Just in time for the Super Bowl, PepsiCo is cutting the price of Doritos, Cheetos, Lay’s, Tostitos, and other snacks by up to 15%.
The move comes after consumers complained the chips were too pricey. “Our customers . . . have been honest with us about how rising everyday costs are making their daily decisions harder. Message received,” PepsiCo said in a statement.
Lowering the suggested retail price reflects our commitment to help reduce the pressure where we can,” PepsiCo Foods U.S. CEO Rachel Ferdinando added.
The new discounted prices roll out this week, ahead of this Sunday’s big game, one of the biggest days for snack purchases. PepsiCo said supermarkets and other retailers ultimately set the prices for the chips, so the savings that shoppers see will depend on the store.
And no, you’re not imagining it: Grocery prices have increased. In one year alone they jumped 2.7%, from August 2024 to August 2025; and they’re up a reported 29% from 2020, according to the Federal Reserve Bank of St. Louis. That’s due to several factors, including inflation, weather events, and ongoing global supply chain issues, coupled with higher labor costs from the COVID-19 pandemic.
The food and beverage giant, like many of its competitors, has raised prices, hiking snacks by 1% and beverages by 7% in North America, which, along with the impact of GLP-1 weight-loss drugs, has only decreased consumer demand.
“This pricing change is part of PepsiCos broader strategy to increase accessibility and offer more choices for consumers,” Ferdinando said. “Were continuing to refine our portfoliofrom thoughtful recipe enhancements, like the removal of artificial flavors and colors from Lays and Tostitos, to packaging updates aligned with evolving consumer preferences.”
PepsiCo had previously agreed to lower prices and revamp its business, after activist investor Elliott Management demanded the changes after disclosing a $4 billion stake in the company this past September, CNN reported.
PepsiCo financials
PepsiCo’s fourth-quarter earnings, released on Tuesday, beat analyst estimates, with quarterly revenue coming in at $29.34 billion versus an expected $28.97 billion, and earnings per share (EPS) of $2.26 adjusted versus an expected $2.24.
Shares of PepsiCo (Nasdaq: PEP) were up over 3% midday on Wednesday, at the time of this writing, after closing nearly 5% higher the previous day.
Rent can eat up an entire paycheck at the start of the month, so a growing number of renters are turning to a financial product that promises relief by letting them split the bill for a price.
So-called rent now, pay later services have emerged over the past few years as housing costs climb and paychecks grow less predictable, particularly for lower-income and gig-economy workers. According to the Bureau of Labor Statistics, rents have jumped nearly 28% in past five years.
Companies such as Flex, Livble, and, more recently, Affirm, say breaking rent into multiple payments can help renters manage cash flow. But consumer advocates warn the products typically function like short-term loans, layering fees onto already strained budgets and, in some cases, carrying triple-digit effective interest rates raising questions about whether they ease financial pressure or deepen it.
Kellen Johnson, 44, started using Flex to split up his rent payments about two years ago. Instead of paying the whole $1,850 of his rent on the first of the month, Johnson would pay $1,350 on that date, and $500 on the 15th. For the service, Flex collected a $14.99 monthly subscription fee, as well as 1% of the total rent, which for Johnson was $18.50, bringing his monthly charges for the app to more than $33.
Johnson said he was willing to pay the extra costs in part because he worked as an independently contracted delivery person for Amazon at the time, and his paychecks could vary.
It was an expense that I was incurring, but I went ahead as it was more convenient, said Johnson, who now works as a driver for senior citizens in Sacramento, California.
Roughly 109 million Americans, or about 42.5 million households, are renters in the United States. The Census Bureau estimated in 2024 that a large share of those households pay 30% or more of their monthly income on rent. The bureau considers such households to be cost burdened, meaning rent consumes so much of their income that they have less ability to plan for future expenses or build wealth.
Rent now, pay later services generally operate the same way: The company pays the landlord the full rent when due, and the renter repays the company in two or more installments over the course of the month. Because rent can be such a large expense, the companies argue that spreading payments out can give renters more cash on hand.
Many of these services come with fees. The fees can be structured differently but should be generally thought of as cost of credit, consumer advocates warn. In Johnsons case, he was paying $33.49 for a two-week loan of $500, for an effective annual percentage rate of 172%, when expressed using standard consumer-lending calculations.
Renters should be skeptical of any financing providers that have partnered with a landlord, and be skeptical of anything that sells itself as no fees or no interest, said Mike Pierce, executive director of Protect Borrowers. Pierce previously worked at the Consumer Financial Protection Bureau and co-authored a report that was released this week on the industry.
Launched in 2019, Flex is one of the largest companies focused on splitting rent payments. The company says its 1.5 million customers now send about $2 billion a month in rent through its system, and several of the countrys largest landlords accept Flex as a payment option.
Flex says most of its customers are lower-income renters with weaker credit profiles. The company reports a median credit score of 604 among its users and says about one in three customers works more than one job to make ends meet. A Flex spokesman says the average customer uses the service three to four times a year. Johnson used it every month.
Livble does not charge a subscription, but charges renters a fee ranging from $30 to $40, according to the companys help page. Depending on how long the renter defers part of the payment, Livbles fees can translate into effective annual percentage rates of roughly 104% to 139%.
The buy now, pay later company Affirm said this month that it is piloting a program allowing some customers to split rent into two payments. The program is being tested in partnership with Esusu, a company that reports rent payments to credit bureaus to help consumers build credit. An Affirm spokesman said the company is not charging renters interest or fees to use the product, but may charge landlords fees.
As another financing option, landlords are increasingly accepting credit cards for rent payments. Bilt, a credit card startup, built its brand around targeting renters when it launched, and some tenants also use credit cards to accumulate rewards or points.
But paying rent by credit card can also be costly. Landlords typically pass the processing fees on to tenants. Depending on the card issuer and payment network, these fees can range from about 2.5% to 3.5% of the rent. For a renter paying $1,500 a month, that translates to roughly $37.50 to $52.50 in fees a monthly cost comparable to what services like Livble and Flex charge.
Economists and renters advocates argue that none of these financing options address the fundamental issue of affordability in the rental market. If credit cards, or flexible rent payment options become more widely used, they worry rents could rise further as landlords start factoring in a potential renters weekly cash flow as opposed to the rental market in the area the building is located in.
Merchants already pass along credit card processing costs to customers in the form of higher prices, and advocates worry that the rental market could adopt similar patterns. For example, Livble is owned by RealPage, which last year settled allegations that its algorithm allowed landlords to collude and push rents higher.
By Ken Sweet and Cora Lewis, The Associated Press
Economics Writer Christopher Rugaber contributed.
Amazon is rolling out a new feature in hopes of retainingor perhaps attractingnew Prime members.
The tech giant announced Wednesday that Alexa+, its AI-powered assistant, is now available for free to all Prime members. Last March, Amazon began offering an early access preview for the new voice assistant that saw an inspiring response, with tens of millions of customers requesting access, according to a statement.
The company has revamped its legacy Alexa product to handle more complex interactionsoffering examples of how users can engage in deep conversations with Alexa+ that may be ongoing over the course of potentially several days, as the technology can remember context, said Daniel Rausch, Amazons vice president of Alexa and Echo, in the statement. Homeowners are also pairing Alexa+ with their Ring cameras to identify unusual patterns around their homes, he said.
People are engaging in two to three times more conversations with Alexa+ than they were previously, Rausch told CNBC. Every week in a customers journey, engagement goes up, and that is really the sign of a hit product, basically.
USERS ARENT SOLD ON ALEXA+
Despite Rauschs cheerful assessment of the product, reviews elsewhere are more tempered.
One Redditor posted Alexa+ is a mess on an Alexa subreddit four months ago, lamenting several of the technologys shortcomings, including that it feels like a huge downgrade from regular Alexa. The post garnered more than 240 responses, with other Redditors sharing similar frustrations. Still other Redditors shared near-daily complaints about lag times or the quality of responses on that subreddit.
One task that Alexa+ may not be programmed to assist with is one thats been on the mind of many Prime members lately. In the U.S., searches for how to cancel Amazon Prime have surged 110% in the past month, reaching the highest level since December 2017, according to Google Trends.
AMAZON PRIME FREEBIES
Since launching Amazon Prime in 2005, the company has steadily added new features for membersincluding Alexa, which debuted with the first Echo device in 2014. While still available on such devices, Alexa is now available as an app and at Alexa.com.
Full access to Alexa+ is available for free only to Prime members, though the company will allow nonmembers to test the new Alexa+ chat functionality at Alexa.com for free. Unlimited access to Alexa+ for people who dont want to pay the $139 annual Prime membership will cost $19.99 per month.
AMAZON SELL-OFF
The Alexa+ news alone wasnt enough to lift Amazon stock, as fears of an AI bubble have fueled a broader sell-off for tech stocks in recent days. Shares of Amazon have tumbled nearly 4% during the last five trading sessions.
Of course, there are other factors at playthe company announced last week that it is slashing 16,000 jobs and will announce earnings on Thursday after the market closes.
For a show that lasts roughly 13 minutes, the Super Bowl halftime performance has fueled decades of conversation.
Sometimes the spark comes from a single moment as it did when Janet Jackson and Justin Timberlakes infamous wardrobe malfunction triggered a broadcast reckoning. Other times, it arrives through imagery and intent, from Jennifer Lopezs 2020 caged children staging that critiqued U.S. immigration policies to children at the U.S.-Mexico border to Kendrick Lamars carefully layered Black storytelling, delivered as Donald Trump watched from his seat inside the Caesars Superdome in New Orleans.
The halftime show magnifies everything fashion choices, choreography, symbolism and invites interpretation on a scale few artists ever experience.
That history forms the backdrop as Bad Bunny prepares to take the halftime stage, a moment that places Latin identity at the center of Americas most-watched television event. The conversation building around his performance extends beyond music, touching on language, culture and how much room one of the worlds biggest stars will have for symbolism and social commentary including past critiques of Trump within a show long shaped by tight NFL oversight.
With that context, here is a look at some of the most talked about halftime moments.
Timberlake and Jackson’s wardrobe malfunction
The most enduring halftime controversy unfolded during the 2004 Super Bowl in Houston, when Jackson performed alongside Timberlake.
In the closing seconds of Rock Your Body, Timberlake tugged at Jacksons costume, briefly revealing her right breast, adorned with a decorative shield. Timberlake later described the moment as an unintended wardrobe malfunction, a phrase that quickly entered pop-culture shorthand.
The reaction was immediate and far-reaching. The incident prompted FCC scrutiny, congressional attention and a reevaluation of live television standards. CBS, which aired the game, was fined $550,000 by the Federal Communications Commission, a penalty later overturned, and broadcasters expanded the use of delays for live events.
The professional fallout, however, was uneven. Jackson was disinvited from the Grammy Awards telecast the following week and largely retreated from the public spotlight, while Timberlakes career continued uninterrupted. Years later, Timberlake said the two had reconciled, but the disparity in their treatment wasn’t forgotten.
When the NFL announced Timberlakes return to the halftime stage in 2018, the decision reignited debate. Critics pointed to what they viewed as a racial and gender double standard, arguing that Jackson, a Black woman, bore the brunt of the consequences while Timberlake, a white man, emerged largely unscathed.
Online, hashtags such as #JusticeForJanet resurfaced, reframing the moment through a broader cultural lens.
Formation: Beyoncé and political symbolism
When Beyoncé performed Formation in 2016, the halftime show became a moment of cultural declaration.
Set in the Bay Area, the performance leaned heavily into Black history and identity. Dancers appeared in Black Pantherinspired attire, raised clenched fists and formed symbolic shapes on the field as Beyoncé delivered lyrics celebrating Black features and pride. The imagery echoed decades of Black activism, from civil rights-era protest to modern calls for social justice.
The performance drew widespread acclaim for its clarity and artistry while also sparking criticism from conservative commentators and some law enforcement groups who accused it of promoting anti-police sentiment. With an audience of more than 110 million viewers, the debate quickly moved beyond the stadium.
Several moments stood out. Dancers briefly formed an X, interpreted by some as a reference to Malcolm X, while raised fists recalled the 1968 Olympic protest by Tommie Smith and John Carlos. The visuals aligned directly with the message of Formation, which centers Black identity and self-definition.
Believe in Love: Coldplay and evolving visibility
Coldplay was the halftime show headliner in 2016, bringing out Beyoncé and Bruno Mars for a performance built around themes of unity, inclusion and joy.
The set featured a rainbow-colored stage, vibrant dancers and a crowd mosaic that spelled out Believe in Love. A pride flag was visible near frontman Chris Martin, and the shows closing imagery emphasized togetherness.
While many viewers praised the message as affirming and timely, some conservative groups criticized the presentation for embracing LGBTQ+ symbolism. The reaction reflected broader cultural conversations around representation and visibility in mainstream entertainment.
When Kendrick Lamar’s symbolism becomes the story
Last year, Kendrick Lamar treated the halftime stage as a controlled narrative space, using choreography, costuming and staging to explore themes of identity, power and perception. His performance unfolded with cinematic precision, opening as dancers spilled out of a Buick GNX in choreographed fashion before Lamar took command of the field.
Dancers dressed in red, white and blue framed the performance, while Samuel L. Jackson appeared as an Uncle Sam figure who interrupted the action with pointed commentary, labeling the spectacle too loud and urging Lamar to play the game. The exchange underscored the tension between expression and expectation that has come to define modern halftime shows.
While the performance stayed within league parameters, it still sparked debate among commentators who scrutinized the imagery and tone. The response reinforced how the halftime show, even without overt rule-breaking, can function as visual storytelling that invites interpretation at a scale unmatched in other live events.
Other defiant moments on a global stage
The NFL has long maintained guardrails around the halftime show, particularly when performances edge toward political commentary.
Still, some artists have chosen to test and at times ignore those limits.
Ahead of Jennifer Lopez’s performance with Shakira in 2020, the NFL raied concerns about a segment widely interpreted as referencing children held in immigration detention facilities. The league asked Lopez to cut a segment featuring children in cages, a critique of U.S. immigration policies. But she refused and moved forward with the imagery, using visual symbolism rather than explicit messaging.
The 2022 show featured a hip-hop celebration with a robust lineup of Dr. Dre, Snoop Dogg, Mary J. Blige, 50 Cent, Eminem and Kendrick Lamar. But as Eminem’s performance of Lose Yourself concluded, the rapper dramatically took a knee, lowering his head in a gesture widely interpreted as a tribute to Colin Kaepernick, whose decision to kneel during the national anthem in 2016 to protest police brutality sparked a nationwide cultural reckoning. Kaepernicks protest was later echoed by other players, and he would soon find himself out of the league.
Reports ahead of the game suggested the NFL had discouraged the gesture, though the league disputed that account. We watched all elements of the show during multiple rehearsals this week and were aware that Eminem was going to do that, NFL spokesman Brian McCarthy said at the time.
During Madonna’s halftime set in 2012, M.I.A. flashed a middle finger toward the camera, a split-second gesture that immediately drew regulatory attention and replay-heavy coverage. The action prompted an immediate fine and a legal dispute with the NFL. The sides later reached a settlement, ending a multimillion-dollar case over the incident.
By Jonathan Landrum Jr., AP entertainment writer
The Trump administration is reducing the number of immigration officers in Minnesota but will continue its enforcement operation that has sparked weeks of tensions and deadly confrontations, border czar Tom Homan said Wednesday.
About 700 federal officersroughly a quarter of the total deployed to Minnesotawill be withdrawn immediately after state and local officials agreed over the past week to cooperate by turning over arrested immigrants, Homan said.
But he did not provide a timeline for when the administration might end the operation that has become a flashpoint in the debate over President Donald Trumps mass deportation efforts since the fatal shootings of U.S. citizens Renee Good and Alex Pretti in Minneapolis.
Immigration and Customs Enforcement officials have said the surge in Minnesota that ramped up dramatically in early January is its largest immigration operation ever. Masked, heavily armed officers have been met by resistance from residents who are upset with their aggressive tactics.
A widespread pullout, Homan said, will only occur after protesters stop interfering with federal agents carrying out arrests and setting up roadblocks to impede the operations. About 2,000 officers will remain in the state after this week’s drawdown, he said.
Given this increase in unprecedented collaboration, and as a result of the need for less public safety officers to do this work and a safer environment, I am announcing, effective immediately, well draw down 700 people effective today 700 law enforcement personnel, Homan said during a news conference.
He didn’t say which jurisdictions have been cooperating with the Department of Homeland Security.
Trump administration pushed for cooperation in Minnesota
Trump’s border czar took over the Minnesota operation in late January after the second fatal shooting by federal officers and amid growing political backlash and questions about how the operation was being run.
Homan said right away that federal officials could reduce the number of agents in Minnesota, but only if more state and local officials cooperate. He pushed for jails to alert ICE to inmates who could be deported, saying transferring such inmates to ICE is safer because it means fewer officers have to be out looking for people in the country illegally.
The Trump administration has long complained that places known as sanctuary jurisdictionsa term generally applied to local governments that limit law enforcement cooperation with DHShinder the arrest of criminal immigrants.
Minnesota officials say its state prisons and nearly all of the county sheriffs already cooperate with immigration authorities.
But the county jails that serve Minneapolis and St. Paul and take in the most inmates had not previously met ICEs idea of full cooperation, although they both hand over inmates to federal authorities if an arrest warrant has been signed by a judge. It wasnt immediately clear after Homans remarks whether those jails have since changed their policies.
Border czar calls Minnesota operation a success
Homan said he thinks the ICE operation in Minnesota has been a success, checking off a list of people wanted for violent crimes who were taken off the streets.
I think its very effective as far as public safety goes, he said Wednesday. Was it a perfect operation? No.
He also made clear that pulling a chunk of federal officers out of Minnesota isn’t a sign that the administration is backing down. We are not surrendering the presidents mission on a mass deportation operation, Homan said.
Youre not going to stop ICE. Youre not going to stop Border Patrol, Homan said about the ongoing protests. The only thing youre doing is irritating your community
Steve Karnowski, Associated Press
Associated Press reporters Corey Williams and John Seewer contributed to this report.
Jeffrey Epstein’s network of money and influence often intersected with scientific and academic communities. The disgraced financier spent years cultivating relationships with researchers at elite universities, frequently dangling the promise of funding. Some of the work he supported has had, and may still have, direct and indirect impacts on Silicon Valleys most powerful technologies.
Epstein was first convicted in 2008 on charges of soliciting a minor for prostitution, yet he continued to maintain a web of relationships across the worlds of technology and academia until he was indicted on federal sex-trafficking charges in 2019. The Department of Justices latest release of the Epstein files includes emails that reveal new names and details about those connections that had not previously been made public.
Joscha Bach
One striking example is Epsteins patronage of German AI scientist and executive Joscha Bach. Known in academic and AI circles for his work on cognitive architectures, computational models that aim to replicate aspects of human cognition, Bach received extensive financial support from Epstein while completing postdoctoral work at MIT.
According to emails reviewed by SFGate, Epstein covered Bachs rent, flights, medical bills, and even private school tuition for his children in Menlo Park between 2013 and 2019.
Bach is now the executive director of the California Institute for Machine Consciousness, a small, independent research organization focused on whether machines could ever become conscious. According to SFGate, Epstein met Bach through other AI and psychology researchers and began financing his work at the MIT Media Lab and the Harvard Program for Evolutionary Dynamics in 2013.
The files show no indication of sexual impropriety on Bachs part, and he has never been accused of such conduct. Bach told SFGate that MIT approved the funding and said many prominent scientists maintained relationships with Epstein. He added, The prevailing view was that Epstein, having served his sentence, was complying with the law.
Antonio Damasio
Epstein also corresponded with Antonio Damasio, the director of USCs Brain and Creativity Institute. In 2013, Damasio asked Epstein to fund a new line of robotics and neuroscience research. Damasio, the Dornsife Chair in Neuroscience, and another USC researcher hoped to study the origins of emotion in the brain, and sought a nontraditional funding source so they could retain greater control over the direction of the work.
Damasio presented the proposal to Epstein in February 2013 at Epsteins New York City home, but Epstein ultimately declined to fund the research. Damasio told Annenberg Media that he did not know Epstein was a convicted sex offender at the time, and said he would never have contacted him had he known. I was looking for a prestigious philanthropist, not a criminal, Damasio said.
Damasios primary field is neurobiology, though he also teaches psychology and philosophy, with a focus on the neural systems that underlie emotion, decision-making, memory, language, and consciousness. He is best known for an influential theory arguing that emotions and their biological foundations, not just reason, play a central role in decision-making, even when the decision-maker is not consciously aware of it. He also theorized that emotions provide the scaffolding for social cognition, shaping how people process, store, and apply information about others and social situations.
Damasio argues that current AI models that power robots lack a sense of biological “vulnerability” that drives survival instincts and intelligence in living organisms. He theorises that training a robot to be “concerned” about its own preservation might help the robot solve problems more creatively.
David Gelernter
The DOJ document release also revealed that Epstein corresponded between 2009 and 2015 with Yale computer science professor David Gelernter, an early pioneer of concepts now associated with digital twins and metaverse-style overlays, which he calls computed worlds. Gelernter is the author of the book Mirror Worlds, which outlines much of that research. In 2001, Gelernter helped found a company called Mirror Worlds Software based on those ideas, but the venture failed to gain traction and shut down in 2004.
In his correspondence with Epstein, Gelernter sought business advice rather than research funding, according to the New Haven Register. The files also revealed no evidence of wrongdoing by Gelernter. He has said he did not know Epstein was a convicted sex offender and was never aware of Epsteins sex-trafficking operation.
In 1993, Gelernter was severely injured by a mail bomb sent by the Unabomber, Ted Kaczynski, which destroyed four of his fingers and permanently damaged one of his eyes. He is also known for controversial views, including claims that liberal academia has a destructive influence on American society, that women, especially mothers, should not work outside the home, and for rejecting the scientific consensus that humans are driving climate change.
Marvin Minsky
The most direct link between Epstein and the AI world ran through MIT professor and pioneer Marvin Minsky, who died in 2016. Minsky helped establish artificial intelligence as a formal research discipline in the 1950s and later co-founded the field at MIT with John McCarthy, training generations of AI scientists.
Epstein donated $100,000 to MIT to support Minskys research in 2002, before Epsteins first criminal conviction. That gift was the first in a series of donations to MITs Media Lab that ultimately totaled $850,000 between 2002 and 2017. Minsky died in 2016.
In 2019, court documents from a deposition by victim Virginia Giuffre were unsealed, revealing her allegation that Ghislaine Maxwell directed her to have sex with Minsky during a visit to Epsteins compound. Minskys wife said the allegation was impossible because she was with him the entire time they were on the island. Minsky never faced charges, but the revelations placed his name at the center of a reckoning at MITs Media Lab over the influence of Epsteins money on the labs work.
A gray zone
In many ways, Jeffrey Epstein operated in a gray zone created by shifting funding models for AI research. Long before the current AI boom, private industry had already overtaken the federal government as the primary backer of foundational AI work. In recent years, government funding has become increasingly tied to defense and intelligence priorities, leaving reearchers in less immediately applicable fields with few viable grant options. At the same time, AI research has grown extraordinarily expensive, requiring elite talent and vast computing resources.
As a result, universities and academic labs have become far more dependent on private philanthropy to sustain their work. Funding from wealthy donors often comes with fewer restrictions. It can arrive faster, offer greater flexibility, and require less public disclosure than government grants. This likely explains part of Epsteins appeal to researchers. But the arrangement cuts both ways. Such donations also require little transparency from the donor, meaning beneficiaries may know very little about the source of their funding.
Epsteins case is extreme, but it highlights a broader risk: when public research funding is scarce and the costs of advanced AI are high, private money becomes more attractive, along with the ethical and reputational dangers it can carry. And the problem is not easing. Microsoft chief scientist Eric Horvitz warned that U.S. cuts to National Science Foundation research grants during the Trump administration could undermine the countrys AI leadership, the Financial Times reported, noting that more than 1,600 NSF grants worth nearly $1 billion have been scrapped since 2025.
The Washington Post informed its team on Wednesday morning that it was starting a round of mass layoffs, according to multiple media reports and a memo seen by Fast Company. Multiple sections are being shut down completely, while others are being shrunk significantly.
The papers executive editor, Matt Murray, announced the cuts to the newsroom employees, saying that all sections would be impacted by the layoffs. He said the Post would be making a strategic reset, and is also cutting staff on the business side.
The New York Times reported that approximately 30% of the Posts employees are being laid off, including more than 300 of the around 800 journalists.
News of the layoffs attracted a harsh rebuke from people in the media, including the Post‘s own former editor, who criticized the paper’s owner, Amazon founder Jeff Bezos.
This ranks among the darkest days in the history of one of the worlds greatest news organizations, Marty Baron, executive editor of the Post from 2013 to 2021, said in a statement. The Washington Posts ambitions will be sharply diminished, its talented and brave staff will be further depleted, and the public will be denied the ground-level, fact-based reporting in our communities and around the world that is needed more than ever.”
Reached for comment, a Washington Post spokesperson sent the following statement: The Washington Post is taking a number of difficult but decisive actions today for our future, in what amounts to a significant restructuring across the company. These steps are designed to strengthen our footing and sharpen our focus on delivering the distinctive journalism that sets The Post apart and, most importantly, engages our customers.
Sports and other sections said to be gutted
The sports section will reportedly be eliminated entirely, meaning that Washingtons paper of record will not provide day-to-day coverage for any of the citys professional or college sports teams.
Murray noted that some of the sports reporters will be moved to the features department to cover the culture of sports.
This comes in the wake of controversy surrounding the Posts plans for the Winter Olympics, which start this week. The Times reported on January 24 that the paper axed its plans to send a delegation to the Italy games just two weeks before the opening ceremony, but quickly reversed that decision, sending a team of four after the report came out.
The Olympics arent the only major event looming on the sports calendar, as Super Bowl LX will be played in San Francisco this weekend, NCAA March Madness is just about a month away, and the FIFA World Cup, hosted in North America this year, kicks off in June.
Meanwhile, the Post is reportedly cutting down its Metro desk, which covers Washington, D.C., and its surroundings, from over 40 journalists to well below half of that. The Post is drastically reducing its international coverage as well, although some international bureaus will stay operational
Additionally, the paper is reportedly closing the books section and ending its daily Post Reports podcast.
Weeks of speculation regarding the paper’s future
The announcement comes after weeks of speculation within the newsroom. The Washington Post Guild made a statement last week, directly attacking Bezos, whose holding company, Nash Holdings, bought the paper for $250 million in 2013 and has owned it ever since.
During Donald Trumps first term as president, the Post adopted the slogan Democracy Dies in Darkness and experienced a period of growth thanks to its aggressive coverage of the administration.
In 2023, Bezos hired Will Lewis as publisher of the Post, and these layoffs are just the latest in a line of changes made since then. Notably, the paper did not endorse a candidate in the 2024 presidential election for the first time in 36 years.
In response to the layoffs, the Washington Post Guild released another statement: These layoffs are not inevitable, its first paragraph reads. A newsroom cannot be hollowed out without consequences for its credibility, its reach and its future.
This story has been updated with the Post‘s response to our inquiry.
Low Earth orbit is already getting crowded. Around 14,500 active satellites are circling Earth, and roughly two-thirds of them are run by SpaceX. Now, in filings connected to Elon Musks plan to fold SpaceX and his AI firm xAI together ahead of an IPO, the company has asked the Federal Communications Commission (FCC) for permission to launch up to one million more.
The figure is so large it would dwarf the number of satellites currently in orbit. In fact, it is more than every object ever sent into space by every nation combined. So why is Musk planning it, and what would it mean for the rest of us?
In a public update posted on the SpaceX website as part of the merger process between SpaceX and xAI, Musk wrote that Launching a constellation of a million satellites that operate as orbital data centers is a first step towards becoming a Kardashev II-level civilization. The Kardashev scale is a measure of technological development first outlined in the 1960s by Soviet astronomer Nikolai Kardashev, who died in 2019.
While the scale of the proposal may have impressed Kardashev, many experts are far more skeptical. A million new satellites would represent roughly a 67-fold increase over todays orbital population. Proposals on the scale being discussedup to one million satellitesrepresent a step change that deserves the same level of scrutiny we would apply to any other major global infrastructure project,” says Ruskin Hartley, CEO of DarkSky International, a nonprofit focused on preserving night skies and mitigating the impacts of light pollution.
Satellite deployment at such a scale would have huge knock-on effects. The consequences extend well beyond astronomy, Hartley says. They include cumulative impacts on the night sky, increased atmospheric pollution from satellite launches and re-entries, and a sharply elevated risk of orbital congestion and collision cascades that could impair access to low Earth orbit for all nations. When satellites burn up, they release metals such as aluminum into the upper atmosphere, a process scientists and the U.K. Space Agency warn is still poorly understood but likely accelerating as megaconstellations grow.
There is also the question of safety. Space is already crowded with satellites that power communications, enable GPS navigation, and support countless services we rely on every day. Adding vastly more objects increases the chances of close approaches, which, if not monitored and avoided, can result in collisions and cascading debris. SpaceX will say they can do that stationkeeping successfully, but it doesn’t take many failures to have you end up in a bad situation, says Jonathan McDowell, an astronomer and space sustainability analyst based in London and Boston and formerly at the Center for Astrophysics. The SpaceX satellites will be in the higher part of low Earth orbit where it will take a long time for failed satellites to re-enter.
Hartley, for his part, argues that these risks demand far more scrutiny. Decisions made now will shape the near-Earth environment for generations, he says.
Not everyone believes the million-satellite figure is even realistic. As to the question of if its practical, I would think not, says Caleb Henry, director of research at Quilty Space. Filing for 1 million satellites is probably a way for SpaceX to push the envelope before accepting whatever fraction regulators deem acceptable.
That tactic may already be working. The FCC initially rejected a 2022 SpaceX proposal to launch 30,000 satellites, before later approving it in 25% tranches. The commission authorized another 7,500 satellites this January, for a total approval of 15,000 satellites from that filing, says Henry. SpaceX is also asking the FCC to waive standard deployment milestones, and says the economics of the plan depend on Starship becoming fully reusable, a goal it has not yet reached.
In that sense, the million-satellite request is not a signal of imminent growth, but a bid to stake out spectrum and orbital real estate for a future that Musk is already trying to define.
President Donald Trump’s administration is expected to unveil its grandest plan yet to rebuild supply chains of critical minerals needed for everything from jet engines to smartphones, likely through purchase agreements with partners on top of creating a $12 billion U.S. strategic reserve to help counter China’s dominance.Vice President JD Vance is set to deliver a keynote address Wednesday at a meeting that Secretary of State Marco Rubio is hosting with officials from several dozen European, Asian and African nations. The U.S. is expected to sign deals on supply chain logistics, though details have not been revealed. Rubio met Tuesday with foreign ministers from South Korea and India to discuss critical minerals mining and processing.The meeting and expected agreements will come just two days after Trump announced Project Vault, or a stockpile of critical minerals to be funded with a $10 billion loan from the U.S. Export-Import Bank and nearly $1.67 billion in private capital.Trump’s Republican administration is making such bold moves after China, which controls 70% of the world’s rare earths mining and 90% of the processing, choked off the flow of the elements in response to Trump’s tariff war. The two superpowers are in a one-year truce after Trump and Chinese President Xi Jinping met in October and agreed to pull back on high tariffs and stepped-up rare earth restrictions.But China’s limits remain tighter than they were before Trump took office.“We don’t want to ever go through what we went through a year ago,” Trump said on Monday when announcing Project Vault.
Countering China’s dominance on critical minerals
Other countries might join with the Trump administration in buying up critical minerals and taking other steps to spur industry development because the trade war revealed how vulnerable Western countries are to China, said Pini Althaus, who founded Oklahoma rare earth miner USA Rare Earth in 2019.“They’re looking at setting up sort of a buyers’ club, if you will,” said Althaus, who now is working to develop new mines in Kazakhstan and Uzbekistan as CEO of Cove Capital. “The key producers and key consumers of critical minerals will sort of get together and work on pricing structures, floor pricing and other things.”The government last week also made its fourth direct investment in an American critical minerals producer when it extended $1.6 billion to USA Rare Earth in exchange for stock and a repayment agreement.Seeking government funding these days is like meeting with private equity investors because officials are scrutinizing companies to ensure anyone they invest in can deliver, Althaus said. And the government is demanding terms designed to generate a return for taxpayers as loans are repaid and stock prices increase, he said.
The stockpile strategy
Meanwhile, the U.S. Export-Import Bank’s board this week approved the $10 billion loan the largest in its history to help finance the setup of the U.S. Strategic Critical Minerals Reserve. It is tasked with ensuring access to critical minerals and related products for manufacturers, including battery maker Clarios, energy equipment manufacturer GE Vernova, digital storage company Western Digital and aerospace giant Boeing, according to the policy bank.Bank President and Chairman John Jovanovic told CNBC that the project creates a public-private partnership formula that “is uniquely suited and puts America’s best foot forward.”“What it does is it creates a scenario where there are no free riders. Everybody pitches in to solve this huge problem,” he said.Manufacturers, which benefit the most from the reserve, are making a long-term financial commitment, Jovanovic said, while the government loan spurs private investments.The stockpile strategy may help spark a “more organic” pricing model that excludes China, which has used its dominance to flood the market with lower-priced products to squeeze out competitors, said Wade Senti, president of the U.S. permanent magnet company AML.The Trump administration also has injected public money directly into the sector. The Pentagon has shelled out nearly $5 billion over the past year to help ensure its access to the materials after the trade war laid bare just how beholden the U.S. is to China.
Efforts get some bipartisan support
A bipartisan group of lawmakers last month proposed creating a new agency with $2.5 billion to spur production of rare earths and the other critical minerals. The lawmakers applauded the steps by the Trump administration.“It’s a clear sign that there is bipartisan support for securing a robust domestic supply of critical minerals that both reduces our reliance on China and stabilizes the market,” Sens. Jeanne Shaheen, D-N.H., and Todd Young, R-Ind., said in a joint statement Tuesday.Building up a stockpile will help American companies weather future rare earth supply disruptions, but that will likely be a long-term effort because the materials are still scarce right now with China’s restrictions, said David Abraham, a rare earths expert who has followed the industry for decades and wrote the book “The Elements of Power.”The Trump administration has focused on reinvigorating critical minerals production, but Abraham said it’s also important to encourage development of manufacturing that will use them. He noted that Trump’s decisions to cut incentives for electric vehicles and wind turbines have undercut demand for these elements in America.
Didi Tang, Josh Funk and Matthew Lee, Associated Press
The footage was real, verified, and delightful: a security camera clip of a coyote bouncing on a backyard trampoline in Los Angeles. Days after the video went viral, near-identical kangaroos, bears, and rabbits began circulating too, all generated by AI. Millions shared them, believing theyd captured another glimpse of animals behaving hilariously.
It was an amusing mix-up, but it was also a warning.
AI-generated video tools have moved far beyond producing surreal or obviously manipulated clips. They are now convincingly imitating the formats we instinctively trust most: CCTV, dashcams, police bodycams, wildlife cameras, and handheld eyewitness footage. These are the clips that shape public understanding during protests, disasters, violence, and emergencies. And the fake ones are becoming indistinguishable from the real thing.
AI-generated realism has already entered the news cycle
At Storyful, we verify thousands of real-world videos for newsrooms and brands worldwide. This year we ran a test: we fed real breaking-news headlines from our own platform into one of the newest AI video models.
In seconds, we got clips that mimicked the texture and perspective of eyewitness reporting. Not glossy AI experiments, news-like footage that could plausibly land in a newsroom inbox during a breaking story.
Side by side with the original real clips, even trained journalists needed to slow down and scrutinize the details.
Consider this example, inspired by a verified authentic video posted to social media in the wake of heavy monsoon rains in India:
Firefighters Save Man Clinging to Pole Amid Raging Indian Floods
A man was rescued on Tuesday, September 16, in Indias Uttarakhand state after spending more than four hours clinging onto an electricity pole as deadly floodwaters raged around him, local media reported.
Real:
And this fully synthetic video, created by prompting OpenAIs video generator app Sora with the title of the first video.
Fake:
This is no longer a theoretical future. It is happening right now.
Guardrails are already slipping. Tutorials circulate openly on Reddit explaining how to remove the watermark on videos created by one of the most popular AI-video generators, OpenAIs Sora. Restrictions on certain AI prompts can be bypassedwhen they existor models can be run locally without curbs on highly realistic content. And because these tools can create fake CCTV or disaster footage on demand, the question isnt whether AI can generate convincing videos of things that never happened. Its how far will a convincing fake spread before anyone checks it?
Why AI-generated videos feel believable
The most significant shift in AI-generated video is not just its appearance, but also its behavior.
Real eyewitness footage contains the rough edges that come with real life: a shaky hand, the camera pointed at the ground before the action begins, long stretches of nothing happening, imperfect angles and missed details.
AI does not yet replicate these moments. It goes straight to the action, framed center-perfect, lit cleanly, and paced like a scene built for maximum impact. It offers the moment we expect to see, without the messy human lead-up that usually surrounds it.
The reason is simple. Most models are still trained heavily on cinematic material rather than chaotic, handheld user-generated content. They understand drama better than they understand reality. That gap is what allows verification teams to spot the differencefor now.
As models evolve and prompt-writing improves, these behavioral tells will fade. The training data for these video foundation models includes both shaky bystander videos and slick documentaries, allowing them to ably imitate their style and sense of realism.
Public confidence is already eroding
The Reuters Digital News Report finds that 58% of global audiences fear they can no longer tell real from fake online. That fear used to apply mainly to politics and propaganda. Now it applies to harmless backyard videos.
This marks a deeper psychological shift. Once a viewer starts doubting everyday videos, they dont toggle that skepticism on and off. If they question a dog rescue, they will question a protest. If they question a prank, they will question a war zone.
Trust doesnt collapse in a single dramatic moment. It erodes drip by drip, through thousands of small uncertainties. And as AI-generated video becomes abundant, authentic footage becomes scarce.
How to tell when a video is AI-generated
AI detection tools can be a useful part of your workflow, but they are not a replacement for human verification. According to Storyfuls analysis, current tools achieve 6575% accuracy under ideal conditions, but that accuracy drops below 50% within weeks of a new AI model release. These are the signals Storyful’s verification teams use daily, cues the public can learn to recognize quickly.
AI starts at the climax.Real footage almost always includes dead time or fumbling before the action.
Subjects sit perfectly in the center of the frame.Eyewitnesses rarely capture the chaos of breaking news like cinematographers.
Motion is too smoothReal user-generated content stutters, shakes, refocuses, and slips.
Timestamps, signage, and license plates break down under scrutinyAI often approximates these details instead of rendering thm accurately.
Disaster and wildlife clips look too composed.Real life is uncertain. AI often looks staged.
These cues wont hold forever, but right now they offer critical protection.
Authenticity is now an asset
Tech platforms can add more guardrails to their video generator tools, regulators can update frameworks, detection tools can improve, and so can our own critical faculties. And as newsrooms help audiences navigate through the morass of fakery, the most impactful way they can rebuild trust is to be transparent.
Audiences no longer trust sources say. They want to see how a journalist or a newsroom knows something is real.
More news organizations are adopting verification-forward formats, including BBC Verify and CBS News Confirmed, which integrate open-source and forensic checks into reporting, examining provenance, imagery, metadata patterns, and geolocation when relevant. Storyful Newswire equips all of our partners with these basic but essential details about every video on our platform.
This transparency is becoming the primary differentiator in an environment where AI-generated video is cheap, fast, and everywhere. The more AI-generated footage floods the ecosystem, the more credibility belongs to organizations that make showing their work a key part of the story.
The internet’s most unforgettable videos were never perfect. They were unpredictable, flawed, and human, the kinds of moments AI still struggles to imagine. AI-generated footage can now mimic the visual language of truth. But it cannot yet reproduce the randomness of real life. What’s at stake when it does isn’t simply misinformation. It’s the public’s ability to trust what it sees in the moments that matter most.
James Law is Editor in Chief of Storyful, a news agency used by 70 percent of the top 20 global newsrooms specializing in verifying breaking news and viral video.
