically Sucking Up to Trump</a>

Sample Return on Investment

Seriously, Trump?

Now That NASA Found Signs of Life on Mars, It's Clear Trump Made a Massive Error

Elon Musk's tunnel-digging firm The Boring Company's operations in Las Vegas were suspended after a worker sustained a "crushing injury."
As local fire department and OSHA spokespeople <a href="https://fortune.com/2025/09/11/boring-company-job-site-injury-reported-safety/">told Fortune</a>, authorities received a call about an "industrial/machinery incident" on Wednesday evening. The injured worker had to be lifted out of the tunnel using an on-site crane and transported to a local hospital.
"The patient is reported to be stable," a spokeswoman for the patient told Fortune.
It's yet another example of the poor safety record of Musk's many businesses. Both SpaceX and Tesla have faced significant criticism for <a href="https://www.reuters.com/investigates/special-report/spacex-musk-safety/">soaring injury rates</a> and workers being forced to work in <a href="https://www.thenation.com/article/society/tesla-racism-sexual-harassment/">poor conditions</a>.
An ongoing investigation into the latest Boring Company incident has ground operations to a halt at the company's construction site along Paradise Road in Las Vegas. The company is hoping to expand its tunnel network to the city's airport, as part of its approved plans to dig a 68-mile and 105-station transit system, <a href="https://www.boringcompany.com/vegas-loop">dubbed the "Vegas Loop</a>."
It's a highly ambitious plan, aiming to cut travel time from the Harry Reid International Airport to the Las Vegas Convention Center down to just five minutes.
But considering the company has only completed 3.5 miles of tunnels below the convention center over the last six years, it may not be up to the task.
The Boring Company has <a href="https://sanantonioreport.org/san-antonio-airport-downtown-tunnel-elon-musk-boring-company/">also abandoned</a> <a href="https://www.latimes.com/california/story/2022-07-27/elon-musk-abandons-plans-for-ontario-airport-tunnel">several plans</a> for other transportation systems around the country.
As <a href="https://www.bloomberg.com/news/features/2025-07-31/elon-musk-s-boring-company-is-digging-tunnels-for-spacex-xai-office?sref=YfHlo0rL">Bloomberg reported</a> earlier this year, Musk's firm has fallen far short of his highly ambitious goals of revolutionizing transportation using an <a href="https://futurism.com/elon-musk-just-announced-he-has-approval-to-build-a-29-minute-hyperloop-from-ny-to-dc">ultra-fast underground hyperloop</a>. Its most notable sources of revenue so far have been the <a href="https://www.boringcompany.com/not-a-flamethrower">sale of a "not-a-flamethrower</a>," as well as a perfume that <a href="https://www.boringcompany.com/burnthair">smells like burnt hair</a>.
According to Fortune, the Boring Company still doesn't have permits to dig a tunnel below the iconic Las Vegas Boulevard, a portion of which is better known as the Strip.
Worse yet, instead of shuttling passengers through the narrow tunnels inside high-capacity trains, the Boring Company has opted for a long line of modified Tesla sedans, which have already <a href="https://futurism.com/the-byte/elon-musk-tunnel-traffic-jam">caused traffic jam-like situations underground</a>. That's a far cry from the <a href="https://futurism.com/elon-musk-hyperloop-pedestrians-mass-transit">original renders</a> the company showed off seven years ago, involving "hyperloop" tunnels and futuristic, mass transit-like shuttles.
The company has also employed human drivers to chauffeur each Tesla through the tunnels. While it started testing self-driving vehicles with safety drivers last month, convention center executives <a href="https://fortune.com/2025/08/26/tesla-self-driving-cars-testing-boring-co-tunnels-las-vegas/">told Fortune</a> full autonomy is still "ways off."
More on the Boring Company: <a href="https://futurism.com/the-byte/drivers-lost-elon-musk-tunnel">Confused Drivers Are Somehow Getting Lost in Elon Musk's Goofy Las Vegas Tunnel</a>

Hole Story

"The patient is reported to be stable."

Boring Company Pauses Work After Worker Gets Crushed

Google is depending on thousands of contractors to train the AI behind its flagship chatbot Gemini.
As <a href="https://www.theguardian.com/technology/2025/sep/11/google-gemini-ai-training-humans">The Guardian reports</a>, many of these "AI raters," tasked with instructing the model and <a href="https://futurism.com/the-byte/google-demo-gemini-ai-tech-fail">correcting its many mistakes</a>, are facing poor working conditions and are often exposed to extremely disturbing content.
It's yet again a worrying reminder that despite tech companies attempting to paint their AI models as miraculous, autonomous fountains of knowledge and cognition that could eventually replace human workers, the current reality is the exact opposite: AI relies on the labor of huge numbers of hidden humans to give it the illusion of intelligence.
And it's not just the current crop of large language models — AI raters across the globe are being <a href="https://www.economist.com/international/2025/04/10/there-is-a-vast-hidden-workforce-behind-ai">tasked to label data</a> for AI-based, self-driving car software, and other related applications.
"AI isn’t magic; it’s a pyramid scheme of human labor," Distributed AI Research Institute's Adio Dinika told the Guardian. "These raters are the middle rung: invisible, essential and expendable."
In the realm of large language models like Google's Gemini, raters are being tasked with moderating the output of AI, not only ensuring the accuracy of its responses, but also that it doesn't expose users to inappropriate content.
Rachel Sawyer, one such worker who was contracted as a "generalist rater" for Google last year, told the Guardian that she was "shocked" after being told to work with "distressing content."
"Not only because I was given no warning and never asked to sign any consent forms during onboarding," she added, "but because neither the job title or description ever mentioned content moderation."
According to the newspaper, raters with no special expertise are being asked to verify the accuracy across highly complex domains such as architecture, astrophysics, and even medical guidance.
One rater told the newspaper that she was pressured to work faster, while being tasked with training the AI on sensitive medical topics like chemotherapy treatments for bladder cancer.
"I pictured a person sitting in their car finding out that they have bladder cancer and googling what I’m editing," she said.
In a statement to the Guardian, Google refuted the idea that raters are responsible for making AI models sound smart. A spokesperson argued that "quality raters" were "employed to provide external feedback on our products," which "help us measure how well our systems are working, but do not directly impact our algorithms or models."
Japanese contractor GlobalLogic contracted thousands of raters in the US to train Google's AI. Their hourly wages of $16 to $21 an hour — dwarfed by the <a href="https://www.nytimes.com/2025/07/31/technology/ai-researchers-nba-stars.html">astronomical salaries</a> of AI researchers — tend to still be far higher than their counterparts in Africa and South Africa, per the Guardian's sources. But it's not an easy job, and one that that can leave lasting scars.
"They are people with expertise who are doing a lot of great writing work, who are being paid below what they’re worth to make an AI model that, in my opinion, the world doesn’t need," one rater told the newspaper.
Many raters are also being pressured to work on tight deadlines and cope with rapidly changing guidelines.
"We had no idea where it was going, how it was being used or to what end," a different GlobalLogic-contracted rater told the Guardian.
It's a sobering glimpse behind the scenes, highlighting how much human work goes into tech that's being sold as a revolutionary way to circumvent human labor. Yet despite the raters' best efforts, frontier AI models including Google's Gemini continue to frequently hallucinate, bungling answers to the simplest of prompts.
More on AI: <a href="https://futurism.com/godfather-of-ai-unemployment">Godfather of AI Says His Creation Is About to Unleash Massive Unemployment</a>