A dispute between AI company Anthropic and the Pentagon over how the military can use the companys technology has now gone public. Amid tense negotiations, Anthropic has reportedly called for limits on two key applications: mass surveillance and autonomous weapons. The Defense Department, which Trump renamed the Department of War last year, wants the freedom to use the technology without those restrictions.
Caught in the middle is Palantir. The defense contractor provides the secure cloud infrastructure that allows the military to use Anthropics Claude model, but it has stayed quiet as tensions escalate. Thats even as the Pentagon, per Axios, threatens to designate Anthropic a supply chain risk, a move that could force Palantir to cut ties with one of its most important AI partners.
The threat may be a negotiating tactic. But if carried out, it would have sweeping consequences, potentially barring not just Anthropic but its customers from government work. “That would just mean that the vast majority of companies that now use [Claude] in order to make themselves more effective would all of a sudden be ineligible for working for the government, says Alex Bores, a former Palantir employee who is now running for Congress in New York’s 12th district. It would be horribly hamstringing our government’s ability to get things done. (Palantir did not respond to a request for comment.)
Alex Bores [Photo: John Nacion/Variety via Getty Images]
Anthropic and the Pentagons war of words
Anthropic has, until now, maintained close ties with the military. Claude was the first frontier AI model deployed on classified Pentagon networks. Last summer, the Defense Department awarded Anthropic a $200 million contract, and the companys technology was even used in the recent U.S. operation to capture Nicolas Maduro, the Wall Street Journal reported this week.
But the company’s commitment to certain AI safety principles has irked some people in President Donald Trump’s orbit. (Katie Miller, Stephen Millers wife, has publicly accused the company of liberal bias and criticized its commitment to democratic values.) Unlike rivals xAI and OpenAI, both of which also also have Defense Department contracts, Anthropic is now locked in a fight with the Pentagon that playing out in public.
“Anthropic is committed to using frontier AI in support of US national security. Thats why we were the first frontier AI company to put our models on classified networks and the first to provide customized models for national security customers, a company spokesperson tells Fast Company. Claude is used for a wide variety of intelligence-related use cases across the government, including the DoW, in line with our Usage Policy. We are having productive conversations, in good faith, with DoW on how to continue that work and get these complex issues right.”
The Pentagon has taken a more confrontational tone. Agency officials are reviewing their relationship with Anthropic and have suggested that other contractors may also be required to stop working with the company. The Department of Wars relationship with Anthropic is being reviewed, Chief Pentagon spokesman Sean Parnell tells Fast Company. “Our nation requires that our partners be willing to help our warfighters win in any fight.” (Parnell did not respond to a request for clarification regarding specific concerns about autonomous weapons or surveillance.)
Palantir, the middleman
Palantir occupies a critical position in this ecosystem. A longtime government software provider, it has met a bevy of requirements allowing it to offer cloud services to support classified work. And, as is typical in the dizzying world of government technology contracting, Palantir also has key partnerships with Anthropic.
Two years ago, the companies partnered to bring Anthropics technology to the government, a move that made Claude available to defense and intelligence services through Amazon Web Services. Last April, Anthropic joined Palantirs FedStart program, which expanded the availability of its technology to government customers through Google Cloud.
Government tech contracting is a wonky business, but companies that want to sell software to the government typically need to work with a certified cloud provider like Palantir, or obtain certification themselves. If youve never operated in a classified environment before, you essentially need a vehicle, explains Varoon Mathur, who worked on AI in the Biden administration. Palantir is a defense contractor with deep operational integration. Anthropic is an AI model provider trying to access that ecosystem.
Growing tensions over how the Defense Department might use Claude also raise questions about how much visibility companies like Palantir and Anthropic have into the governments use of their tools. Anthropic and OpenAI offer Zero Data Retention usage, where they dont store the asks made of their AI, notes Steven Adler, a former OpenAI employee and AI safety expert tells Fast Company. Naturally this makes it harder to enforce possible violations of their terms.
A person familiar with the matter said Anthropic does have insight into how its technology is used, regardless of whether its in a classifiedenvironment, and that the company is confident its partners and users have been deploying the tech in line with its policies. In its reporting, the Wall Street Journal cited people familiar with the matter who said an Anthropic employee did reach out to Palantir to ask about Claudes use in the Maduro operation, though Anthropic denied to that outlet that it had spoken with Palantir beyond technical discussions. The Anthropic spokesperson tells Fast Company that the company cannot comment on its technologys use in specific military operations, but said it work[s] closely with our partners to ensure compliance.”
More broadly, the standoff risks chilling relationships between Silicon Valley and Washington at a moment when the government is pushing to adopt AI more aggressively. To state basically that it’s our way or the highway, and if you try to put any restrictions, we will not just not sign a contract, but go after your business, is a massive red flag for any company to even think about wanting to engage in government contracting, says Bores.
From breathtaking jumps to mesmerizing spins, figure skating is one of the most popular sports at the Olympic Winter Games Milano Cortina 2026. In a survey, 56% of 1,000 Americans who planned on watching the winter Olympics said they would be tuning in to watch figure skating, according to market research from Reviews.com.
And all eyes are on the American trio of female skaters known as the ‘Blade Angels,’ on Tuesday with the start of the women’s short program. Amber Glenn, Alysa Liu, and Isabeau Levito are hoping to take home the gold in individual women’s figure skating, something the U.S. women’s team has not done since 2006.
Only the top 24 women skaters in the women’s short program will advance to the compete in the free skate final on Thursday.
Here’s what to know.
Who are the ‘Blade Angels’?
The “Blade Angels” as they call themselves, are three U.S. women’s single figure skaters representing Team USA in this year’s winter Olympics: Amber Glenn, Alysa Liu, and Isabeau Levito.
They have captivated the nation not only with their skating, but their friendship, and lively, non-conformist, authentic personalities. Glenn is the first openly LGBTQ+ woman to compete in womens figure skating at the Winter Olympics. Their bravery and impressive skills have also garnered attention from celebrities like Madonnawho sent Glenn a video saying “Go get that gold”and Taylor Swift, who introduced them in an Olympic video.
The three have an impressive amount of wins among them: Liu is the 2025 World Figure Skating Champion, Glenn a three-time skating champion, and Levito is the 2024 world silver medalist.
“I haven’t seen a U.S. women’s team this strong in 20 years,” Olympic gold medalist and commentator Tara Lipinski told NBC Olympics.
What’s the history of Olympic figure skating?
Figure skating was first introduced in the 1908 Summer Olympics, but didn’t become part of the winter games until 1924. From early on, it was one of the first Olympic sports with a female category, and actually the only winter Olympic sport for women until 1936.
In the years since the 1950s, the U.S. women’s team has dominated the sport, taking home the gold again and againfrom Tenley Albright, who was the first woman to win at the 1956 Cortina dAmpezzo Games, to more recently, Michelle Kwan and Kristi Yamaguchi.
Reality Check: Inside Americas Next Top Model doesnt begin in 2003, when Americas Next Top Model premiered and took television by storm. It doesnt begin in the 1990s, when eventual host Tyra Banks rose to superstardom in the modeling industry. Instead, it begins in 2020, when the pandemic led a new generation to binge early-aughts reality TV, this time watching with a modern lensand, naturally, tearing it to shreds on TikTok.
From there, Netflixs newest docuseries rewinds to tell the full story of Americas Next Top Model, from its pre-production through its 24 scandalous cycles and into its modern-day legacy, featuring interviews with contestants, producers, and judges, including Banks herself. Reality Check leans into the same trend that inspired its creation: reexamining years-old media, like a cult classic reality TV show, with a critical eye.
What viewers casually did on TikTok in 2020, Reality Check does with finesse. It dissects the surface-level controversies fans already know about, from body shaming to the show’s infamous race-swapping photoshoot (and its oft-forgotten sequel four years later).
Shandi Sullivan [Photo: Netflix]
It also brings new revelations that cast the series in an even darker light, like Cycle 2 contestant Shandi Sullivan alleging that she was sexually assaulted on camera, with production doing nothing to step in or help her. Instead, the docuseries details Top Model portraying the incident as Sullivan cheating on her boyfriend, even filming her tearful call home to break the news to him. Through it all, Banks dodges accountability, even teasing that the show could come back for a 25th cycle.
Reality Check is the latest entry in a trend of exposé documentaries around TV that many young adults watched in their childhoods, cashing in on nostalgia by peddling its opposite. In 2024, Quiet on Set: The Dark Side of Kids TV dove into the controversies surrounding Nickelodeon series, particularly those created under Dan Schneiders tenure as a producer and showrunner. (That series built on the fervor generated by Jennette McCurdys best-selling 2022 memoir Im Glad My Mom Died, which included stories from her time starring in iCarly and Sam & Cat.)
Also in 2024, VICE released Dark Side of Reality TV, a 10-part docuseries with each episode focusing on the behind-the-scenes truth of a different reality TV show, including Toddlers & Tiaras, Hells Kitchen, Survivor, andyesAmericas Next Top Model.
Even as the aesthetics of 2000s pop culture are celebrated and embraced, their actual pop culture artifacts are subject to renewed criticism. Everyone loves to binge-watch. Everyone loves to hate-watch. Documentaries like Reality Check provide the perfect crossover.
Snap is hoping to snap up another revenue stream in its quest to reduce its dependency on advertising. The social media company announced on Tuesday that it will begin offering subscriptions to select creators so they can earn income from their most engaged fans.
In a move that supports both creators and its bottom line, Snap will begin testing Creator Subscriptions next week with a group of 15 Snapchat creators that includes Jeremiah Brown, Harry Jowsey, and Skai Jackson. Combined, these three creators have more than 3 million followers on Snap, and the company is betting that some portion of those followers will convert to paid subscribers to receive exclusive content, priority replies, and an ad-free experience.
Creators will have the ability to set the monthly pricing for their subscriptions within Snap-recommended tiers that range from $4.99 to $19.99 per month, according to reporting by CNBC, and creators will receive approximately 60% of subscription revenue. Starting next Monday, Snapchatters will be able to subscribe to participating subscribers so long as they have Apple devices; the company hasnt said when the feature will be available for Android users.
This launch builds on our continued investment in a creator-first monetization ecosystemone designed to help creators strengthen relationships with their communities and build sustainable, scalable businesses on Snapchat, the company said in a statement.
MOVING BEYOND ADVERTISING
By embracing the creator-to-consumer subscription model, Snapchat is hoping to build on the success of longer-running monetization offerings like the Unified Monetization Program and the Snap Star Collab Studio, both designed for creators.
Snap is focused on revenue diversification, as CEO Evan Spiegel has repeatedly emphasized in recent quarters.
Thats seen the Santa Monica, California-based company test the waters to gauge what (and how much) Snapchatters are willing to pay for social media content. The company launched Snapchat+ in 2022 for $3.99 per month, which unclocks some exclusive features, and about four months ago, it announced it would begin charging for storage plans for Memories on the platform.
Both Snapchat+ and the Memories Storage Plans have proven successful, even if theres been some grumbling among users. Thanks to those subscription offerings, and others, Snap ended 2025 with 24 million subscribers, a 71% increase from the same period in 2024, according to the companys fourth-quarter results released earlier this month.
CAN SUBSCRIPTIONS REVIVE THE STOCK?
By looping in creators with the newest monetization features, Snapchat wants to give them the freedom to experiment while fostering connections on the platform.
We want the next step in our long-term creator monetization journey to be one thats really rooted in real relationships, Jim Shepherd, Snaps head of content partnerships, said in an interview with CNBC.
But the social media platform arrives late to whats already become a crowded market for subscriptionsand at a time when subscription creep is becoming a more vexing issue. Whats more, a recent survey found that Gen Z, a core demographic for Snap, is feeling the subscription fatigue with respect to streaming servicesa sentiment that might extend elsewhere.
And investors remain skeptical about Snaps recent efforts to diversify its revenue stream. The stock fell nearly 2% in mid-day trading on Tuesday, extending a year-to-date selloff of more than 41%.
One of the companies best known for cranking out ultracheap goods is facing a serious investigation in Europe over concerns about illegal products and predatory business practices.
The EUs European Commission said Tuesday that it has opened formal proceedings against Shein under the Digital Services Act, which sets ground rules for online services that Europeans use. In the announcement, the commission says it is targeting Shein over worries that the shopping platform is addictive by design, powered by opaque algorithms, and engages in the sale of illegal goods, including weapons and child sexual abuse material in the form of child-like sex dolls.
Late last year, French watchdog agency the Directorate General for Competition, Consumer Affairs and Fraud Control flagged the Chinese online retailer to authorities after finding sex dolls constituting child sexual abuse material for sale along with other pornographic content not restricted by an age gate. These acts fall within the scope of serious criminal offences under French law, the regulator wrote at the time, noting that the violations could be punishable by imprisonment and a 100,000 fine under the countrys criminal code.
Based on the findings, French authorities initiated a criminal investigation into Shein over the sale and distribution of child sexual abuse material and kicked off a coordinated European investigation under the Digital Services Act. The French consumer protection agency found childlike sex dolls for sale on Chinese e-commerce site AliExpress, owned by Alibaba Group.
In a parallel investigation, French customs agents inspected 200,000 Shein packages for compliance with French laws and found that eight out of 10 products it examined potentially ran afoul of the law, including cosmetics containing banned ingredients and unsafe childrens toys.
In response, Shein said that it would restrict the sale of sex dolls and permanently ban “all seller accounts linked to illegal or non-compliant sex-doll products” on its platform. The fight against child exploitation is non-negotiable for Shein, Shein Executive Chairman Donald Tang said in a statement addressing the controversy. These were marketplace listings from third-party sellersbut I take this personally.
Regulators catch up to fast fashion
Fast fashion retailers like Shein and Temu, which ship lightning fast from China and offer hundreds of thousands of designs, have exploded in recent years. Regulators are only beginning to catch up to the controversial business model, which has seen Shein sprint toward $2 billion in revenue in 2025, in spite of the companys many headwinds.
The online shopping frenzy over trendy, ultracheap clothes took off during the pandemic and got a massive boost from TikTok, where Gen Z influencers reveal and review their clothing shipments in haul videos.
Shein and other fast fashion retailers rely on a test and repeat model that throws many thousands of clothing designs at the wall to see what sticks, producing small batches of 50 to 100 items. Designs that flop are swiftly retired and if a design takes off, its production scales up to meet demand.
While the fast fashion trend keeps TikTok creators well-stocked with fresh content, the phenomenons major players have faced an array of serious concerns during their rise. The fast fashion worlds quick cycles and frequent returns create vast amounts of waste destined for the landfill, not to mention the emissions consequences of shipping so many small packages around the globe on short notice.
Beyond the steep environmental price of cheap goods, investigations have found that the laborers constantly sewing new designs for Shein and its ilk often work grueling shifts in difficult conditionsand sometimes those workers arent even old enough to legally be there.
Shein also got wrapped up in Trumps tariff wars last year, when the president ended the de minimis loophole that made it possible for Chinese companies to ship small, low-value packages into the U.S. without paying tariffs and extra duties. EU finance ministers followed suit late last year, announcing that Europe would begin to impose customs duties on low value packages shipped into Europe some time in 2026.
Right now, criminal and state-sponsored hackers are intercepting and storing encrypted data they cannot yet decode. Likely targets include everything from corporate secrets and medical records to legal agreements and military communications. Why would these actors bother to steal data they cant read? Because they are betting on developments in quantum computing that will eventually let them crack this encrypted data wide open.
This isnt a fringe theory. The NSA (National Security Agency), NIST (National Institute of Standards and Technology), and ENISA (European Agency for Cybersecurity) are all treating this harvest now, decrypt later scenario as a live threat that is serious enough to demand immediate action. The NSA has mandated that all U.S. national security systems must transition to quantum-resistant cryptography by 2035with new acquisitions required to be compliant by 2027. In Europe, ENISA issued updated guidance in April 2025 warning that the threat is sufficient to warrant caution, and to warrant mitigating actions to be taken, and recommending that organizations begin deploying post-quantum cryptography immediately. NIST has launched a parallel global effort to develop the new cryptographic standards on which these transitions will depend.
The message from all three bodies is the same: organizations are running a grave risk if they wait until quantum computers can break current encryption standards to begin upgrading. That is the reason business leaders need to pay attention to quantum computing nownot because the technology is ready, but because the risk is grave, and the cost of preparation is trivial compared to the cost of being caught flat-footed.
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/creator-faisalhoque.png","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/faisal-hoque.png","eyebrow":"","headline":"Ready to thrive at the intersection of business, technology, and humanity?","dek":"Faisal Hoques books, podcast, and his companies give leaders the frameworks and platforms to align purpose, people, process, and techturning disruption into meaningful, lasting progress.","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"https:\/\/faisalhoque.com","theme":{"bg":"#02263c","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#ffffff","buttonHoverBg":"#3b3f46","buttonText":"#000000"},"imageDesktopId":91420512,"imageMobileId":91420514,"shareable":false,"slug":""}}
Quantum Computing 101
Classical computers store and process information as bits, where each bit is either a 0 or a 1. Quantum computers, by contrast, exploit the properties of quantum mechanics, working instead with qubits, which can exist in multiple states simultaneously. Tapping into the unusual features of quantum states in this way allows quantum computers to explore vast numbers of possibilities in parallel rather than working through them one by one.
This doesnt mean that quantum computers are generally better than, or a replacement for, classical computers. Rather, quantum computers are a specialist tool for handling a specific class of problems that involve enormous combinatorial complexitythe kind of problems where the number of possible solutions explodes so fast that even the most powerful classical supercomputers cant meaningfully explore them. In areas like these, quantum computers have the potential to offer not just incremental improvements on classical computing, but to redefine what is computationally possible in whole fields.
Logistics optimization, financial modeling, drug discovery, and cryptography are all examples of fields that involve exactly the kind of combinatorial complexity that quantum computers are built to handle. Of these, it is cryptography that demands the most immediate attention.
Hype and Reality
Disentangling the reality from the hype about quantum computing is genuinely difficultand not just for casual observers. In January 2025, Nvidia CEO Jensen Huang suggested that useful quantum computers could be decades away, sending stocks in quantum-related companies into freefall. By mid-2025, he was far more bullish, describing the field as being on the cusp of an inflection point. If one of the most technically informed CEOs on the planet can shift his assessment that dramatically in six months, the rest of us should be humble about our ability to call the timing.
As is often the case with new technologies, there is real momentum on both sides. On the bullish side, Google announced in late 2024 that its Willow quantum chip solved a problem in five minutes that they claimed would have taken a classical supercomputer ten septillion years. In February 2025, Microsoft unveiled its Majorana 1 chip, claiming that they had implemented a new approach to building qubits that could scale faster than competing designs. IBM continues to publish ambitious roadmaps.
Credible researchers such as Nathalie de Leon, an experimental quantum physicist at Princeton, say that there has recently been a vibe shift in the fielda growing sense that useful quantum machines could arrive within ten years rather than thirty. I am much more certain that quantum computation will be realized, and that the timeline is much shorter than people thought, Dorit Aharonov, a computer scientist at Hebrew University in Jerusalem, told Nature. Capital markets are paying attention too.
But the bear case is also serious. Quantum computing stocks like Rigetti and D-Wave have traded at more than 500 times estimated saleswith almost no real-world revenue and few practical applications. The machines remain fragile, error-prone, and require operating temperatures near absolute zero. There is a persistent and uncomfortable pattern in the research: researchers working on quantum computing announce a speedup, and classical computing researchers almost immediately find ways to match it. Quantum computing could be the next big thingor it could be the next hot air balloon.
The point isnt to resolve this debate. The point is that smart, informed people disagree sharply about whenor whetherquantum computing will deliver on its promise. And that disagreement should sound uncomfortably familiar.
Weve Been Here Before
When ChatGPT launched in November 2022, it became the fastest-growing consumer application in history. Within two months, it had 100 million users. Enterprises scrambled to adapt. Boards demanded AI strategies overnight. It felt like a bolt from the blue.
But it wasnt. Machine learning as a discipline dates back to the 1950s. Neural networks were being explored seriously in the 1980s. Even looking at the more recent past, the technology had been visibly advancing for over a decade. In 2016, AlphaGo defeated world champion Lee Sedol 41. In 2017, transformer architecture was introduced, which became the foundation for modern large language models.
This is not a story about AI. Rather, its a long-running, frequently recurring story about institutional blindness to technological disruption. With AI, the technology was visibly coming, and still most companies were caught without a plan, a team, or any institutional understanding of what was happening. Quantum computing is following a similar pattern: once again we see real scientific progress, genuine uncertainty about timescales, sharp disagreement among experts, and a business community that is mostly not paying attention.
Those who cannot remember the past are doomed to repeat it, said the philosopher George Santayana. So let us remember the general state of unpreparedness around AI, and try to do better with quantum computing.
What Businesses Should Do Today
The whole point of learning from the AI experience is that preparation doesnt have to be expensiveit just has to start early.
1. Develop organizational literacy. You dont need to hire quantum physicists. You need a small number of peoplein strategy, technology, and risk managementwho can follow developments, read past the hype, and flag when something becomes relevant to your business. The goal is to ensure that when a headline lands about a qubit milestone or a new standard, someone in your organization can tell you whether it matters and why.
2. Identify your exposed workflows. The potential impact of quantum computing on your business extends well beyond cryptography. Which of your core operations involve the kind of complex optimization, simulation, or modeling processes that could be disruptedor where a competitor with quantum capabilities could leapfrog you? You dont need to solve for this today. You need to know where to look when the time comes.
3. Define your trigger conditions. What specific developmentsa demonstrated commercial application in your sector, a regulatory mandate, a breakthrough in error correctionwould move you from monitoring to investing? Set these thresholds now, so that when news breaks, youre executing a plan rather than reacting to a headline.
4. Get your cryptographic house in order. This is the most concrete and most urgent action. The NSA, ENISA, and NIST are all moving toward post-quantum cryptographic standards, but those standards are still evolving. That means you need two things. First, you need an understanding of where encryption actually sits in your organizationwhich of your systems depend on which cryptographic standards and where does encrypted data flow to third parties whose security posture you dont control? Second, you need architecture that lets you swap cryptographic components independently when the standards settle, without forcing a rebuild of everything around them. Engineers call this crypto-agility. Think of it as future-proofing your security not against a specific threat, but against the certainty that the regulatory and threat landscape will keep shifting.
None of this is to say that quantum computing is certain to become practically relevant to your organization in the near future or even in any future. Experts themselves disagree over this question. The point is that businesses cannot afford to wait for the experts to reach consensus. If quantum computing becomes practically useful within the next decade, the implications could be enormous. The cost of paying attention is low; the cost of being caught flat-footed could be devastating.
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/creator-faisalhoque.png","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/faisal-hoque.png","eyebrow":"","headline":"Ready to thrive at the intersection of business, technology, and humanity?","dek":"Faisal Hoques books, podcast, and his companies give leaders the frameworks and platforms to align purpose, people, process, and techturning disruption into meaningful, lasting progress.","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"https:\/\/faisalhoque.com","theme":{"bg":"#02263c","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#ffffff","buttonHoverBg":"#3b3f46","buttonText":"#000000"},"imageDesktopId":91420512,"imageMobileId":91420514,"shareable":false,"slug":""}}
Hotel magnate Thomas Pritzker will step down as the executive chairman of Hyatt Hotels after details of his affiliation with Jeffrey Epstein were revealed in documents related to the burgeoning investigation of ties between the notorious sex trafficker and the elite and powerful.
Pritzker, in a prepared statement, said he deeply regrets his association with Jeffrey Epstein and Ghislaine Maxwell, a long time associate of Epstein who is serving a 20-year sentence for sex trafficking.
I exercised terrible judgment in maintaining contact with them, and there is no excuse for failing to distance myself sooner, Pritzker said in a statement. “I condemn the actions and the harm caused by Epstein and Maxwell and I feel deep sorrow for the pain they inflicted on their victims.
There are numerous emails between Pritzker and Epstein included in a cache of Epstein-related documents recently released by the U.S. Department of Justice, with several detailing attempts for dinner meet ups and invitations to various functions.
Interactions between the two continued even after Epstein pleaded guilty in 2008 to soliciting prostitution from an underage girl. Emails made public late last year show the crime did little to diminish the desire of that network to stay connected to the financier.
Epstein died by suicide while incarcerated in 2019.
Pritzker, 75, who is the cousin of Illinois Governor JB Pritzker, was the executive chairman at Hyatt for more than 20 years. His retirement is effective immediately. Hyatt CEO Mark Hoplamazian will succeed Pritzker as chairman.
Hyatt has more than 1,500 hotels and all-inclusive resorts in more than 83 countries.
Revelations of ties to Epstein have led to the departure or ousting of multiple high-profile individuals in recent days.
Dubai announced last week that it was replacing the chairman of one of the worlds largest logistics companies, DP World, because of his ties to Epstein.
Also last week, Kathy Ruemmler, the top lawyer at storied investment bank Goldman Sachs and former White House counsel to President Barack Obama, announced her resignation after emails between her and Jeffrey Epstein showed a close relationship where she described him as an older brother and downplayed his sex crimes.
Brad Karp resigned as chairman of one of the most prestigious U.S. law firms earlier this month, saying news coverage of his exchanges with Epstein had created a distraction.
Karp had served as chairman of Paul, Weiss, Rifkind, Wharton & Garrison since 2008. The New York firm has advanced the cause of civil rights, handled the legal affairs of corporate power brokers and grown into a multibillion-dollar global enterprise.
Late last year, King Charles III striped his brother, formerly Prince Andrew, of all his titles and honors, for his relationship with Epstein. This month, King Charles said that he is ready to support UK police examining claims that his brother gave confidential information to Epstein.
Michelle Chapman, AP business writer
Restaurant operators have been automating customer service processes for years. Implementing kiosks, self-checkout, and mobile ordering has helped margins and cut labor costs. But now there’s a problem. Friendliness scores dropped 12 points in just one year. Thirty-three percent of customers actively avoid restaurants that feel too automated. And AI is about to flood the market.
Here’s the choice operators face: double down on customer-facing automation and watch friendliness scores keep falling or use AI differently. Its time to stop automating what customers value and instead start automating what they don’t see.
Smart operators recognize that having AI take orders is not the win. The win is using AI to orchestrate back-of-house operations. That includes tasks like:
Making sure your kitchen doesn’t run out of the appetizer everyone wants on Friday night.
Triggering loyalty promos when inventory starts piling up.
Adjusting labor schedules in real time when online orders spike.
Thats AI as a conductor, coordinating information across your tech stack so your staff can focus on what drives loyalty, making guests feel seen.
THE NEXT AI REVOLUTION
There IS a consumer-facing AI revolution coming. Just not the one operators expect. It’s not in the dining room. It’s on phones, in cars, and through smart speakers.
Half of consumers already use AI-powered search for buying decisions. Soon they’ll place orders directly through ChatGPT or voice assistants or their car dashboard.
Before long, customers will say, “Hey Siri, order our usual from [Restaurant]” while they are driving. Soon they will be able to order a pizza from the ad they see while watching a game. This isn’t science fiction. The infrastructure is already in everyone’s pockets and living rooms.
So now you’ve got a new front door. Great. Except most operators can’t even walk through it. Why? Because their tech stack is a mess:
POS doesn’t talk to inventory.
The loyalty platform won’t integrate with voice ordering.
Menu data is scattered.
Operators deploying chatbots to replace hostesses? They’re building on sand. The ones investing in integrated platforms that let AI coordinate inventory, labor, loyalty, and ordering? They’re building the foundation. When voice ordering becomes default, operators can capture more market share.
3 AUTOMATION STRATEGIES
Successful automation requires going back to the basics.
1. Audit your tech stack. Look for integration gaps. If your systems can’t share data, you’re not ready.
2. Stop automating guest interactions. Kill pilot programs. Instead, shift the budget to technology that supports operational intelligence, such as predictive inventory, dynamic scheduling, loyalty engines that learn.
3. Clean up your menu data. Make it structured with consistent item names, modifiers, and rules that machines can understand. Make it API-ready, so other systems can reliably query whats available and order it correctly. When voice ordering goes mainstream in 2026, restaurants that can’t be found by AI agents will be invisible. Simple as that.
Here’s what it comes down to. AI isn’t replacing your people. It’s making them better. And it’s making sure you show up when someone says “order pizza” to their car. Miss that, and you’re invisible.
Savneet Singh is the president and CEO of PAR Technology Corp.
Nothing is certain, they say, but death and taxes. But a new idea from Meta could add social media to that list.
The tech giant was granted a patent in December that would allow it to simulate a user via artificial intelligence when he or she is absent from the social network for extended periods, including, “for example, when the user takes a long break or if the user is deceased.”
The patent covers a bot that could simulate your activity across Metas products, including Facebook, Instagram, and Threadsmaking posts, leaving comments, and interacting with other users. It could even, potentially, communicate directly with people via chats or video calls, the patent reads.
Andrew Bosworth, Metas chief technology officer, is listed as the primary inventor, and the patent was first filed in November 2023. A Meta spokesperson tells Fast Company the company has “no plans to move forward with this example.”
Withdrawing from a social media platform can affect “the user experience of several users,” the patent reads. “The impact on the users is much more severe and permanent if that user is deceased and can never return to the social networking platform.”
Creepy? Sure seems it. Unprecedented? Not as much as you might think.
In 2021, Microsoft obtained a patent for a chatbot that would let you talk with dead people, both loved ones and celebrities. Like Meta, Microsoft said it had no plans to use the technologyand Tim OBrien, Microsofts general manager of AI programs at the time, said in a social media post he agreed it was “disturbing.” Meanwhile, startups like Eternos and HereAfter AI let people create a “digital twin” that can engage with loved ones after they have passed away.
Meta first publicly discussed the concept of a chatbot for the dead about two-and-a-half years ago, when founder Mark Zuckerberg, in an interview with podcaster Lex Fridman (in the Metaverse, of course), said, If someone has lost a loved one and is grieving, there may be ways in which being able to interact or relive certain memories could be helpful. Zuckerberg did note, however, that the technology could become “unhealthy.”
Metas take on a postmortem chatbot would analyze user-specific data, including posts, voice messages, chats, comments, and likes, to build a sense of who the person was. It would amalgamate that data into a digital persona designed to mimic the users activity.
The bot would identify that any responses were not actually generated by the user, the patent says, but rather were the result of a simulation.
Now, there are some hurdles Meta doesnt mention in the patent. What people say in a direct message to a close friend or loved one isnt necessarily meant for wider consumption. Picture, for instance, one spouse venting to the other about how frustrated they were with their child after some “terrible twos” or teenage incidentonly for that child to later be told by the bot how much they annoyed their now-dead loved one.
After all, AI has yet to grasp social niceties, or when silence or a white lie is better than the truth.
Presently, when someone dies, Meta offers several options for survivors. The page can be permanently removed (assuming you have the necessary paperwork, such as a death certificate), or it can be turned into a memorial, where people can read past posts and leave messages of their own.
As unpleasant as the topic is, Meta has good reason to think about death. One study predicts that by 2050, the number of dead users on Facebook will outnumber the living. By 2100, there could be more than 4.9 billion dead profiles on the platform.
Variant, a generative design tool that promises endless UI exploration, recently introduced a feature most creative people and designers have used for decades: the eyedropper. In Variant, the tool picks vibes: It lets you click on one AI-generated interface and inject its aesthetic DNAtypography, spatial relationships, and color palettesinto another. After so much hype around vibecoding and its text-based imprecision, seeing a familiar, direct manipulation tool applied to generative AI feels great.
The new AI modality takes a nice step to close the gap between the impenetrable ways of large language model black boxes and the tools designers actually use with their eyes and hands. Adopting a universally understood tool to control AI in any way other than words is exactly the kind of innovation the sector needs now.
Its just too bad that Variant itself is the vessel for it. The tools underlying AI engine suffers from a distinct lack of differentiation. Everything it makes looks flat and same-y, so the new style absorb-and-drop tool is not really that useful. Yes, the transformed UI changes, but the results already looked very similar anyway (except for the color palettes).
That said, the implementation is cute. When you click on a previously generated UI, the eyedropper animates the design as it is sucking its soul. You then move the eyedropper, click on another generated UI, and the new style spills over it, rearranging it to match the source. Its a satisfying bit of UI theater, an illusion broken by the fact that you have to wait a little to see the results, as the AI works it all out.
The problem is the little variance in Variant. You cant eyedrop a bitmap image or a Figma project and tell the AI, make this new app UI look like this. Currently, Variants eyedropper feels like trying to paint in Photoshop when your palette only contains five shades of beige.
A for effort
Thats too bad, considering the eyedropper is one of the most resilient and powerful metaphors in computing history. The concept dates back to SuperPaint in 1973, which introduced the ability to sample hue values from a digital canvas. While MacPaint popularized digital painting tools in 1984, it was Adobe Photoshop 1.0 in 1990 that locked the eyedropper icon as the standard for color sampling.
Then, in 1996, Adobe Illustrator 6.0 evolved the tool into a style thief. It allowed designers to absorb entire sets of attributesstroke weights, fill patterns, and effectsand inject them into other objects. Now Variant is effectively trying to take this to its UI design arsenal. The difference is that Adobes tools offered precision. You knew exactly what you were getting. With Variant, you are making a visual suggestion to a probabilistic engine and hoping for the best.
But it is a good change that highlights why we need more tools like this eyedropper and fewer text prompts. Unlike the latest generation of multi-modal video generative AIs, the lack of precision in vibecoding tools is unnerving to me. It reminds me of an exercise I did in communication design class, back in college: A professor made us play a game where one student built a shape with Tangram pieces and had to verbally describe to a partner how to reproduce it with another Tangram set. It was impossible to match it.
We are humans, orders of magnitude better semantic engines than any AI, and even we fail at describing visuals with words. We need interfaces that allow for direct, exact manipulation, not just crossing fingers and hoping for the best. Variants eyedropper shows us the way. Generative AI tool makers, more of this, please. Stop forcing designers to talk to the machine, and let us show what we want.
We made a tool that lets you absorb the vibe of anything you point it at and apply it to your designsIt's absurd and it just worksStyle Dropper, now available in @variantui pic.twitter.com/B3eXDntYtw— Ben South (@bnj) February 10, 2026
