AI hiring is here. It’s making companies — and job seekers — miserable. As America’s labor market slows, AI-led interviews and auto-generated cover letters are dramatically changing the process of getting a job. And maybe not for the better.

More than half of the organizations surveyed by the Society for Human Resource Management used AI to recruit workers in 2025. And an estimated third of ChatGPT users reportedly leaned on the OpenAI chatbot to help with their job search.

However, recent research found that when job seekers use AI during the process, applicants are less likely to be hired. Meanwhile, companies are fielding an increased volume of applications.

“The ability (for companies) to select the best worker today may be worse due to AI,” said Anaïs Galdin, a Dartmouth researcher who co-authored a study looking at how large language models (LLMs) have impacted cover letters.

Galdin and her co-author, Jesse Silbert at Princeton, analyzed cover letters for tens of thousands of job applications on Freelancer.com, a jobs listing site.

The researchers found that after the introduction of ChatGPT in 2022, the letters all got longer and better-written, but companies stopped putting so much stock in them. That made it harder to distinguish a qualified hire from the rest of the applicant pool, and the rate of hiring dropped as did the average starting wage.

“If we do nothing to make information flow better between workers and firms, then we might have an outcome that looks something like this,” said Silbert, referring to the results of his study.

And with more applications to review, employers are automating the interview itself.

A majority (54 per cent) of the US job seekers surveyed by recruiting software firm Greenhouse in October said they’ve had an AI-led interview. Virtual interviews exploded in popularity during the pandemic in 2020. Many companies now use AI to ask the questions, but that hasn’t made the process any less subjective.

“Algorithms can copy and even magnify human biases,” said Djurre Holtrop, a researcher who has conducted studies about the use of asynchronous video interviews, algorithms, and LLMs in hiring. “Every developer needs to be wary of that.”

Daniel Chait, CEO of Greenhouse, warned that with AI infiltrating hiring – from applicants using the tool to apply to hundreds of jobs and employees automating the process in response – it has created a “doom loop” making everyone miserable.

“Both sides are saying, ‘This is impossible, it’s not working, it’s getting worse,’” Chait told CNN.
Pushing back

Employers are embracing the technology — one estimate projects the market for recruiting technology will grow to US$3.1 billion by the end of this year. But state lawmakers, labor groups and individual workers have begun pushing back over fears that AI could discriminate against workers.

Liz Shuler, president of the AFL-CIO labor union, called the use of AI in hiring “unacceptable.”

“AI systems rob workers of opportunities they’re qualified for based on criteria as arbitrary as names, zip codes, or even how often they smile,” Shuler said in a statement to CNN.

States such as California, Colorado, and Illinois are enacting new laws and regulations aimed at creating standards for the technology’s use in hiring, among other areas.

A recent executive order signed by President Donald Trump threatens to undermine state-level AI regulations. Samuel Mitchell, a Chicago-based lawyer who argues employment cases, said that the order can’t “preempt” state law but does add to the “ongoing uncertainty” around new regulations on the tech.

However, he added that existing anti-discrimination laws still apply to hiring, even if a company uses AI. And lawsuits are already being filed.

In a case backed by the American Civil Liberties Union, a deaf woman is suing HireVue (an AI-powered recruiting company) over claims an automated interview she was subject to did not meet accessibility standards required by law.

HireVue denied the claim and told CNN that its technology works to reduce bias through a “foundation of validated behavioral science.”

But despite initial challenges, AI hiring seems here to stay. And to be sure, new developments in AI have led to more sophisticated ways to analyze resumes, opening doors for candidates who may have otherwise been overlooked.

But those who value the “human touch” in hiring are left wanting.

Jared Looper, an IT project manager based in Salt Lake City, Utah, began his career as a recruiter. As part of his current job search, he was interviewed by an AI recruiter.

#Paraplegic engineer becomes the first wheelchair user to blast into space.

A paraplegic engineer from Germany blasted off on a dream-come-true rocket ride with five other passengers Saturday, leaving her wheelchair behind to float in space while beholding Earth from on high.

Severely injured in a mountain bike accident seven years ago, Michaela Benthaus became the first wheelchair user in space, launching from West Texas with Jeff Bezos’ company Blue Origin. She was accompanied by a retired SpaceX executive also born in Germany, Hans Koenigsmann, who helped organize and, along with Blue Origin, sponsored her trip. Their ticket prices were not divulged.

An ecstatic Benthaus said she laughed all the way up and tried to turn upside down in space.

“It was the coolest experience,” she said shortly after landing.

The 10-minute space-skimming flight required only minor adjustments to accommodate Benthaus, according to the company. That’s because the autonomous New Shepard capsule was designed with accessibility in mind, “making it more accessible to a wider range of people than traditional spaceflight,” said Blue Origin’s Jake Mills, an engineer who trained the crew and assisted them on launch day.

Among Blue Origin’s previous space tourists: those with limited mobility and impaired sight or hearing, and a pair of 90-year-olds.

For Benthaus, Blue Origin added a patient transfer board so she could scoot between the capsule’s hatch and her seat. The recovery team also unrolled a carpet on the desert floor following touchdown, providing immediate access to her wheelchair, which she left behind at liftoff. She practiced in advance, with Koenigsmann taking part with the design and testing. An elevator was already in place at the launch pad to ascend the seven stories to the capsule perched atop the rocket.

Benthaus, 33, part of the European Space Agency’s graduate trainee program in the Netherlands, experienced snippets of weightlessness during a parabolic airplane flight out of Houston in 2022. Less than two years later, she took part in a two-week simulated space mission in Poland.

“I never really thought that going on a spaceflight would be a real option for me because even as like a super healthy person, it’s like so competitive, right?” she told The Associated Press ahead of the flight.

Her accident dashed whatever hope she had. “There is like no history of people with disabilities flying to space,” she said.

When Koenigsmann approached her last year about the possibility of flying on Blue Origin and experiencing more than three minutes of weightlessness on a space hop, Benthaus thought there might be a misunderstanding. But there wasn’t, and she immediately signed on.

It’s a private mission for Benthaus with no involvement by ESA, which this year cleared reserve astronaut John McFall, an amputee, for a future flight to the International Space Station. The former British Paralympian lost his right leg in a motorcycle accident when he was a teenager.

An injured spinal cord means Benthaus can’t walk at all, unlike McFall who uses a prosthetic leg and could evacuate a space capsule in an emergency at touchdown by himself. Koenigsmann was designated before flight as her emergency helper; he and Mills lifted her out of the capsule and down the short flight of steps at flight’s end.

“You should never give up on your dreams, right?” Benthaus urged following touchdown.

Benthaus was adamant about doing as much as she could by herself. Her goal is to make not only space accessible to the disabled, but to improve accessibility on Earth too.

While getting lots of positive feedback within “my space bubble,” she said outsiders aren’t always as inclusive.

“I really hope it’s opening up for people like me, like I hope I’m only the start,” she said.

U.S. mulls letting Nvidia sell H200 chips to #China, sources say. The Trump administration is considering greenlighting sales of Nvidia’s H200 artificial intelligence chips to China, people familiar with the matter said, as a bilateral detente boosts prospects for exports of advanced U.S. technology to China.

The U.S. Commerce Department, which oversees U.S. export controls, is reviewing a change to its policy of barring sales of such chips to China, the sources said, stressing that plans could change.

The White House and the U.S. Commerce Department did not immediately respond to requests for comment. Nvidia did not comment directly on the review but said current regulation does not allow the company to offer a competitive AI data center chip in China, leaving that massive market to its rapidly growing foreign competitors.

The possibility signals a friendlier approach to China, after U.S. President Donald Trump and Chinese leader Xi Jinping brokered a trade and tech war truce in Busan last month.

China hawks in Washington are concerned that shipments of more advanced AI chips to China could help Beijing supercharge its military, fears that prompted the Biden administration to set limits on such exports.

Faced with Beijing’s muscular use of export controls on rare earth minerals, critical for producing a raft of tech goods, Trump this year has threatened new restrictions on tech exports to China, but ultimately rolled them back in most cases.

The H200 chip, unveiled two years ago, has more high-bandwidth memory than its predecessor the H100, allowing it to process data more quickly.

It is estimated to be twice as powerful as Nvidia’s H20 chip, the most advanced AI semiconductor that can legally be exported to China, after the Trump administration reversed its short-lived ban on such sales earlier this year.

Earlier this week, Nvidia CEO Jensen Huang, whom Trump has described as a “great guy,” was among the guests at the White House during Saudi Crown Prince Mohammed bin Salman’s visit.

The U.S. Commerce Department announced this week it had approved shipments of the equivalent of up to 70,000 Nvidia Blackwell chips, Nvidia’s next-generation AI chip, to Saudi Arabia’s Humain and G42 of the United Arab Emirates.

CAE signs new agreement with Sweden-based Saab to provide training services, devices.

MONTREAL — CAE Inc. has announced an agreement to be the preferred supplier of certain training and simulation equipment for Sweden-based Saab’s airborne early warning system.

The Montreal-based company says the agreement will leverage CAE’s expertise in advanced training systems and the delivery of flight training devices.

Under the agreement, CAE will provide simulation-based training to support Saab’s airborne early warning system.

CAE says it will also provide pilot and maintenance training services.

CAE CEO Matt Bromberg says in a press release that training and simulation expertise is critical to defence forces in Canada and around the world, while the agreement with Saab responds to an evolving geopolitical landscape that requires stronger partnerships.

CAE’s announcement comes after Sweden’s King Carl XVI Gustaf and Queen Silvia arrived in Ottawa on Tuesday, with Prime Minister Mark Carney saying the two countries were signing a strategic partnership.

Big tech platforms, including TikTok and Netflix, are all looking to add video podcasts to their content, a rapidly growing format that attracts a young audience prized by advertisers and where YouTube dominates.

Roman Wasenmüller, the head of podcasts at Spotify, said that the development “marks a new chapter for podcasting,” as he announced a partnership with Netflix in mid-October.

Google-owned YouTube is now the leading destination for podcasts in the United States, with 33 per cent market share according to Edison Research, and more than one billion consumers worldwide.

YouTube even offers audio show producers the ability to generate video using artificial intelligence to better illustrate their show on the platform.

Spotify is now moving in, and by the end of September, 390 million users of the Swedish streaming service had watched at least one video-version of a podcast.

“Podcasting is correctly seen as a medium that is growing in usage considerably faster than a lot of legacy media, and that’s attractive to investors,” said Martin Spinelli, a podcasting professor at the University of Sussex.
Connect with Gen Z

The rise of the podcast is particularly pronounced among young people, noted Yoram Wurmser, an analyst at Emarketer.

“Gen Z are heavy podcast listeners and viewers, so it’s good to connect with them” as much as possible, Wurmser said.

Donald Trump’s podcast tour during his U.S. presidential campaign helped increase his popularity among young adults.

Democrat Zohran Mamdani, recently elected mayor of New York, also gave significant attention to podcasters.

A latecomer to this space, Netflix has already announced the release, in early 2026, of a dozen programs licenced from Spotify.

But according to Business Insider, Netflix aims to quickly offer more than 50 and eventually up to 200 in total that will include new productions made for Netflix.

TikTok announced Monday a collaboration with American radio giant iHeartMedia, aimed at launching up to 25 programs hosted by influencers.

“We’re combining our vast networks to deliver relevant content on a massive scale,” said Rich Bressler, iHeartMedia’s president. “It’s a win for creators, fans, and brands alike.”

TikTok will not broadcast entire episodes but excerpts, similar to what most successful podcasts already make available, some garnering several million views each time.

The Chinese-owned social network doesn’t produce its own videos, any more than YouTube does, but Martin Spinelli expects streaming services to “amplify the content that they’ve already monetized” using podcasts.

Even before launching its offensive, Netflix had already produced audio extras for several of its hit series, from “The Crown” to “Heartstopper.”

Disney has also heavily invested in this niche and announced in early September the arrival of podcasts spun off from series like “Only Murders in the Building” and “The Secret Lives of Mormon Wives.”

For existing podcasts, access to new audiences using video should increase revenues, said Emarketer’s Wurmser, whether through advertising, subscriptions or merchandise.

Spinelli sees it as a victory for independent podcasts, which “will be able to expand their audience.”

“It’s much easier to find” a podcast based on its content “on YouTube than on Apple Podcasts,” said Spinelli, who himself hosts the program “For Your Ears Only,” “and I think it will be the same for Netflix and TikTok.”

--

By Thomas Urbain

James Watson, co-discoverer of the double-helix shape of #DNA, has died at age 97.

James D. Watson, whose co-discovery of the twisted-ladder structure of DNA in 1953 helped light the long fuse on a revolution in medicine, crimefighting, genealogy and ethics, has died. He was 97.

The breakthrough -- made when the brash, Chicago-born Watson was just 24 -- turned him into a hallowed figure in the world of science for decades. But near the end of his life, he faced condemnation and professional censure for offensive remarks, including saying Black people are less intelligent than white people.

Watson shared a 1962 Nobel Prize with Francis Crick and Maurice Wilkins for discovering that deoxyribonucleic acid, or DNA, is a double helix, consisting of two strands that coil around each other to create what resembles a long, gently twisting ladder.

That realization was a breakthrough. It instantly suggested how hereditary information is stored and how cells duplicate their DNA when they divide. The duplication begins with the two strands of DNA pulling apart like a zipper.

Even among non-scientists, the double helix would become an instantly recognized symbol of science, showing up in such places as the work of Salvador Dali and a British postage stamp.

The discovery helped open the door to more recent developments such as tinkering with the genetic makeup of living things, treating disease by inserting genes into patients, identifying human remains and criminal suspects from DNA samples and tracing family trees. But it has also raised a host of ethical questions, such as whether we should be altering the body’s blueprint for cosmetic reasons or in a way that is transmitted to a person’s offspring.

“Francis Crick and I made the discovery of the century, that was pretty clear,” Watson once said. He later wrote: “There was no way we could have foreseen the explosive impact of the double helix on science and society.”

Watson never made another lab finding that big. But in the decades that followed, he wrote influential textbooks and a best-selling memoir and helped guide the project to map the human genome. He picked out bright young scientists and helped them. And he used his prestige and contacts to influence science policy.

Watson died in hospice care after a brief illness, his son said Friday. His former research lab confirmed he had passed away a day earlier.

“He never stopped fighting for people who were suffering from disease,” Duncan Watson said of his father.

Watson’s initial motivation for supporting the gene project was personal: His son Rufus had been hospitalized with a possible diagnosis of schizophrenia, and Watson figured that knowing the complete makeup of DNA would be crucial for understanding that disease -- maybe in time to help his son.

He gained unwelcome attention in 2007, when the Sunday Times Magazine of London quoted him as saying he was “inherently gloomy about the prospect of Africa” because “all our social policies are based on the fact that their intelligence is the same as ours -- where all the testing says not really.” He said that while he hopes everyone is equal, “people who have to deal with Black employees find this is not true.”

He apologized, but after an international furor he was suspended from his job as chancellor of the prestigious Cold Spring Harbor Laboratory in New York. He retired a week later. He had served in various leadership jobs there for nearly 40 years.

In a television documentary that aired in early 2019, Watson was asked if his views had changed. “No, not at all,” he said. In response, the Cold Spring Harbor lab revoked several honourary titles it had given Watson, saying his statements were “reprehensible” and “unsupported by science.”

Watson’s combination of scientific achievement and controversial remarks created a complicated legacy.

He has shown “a regrettable tendency toward inflammatory and offensive remarks, especially late in his career,” Dr. Francis Collins, director of the National Institutes of Health, said in 2019. “His outbursts, particularly when they reflected on race, were both profoundly misguided and deeply hurtful. I only wish that Jim’s views on society and humanity could have matched his brilliant scientific insights.”

Long before that, Watson scorned political correctness.

“A goodly number of scientists are not only narrow-minded and dull, but also just stupid,” he wrote in “The Double Helix,” his bestselling 1968 book about the DNA discovery.

For success in science, he wrote: “You have to avoid dumb people. ... Never do anything that bores you. ... If you can’t stand to be with your real peers (including scientific competitors) get out of science. ... To make a huge success, a scientist has to be prepared to get into deep trouble.”

It was in the fall of 1951 that the tall, skinny Watson -- already the holder of a Ph.D. at 23 -- arrived at Britain’s Cambridge University, where he met Crick. As a Watson biographer later said, “It was intellectual love at first sight.”

Crick himself wrote that the partnership thrived in part because the two men shared “a certain youthful arrogance, a ruthlessness, and an impatience with sloppy thinking.”

Together they sought to tackle the structure of DNA, aided by X-ray research by colleague Rosalind Franklin and her graduate student Raymond Gosling. Watson was later criticized for a disparaging portrayal of Franklin in “The Double Helix,” and today she is considered a prominent example of a female scientist whose contributions were overlooked. (She died in 1958.)

Watson and Crick built Tinker Toy-like models to work out the molecule’s structure. One Saturday morning in 1953, after fiddling with bits of cardboard he had carefully cut to represent fragments of the DNA molecule, Watson suddenly realized how these pieces could form the “rungs” of a double helix ladder.

His first reaction: “It’s so beautiful.”

Following the discovery, Watson spent two years at the California Institute of Technology, then joined the faculty at Harvard in 1955. Before leaving Harvard in 1976, he essentially created the university’s program for molecular biology, scientist Mark Ptashne recalled in a 1999 interview.

Watson became director of the Cold Spring Harbor lab in 1968, its president in 1994 and its chancellor 10 years later. He made the lab on Long Island an educational center for scientists and non-scientists, focused research on cancer, instilled a sense of excitement and raised huge amounts of money.

He transformed the lab into a “vibrant, incredibly important center,” Ptashne said. It was “one of the miracles of Jim: a more disheveled, less smooth, less typically ingratiating person you could hardly imagine.”

Training your #brain may improve focus and attention. Many people have started playing Wordle or doing crossword puzzles, but those brain ticklers don’t appear to improve overall cognition, CNN Chief Medical Correspondent Dr. Sanjay Gupta recently told viewers on CNN.

“What crossword puzzles and word games are probably really good at is making you better at crossword puzzles and word games,” Gupta said. “People often use brain-training exercises with the hopes of reducing their risk (of dementia). Truth is, there’s not a lot of data on this to suggest it actually reduces your risk for dementia.”

But there’s a new twist to that scientific conundrum: According to a new clinical trial, the success of brain training in slowing cognitive decline may depend on the type of game and how it affects certain neurotransmitters in the brain.

Brain games that focus on boosting attention and improving processing speed — like BrainHQ’s Double Decision and Freeze Frame — appear to preserve acetylcholine, a key neurotransmitter, according to the new research.

Acetylcholine is an “excitatory” neurotransmitter and neuromodulator, which “functions like a switch to make the brain more awake, more focused and attentive,” said senior study author Etienne de Villers-Sidani, an associate professor in the Department of Neurology and Neurosurgery at McGill University in Montreal.

When acetylcholine is activated, it changes the activity of the entire brain, said Dr. Michael Merzenich, professor emeritus at the University of California, San Francisco and co-founder and chief scientist of Posit Science, which makes BrainHQ, a for-profit brain-training company.

An elder statesman in the field of neuroplasticity, Merzenich and two other scientists received the prestigious Kavli Award in Neuroscience in 2016 for their groundbreaking discoveries that the adult brain could change, adapt and create new neural connections throughout life. Prior to their discoveries, it was thought that the brain was unable to change or regenerate after a certain point in early adulthood.

“This is the first human study to document an upregulation of acetylcholine, which is absolutely crucial for sustaining brain plasticity in aging,” Merzenich said.

Upregulation causes a cell to add more receptors for a neurotransmitter, thus boosting its ability to respond.

“This is an important study because the training had a brain-wide impact — it’s not limited to the very narrow set of processes that people were trained on,” Merzenich said. “We’re talking about a fundamental physical chemical change that we know really matters as a contributor to brain health.”

The discovery adds to existing knowledge about how to prevent cognitive decline, said preventive neurologist Dr. Richard Isaacson, director of research at the Institute for Neurodegenerative Diseases in Boca Raton, Florida.

Eating a well-balanced diet, improving sleep and getting regular exercise are all proven ways to enhance brain power and overall vitality. Research has also showed that engaging the brain in new ways builds cognitive reserve, which is how the brain can maintain its functioning in the face of aging, damage or early stages of disease.

Brain training can be one of the many ways to build cognitive reserve, Isaacson said.

“There is no one magic pill to prevent dementia, but a combination of interventions can help people take control in the fight against Alzheimer’s,” he said. “Because of the science, I’ve suggested these exact BrainHQ tests as a component of a cognitive engagement plan, along with learning a new language, playing a new instrument or taking up a new hobby, like dancing or photography.”
The games they played

The study, published Tuesday in the peer-reviewed gaming journal JMIR Serious Games, randomized 92 relatively healthy older adults from Quebec into two groups. Each group was asked to do 30 minutes of daily brain training for 10 weeks.

The control group played the card game Solitaire and Bricks Breaking Hex — which requires the user to break bricks in groups of the same color — at their own pace. The intervention group played BrainHQ’s Double Decision and Freeze Frame modules, which became progressively harder as players improved.

Freeze Frame displays a target image and then a series of other images, asking the user to click “no” for every wrong image. Double Decision briefly shows the user one of two cars in a desert, along with a Route 66 sign that can appear anywhere on the screen. To do the training correctly, the player must quickly click on the correct car and the location of the sign.

An earlier version of Double Decision was used in the 2001 ACTIVE trial (Advanced Cognitive Training for Independent and Vital Elderly), which found that cognitive gains from the training were still present at a five-year follow-up.

Cognitive and other testing was done before and after the training and at a three-month follow-up. Acetylcholine was measured by PET scans.

According to the scans, the intervention group had a 2.3 per cent increase in the upregulation of acetylcholine after the 10-week high speed training. The improvement was in key areas of the brain responsible for memory and decision-making, said functional imaging specialist Dr. Raj Attariwala, founder and medical director of AIM Medical Imaging in Vancouver, who was not involved in the study.

This improvement nearly countered the average 2.5 per cent decline in acetylcholine that occurs naturally in each decade of life. The control group, however, had no significant benefit.

Although the study advances science in the field, it’s too soon to draw conclusions because the “work is early stage (and) effect sizes are small,” said brain game researcher Aaron Seitz, a joint professor of psychology, game design, and physical therapy, movement and rehabilitation services at the Northeastern University Bouvé College of Health Sciences in Boston. He was not involved with the study.

“It will be important for others to replicate these findings before we can reliably conclude that acetylcholine production is upregulated by these types of computerized exercises,” Seitz said.

By Sandee LaMotte, CNN

#Tech billionaire Thiel says Greta Thunberg servant of ‘#Antichrist’, The eight hours of lectures fused religious beliefs with warnings against technology regulation, according to recordings reviewed by The Washington Post.

In four roughly two-hour talks delivered over the past month at the Commonwealth Club in San Francisco, Thiel argued that those proposing limits on technology development threaten to bring about the destruction of the United States and an era of global totalitarian rule, the Post reported Friday.

Thiel, part of the so-called PayPal mafia that also includes Elon Musk, has a net worth that stands at around USA$27 billion and has close ties to the Trump administration, including Vice President JD Vance, a former associate.

He was also the only major figure from Silicon Valley that supported Donald Trump in his 2016 campaign to be president.

In Thiel’s view, the biblical Antichrist is not an evil tech genius but someone who warns of existential risk nonstop and calls for strong regulation in innovative sectors.

“In the 21st century, the Antichrist is a Luddite who wants to stop all science. It’s someone like Greta or Eliezer,” Thiel said in his September 15 opening lecture, referring to Thunberg and AI critic Eliezer Yudkowsky, according to the recordings.

In his talk, the billionaire criticized financial regulations as signs that a world government has begun to emerge that could be taken over by an Antichrist figure.

“It’s become quite difficult to hide one’s money,” Thiel said, describing an “incredible machinery of tax treaties, financial surveillance, and sanctions architecture” that gives wealthy people only an “illusion of power and autonomy.”

The lectures come amid rising Christian nationalism in the United States and as Silicon Valley leaders escalate their fight against AI regulation under U.S. President Trump’s second term.

Tech leader John Ruffolo hopeful #Canada will take digital sovereignty seriously

OpenAI’s Sora joins Meta in pushing AI-generated videos. Some are worried about a flood of ‘AI slop’.

If the future of the internet looks like a constant stream of amusing videos generated by artificial intelligence, then OpenAI just placed its stake in an emerging market.

The company behind ChatGPT released its new Sora social media app on Tuesday, an attempt to draw the attention of eyeballs currently staring at short-form videos on TikTok, YouTube or Meta-owned Instagram and Facebook.

The new iPhone app taps into the appeal of being able to make a video of yourself doing just about anything that can be imagined, in styles ranging from anime to highly realistic.

But a scrolling flood of such videos taking over social media has some worried about “AI slop” that crowds out more authentic human creativity and degrades the information ecosystem.

“These things are so compelling,” said Jose Marichal, a professor of political science at California Lutheran University who studies how AI is restructuring society. “I think what sucks you in is that they’re kind of implausible, but they’re realistic looking.”

The Sora app’s official launch video features an AI-generated version of OpenAI CEO Sam Altman speaking from a psychedelic forest, and later, the moon and a stadium crowded with cheering fans watching rubber duck races. He introduces the new tool before handing it off to colleagues placed in other outlandish scenarios. The app is available only on Apple devices for now, starting in the U.S. and Canada.

Meta launched its own feed of AI short-form videos within its Meta AI app last week. In an Instagram post announcing the new Vibes product, Meta CEO Mark Zuckerberg posted a carousel of AI videos, including a cartoon version of himself, an army of fuzzy, beady-eyed beings jumping around and a kitten kneading a ball of dough. Both Sora and Vibes are designed to be highly personalized, recommending new videos based on what people have already engaged with.

Marichal’s own social media feeds on TikTok and other sites are already full of such videos, from a “housecat riding a wild animal from the perspective of a doorbell camera” to fake natural disaster reports that are engaging but easily debunked. He said you can’t blame people for being hard-wired to “want to know if something extraordinary is happening in the world.”

What’s dangerous, he said, is when they dominate what we see online.

“We need an information environment that is mostly true or that we can trust because we need to use it to make rational decisions about how to collectively govern,” he said.

If not, “we either become super, super skeptical of everything or we become super certain,” Marichal said. “We’re either the manipulated or the manipulators. And that leads us toward things that are something other than liberal democracy, other than representative democracy.”

OpenAI made some efforts to address those concerns in its announcement on Tuesday.

“Concerns about doomscrolling, addiction, isolation, and (reinforcement learning)-sloptimized feeds are top of mind,” it said in a blog post. It said it would “periodically poll users on their wellbeing” and give them options to adjust their feed, with a built-in bias to recommend posts from friends rather than strangers.

Matt O’brien, The Associated Press

AP Technology Writer Barbara Ortutay contributed to this report.