Deranged Video Shows AI Job Recruiter Absolutely Losing It During an Interview

Looking for work is already arduous enough — but for one job-seeker, the process was made worse by an insane AI recruiter.

Looking for work is already arduous enough — but for one job-seeker, the process became something out of a deleted "Black Mirror" scene when the AI recruiter she was paired with went veritably insane.

In a buckwild TikTok video, the job-seeker is seen suffering for nearly 30 seconds as the AI recruiter barked the term "vertical bar pilates" at her no fewer than 14 times, often slurring its words or mixing up letters along the way.

@its_ken04

It was genuinely so creepy and weird. Please stop trying to be lazy and have AI try to do YOUR JOB!!! It gave me the creeps so bad #fyp

? original sound - Its Ken ?

The incident — and the way it affected the young woman who endured it — is a startling example not only of where America's abysmal labor market is at, but also of how ill-conceived this sort of AI "outsourcing" has become.

Though she looks nonplussed on her interview screen, the TikToker who goes by Ken told 404 Media that she was pretty perturbed by the incident, which occurred during her first (and only) interview with a Stretch Lab fitness studio in Ohio.

"I thought it was really creepy and I was freaked out," the college-aged creator told the website. "I was very shocked, I didn’t do anything to make it glitch so this was very surprising."

As 404 discovered, the glitchy recruiter-bot was hosted by a Y Combinator-backed startup called Apriora, which claims to help companies "hire 87 percent faster" and "interview 93 percent cheaper" because multiple candidates can be interviewed simultaneously.

In a 2024 interview with Forbes, Apriora cofounder Aaron Wang attested that job-seekers "prefer interviewing with AI in many cases, since knowing the interviewer is AI helps to reduce interviewing anxiety, allowing job seekers to perform at their best."

That's definitely not the case for Ken, who said she would "never go through this process again."

"If another company wants me to talk to AI," she told 404, "I will just decline."

Commenters on her now-viral TikTok seem to agree as well.

"This is the rudest thing a company could ever do," one user wrote. "We need to start withdrawing applications folks."

Others still pointed out the elephant in the room: that recruiting used to be a skilled trade done by human workers.

"Lazy, greedy and arrogant," another person commented. "AI interviews just show me they don't care about workers from the get go. This used to be an actual human's job."

Though Apriora didn't respond to 404's requests for comment, Ken, at least, has gotten the last word in the way only a Gen Z-er could.

"This was the first meeting [with the company] ever," she told 404. "I guess I was supposed to earn my right to speak to a human."

More on AI and labor: High Schools Training Students for Manual Labor as AI Looms Over College and Jobs

The post Deranged Video Shows AI Job Recruiter Absolutely Losing It During an Interview appeared first on Futurism.

Visit link:
Deranged Video Shows AI Job Recruiter Absolutely Losing It During an Interview

Crypto Prediction Platform Polymarket Was Utterly Wrong About the New Pope

Cardinal Robert Prevost shocked the world, succeeding Pope Francis to become the first American Pope. Polymarket bettors were surprised, too.

Papal Predictor

Polymarket, the crypto-based prediction market where users can place bets on events ranging from national elections to natural disasters, got the Pope odds way wrong.

Those forecasts were reflected in Polymarket as well as the similar platform Kalshi, which both took a hard turn toward Parolin as the public caught sight of white smoke, according to Axios. Meanwhile, market betters had Prevost hovering around a one-to-two percent chance of becoming the next Pope.

And yet, in a stunning twist of fate, it was the American-born Robert Prevost — a graduate of Villanova and the conclaver with the highest likelihood of ever consuming a hot dog and/or $1.99 pizza slice from Costco — who clinched the title. According to a screenshot posted to X-formerly-Twitter by Brew Markets, Prevost was sitting at around 0.03 percent when his shock election was announced.

In short, Polymarketers were absolutely blindsided — highlighting how Polymarket's oft-exalted predictive prowess isn't always all it's cracked up to be.

Robert F. Prevost has been elected the first American pope in history.

Polymarket gave him just a 0.3% chance of being chosen. pic.twitter.com/QzD41JYpU9

— Brew Markets (@brewmarkets) May 8, 2025

Vatican Bombshell

The Vatican shocked the world today when the Chicago-born American and Peruvian was named Pope Leo XIV before taking to the balcony of St. Peter's Basilica to greet the world as Holy Father.

Prevost is the first American Pope, and appears to be the first Pope to retweet Catholic Snoopy posts.

Though Prevost was an ally of his predecessor, the late Pope Francis, he was by no means seen as the frontrunner; most expected either Italian Cardinal Pietro Parolin or Filipino Cardinal Luis Antonio Tagle to ascend to the papacy.

The hard swing towards Parolin in the moments before the announcement broke was likely due to the conclave's speediness, as Axios pointed out.

Anyway, if you're a Polymarket who bet on the American Prevost to win the papacy, email us!

More on Polymarket: Crypto Platforms Like Polymarket Now Taking Bets on Los Angeles Fire Devastation

The post Crypto Prediction Platform Polymarket Was Utterly Wrong About the New Pope appeared first on Futurism.

The rest is here:
Crypto Prediction Platform Polymarket Was Utterly Wrong About the New Pope

Scientists Successfully Grow Human Tooth in Lab, With Aim of Implanting in Humans

Scientists at King's College London, UK, say they've successfully grown a human tooth in a lab for the first time.

Scientists at King's College London say they've successfully grown a human tooth in a lab for the first time.

As detailed in a paper published in the journal ACS Macro Letters, the team said it uncovered a potential way to regrow teeth in humans as a natural alternative to conventional dental fillings and implants, research they say could "revolutionize dental care."

The researchers claim they've developed a new type of material that enables cells to communicate with one another, essentially allowing one cell to "tell" another to differentiate itself into a new tooth cell.

In other words, it mimics the way teeth grow naturally, an ability we lose as we grow older.

"We developed this material in collaboration with Imperial College to replicate the environment around the cells in the body, known as the matrix," explained author and King’s College London PhD student Xuechen Zhang in a statement. "This meant that when we introduced the cultured cells, they were able to send signals to each other to start the tooth formation process."

"Previous attempts had failed, as all the signals were sent in one go," he added. "This new material releases signals slowly over time, replicating what happens in the body."

However, porting the discovery from the lab, and transforming it into a viable treatment will require years of research.

"We have different ideas to put the teeth inside the mouth," Xuechen said."We could transplant the young tooth cells at the location of the missing tooth and let them grow inside mouth. Alternatively, we could create the whole tooth in the lab before placing it in the patient’s mouth."

While we're still some ways away from applying the findings to human subjects, in theory the approach could have some significant advantages over conventional treatments like fillings and implants.

"Fillings aren’t the best solution for repairing teeth," said Xuechen. "Over time, they will weaken tooth structure, have a limited lifespan, and can lead to further decay or sensitivity."

"Implants require invasive surgery and good combination of implants and alveolar bone," he added. "Both solutions are artificial and don’t fully restore natural tooth function, potentially leading to long-term complications."

The new approach, in contrast, could offer a better long-term solution.

"Lab-grown teeth would naturally regenerate, integrating into the jaw as real teeth," Xuechen explained. "They would be stronger, longer lasting, and free from rejection risks, offering a more durable and biologically compatible solution than fillings or implants."

While nobody knows whether lab-grown teeth will become a viable dental treatment, experts remain optimistic.

"This new technology of regrowing teeth is very exciting and could be a game-changer for dentists," King's College clinical lecturer in prosthodontics Saoirse O'Toole, who was not involved in the study, told the BBC. "Will it come in my lifetime of practice? Possibly. In my children's dental lifetimes? Maybe. But in my children's children's lifetimes, hopefully."

More on lab teeth: Scientists Grow Living "Replacement Teeth" for Dental Implants

The post Scientists Successfully Grow Human Tooth in Lab, With Aim of Implanting in Humans appeared first on Futurism.

See original here:
Scientists Successfully Grow Human Tooth in Lab, With Aim of Implanting in Humans

California Nuclear Power Plant Deploys Generative AI Safety System

America's first nuclear power plant to use artificial intelligence is, ironically, the last operational one in California. 

America's first nuclear power plant to use artificial intelligence is, ironically, the last operational one in California.

As CalMatters reports, the Diablo Canyon power plant is slated to be decommissioned by the end of this decade. In the interim, the plant's owner, Pacific Gas & Electric (PG&E), claims that it's deploying its "Neutron Enterprise" tool — which will be the first nuclear plant in the nation to use AI — in a series of escalating stages.

Less than 18 months ago, Diablo Canyon was hurtling headlong toward a decommissioning that would have begun in 2024 and ended this year. In late 2023, however, the California Public Utility Commission voted to stay its execution for five years, kicking the can on the inevitable to 2029 and 2030, respectively.

Just under a year after that vote, PG&E announced that it was teaming up with a startup called Atomic Canyon, which was founded with the plant in mind and is also based in the coastal Central California town of San Luis Obispo. That partnership, and the first "stage" of the tool's deployment, brought some of Nvidia's high-powered H100 AI chips to the dying nuclear plant, and with them the compute power needed for generative artificial intelligence.

Running on an internal server without cloud access, Neutron Enterprise's biggest use case, much like so-called AI "search engines," is summarizing a massive trove of millions of regulatory documents that have been fed into it. According to Atomic Canyon CEO and cofounder Trey Lauderdale, this isn't risky — though anyone who has used AI to summarize information knows better, because the tech still often makes factual mistakes.

Speaking to CalMatters, PG&E executive Maureen Zalawick insisted that the AI program will be more of a "copilot" than a "decision-maker," meant to assist flesh-and-blood employees rather than replace them.

"We probably spend about 15,000 hours a year searching through our multiple databases and records and procedures," Zalawick explained. "And that’s going to shrink that time way down."

Lauderdale put it in even simpler terms.

"You can put this on the record," he told CalMatters. "The AI guy in nuclear says there is no way in hell I want AI running my nuclear power plant right now."

If that "right now" caveat gives you pause, you're not alone. Given the shifting timelines for the closure of Diablo Canyon in a state that has been painstakingly phasing out its nuclear facilities since the 1970s over concerns about toxic waste — and the fact that Lauderdale claims to be talking to other plants in other states — there's ample cause for concern.

"The idea that you could just use generative AI for one specific kind of task at the nuclear power plant and then call it a day," cautioned Tamara Kneese of the tech watchdog Data & Society, "I don’t really trust that it would stop there."

As head of Data & Society's Climate, Technology, and Justice program, Kneese said that while using AI to help sift through tomes of documents is worthwhile, "trusting PG&E to safely use generative AI in a nuclear setting is something that is deserving of more scrutiny." This is the same company whose polluting propensities were exposed by the real-life Erin Brokovich in the 1990s, after all.

California lawmakers, meanwhile, were impressed by the tailored usage Atomic Canyon and PG&E propose for the program — but it remains to be seen whether or not that narrow functionality will remain that way.

More on AI and energy: Former Google CEO Tells Congress That 99 Percent of All Electricity Will Be Used to Power Superintelligent AI

The post California Nuclear Power Plant Deploys Generative AI Safety System appeared first on Futurism.

See more here:
California Nuclear Power Plant Deploys Generative AI Safety System

Astronaut Insists the Mushrooms He’s Growing in Space Are "Not the Ones You’re Thinking"

SpaceX's new Fram2 mission is experimenting with growing mushrooms in microgravity — but not the magic kind.

A crypto billionaire and a filthy rich ketamine user have launched a trip to the stars — stop us if you've heard this one before.

We promise it's not quite as Silicon Valley as it sounds, though as Australian explorer and freshman astronaut Eric Phillips told Ars Technica, there are shrooms involved.

Alongside Norwegian filmmaker Jannicke Mikkelsen, German roboticist Rabea Rogge, and Chinese crypto billionaire Chun Wang, Philips is a member of SpaceX's Fram2 mission. The first private flight of its kind, the four-person team launched in a Crew Dragon capsule atop a Falcon 9 rocket for the first-ever civilian mission flying over Earth's poles.

Chartered by Chun — and, of course, greenlit by SpaceX owner and resident White House psychonaut Elon Musk — the four-person crew launched on March 31 and are currently in orbit, working on nearly two dozen scientific experiments they have planned for their short journey.

Among them, as Ars noted, is the plan to become the first mushroom growers in space — but "they’re not the ones you’re thinking," Philips told the website. Instead, per a Fram2 statement released ahead of the launch, they'll be growing delectable oyster mushrooms.

FOODiQ Global, the Australian company behind the "Mission MushVroom" experiment aboard Fram2, said in the press release that "oyster mushrooms are the perfect space crop" because they grow rapidly and have tons of nutrients. They even have "the unique ability to make vitamin D," the statement noted.

Along with all those nutritional benefits, those yummy shrooms will almost certainly taste better than space food — if top space minds can figure out a way to cook them in orbit, that is.

In an op-ed for Business Insider, FOODiQ founder and CEO Flávia Fayet-Moore said that she identified mushrooms as an ideal in-orbit crop, particularly for years-long missions to Mars and other planets.

"Can you imagine eating thermostabilized, dehydrated food for five years?" the space nutritionist — yes, that is apparently a real thing — wrote. "I can't."

We won't know how well the shrooms grew in microgravity until Fram2 gets back to Earth this week.

More on space life: Boeing's Starliner Disaster Was Even Worse Than We Thought, Astronaut Reveals

The post Astronaut Insists the Mushrooms He's Growing in Space Are "Not the Ones You’re Thinking" appeared first on Futurism.

See the original post here:
Astronaut Insists the Mushrooms He's Growing in Space Are "Not the Ones You’re Thinking"

Strange Signal Coming From Dead Galaxy, Scientists Say

Astronomers say they've detected a mysterious type of signal known as a fast radio burst coming from an ancient, dead galaxy.

Radio Star

Astronomers say they've detected a mysterious type of signal known as a fast radio burst coming from an ancient, dead galaxy billions of light years away. Figuratively speaking, it makes for one hell of a sign of life. 

The findings, documented in two studies published in The Astrophysical Journal Letters, upends the long held belief that FRBs — extremely powerful pulses of energy — originate exclusively from star-forming regions of space, as dead galaxies no longer support the birth of new stars. 

Adding to the seeming improbability of the FRB's origin, the researchers believe that the signal's source came from the furthermost outskirts of the galaxy, about 130,000 light years from its center, with only moribund stars at the end of their stellar evolution for company.

"This is both surprising and exciting, as FRBs are expected to originate inside galaxies, often in star-forming regions," said Vishwangi Shah, lead author of one of the studies and an astronomer at McGill University, said in a statement about the work"The location of this FRB so far outside its host galaxy raises questions as to how such energetic events can occur in regions where no new stars are forming."

Quick and the Dead

Though they're often only milliseconds in duration, FRBs are so powerful at their source that a single pulse emits more energy than our Sun does in an entire year. 

What could cause such staggering outbursts? Astronomers have speculated that they originate from magnetars, a type of collapsed, extremely dense stellar object called a neutron star that maintains an unfathomably potent magnetic field, perhaps trillions of times stronger than Earth's.

But that theory is now being challenged by this latest FRB, designed FRB 20240209A, because there are no young stars in the 11.3 billion year old galaxy that could form magnetars. Only extremely massive stars, which have short lifespans as a consequence of their size and thus would need to have been recently formed, possess enough mass to collapse into neutron stars in the first place. 

Outcasts Together

FRB 20240209A isn't the first to be found in such a remote location. In 2022, astronomers detected another signal originating from the outskirts of its galaxy, Messier 81, where no active star formation was taking place.

"That event single-handedly halted the conventional train of thought and made us explore other progenitor scenarios for FRBs," said Wen-fai Fong, a coauthor of both studies and an astrophysicist at Northwestern University, in the statement. "Since then, no FRB had been seen like it, leading us to believe it was a one-off discovery — until now."

Crucially, the M81 FRB was found in a dense conglomeration of stars called a globular cluster. Given their similar circumstances, it led the astronomers to believe that FRB 20240209A could be residing in a globular cluster, too. To confirm this hunch, they hope to use the James Webb Telescope to image the region of space around the FRB's origins.

More on space: Scientists Intrigued by Planet With Long Tail

The post Strange Signal Coming From Dead Galaxy, Scientists Say appeared first on Futurism.

Here is the original post:
Strange Signal Coming From Dead Galaxy, Scientists Say

NASA’s Lunar Space Station Just Took a Massive Step Towards Launching

A core component of NASA's Gateway lunar space station just passed a grueling round of pressure tests, a big win for the project.

Under Pressure

NASA announced yesterday that its forthcoming Gateway lunar space station — an outpost designed to house astronauts in the Moon's orbit — just passed a critical milestone.

According to the agency, Gateway's Habitation and Logistics Outpost (HALO) successfully passed a grueling round of "static load testing," defined by NASA as a "rigorous stress test of how well the structure responds to the forces encountered in deep space."

In other words, HALO won't crumble or crack under the extreme conditions it'll face in lunar orbit.

"Static load testing is one of the major environmental stress tests HALO will undergo," NASA continues in its announcement, adding that HALO, which is currently in Italy, will be transferred to Arizona "once all phases of testing are complete." There, NASA contractor Northrop Grumman will add HALO's finishing touches.

HALO is one of "four pressurized Gateway modules where astronauts will live, conduct science, and prepare for missions to the lunar South Pole region," per NASA's announcement.

It's an exciting mile marker for Gateway, which stands to mark the first sustained human presence on and around our Moon — one of the core goals of NASA's ongoing Artemis program, and perhaps a stepping stone in humanity's efforts to send humans to Mars.

Looking Ahead

While the stress test was a key breakthrough for the Gateway mission, it's still a ways off from lift-off.

The outpost will launch in pieces, and the first components to take flight — HALO and the Power and Propulsion Element (PPE) — are slated for launch aboard a SpaceX Falcon Heavy rocket in December 2027 at the earliest. By conservative estimates, Gateway is not expected to be inhabited until 2028.

It's an ambitious plan and there's always a chance of delays. In the meantime, it's heartening to see NASA's Gateway, piece by piece, move forward.

"Gateway is humanity's first lunar space station supporting a new era of exploration and scientific discovery as part of NASA's Artemis campaign that will establish a sustained presence on and around the Moon," said NASA of the achievement, "paving the way for the first crewed mission to Mars."

More on the Artemis missions: NASA's Moon Launcher Is in Big Trouble

The post NASA’s Lunar Space Station Just Took a Massive Step Towards Launching appeared first on Futurism.

Read the original:
NASA’s Lunar Space Station Just Took a Massive Step Towards Launching

NASA’s Lunar Rover Prototype Looks Like If a Tractor and a Golf Cart Had a Baby

In preparation for the first crewed Moon mission in 50 years, NASA is prototyping a new Moon rover that looks a lot like a tractor/golf cart.

Keep Roving

In preparation for the first crewed lunar mission in half a century, NASA is prototyping a new Moon rover — and it looks like some pretty distinctly Earth-bound vehicles.

Announced in a NASA press release, the new Ground Test Unit (GTU) is currently in development at the agency's Johnson Space Center in Houston, Texas.

As photos of the prototype in action show, it's already been taken for a joyride by some big names: astronaut Kate Rubins, the first person to ever sequence DNA in space, and Apollo 17 pilot-turned-senator Harrison "Jack" Schmitt.

Those photos also show that the lunar terrain vehicle (LTV) looks, for some reason, a lot like a mix between a tractor and a golf cart — and the agency didn't mention why, exactly, it was built to look that way.

Private Partnership

Though there's little explanation about those visual references, NASA did explain how this LTV prototype, which will never be sent to the Moon, was built.

Earlier this year, the American space agency contracted three private companies — Intuitive Machines, Lunar Outpost, and Venturi Astrolab — to create components for the GTU.

"The Ground Test Unit will help NASA teams on the ground, test and understand all aspects of rover operations on the lunar surface ahead of Artemis missions," NASA engineering lead Jeff Somers explained in the press release. "The GTU allows NASA to be a smart buyer, so we are able to test and evaluate rover operations while we work with the LTVS contractors and their hardware."

Designed to carry two astronauts, the GTU also has additional capabilities allowing for it to be operated remotely, which makes it sort of sound like it can be "summoned" a la Tesla.

Because we're curious, Futurism has reached out to NASA to ask if either tractors or golf carts were used as references for this GTU prototype — and we're hoping the answer will be yes to both.

More on NASA's lunar ambitions: Scientists Outraged at Canceled NASA Moon Mission Plead Congress to Reconsider

The post NASA’s Lunar Rover Prototype Looks Like If a Tractor and a Golf Cart Had a Baby appeared first on Futurism.

Follow this link:
NASA’s Lunar Rover Prototype Looks Like If a Tractor and a Golf Cart Had a Baby