Tech critic David Phelan has been testing the nearly sold-out Apple iPhone Duo ahead of its launch

© David Phelan/The Standard

Il 24 ottobre, l’Ucraina ha annunciato che i suoi sistemi Starlink e Starshield erano di nuovo pienamente operativi dopo oltre una settimana di restrizioni. Nello specifico, il Ministero della Difesa ucraino ha rilevato che la connessione è tornata stabilmente a velocità superiori a 400 km/h, a fronte dei 160 km/h a cui era precipitata giorni prima.
La limitazione, hanno spiegato diversi funzionari ucraini, aveva compromesso la capacità ucraina di intercettare i droni russi in fase di attacco, costringendo le forze armate di Kiev ad annullare alcune operazioni già pianificate.

Segue nostro Telegram.

Tuttavia, a brevissima distanza dal ripristino della piena operatività dei sistemi Starlink a disposizione di Kiev, Elon Musk, amministratore delegato di Space-X, ha accusato l’esercito ucraino di aver violato i termini di servizio.

In un post sul suo profilo X, il magnate di origini sudafricane ha affermato che: «l’uso di Starlink, un servizio commerciale civile, per attacchi con droni offensivi volti a uccidere persone non è mai stato consentito, né dovrebbe esserlo. Starshield è un sistema diverso ed è controllato dal governo degli Stati Uniti. Ciò che decidono di fare spetta a loro».

L’esercito ucraino «ha violato i termini di servizio di Starlink e lo ha utilizzato per attacchi con droni offensivi volti a uccidere persone. Quella falla è stata chiusa la settimana scorsa».

Starlink «rimane cruciale per la difesa dell’Ucraina, poiché è l’unico sistema che non può essere disturbato o distrutto, almeno per ora. Se Starlink venisse spento, il fronte ucraino collasserebbe immediatamente. Questa non è una minaccia, è semplicemente un dato di fatto.

Il sistema si è rivelato cruciale per «impedire alla Russia di sopraffare l’Ucraina nei primi giorni della guerra, poiché tutti gli altri mezzi di comunicazione erano stati distrutti o disturbati dalla Russia».

Space-X «ha fatto molto di più per difendere l’Ucraina di qualsiasi altra azienda privata sulla Terra. Detto questo, non provo che disprezzo assoluto per i guerrafondai che fanno la morale e spingono per un massacro senza fine, senza offrire alcuna soluzione realistica al conflitto. Che possano bruciare all’inferno. Fine della storia».

Starlink è un sistema internet satellitare gestito da Space-X, mentre Starshield è una versione che si avvale della tecnologia Starlink per scopi militari gestita dal governo degli Stati Uniti.

Questa integrazione rende quantomeno ambigua la posizione di Starlink, mettendo in seria discussione la natura privata della casa madre Space-X.

Quella del connubio pubblico-privato negli Stati Uniti è una storia vecchissima, ma che trova negli anni Cinquanta uno dei suoi momenti decisivi.

All’epoca, in piena Guerra Fredda, lo spauracchio sovietico, agitato con cieca ossessività specialmente in seguito allo shock generato dal lancio dello Sputnik (1957) e all’entrata in servizio del bombardiere strategico Tupolev-95, si rivelò determinante a costruire un robusto consenso politico attorno alla decisione dell’amministrazione Eisenhower di destinare una quota crescente di risorse al rafforzamento della marina e al potenziamento dei settori di ricerca e sviluppo.

L’onere di gestire le risorse e coordinare gli sforzi in questo senso ricadde sulla Defense Advanced Research Projects Agency (Darpa), un organismo facente capo al Pentagono fondato, al pari della National Aeronautics and Space Administration (Nasa), nel 1958 in conformità all’esigenza di colmare il divario tecnico-scientifico con l’Unione Sovietica.

Mentre il governo moltiplicava le misure di stimolo alle imprese (assegnazione di commesse pubbliche agli atenei, concessione di agevolazioni fiscali, ecc.), la Darpa elaborava una rete di ricerca e condivisione a maglie strettissime, in grado di interconnettere decine di università, start-up e laboratori di ricerca e portare gli Stati Uniti all’avanguardia nel campo cruciale della scienza. Il lancio dei programmi Gemini e Apollo, lo sviluppo dei missili balistici a medio e lungo raggio, la realizzazione dei sottomarini e delle portaerei a propulsione nucleare si devono al contributo determinante apportato dalla Darpa.

Nel 1982, il Congresso approvò il programma di sostegno alle start-up Small Business Innovation Research (Sbir), affidandone la gestione alla Small Business Administration, predisponendo parallelamente la trasformazione della Darpa nel fondamentale canale di collegamento tra imprese e dipartimenti universitari, in grado di assolvere ai compiti di mediazione tecnologica e imprenditoriale necessari alla commercializzazione delle innovazioni che venivano realizzate nei settori di punta (informatica, telecomunicazioni, biotecnologia, semiconduttori, ecc.) dalle società hi-tech a cui il complesso militar-industriale aveva subappaltato parte delle mansioni legate ai comparti di ricerca e sviluppo, nell’ambito della celeberrima Strategic Defense Initiative (Sdi).

Il progetto, concepito dall’amministrazione Reagan e annunciato urbi et orbi nel marzo 1983, puntava alla creazione di un sistema di difesa aerea multistratificato basato su intercettori spaziali in grado di colpire i missili balistici nemici prima che colpissero il suolo statunitense, “superando” così il principio della mutua distruzione assicurata su cui si era imperniata la Guerra Fredda fino a quel momento.

La Sdi, recita un documento del dipartimento di Stato, «si basava su tecnologie avveniristiche, inclusi sistemi laser spaziali non ancora realizzati», il cui sviluppo assorbì enormi risorse pubbliche in larga parte ridistribuite a beneficio delle aziende private sotto il coordinamento della Darpa.

Una volta convertite all’uso civile, le nuove tecnologie ad elevata intensità di capitale messe a punto contestualmente alla Sdi assunsero rapidamente un grado di pervasività – inteso come capacità di trovare applicazione in un numero sbalorditivo di campi sconvolgendone il funzionamento e i rendimenti – tale da scompaginare il vecchio sistema di produzione fordista ad elevata intensità di manodopera riducendo drasticamente la presenza umana sia nella manifattura che nel settore di servizi.

«Non si tralasciò occasione – osserva Giuseppe Guarino – che si prestasse per la stimolazione e la promozione della Information Technology per finalità militari. Ognuna delle sue fasi essenziali, con successiva trasformazione e utilizzazione dei prodotti a fini commerciali, è stata in pratica resa possibile dal sostegno del Dipartimento della Difesa, dalla Nasa e degli altri dipartimenti militari, compresa la Cia […]. Il merito dell’intervento pubblico non è di aver assorbito l’intera produzione della Information Technology, ma di aver contribuito in maniera determinante con i propri acquisti iniziali a far scendere i prezzi a livelli che avrebbero consentito la diffusione civile e commerciale delle innovazioni, nonché con quelli successivi che hanno favorito l’utilizzazione delle nuove tecnologie nell’organizzazione pubblica, anche civile, e la diffusione delle stesse» (1).

La progressiva affermazione del nuovo modello in tutti i comparti dell’economia ha accelerato il processo di “terziarizzazione” dell’economia statunitense, e palesato il ruolo cruciale giocato dallo Stato rispetto al finanziamento dello sviluppo tecnologico realizzato da aziende private.

Negli anni ’90, sotto la direzione del direttore George Tenet, la Cia iniziò a finanziare generosamente le piccole aziende intente a sviluppare tecnologie applicabili a fini di intelligence aggirando le lungaggini che caratterizzano la mastodontica macchina burocratica statunitense. A questo scopo, la Casa Bianca e il Congresso approvarono la creazione di In-Q-Tel, oscura società di venture capital senza scopo di lucro incaricata di finanziare la messa a punto di tecnologie d’avanguardia per conto – o quasi – della Cia. Come riporta il «Wall Street Journal»:  «i vertici In-Q-Tel considerano l’azienda come indipendente, nonostante i legami molto stretti con la Cia e il fatto che quasi tutte le decisioni di investimento sono gestite dall’agenzia di spionaggio […].La presenza di intricati legami è endemica nel venture capital, in cui la conoscenza intima del settore è essenziale per il successo. Le altre società, tuttavia, giocano con il proprio denaro, o quello di privati».

In-Q-Tel «utilizza denaro pubblico, soggetto a severe norme sul conflitto di interessi e, secondo fonti vicine, si tratta di almeno 120 milioni di dollari all’anno. A volte impegna questo capitale secondo modalità che, anche se non intenzionalmente, avvantaggiano gli stessi amministratori dell’azienda in virtù di altre cariche nel settore hi-tech».

Quasi sempre, le scelte di In-Q-Tel «attraggono altri finanziamenti. Ogni dollaro investito in una piccola impresa in genere è accompagnato da 15 dollari provenienti da altre fonti. Questo aumenta le probabilità di successo del piccolo business in questione e rende le sue stock option più preziose».

Nel 2000, In-Q-Tel ha investito somme piuttosto rilevanti in alcune imprese che producevano satelliti e che si occupavano di raccogliere, classificare e analizzare dati sensibili. In breve tempo, la società ha accumulato quote azionarie di maggioranza di decine compagnie hi-tech, così da dettarne gli indirizzi e allinearle alle esigenze di Washington. Opportunamente orientata dai fondi di In-Q-Tel, una compagnia specializzata nell’elaborazione di soluzioni chimiche adatte alla fabbricazione di tappeti sviluppò in breve tempo una specie di sonda in grado di rilevare la presenza di veleni ed altre sostanze chimiche letali in ambienti chiusi, poi sistematicamente impiegata nei teatri di guerra afghano e iracheno.

Parallelamente, In-Q-Tel investì in un’azienda operante nel settore delle antenne satellitari, ricavando sistemi di localizzazione portatili di cui le truppe e le forze speciali Usa si sono servite per segnalare la propria posizione ai centri di comando.

Più recentemente, i capitali di In-Q-Tel hanno finanziato aziende cruciali come Anduril e, soprattutto, Palantir, saldamente integrata nelle macchine militari statunitense, israeliana e ucraina oltre che nella pubblica amministrazione britannica e in molti altri comparti critici.

L’altissimo livello di “collaborazione pubblico-privato” raggiunto negli Stati Uniti grazie a entità come In-Q-Tel è emblematizzato dal cosiddetto “scandalo Datagate”, scoperchiato nel 2013 grazie alle rivelazioni dello “spifferatore” Edward Snowden. In qualità di tecnico impiegato per una società appaltatrice della National Security Agency, Snowden ha documentato con dovizia di particolari l’esistenza di una rete di sorveglianza capillare e massiccia sui miliardi di comunicazioni sia interne che esterne agli Stati Uniti gestita dalla Nsa stessa attraverso Prism. Vale a dire un sistema in grado di assicurare all’intelligence nazionale la possibilità di accedere in maniera diretta ai server di cui si servivano le aziende operanti nel settore dell’alta tecnologia per ottenere qualsiasi tipo di informazione (messaggi, fotografie, e-mail, ecc.).

Entrato in vigore nel 2007 e beneficiando nel corso degli anni della collaborazione di colossi del calibro di Microsoft, Yahoo, Google, Facebook, PalTalk, Skype, AoL, YouTube e Apple, Prism ha posto la Nsc e anche il Gchq britannico nelle condizioni di acquisire i metadati in transito attraverso i cavi in fibra ottica collegati ai server centrali di Yahoo e Google.

Space-X non ha beneficiato dei fondi di In-Q-Tel, ma del subappalto sistematico ad opera di agenzie statali come la Nasa o dello stesso Pentagono, che come contropartita gode di libero e completo accesso alla tecnologia  Starlink. Al punto da garantirlo a Paesi alleati come l’Ucraina.

La decisione di ridurre prima e ripristinare poi il servizio Starlink a Kiev non fa quindi capo a Elon Musk, ma sui vertici del potere pubblico statunitense a cui Space-X – così come le altre grandi aziende hi-tech statunitensi – rispondono più o meno indirettamente.

NOTE

1) Guarino, Giuseppe, I soldi della guerra. Gli Stati Uniti: spesa militare, innovazione, economia globale, Mondadori, Milano 2003, pp. 67-80.

• OpenAI CEO Sam Altman is conflicted about how AI models are changing fields like mathematics.
• Altman said that he had some "sadness and melancholy" when he saw what OpenAI's models could do.
• Prominent mathematicians have blasted AI companies for pursuing some of the field's most complex problems.

OpenAI CEO Sam Altman said he feels conflicted about OpenAI solving some of math's biggest problems.

"As I said, I don't think it was bad for AI," Altman told Stripe President John Collison during an episode of "Cheeky Pint" that was released on Thursday. 'And I think, on net, the world deserves to understand how the universe works. But I do have empathy for the method."

A number of prominent mathematicians have blasted AI companies for pushing frontier models to solve some of the most complex math problems. OpenAI pulled out of a Caltech "math hackathon" after current and former Caltech mathematicians and others in the field said the event was "likely to have destructive impacts for the mathematical community."

Separately, 28 Fields Medal winners published their own open letter, warning that "the push by AI companies to solve mathematical problems as a benchmark is detrimental to the science of mathematics, and to the mathematical community."

OpenAI received considerable attention after announcing that its models had solved a 90-year-old, unresolved problem related to the Navier-Stokes equations, a set of formulas that predict how liquids and gases move. The Clay Mathematics Institute had offered $1 million to anyone who could determine whether those formulas would always behave as we would expect in the physical universe, part of a series of difficult mathematical questions known as the "Millennium Problems." OpenAI said that its AI models found an example where the equations do not, and end up with a "singularity" where a vortex starts spinning infinitely quickly.

Altman said OpenAI is "working through" how to continue scientific research while also making sure those in highly specialized fields "get the best version of the future that they can have."

"I would like to figure out a way to do this that is as sensitive to the communities that have driven these fields forward for centuries as we can," he said, adding, "It's a hard thing."

Collison tried to assuage Altman's feelings, reminding the OpenAI CEO that "the longshoremen union weren't happy about the port automation."

Altman said part of the reason he feels differently about the whole experience is based on an OpenAI Slack channel that was named "Model Name-Hold My Beer" after the meme about if you think this is crazy just wait. (He did not explain if this was a released model or not.)

"People were just throwing that at the hardest problems they could," Altman said. "And people didn't think the model was going to do these things."

The AI model tackled "people's hardest research problems from their academic career." Altman said he was so excited, "I couldn't go to bed."

In addition to the euphoria, he said other feelings set in.

"I also had much more sadness and melancholy, and a feeling of something ending than I was expecting," Altman said.

The OpenAI CEO said it was "beautiful" that any mathematician could solve even the most formidable problem if only they worked hard enough. Altman said AI is changing that equation.

"I mean, certainly, most mathematicians never solve a Millennium Prize problem," he said. "But the idea that you might be able to."

I moved to Poland from London with my Polish husband, and we are raising our two-year-old  son here. My son will grow up and live his entire life within the digital infrastructure that the government is building today. The fact that I was not born in Poland does not mean I don’t care about its future, as a mother, I have an obligation to fight for my son’s rights as a Polish  citizen. 

Recently, I saw that the Polish Ministry of National Defence has signed a  Letter of Intent with the American analytics company Palantir Technologies.

Having lived  through what happened in the United Kingdom when our government signed similar agreements, I believe that Polish people deserve to know what happens when public systems are handed over to a foreign defence contractor.

Among the company’s founding investors was In-Q-Tel, the CIA’s venture capital arm. 

In London, I was part of a campaign group fighting against Palantir’s contract with our National Health Service (NHS).

Under that agreement, the US defence contractor gained deep access to NHS data systems, including mental health records, reproductive health data, and medication history.

Heavily redacted contracts

Over 100 pages of the contract remain heavily redacted and hidden from the public. Because Palantir is an American corporation, citizens cannot easily hold them to account or use basic data access requests in the way they could with an EU-based company.

In the UK, two parliamentary committees and the British Medical Association are calling on the government to exit the contract before the break clause in  2027. 

Poland is not the first European country Palantir has approached, but other allied nations  have already drawn the line.

In Switzerland, the Swiss armed forces rejected Palantir’s offers nine times. The military’s conclusion was that it was technically impossible to guarantee that  sensitive national data would not reach American intelligence services.

When a Swiss  investigative outlet reported on this, Palantir sued them and lost on 22 of 23 counts. In Germany, the Bundeswehr explicitly excluded Palantir from a military cloud procurement on  grounds of data sovereignty, with senior military officials stating that granting external  industry employees’ access to the national database was completely unacceptable. 

Poland is now preparing to deepen its cooperation with a company that the Swiss and German militaries both rejected on security and sovereignty grounds.

It is our children who  will have to face the long-term consequences of today’s decisions. I have seen what happens when people in the United Kingdom did not ask questions until it was too late, and I cannot stay silent and watch exactly the same thing happen in Poland.

If you care about data sovereignty, public transparency, and protecting the next generation,  join us at the public protest against Palantir’s contracts in Poland, organised by Nasze Dane, on 8 November at Plac Konstytucji in Warsaw at 3pm.

• Halloween is the busiest day for Uber and Lyft drivers.
• Lyft's CEO called it "our biggest day" in 2025.
• It can be a lucrative day for gig workers — if they can stand rowdy passengers.

The busiest day of the year for Uber and Lyft drivers is almost here.

While you might guess that New Year's Eve or St. Patrick's Day would draw the most rider demand as people go out partying, Halloween actually wins the prize for ride-hailing drivers in the US, according to ride volume data and comments from executives.

The culmination of spooky season is likely to be big again this year, with Halloween falling on a Saturday, giving people more time to party — and creating more demand for rides.

Last year, the most ride-hailing trips occurred on October 31, according to Gridwise, a data analytics company. Ride volumes were 23% higher than median rideshare volume for the year, adjusted for day of the week and quarterly seasonal variation, Gridwise said.

The next-busiest days were November 1 and July 3, the day before Independence Day. Ride-hailing volumes were about 20% above the median on each day, Gridwise said. The day after Halloween may be big because rides run late into the night.

Halloween kicks off the busiest season of the year for Uber and Lyft. Besides getting a ride home from Halloween parties, riders also request trips to the airport for flights home around Thanksgiving and Christmas.

Earlier this month, Uber offered some drivers hundreds of dollars in bonuses for driving more — a potential sign that the company needs more drivers working as ride demand ramps up.

For Lyft, Halloween 2025 was "our biggest day" and "our biggest week, and not by a little bit," CEO David Risher said a few days after Halloween 2025 on an earnings call last November.

A company spokesperson declined to say how many trips Lyft drivers completed on Halloween but said that it amounted to "millions of rides."

Uber, meanwhile, "hit a new record over Halloween weekend" in 2025, with about 130 million ride-hailing and delivery trips worth $2 billion, CEO Dara Khosrowshahi said on an earnings call last November.

While lucrative, working on Halloween also comes with risks for drivers.

Raymundo Rodriguez, who drives for Uber on the island of Oahu in Hawaii, said he started driving for the app around Halloween two years ago.

There were plenty of riders, he said, but some trips involved dealing with rowdy or drunk passengers. Costumes can also pose risks: One rider dressed as a stripper left glitter on the black leather seats of his car after she got out — a mess that Rodriguez said he had to clean up himself.

This year, Rodriguez said, he's planning to skip driving for the app and take his kids trick-or-treating.

"I don't want nobody throwing up in my car," he said.

Do you have a story idea about Uber? Contact this reporter at abitter@businessinsider.com or via encrypted messaging app Signal at 808-854-4501. Use a personal email address, a nonwork WiFi network, and a nonwork device; here's our guide to sharing information securely.

With the midterm elections approaching, data centers have become an increasingly popular political target. Politicians are raising alarms about their impact on communities, with some calling for construction moratoriums or outright bans.

There's been plenty of fearmongering. But, the more Americans learn about data centers, and the more communities experience their benefits firsthand, the stronger the case for continued development becomes.

Data centers have existed for decades, but the rapid growth of artificial intelligence has dramatically increased the demand for computing power. Americans understandably have questions about what this development means for electricity prices, water supplies, and their quality of life. Newer facilities are increasingly being designed with technologies that can address these concerns, including closed-loop cooling systems and quieter operations.

DATA CENTERS COULD BE THE SLEEPER ISSUE OF THE 2026 MIDTERM ELECTIONS

Yet much of the national debate overlooks what data centers are already delivering to communities across America.

Take Ellendale, North Dakota, a town of 1,100 people. Before Applied Digital's new $11 billion data center development, the town collected roughly $400,000 annually in sales taxes. In the first seven months of 2026 alone, that figure jumped to $3.5 million.

THIS TINY FARMING TOWN YOU’VE NEVER HEARD OF IS HELPING MICROSOFT SELL AMERICA ON DATA CENTERS

That revenue is helping the city repave its roads and renovate its senior center. A historic opera house is being restored. For a small town, this kind of investment can transform the community for generations.

And Ellendale is far from alone.

Schools across America are seeing enormous benefits from the data center boom.

In Richland Parish, Louisiana, Meta's massive investment has helped fund an increase in annual teacher bonuses ranging from $10,000 to $50,000.

Think about what that means for a working family. Teachers who have dedicated their careers to educating children are receiving a life-changing financial benefit, while the school district becomes an attractive destination for talented educators.

It's hard to imagine a politician campaigning against higher teacher pay or better schools. Yet those are precisely the kinds of benefits communities stand to lose when development gets blocked.

PROPOSAL WOULD PAY HOMEOWNERS 'DATA CENTER DIVIDENDS' TO COUNTER GROWING BACKLASH

The benefits extend to homeowners struggling with the rising cost of living.

In Loudoun County, Virginia, data centers generate 45 percent of county revenue. That enormous tax base has allowed local officials to cut homeowner property tax rates every year for a decade.

TO BEAT CHINA IN AI, AMERICA MUST WORK WITH COMMUNITIES — NOT AROUND THEM

In neighboring Henrico County, local leaders used data center revenue to fund a $60 million Affordable Housing Trust Fund, helping young families and first-time buyers purchase homes.

At a time when housing affordability is dominating political conversations, building data centers is helping some communities make the American Dream a little easier to achieve.

Electricity prices are another major concern, and understandably so. Families are already struggling with higher utility bills, and the arrival of a major new industrial customer can raise legitimate questions about who pays for new infrastructure.

But the enormous, predictable electricity demand from data centers also creates opportunities to finance new generation and grid investments, spreading infrastructure costs across a larger customer base.

In Georgia, agreements involving new large electricity customers, including OpenAI, are projected to lower residential power bills by $180 annually by 2029.

With the right arrangements in place, the arrival of data centers can mean lower costs for the families already living in these communities.

And the benefits extend well beyond tax revenue and household expenses.

Trade unions have been among the most vocal supporters of data centers because of the good-paying jobs they create. Building these facilities requires years of work for skilled tradesmen, with new projects creating opportunities for the next generation of American workers.

In Baltimore, Maryland, IBEW Local 24 expanded its membership by 50 percent alongside data center construction. In Luzerne County, Pennsylvania, IBEW Local 163 doubled its ranks following construction of an Amazon data center.

Politicians have spent decades promising to bring back good-paying blue-collar jobs. Here is an industry creating enormous demand for exactly those workers.

Of course, developers have a responsibility to work with residents and local governments to address legitimate concerns. Communities deserve clear answers about electricity costs and water use. They should also understand how the investments taking place in their backyards will benefit local families.

But, when data centers are developed responsibly, the economic benefits can be enormous. Communities are rebuilding roads and improving schools. Homeowners are seeing lower tax burdens, and skilled workers are finding new opportunities.

For decades, politicians have campaigned on championing the Forgotten Man and bringing economic opportunity back to communities that have struggled to attract investment.

Now, data centers are delivering billions of dollars in private investment to many of those very places.

We're watching a new industrial boom take shape across America, creating jobs and revitalizing communities that Washington has spent decades promising to help. Before politicians rush to stop it, they ought to take a serious look at what those communities are gaining.

CLICK HERE TO DOWNLOAD THE FOX NEWS APP

And they should consider what their constituents stand to lose.

• Microsoft offered a buyout to 60,000 employees this week.
• The offer spooked the tech industry and stoked chatter and concern inside the company.
• The key question: How many employees Microsoft is prepared to lose?

Microsoft's huge employee buyout stunned the industry this week, raising urgent questions about why the software giant might want so many people to leave.

Inside the company, concerned staff debated the meaning of the move. Why might Microsoft want to shed 60,000 workers? Will roles be replaced by AI? Are more layoffs ahead?

Business Insider spent recent days speaking with Microsoft insiders to find out what's really happening. The people asked not to be identified discussing sensitive matters.

The main takeaway so far is that Microsoft seems to have unnecessarily spooked employees with the sheer size of its buyout offer, leaving leaders scrambling to calm the troops.

Chief Financial Officer Amy Hood in an internal memo explained the offer was intended to "offer employees agency if they choose to leave the company on their own terms" and said the company expected "very few people" to accept.

Microsoft doubled down later and told Business Insider it anticipates 3,000 or fewer employees will take the buyout. That's only about 5%, or less, of those eligible.

Layoff concerns

Earlier this year, the company offered a retirement buyout to employees based on age and tenure. About one-third accepted.

Soon after, Microsoft did another round of layoffs. Some employees who didn't accept the retirement offer ended up losing their jobs.

That's led to speculation that more layoffs may come in the wake of this week's bigger buyout offer.

"Future workforce needs will continue to be based on business needs and priorities," Microsoft said in a statement.

Misconceptions and differences

Business Insider spoke to Microsoft executives who said there are misconceptions about the current buyout offer.

These executives could only speak about their own teams, and they were hesitant to definitively say layoffs aren't coming. However, they pointed out a few differences between the current buyouts and the retirement offer.

"Microsoft doesn't want to shrink by 60,000 people," one executive said, while asking not to be identified. "We are not replacing everyone with AI."

This person thinks Microsoft made latest buyout offer because some employees complained that they weren't eligible for the retirement buyout earlier this year.

And in the case of the retirement buyout, when employees left, Microsoft didn't fill those roles — the headcount went away.

This week's buyout works differently. In some cases, when staff leave this time their roles will be backfilled with new hires.

Why the latest buyout looks so big

One reason the latest buyout looks so huge: Some leaders decided it would be simpler to offer it to their entire organization, rather than try to pick which specific teams or individuals should be included.

"It's easier to opt-in giant organizations," one Microsoft executive told Business Insider.

Charles Lamanna, the executive in charge of the Copilot Platform, AI agents, Office and Word, decided to offer it to everyone in his organization, one of the people said. That includes many people in critical positions. The organization is prepared to backfill roles if any of those critical employees accept the buyout, this person said.

The Cloud + AI group took a different approach

Microsoft's key Cloud + AI division took a slightly different approach, according to another executive from the company.

Two teams — Azure Hardware Systems and Infrastructure and Cloud Operations + Innovation teams — decided not to offer the latest buyout to staff.

Most other groups within the Cloud + AI division did offer the buyout to employees. And this division can backfill some roles, this executive told Business Insider.

Cutting costs and avoiding layoffs

Backfilling roles could still help Microsoft cut labor costs. One executive at the company said if a senior person were to leave, they would likely backfill the role with someone more junior who would potentially be paid less. A Microsoft spokesperson said that's not an official company position.

If all this seems overly complex, compared to layoffs, there may be a reason: Microsoft has been under fire for cutting American staff while employing foreign workers.

The Trump administration recently suspended Microsoft from a federal program that sponsors foreign workers for permanent residency, accusing the company of laying off American employees while hiring lower-paid foreign workers. Microsoft has denied those claims.

"Never say never, but there's a lot of desire not to do big layoffs, especially as an American company," one of the Microsoft executives told Business Insider.

Have a tip? Contact this reporter via email at astewart@businessinsider.com or Signal at +1-425-344-8242. Use a personal email address and a non-work device; here's our guide to sharing information securely.

• I schmoozed with CEOs, founders, and investors at Business Insider's The Long Play in New York.
• Attendees heard from speakers ranging from Barry Diller to Ashley Graham — and spent a whole lot of time networking.
• Some attendees tried pitching deals. Others debated the future of work.

The New York venue was bustling with executives, some of whom governed companies with thousands of employees.

I met my first one within seconds of walking in.

On Thursday, I attended Business Insider's The Long Play event at Perelman Performing Arts Center. (Full disclosure: Business Insider is my employer, and you're probably reading this on its website.) I entered the event with a mission: to chat, network, and maybe learn a thing or two about my own career.

Mingling among the crowd, I chatted with the founders of AI startups and communications chiefs. They debated remote work, asked about management styles, and discussed working together. "I've got to work on my elevator pitch," one CEO told me.

I sat down for lunch with some attendees to discuss a question: Is workplace loyalty dead? Most said no, or that it was more complicated. One brave soul, the content creator Colin Rocker, said yes.

"I still think it's dead," he told me after the lunch chat. "But, I will say, I got more context that it's not necessarily working across the org chart."

He added: "That hurts managers, too."

Others at our table traded mentorship techniques and thoughts on leading an office. Nitzan Shapira said that, while working at Intel and Cisco, meetings were crucial. He has more flexibility while leading his AI startup, Harmony. "I'm kind of allergic to meetings," he said.

Munching on my plate of salmon, salad, and a caprese sandwich, I continued to chat with the crowd. Then came the dimming of the lights and the voice on the loudspeaker. The interviews were about to begin.

A who's who of the audience

You can tell a lot about the crowd from when people raise their hand — and when they don't.

I learned, for example, that much of the crowd had played "Grand Theft Auto." What better way to blow off some steam after a career-building conference than with some virtual vice? The attendees didn't consider themselves avid gamers, though; only a few hands shot up when Take-Two CEO Strauss Zelnick asked.

There were also a handful of investors in the crowd. Supermodel Ashley Graham asked for a show of hands, and I spotted three. During one of the breaks between talks, I talked shop with yet another venture capitalist.

As for GLP-1s, the crowd was less eager. Senior correspondent Melia Robinson asked while interviewing Ro CEO Zach Reitano, and she got only a few raised hands. (I bet, if we swapped the public hand-raising for a blind survey, that number would go up.)

The attendees around me were rapt. My seatmate, who worked for an AI startup, took notes with pen and paper. I saw some others jotting down learnings in the Notes app on their phones.

WeWork CEO John Santora described upping the once-freewheeling company's dress code. When he got there, Santora spotted pajamas and shorts; he now spots some salespeople in sport coats.

It prompted me to look around and check out what the audience was wearing. There were lots of blazers, button-ups, and slacks. A handful went more casual with their denim jackets and colorful blouses.

There were belly laughs, like when my talented colleague Sarah Needleman flubbed "how do you define success" with "how do you define sex" while interviewing Zelnick. He chuckled; the crowd joined in, and it bonded the audience.

And there were sobering moments, like when Noa Argamani described her drive to fund Israeli startups after spending 246 days in Hamas captivity.

The list goes on: There were chats with media mogul Barry Diller, banking entrepreneur Cristina Junqueira, and Nadia Carlsten, the CEO of the sneaker-company-turned-AI-infrastructure-provider Smartbird.

The event was really about meeting other people, though, and being inspired. I endlessly looped around the room, talking to C-suiters at startups and public relations agents. You could always hear the hum of new connections coming from the back of the room.

It ended with glasses of wine and hors d'oeuvres. After biting into a surprisingly delicious slider — Graham talked about pairing her wine with Korean barbecue and lasagna, and it made me hungry! — I had officially reached my schmoozing limit.

I ended the day with a full heart and a handful of new phone numbers. If you gave me yours, expect a text soon with plans to get coffee.

Pechino Express torna nel 2027 con dieci nuove coppie e una lunga avventura attraverso il Sud America. Sky ha svelato il cast della prossima edizione, intitolata “La rotta dei due Oceani”, che porterà i concorrenti dal Cile al Brasile passando per l’Argentina. Alla guida del programma sarà ancora una volta Costantino della Gherardesca. La nuova stagione dello show Sky Original prodotto da Banijay Italia sarà trasmessa in esclusiva su Sky e disponibile in streaming su NOW nel 2027.

Pechino Express 2027, tutte le coppie del cast

Tra i nomi più attesi ci sono Rocco Siffredi e Rózsa Tassi, marito e moglie dal 1993. Lei, ex Miss Ungheria, ha mantenuto negli anni un profilo decisamente più riservato rispetto al marito. Per la coppia Pechino Express rappresenterà una nuova esperienza insieme dopo oltre trent’anni di matrimonio.

Spazio anche a Franceska Nuredini e Ginevra Festa, ballerine professioniste e amiche. Nuredini ha lavorato in numerosi tour e produzioni televisive ed è nota anche per la relazione con Elodie, mentre Festa è stata tra le Veline di Striscia la Notizia.

In gara ci saranno poi Benji e Fede, Benjamin Mascolo e Federico Rossi. Dopo la separazione artistica del 2020 e la reunion iniziata nel 2024, il duo pop affronterà questa volta una sfida lontana dai palchi.

La coppia dei creator sarà formata da Alessia Lanza e Andrea Fratino. Lei è una delle influencer italiane più seguite della sua generazione e conduce il podcast Mille Pare; lui crea soprattutto contenuti legati al calcio ed è stato tra i concorrenti de La Caserma.

Visualizza questo post su Instagram

Un post condiviso da NOW (@nowtvit)

Pechino Express 2027, dagli sportivi alle Bambole di Pezza

Il mondo dello sport sarà rappresentato da Andrew Howe e Tommaso Marino. Howe è stato vicecampione del mondo e campione europeo nel salto in lungo e detiene il record italiano di 8,47 metri. Marino ha invece alle spalle oltre vent’anni da cestista professionista ed è oggi commentatore di basket.

A rappresentare la musica rock saranno Cleo Ungarelli e Dani Piccirillo, rispettivamente cantante e chitarrista delle Bambole di Pezza.

Una delle coppie più riconoscibili sui social sarà invece quella formata da Angela “Bella Fisica” Lin e Alessandro Zhu, madre e figlio. Angela è diventata virale grazie ai video realizzati nel suo negozio romano e ai suoi tormentoni, mentre Alessandro è spesso protagonista dei contenuti insieme alla madre.

Nel cast anche Danilo da Fiumicino e Alessandro Pieravanti, amici, autori e podcaster alla guida di De Core. Danilo Contaldo è inoltre speaker radiofonico, mentre Pieravanti ha lavorato per anni anche come musicista e autore.

Pechino Express 2027, dai divulgatori Geopop alle sorelle Gregoraci

Completano il cast Riccardo De Marco, conosciuto come DeNa, e Maria Bosco, MeriBì. Entrambi lavorano per Geopop: lui è chimico e divulgatore scientifico, lei ha una formazione in matematica e ingegneria matematica e si occupa di divulgazione su scienza, tecnologia e temi di attualità.

Ultime, ma tra le coppie destinate ad attirare maggiore attenzione, Elisabetta e Marzia Gregoraci. Le due sorelle sono molto legate ma hanno percorsi professionali differenti: Elisabetta è da anni un volto della televisione italiana, mentre Marzia lavora nel settore immobiliare e delle ristrutturazioni.

La nuova rotta partirà dai deserti del Cile e dalle Ande, proseguirà attraverso le pianure argentine e le cascate di Iguazú, per poi entrare in Brasile. I concorrenti attraverseranno il Pantanal e le zone legate alle antiche miniere d’oro fino al traguardo di Rio de Janeiro. Un viaggio da ovest a est che unirà quindi Oceano Pacifico e Oceano Atlantico, da cui il titolo scelto per l’edizione: “La rotta dei due Oceani”.

Visualizza questo post su Instagram

Un post condiviso da Pechino Express (@pechinoexpress)

CLICCA QUI PER RESTARE AGGIORNATO SUI TUOI PROGRAMMI TV PREFERITI

L'articolo Pechino Express 2027, dieci coppie dal Cile al Brasile: ecco chi sono i nuovi concorrenti proviene da Affaritaliani.it.

It has been described as the biggest hack in history

© Getty/iStock

It has been described as the biggest hack in history

© Getty/iStock

Veronica Bruganelli è la seconda eliminata dei Live di X Factor 2026. La concorrente della squadra di Jake La Furia ha perso il ballottaggio contro Lenanie, del team di Irama. Il verdetto è arrivato con tre voti contro uno: Irama, Francesco Gabbani e Paola Iezzi hanno scelto di eliminare Veronica, mentre Jake La Furia ha votato contro Lenanie.

X Factor 2026, chi è finito al ballottaggio

Il secondo Live è stato diviso in due manche da sei concorrenti. Al termine della prima il meno votato dal pubblico è stato Lenanie, che aveva interpretato C’è tempo di Ivano Fossati con una riscrittura personale del testo. La scelta aveva provocato anche una discussione tra i giudici, soprattutto tra Francesco Gabbani e Irama, sull’opportunità di modificare così profondamente una cover.

Nella seconda manche è stata invece Veronica a ottenere meno voti. La cantante aveva affrontato Total Eclipse of the Heart di Bonnie Tyler.

X Factor 2026: Veronica contro Lenanie, il voto dei giudici

Nel confronto finale Lenanie ha cantato Cambia un uomo di Marco Mengoni. Veronica ha scelto invece il proprio cavallo di battaglia, You Stole the Show di Sienna Spiro. Jake La Furia ha difeso la propria concorrente votando per eliminare Lenanie. Irama ha fatto la scelta opposta.

A decidere il risultato sono stati quindi Gabbani e Paola Iezzi, che hanno entrambi votato contro Veronica. Con tre voti su quattro la concorrente ha dovuto lasciare il talent e non è stato necessario ricorrere al Tilt, il televoto rapido previsto in caso di parità.

Visualizza questo post su Instagram

Un post condiviso da X Factor Italia (@xfactoritalia)

X Factor 2026: chi si è salvato

Lenanie resta quindi nella squadra di Irama insieme a Patrizia e Le Distanze. Jake La Furia prosegue invece con ALÌA e Alex Fontana. Veronica era già finita al ballottaggio durante il primo Live, quando era riuscita a salvarsi contro Alice Fregoso. Alla seconda occasione il verdetto è stato opposto.

Tutti gli altri concorrenti hanno superato direttamente il televoto. Tra le esibizioni più apprezzate della serata quelle di camilla con As the World Caves In e Alex Fontana con Street Spirit (Fade Out) dei Radiohead.

X Factor 2026, MEEK ospite del secondo Live

Superospite della puntata è stata MEEK, progetto della cantautrice e producer britannica Georgia Meek. L’artista si è esibita con Fabulous, brano che negli ultimi mesi ha ottenuto forte diffusione anche in Italia. Dopo due Live sono quindi due le concorrenti eliminate: Alice Fregoso nella prima serata e Veronica nella seconda. Il prossimo appuntamento è previsto per giovedì 5 novembre.

Visualizza questo post su Instagram

Un post condiviso da X Factor Italia (@xfactoritalia)

CLICCA QUI PER RESTARE AGGIORNATO SUI TUOI PROGRAMMI TV PREFERITI

L'articolo X Factor 2026, secondo Live: chi è stato eliminato e chi invece si è salvato proviene da Affaritaliani.it.

Patrick Stewart è morto a 86 anni nella sua casa di Los Angeles, circondato dalla moglie Sunny Ozell, dai familiari e dagli amici. L’attore britannico, protagonista di Star Trek e della saga degli X-Men, aveva affrontato negli ultimi anni l’Alzheimer. La famiglia ha annunciato la morte il 29 ottobre.

Patrick Stewart: da Shakespeare al capitano Picard

Patrick Stewart era nato il 13 luglio 1940 a Mirfield, nello Yorkshire. Prima della fama televisiva aveva costruito una lunga carriera nel teatro britannico. Dopo la formazione alla Bristol Old Vic Theatre School entrò nella Royal Shakespeare Company, dove rimase per circa 14 anni interpretando numerosi personaggi del repertorio classico.

La svolta internazionale arrivò nel 1987, quando ottenne il ruolo del capitano Jean-Luc Picard in Star Trek: The Next Generation. Interpretò Picard per 178 episodi fino al 1994 e tornò successivamente nel personaggio nei film Star Trek: Generations, Primo contatto, L’insurrezione e La nemesi. Decenni più tardi tornò ancora una volta nell’universo di Star Trek con la serie Star Trek: Picard.

Patrick Stewart e il professor Xavier negli X-Men

Per un’altra generazione Stewart è soprattutto Charles Xavier, il Professor X. Il debutto avvenne con X-Men nel 2000. Tornò nel ruolo in X-Men 2, X-Men: Conflitto finale, X-Men – Giorni di un futuro passato e Logan, oltre ad altre apparizioni legate all’universo Marvel.

Il rapporto sullo schermo con Ian McKellen, interprete di Magneto e suo grande amico anche nella vita, diventò uno degli elementi più riconoscibili della saga. Stewart riprese ancora una versione di Xavier in Doctor Strange nel Multiverso della Follia.

Patrick Stewart: da Dune a Robin Hood

La carriera cinematografica era però molto più ampia. Nel 1981 partecipò a Excalibur di John Boorman. Nel Dune di David Lynch del 1984 interpretò Gurney Halleck. Nel 1993 apparve invece nella commedia Robin Hood – Un uomo in calzamaglia di Mel Brooks, interpretando Re Riccardo.

La sua attività teatrale non si interruppe con la fama hollywoodiana. Nel West End e a Broadway fu protagonista, insieme a Ian McKellen, anche di Aspettando Godot di Samuel Beckett.

CLICCA QUI PER ALTRE NOTIZIE DI SPETTACOLI E COSTUME

L'articolo Patrick Stewart morto a 86 anni: da Picard al Professor X, chi era e quali film ha fatto proviene da Affaritaliani.it.

• Once experimental tech, AI turrets are now becoming an "integral part of air defense" in Ukraine's war.
• These are guns that can detect and lock onto targets by calculating speed and trajectory.
• Ukraine has deployed over a dozen of these and is looking to surge their production and use.

Ukraine is using machine gun turrets controlled by artificial intelligence in battle, and they're working well enough that Kyiv wants to roll them out at a faster pace.

While Ukraine began revealing prototypes of these systems last year, Brave1, the country's defense innovation platform, said on Thursday that they've now become "an integral part of air defense" for its forces.

"In the last month alone, automated turrets have destroyed dozens of enemy drones over Kyiv," Brave1 said, adding that it started work on these weapons in 2024 as experimental technology.

Not much is publicly known about the core capabilities of these turrets, such as their range or how their software works, but they've primarily been described as an air defense tool against Russian drones.

The idea is that artificial intelligence helps the turret detect, acquire, and track a target, making it more accurate than human operators at engaging high-speed air targets such as jet-powered Gerans.

More than a dozen such systems are now deployed in the Kyiv area, with another eight planned for the region, Ukrainian President Volodymyr Zelenskyy said in early October.

In a sign that Ukraine plans to lean further into the technology, Zelenskyy later praised the systems further, saying on October 19 that AI-powered turrets have "proven highly effective against shaheds."

"We are increasing orders, and it is important to accelerate every process as much as possible — organizational, financial, and production-related," Zelenskyy said.

One such system, the Sky Sentinel, is a .50-cal M2 Browning turret developed and operated by private air defense firm Carmine Sky. It can spin 360 degrees and lock onto a target while firing by calculating speed and trajectory.

Each one costs about $150,000, a Ukrainian commander told Business Insider's Jake Epstein last summer, and can distinguish between birds and drones. The turret still requires a human operator to authorize when to fire, the commander said.

Brave1 released a montage on Thursday indicating that Ukraine is also fielding other types of automated turrets, showcasing clips of several designs.

It said turrets had shot down at least 60 targets under its Private Air Defense Project, a pilot that allows private Ukrainian firms to buy air defense equipment and train civilian staff to use them against drones threatening factories, warehouses, and other important sites.

Notably, Brave1 said these turrets took down Russian Geran-5s and Banderols — jet-powered drones and cruise missiles that can reach speeds of 375 to 400 mph.

A similar claim was made by Yuriy Ihnat, the Ukrainian air force's spokesperson, in early October, when he said that AI-powered turrets had shot down Russian jet-powered drones for the first time. Ihnat did not say which system was used.

Kyiv has eagerly integrated AI into much of its war effort, using machine learning models for other functions, such as predicting enemy drone strikes and autonomously conducting supply runs with robotic buggies.

The turrets do have limited range, making them more useful as a final defense against drones or against low-flying targets.

They'll likely be one of several tools Ukraine is developing to counter Russian jet-powered threats.

Over the last few months, the Kremlin has been ramping up production and deployment of cheap cruise missiles or drones equipped with turbojet engines. These weapons, flying at nearly half the speed of sound and launched by the thousands every month, have become a major challenge for Ukrainian air defenses.

• Reddit is testing a departure from its open-internet roots.
• The platform will soon require users to log in to view its content to prevent content scraping.
• CEO Steve Huffman said Reddit can also better target ads to logged-in users.

Your lurking days on Reddit may be numbered.

Reddit plans to begin testing a primarily logged-in experience, a shift that could make it harder for casual visitors to browse the site without an account.

CEO and cofounder Steve Huffman said on Thursday's earnings call that Reddit will test the change in select markets outside the US in the fourth quarter. The company has not said when it will roll out a fully logged-in experience globally.

The tests will help Reddit determine how much content logged-out visitors can still access, including through shared links, search, and its front page.

This is a departure from the open platform it has been for over two decades, where internet users could view its content without logging in.

"It's a big one for us, and it's a big one for me," Huffman said to investors about the change. He gave a few reasons for the shift.

First, logged-in users mean Reddit can target ads more effectively, as they provide the platform with more information to make ad decisions. COO Jen Wong added that more than 80% of Reddit's monetization "sits on the shoulders of logged-in and app users," calling them the company's "bedrock."

Second, the move aims to protect Reddit content from third-party scraping.

"The vast majority of the scraping and stealing of Reddit happens in our logged-out user base," Huffman said. He did not specify who the third parties involved in content scraping were, AI agents or otherwise.

Reddit addressed similar concerns in 2024, updating its "Robots Exclusion Policy" to put guardrails in place against unknown bots crawling its site. It also sued Anthropic, Perplexity, and several other data-mining companies last year, accusing them of scraping Reddit comments for AI training.

Reddit reported a 53% increase in third-quarter revenue, from $584 million in 2025 to $895 million in 2026. Its global daily active unique users in the quarter rose 12% to 130.3 million, even as US users declined 2%.

Huffman said Reddit's search-referral traffic was down in the third quarter. In October, search traffic fell by more than 1 million referrals per week on average, he said. If that rate continues, he said Reddit could lose a significant percentage of its current 130 million daily active unique users in the fourth quarter.

The company's stock dropped more than 17% in after-hours trading on Thursday.