In December of last year, Microsoft told thousands of its engineers, product managers and designers that they could use Claude Code, Anthropic’s command-line coding agent, on the company dime.
By spring, the tool had spread well beyond engineering: into the kind of non-technical roles that, in earlier waves of enterprise software, would have waited years for a seat. Inside Microsoft, the rollout was framed as a learning exercise. Outside it, the surface signal was simpler.
The world’s largest software company, the one with its own foundation models and its own coding assistant, had just paid a competitor to put a rival product in front of its workforce.
Six months later, that experiment is being wound down. According to reporting in Windows Central and other outlets following The Verge’s original scoop, Microsoft is cancelling most direct Claude Code licences inside its Experiences and Devices group, the division that builds Windows, Microsoft 365, Outlook, Teams and Surface.
Advertisement
Affected engineers have been told to migrate to GitHub Copilot CLI by 30 June, the last day of Microsoft’s fiscal year. The official reason is toolchain unification. The unofficial reason is in the calendar.
The 💜 of EU tech
The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!
The Claude pullback is the most credible signal yet that the unit economics of enterprise AI coding do not, at current token prices, work. Not because the tools are bad. The opposite: they are good enough that engineers use them constantly, and the constant use is what breaks the maths.
Advertisement
The clearest evidence is at Uber, which is not Microsoft and does not have Microsoft’s financial cushion. Praveen Neppalli Naga, Uber’s chief technology officer, told The Information in April that the company had burned through its entire planned 2026 AI coding budget in four months.
By March, Naga’s own figures had Claude Code use jumping from 32 per cent to 84 per cent of his roughly 5,000-engineer organisation. Individual engineers were spending between $500 and $2,000 a month on tokens. Around 70 per cent of code committed at Uber now originates with AI, and on the order of one in ten live backend updates is shipped by an agent with no human in the loop.
“I’m back to the drawing board,” Naga said, “because the budget I thought I would need is blown away already.”
That sentence is the whole story in miniature. The forecast was wrong because the variable being forecast, token consumption, behaves nothing like the licences and seats that finance teams know how to model. A traditional enterprise software deal is denominated in users.
Advertisement
A token-priced deal is denominated in how much the model has to think. Agentic coding makes the model think a lot. Sessions run for hours, spawn parallel threads and generate volumes of context that bear no resemblance to the autocomplete interactions that shaped the original pricing structure.
We have been tracking this fracture for months. In November, GitHub paused new Copilot Pro and Pro+ sign-ups because the agentic workloads of paying customers were generating costs that exceeded their monthly plan price.
Cost structures built for lightweight assistance, the company conceded, no longer held.
This is not an Uber problem or a Microsoft problem. It is an industry condition. Bryan Catanzaro, vice-president of applied deep learning at Nvidia, told Axios in April that, for his team, the cost of compute is now far beyond the cost of the employees using it.
Advertisement
This is the chip company saying it. Fortune followed in May with reporting that token-based AI tooling, when used heavily, can cost more per task than the human engineer it was supposed to augment.
A 2024 MIT analysis circulated widely in finance circles since then suggests that, on current pricing, AI automation pencils out as cheaper than human labour for roughly a quarter of the jobs people thought it would replace.
Set that against the spend forecasts. Gartner expects worldwide AI spending to reach $2.5 trillion this year, up 69 per cent on 2025.
The same firm now places generative AI squarely in what it calls the trough of disillusionment, predicting in a May press release that 25 per cent of planned 2026 AI budget will slip into 2027 as proofs of concept die in the procurement pipeline.
Advertisement
A separate Gartner read from April found that only 28 per cent of AI infrastructure projects fully deliver against their business case. That is not the curve of a technology going through an awkward adolescence. That is the curve of a market repricing itself.
Microsoft’s retreat lands inside this repricing, and not by accident. There are two ways to read the move. The first is the one Microsoft has briefed: that Copilot CLI is the strategic destination, that engineers will continue to have access to Claude models inside Copilot, and that the company simply wants a product it can shape directly with GitHub. That story is true.
It is also a story that Microsoft could have told at any point in the past six months and chose not to. What changed was not the strategic logic. What changed was the bill.
The second reading is harder to discount. Microsoft is uniquely positioned to know what enterprise-scale Claude usage actually costs, because its own engineers were the heaviest users outside Anthropic’s customer base. Inside Experiences and Devices, Claude Code had become, by several accounts, the preferred tool.
Advertisement
If the maths had improved with scale, this would be the moment Microsoft locked in a multi-year deal at favourable terms. Instead, it is unwinding the experiment in a window that conveniently closes the books on a fiscal year.
When the company with the most leverage in the room walks away from a vendor whose product its own staff prefer, the signal is not about preference.
Whether this constitutes a bubble depends on definitions. Token-level pricing will fall, as it has fallen at roughly a factor of ten every eighteen months for the past three years. The more interesting question is whether per-task token consumption falls faster than per-token cost.
The evidence so far runs the other way. Each generation of agentic system, by design, consumes more tokens per unit of work, because it reasons longer, plans more elaborately and verifies itself against the world.
Advertisement
Anthropic’s own infrastructure team has spoken publicly about reasoning workloads generating order-of-magnitude more compute per query than chat. That is the bet baked into the next twelve months of model releases. It is also the bet that put Uber back at the drawing board.
There is a worked example in TNW’s own coverage. In April, Anthropic banned a popular open-source agentic framework called OpenClaw from running on consumer Claude subscriptions, after discovering that single instances could chew through the equivalent of $1,000 to $5,000 in API costs in a day of autonomous operation. The framework was running on a $200-a-month Max plan.
The economic transfer was so blatant that Anthropic had to write a new clause into its terms of service. Multiply that pattern across a Fortune 500 engineering organisation, and you have the Uber budget memo.
The counterargument is real and worth stating. The cost of a working AI coding agent compared to the cost of an additional senior engineer is, even at current prices, often favourable on a per-feature basis. The productivity uplift is documented; the substitution is happening. What is breaking is not the value proposition.
Advertisement
It is the procurement model. Companies that signed up for a productivity tool are discovering they signed up for a metered utility, and the meter runs when nobody is looking. The fix may be straightforward: capped budgets per engineer, tiered access for high-leverage roles, agent runtime quotas.
Many of the larger buyers are already there. But the implication is that the era of “give every employee a Claude Code seat” is closing, and what replaces it will look more like AWS billing than like Office licences.
That is what Microsoft’s quiet email to its Windows and Surface teams really announces. Not the end of AI coding. Not even the end of Anthropic at Microsoft, given that Claude models will continue to be reachable through Copilot CLI.
It announces the end of the experimental phase, the phase in which the world’s largest software companies were willing to absorb arbitrary token costs in exchange for learning. The learning is done.
Advertisement
What comes next is the harder part. Enterprises will keep buying AI coding tools, because the productivity is real and the competitive pressure is unforgiving. But they will buy them the way they buy electricity, with usage caps, with shadow meters, with a finance team in the room.
Somewhere in a Microsoft conference room earlier this spring, someone looked at a Claude Code invoice and did the arithmetic against a Copilot CLI roadmap, and made a decision.
The same arithmetic is now being done in every CFO’s office that bought into the December 2025 rollout. The retreat will not be loud. It will be a series of fiscal-year-end emails, sent on a deadline nobody noticed until the budget was already gone.
There are a surprising number of experiments an amateur nuclear physicist can perform, from making a Geiger counter to fusing hydrogen atoms in a fusor. One project which we haven’t seen before is a neutron generator, such as the benchtop neutron generator made by [Rapp Instruments] (translated).
This particular generator takes a feedstock of pure deuterium, which it ionizes and accelerates into a titanium target. The first deuterium nuclei to hit the target react with it to form titanium deuteride, immobilizing them until more ions strike them and they undergo nuclear fusion. The fusion reaction mostly forms helium-4, but sometimes forms helium-3 and a free neutron, which is radiated away. The radiated neutrons are slowed down by a block of high-density polyethylene, and a portion of them strike a silver or indium foil wrapped around a Geiger counter tube. The neutrons activate the silver or indium, and the Geiger counter detects the resultant increase in radioactivity.
The design is a linear particle accelerator built inside an evacuated glass tube. It uses two high-voltage power supplies: a 20 kV supply which ionizes the deuterium gas fed into the tube, and a 100 kV supply which accelerates ions emitted from the source into the target. The target itself is surrounded by a cup-shaped electrode to capture secondary electrons emitted during impact. To prevent arcing, the tube needs to be at a very low pressure, reached by extensive use of an oil diffusion pump.
Advertisement
Radioactivity measurements of the silver and indium foils showed that the generator did work; when irradiating the silver foil for five minutes, it generated 175 counts per second after the neutron source was turned off. Plotting the count rate versus time suggested that a mixture of two silver isotopes was being generated, Ag-110 and Ag-108, based on their half-lives. Irradiation of indium produced a similar exponential decay in radiation.
We recommend checking out the rest of the site; it’s a gold mine of projects, such as this mass spectrometer. For more background on neutron generators, we’ve covered their theory and some of the more common varieties.
Homegrown jewellery brand State Property is now sold beside luxury giants worldwide
Singapore is not short of jewellery stores.
From the gleaming counters of Cartier and Tiffany at Ion Orchard to the indie labels that have quietly multiplied over the past decade, the options are vast.
But in 2015, when Afzal Imram and his wife Lin Ruiyin launched State Property, there was a specific gap they felt no one was filling—fine jewellery that led with design rather than gemstones.
10 years on, that conviction has taken them from a home workshop to retail shelves in the US, the Middle East, the UK, Japan and Hong Kong. Their pieces have even been worn by Michelle Obama, Taylor Swift and Robert Downey Jr.—a roster that would be implausible for most independent jewellers anywhere in the world, let alone one founded by two Singaporeans straight out of university with no industry connections.
Advertisement
“We’re as good as anyone else from any other part of the world,” Afzal, 37, told Vulcan Post. “I don’t think we should see ourselves as being held back in some way.”
We spoke with Afzal to find out how the couple built State Property into an internationally recognised indie fine jewellery brand.
A shared obsession with design
State Property’s husband and wife duo, Afzal and Ruiyin./ Image Credit: State Property
Afzal and Ruiyin, 36, met around 2010 through mutual friends. He was studying industrial design at NUS, while she was at Central Saint Martins in London, completing a degree in jewellery design.
The uncanny geographic distance closed when Afzal went on an exchange to Paris, where the pair spent the time collaborating on each other’s projects, each covering what the other lacked.
“What she’s good at, I’m not good at, and what I’m good at, she’s not so good at,” Afzal said. “We managed to kind of help each other out.”
Advertisement
Both were drawn to architecture and geometry—a design sensibility that would later define State Property. Fine jewellery felt like a natural medium for the couple, who wanted to create pieces that could endure for years, even generations.
“With fine jewellery, it’s not as throwaway as fashion accessories. The trends don’t move so fast, and people keep the pieces for much longer. For generations, sometimes,” Afzal explained
After graduating from university in 2014, the duo began taking on commissions and design consulting work from home through a studio they called Proper People—which still operates today and handles State Property’s branding—to fund State Property’s first collection.
A year later, they launched State Property.
Advertisement
Crafting jewellery across continents
(Left): The Holmes Earrings from State Property’s first collection, Substate; (Right): The Edessa Epaulette Enchantress Bracelet from its fifth collection, Arcane. /Image Credit: State Property
Where traditional jewellers start with a beautiful stone and design around it, State Property does the opposite and prioritises the concept of the piece, then sees what material best fits.
Their vision, as the founders put it, is to strike a balance between minimalism and maximalism—inspired by the 20th-century design style Art Deco, which is characterised by modern materials in symmetry.
Every step of the process for each product is drawn and made by hand./ Image Credit: State Property
A new collection can take anywhere from six months to a year to develop, with each treated as a living body of work. New pieces tied to the same theme are introduced over subsequent seasons, while designs that no longer fit are gradually retired.
Today, State Property has seven collections spanning fine jewellery for both men and women.
Prices start at S$750 for a single 14-karat gold and diamond earring, while pieces from the brand’s anniversary Story of Everything collection range from S$4,980 for a toadstool pendant to S$11,650 for the Railroad Diamond Bracelet. Bespoke commissions and special edition pieces can cost significantly more.
The creative process involves collaborative brainstorming by the duo and prototyping, which ranges from Ruiyin’s sketches to creating wax models worn for scale and comfort, before anything goes up for final production.
Advertisement
Eventually, all final pieces are made in 18-karat gold, with production split across a trusted network of workshops in Hong Kong, Sri Lanka, Italy, and elsewhere, each chosen for specific techniques.
Diamonds are cut in India; gemstones are sourced from Sri Lanka and Brazil. For bespoke commissions involving a client’s own stone, the casting, enamelling, and stone-setting may each happen in different locations—with the final setting done by local artisans in Singapore, so the founders maintain direct oversight of the gemstone at every stage.
Every collection balances accessible everyday pieces with more experimental designs that Afzal knows may be slower to sell—but those, he’s found, generate the most excitement and allow the duo to push their creative boundaries.
“That’s what keeps the brand moving the needle.”
Advertisement
Building credibility from scratch
Designing is done in-house, while other parts of production take place all over the world, including Singapore./ Image Credit: State Property
For the first couple of years, growth was slow and almost entirely organic. The two knew how to design, and they knew how to make a brand look good. Everything else, from pricing to strategy and managing a team, had to be learned on the job.
Pricing, in particular, was a sticking point early on. The most expensive pieces from that first 2015 collection went for around S$3,000. Even that felt bold at the time.
“Asking someone to pay S$3,000 for our work? Who wants to spend that here?” Afzal recalled thinking.
The brand’s first real turning point came in 2017, when State Property won the Emerging Accessories Designer of the Year at the Singapore Fashion Awards.
One of the judges was Tina Tan-Leo, a veteran of Singapore’s fashion retail industry, who offered some perspective that neither founder had considered. “She basically said, ‘Guys, your stuff is good to go international. Why aren’t you there?’”
Advertisement
Tina went on to mentor the State Property for six months, reworking pricing, laying out processes for export, and crucially introducing them to the wider world of independent designer jewellers.
“She was the key for us to put State Property on this international trajectory,” Afzal said.
Ruiyin had described the moment as “surreal”: seeing something that started as an idea in a sketchbook worn by one of the world’s most recognisable women on a red carpet.
Advertisement
From there, the list grew to a degree that still seems improbable for a Singapore business: Rihanna, Gigi Hadid and Florence Pugh, among others.
For a label with no family connections in the industry and no established country reputation to lean on, the cultural legitimacy this provided was transformative.
“If you say you’re a jewellery brand from Singapore, people internationally aren’t sure what that means,” Afzal said. “We had to build that trust ourselves.”
In 2021, that trust was further cemented when State Property became the first Singaporean fine jewellery label on global luxury e-commerce Net-A-Porter.
Advertisement
Getting onto the main street
(Left): Jenna Ortega wearing State Property’s Markeli Twin Link Necklace and Darro Diamond Necklace; (Right): Catherine O’hara wearing the Bering Ring./ Image Credit: State Property
For most of State Property’s early years, discovery came through word-of-mouth, Instagram, and appointments only.
The brand participated in Boutique Fairs starting in 2018, and had a consignment presence at Tangs Plaza from 2019 to 2021—both useful for visibility, but limited in reach.
However, the audiences at these places were accustomed to affordably priced goods, not 18-karat gold and diamond fine jewellery.
That changed with brick-and-mortar retail. State Property opened its Armenian Street atelier in early 2021, followed by a boutique at Takashimaya Shopping Centre in Sep 2022.
The foot traffic proved what no amount of digital marketing could replicate for the luxury jewellery brand. Back then, local and international sales were roughly even. Today, Singapore has become the stronger half.
Advertisement
Internationally, State Property now sits in approximately 20 retailers across Hong Kong, Japan, North America, the UK, and the Middle East, including Dover Street Market London, Moda Operandi, Goop, and most recently a trunk show at Bergdorf Goodman in New York.
Cracking those markets wasn’t just about getting stocked, it required genuinely understanding how people in different climates and cultures dress.
“In Singapore, we don’t have seasons that affect how we dress,” Afzal said. “But in temperate climates, there are considerations that don’t come to us naturally—what earrings work when you’re wearing a scarf, whether you’d even wear a bracelet if you need gloves.”
The research meant travelling to each market, observing what people were actually buying and wearing, and only partnering with retailers whose values aligned with the brand’s. “We’re fortunate to have caught the eyes of key retailers like Saks Fifth Avenue and Goop,” Afzal said.
Advertisement
The team today numbers about 10, and has grown to become a family affair: Afzal’s brother handles the finances, his sister-in-law manages retail partnerships, and a childhood friend of Ruiyin’s oversees manufacturing.
Building a premium brand in Singapore
Some bespoke rings designed by Afzal and Ruiyin./ Image Credit: State Property
State Property’s clientele is made up of people typically in their late 30s to 40s, who tend to be culturally driven and already well-versed in luxury jewellery. These are customers who own pieces from Cartier, Van Cleef, and Tiffany, and are looking for something that sits alongside that collection rather than duplicating it.
Most of its business also comes from returning clients. The brand does not chase volume. “We celebrate every sale,” Afzal has said, “because each one means so much to us and goes a long way in helping us build the brand we envision.” It is built, deliberately, on relationships.
State Property’s pearl necklace, pearl earrings and rabbit chair auctioned for charity in 2019./ Image Credit: Indesign Live, State Property
In Jul, State Property will take part in the Singapore International Jewellery Expo, along with Singapore Diamond and Jewellery Week and the World Diamond Congress. The brand will mark the event with new designs and an art collaboration with local mushroom growers, using mycelium to create jewellery displays.
Afzal and Ruiyin are also interested in exploring adjacent categories.
In 2019, State Property was invited to customise a Qeeboo Rabbit Chair for a group installation at design fair Saturday Indesign. They styled it with a pearl necklace and earring, keeping true to their jewellery aesthetic even on a piece of furniture. The chairs were later auctioned to raise funds for the Children’s Cancer Foundation.
Advertisement
Their measure of success has shifted over time.
“When we started, success looked like recognition—getting into the right stores, having press coverage, being seen,” Afzal said. “But now success is when a piece we made becomes part of someone’s life, when we become the go-to brand for a client’s jewellery needs, or when clients come to us to commemorate deeply personal moments.”
For founders looking to build something premium from Singapore, Afzal advised not to shortchange any part of it.
“The brand has to be built holistically—from how it’s presented online all the way to your after-sales service,” he said. “There is nowhere within that where you can drop the ball.”
The FBI is warning about the Kali365 phishing-as-a-service platform (PhaaS) that is used to hijack Microsoft 365 accounts by abusing OAuth device code authentication to steal session tokens and bypass multi-factor authentication (MFA).
According to the FBI PSA, Kali365 first emerged in April 2026 and is distributed via Telegram channels for cybercriminals seeking an easier way to compromise Microsoft 365 accounts without stealing passwords or intercepting MFA codes.
This authentication method was created to allow devices with limited input capabilities, such as smart TVs, conference room systems, streaming devices, printers, and IoT devices, to authenticate via another device using a short code at Microsoft’s device code login portal, http://microsoft.com/devicelogin.
Advertisement
Device code authentication form Source: BleepingComputer
In February, BleepingComputer reported that extortion gangs, including the ShinyHunters cybercrime group, were targeting Microsoft Entra accounts via device-code and voice phishing.
In these attacks, threat actors initiate the device authorization process themselves to generate a code, then trick targets into entering it on Microsoft’s login page via phishing and social engineering.
Once the victim enters the code and completes MFA, Microsoft issues an OAuth access token that grants the threat actor full access to their account without requiring them to solve any MFA challenges.
The threat actors now have full access to all applications the user normally has access to via their single-sign-on account, including Microsoft 365, Salesforce, or any other cloud SaaS platforms, which are then used to steal data.
The FBI warns that Kali365 gives even low-skilled attackers access to advanced phishing capabilities, including AI-generated phishing lures, automated campaign templates, real-time victim-tracking dashboards, and token-capture functionality.
Advertisement
Security researchers at Arctic Wolf reported on Kali365 activity in April after observing a widespread campaign targeting organizations worldwide.
The researchers said that the campaigns primarily targeted Microsoft 365 environments using phishing emails that directed victims to Microsoft’s device code login portal, where they unknowingly authorized attackers to access their accounts.
The researchers said the resulting attacks gave the hackers access to their mailboxes, where they created malicious inbox rules designed to hide their activity.
In some of the attacks, attackers also registered new devices in victims’ Microsoft environments, further extending their access to the breached network.
Advertisement
Arctic Wolf found that Kali365 operates as a business, with admins who manage product development, resellers who promote the service to other threat actors, and affiliates who conduct phishing attacks.
The researchers say the platform offers two separate attack modes, with the first being device code phishing and the second being an adversary-in-the-middle (AitM) mode named “Cookie Link.”
Cookie Link proxies victims through attacker-controlled infrastructure that captures authenticated browser sessions, session cookies, and tokens after targets log in and solves MFA challenges.
The FBI recommends companies restrict or completely block device code authentication flows using Conditional Access policies where possible, audit existing device code usage, and block authentication transfer policies that allow authentication sessions to move between devices.
Advertisement
The agency also urged impacted organizations to report incidents to the Internet Crime Complaint Center and preserve phishing emails, suspicious login information, and unauthorized device registrations.
Device code phishing has seen widespread adoption in 2026, with other threat actors and platforms now using it as part of their phishing campaigns and attacks.
This adoption includes the EvilTokens PhaaS and Tycoon2FA, which are also using it to compromise Microsoft 365 and Entra accounts.
Automated pentesting tools deliver real value, but they were built to answer one question: can an attacker move through the network? They were not built to test whether your controls block threats, your detection rules fire, or your cloud configs hold.
This guide covers the 6 surfaces you actually need to validate.
The strategy outlines plans to present Ireland as a global leader in engineering by broadening the talent pool and accelerating learning velocity.
Engineering Skillnet and business representative body Ibec have unveiled a new strategic plan, ‘Engineering a Skills-First Future 2026-2029’, which aims to address industry skill issues.
The plan was developed in collaboration with industry leaders, learners and the Engineering Skillnet steering committee to build on a robust approach to identifying talent gaps and mapping emerging technologies to ensure Ireland’s engineering workforce remains agile, productive and globally competitive.
The strategy outlines plans to present Ireland as a global leader in engineering by broadening the talent pool and accelerating learning velocity, according to its creators.
Advertisement
According to research underpinning the strategy, 84pc of engineering employers are experiencing challenges in the identification of skills. Meanwhile, 54pc of all staff were found to be in need of major skilling or upskilling.
To combat these challenges, the strategy found that there needs to be a fundamental shift from a traditional focus on job roles and fixed titles towards a more human-centric, skills-first approach, as a means of further developing the workforce.
Commenting on the research and strategy, Pauline O’Flanagan, the director of Engineering Industries Ireland, noted a critical alignment with national competitiveness.
“In 2026, global competitiveness is defined not by hardware or capital investment alone, but by the skills agility of our workforce to operate and optimise increasingly complex, interconnected systems,” she said.
Advertisement
“To remain a global leader, Ireland must look beyond traditional job classifications and embrace a skills-first approach.”
Fiona Fennell, the network manager of Engineering Skillnet, added, “Work is changing at a rapid pace, with jobs fragmenting and many skills now expiring quickly. Ireland has a uniquely multi-skilled workforce, but how we identify and develop talent needs to evolve.
“We believe that a ‘skills-first’ blueprint is essential to secure Ireland’s economic future. This approach moves away from viewing talent purely through titles or credentials, prioritising instead the balance of confidence and competence needed to perform work well.
“Given shifting geopolitics and demographics, this strategic plan represents a profound commitment to future-proofing skills mobility, starting right from the factory floor.”
Advertisement
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
Tech industry layoffs may be worse at large tech companies than the rest of the IT industry. The New York Times argues those layoffs have now shifted the culture at Big Tech companies, after interviewing more than two dozen of their workers. “Cooperation and collegiality are on the wane; chumminess between employees and managers has cooled as mutual suspicion pervades their relationships; and a throbbing economic anxiety infects almost every conversation.
“Perhaps no site on the internet reflects this transformation more vividly than Blind, where users can post in private channels restricted to employees of a single company, or public channels visible to anyone…”
Since 2022, large tech companies have collectively laid off more than 150,000 workers, unraveling what many tech workers once perceived as a guarantee of affluence and employability. The threat of being replaced by artificial intelligence has loomed over those who remain. This year alone, Amazon has indicated that it is laying off more than 15,000 workers, Block 4,000, Meta 8,000 and Oracle an estimated 30,000… By most measures, the sentiments that Blind tracks have taken a turn for the worse. During the nearly four years before tech companies began major layoffs in the fall of 2022, Meta and Microsoft employees posted about career success — topics like how to maximize their salary or win promotions — more than four times as often as they posted about job insecurity, according to Blind. Since then, the ratios have lurched in the opposite direction: Meta and Microsoft employees have posted about job insecurity roughly 1.5 times as often as they post about success…
The shift has had practical effects. A Meta employee said in an interview that some workers on her team now used less vacation time and that, in a break with custom, people frequently checked on their projects while on vacation. They increasingly worry about getting a poor performance review or losing their job if they aren’t constantly available. The employee, who declined to be identified for fear of retribution, said she and many of her colleagues frequently checked Blind because it could be comforting to see how many other Meta workers shared their anxieties. Employees at several companies said in interviews that their morale was further undermined by the feeling that the layoffs were abrupt and arbitrary, and executed with little empathy.
Advertisement
Several tech workers said it was the scarcity of information about possible layoffs that raised their cortisol levels and made it difficult to focus on their jobs. They often fill the vacuum by turning to Blind, which, in addition to posts by workers, features a “tech layoff tracker” that lists both layoff rumors and those it has confirmed. “I was on Blind five days a week,” said Faith Wilkins El, a software engineer who was laid off from Oracle in late March, after more than four years at the company. Wilkins El, who is part of the Oracle Workers Collective, a group seeking better severance agreements with the company, said navigating Blind was sometimes stressful because it was hard to know what was true or false. (Blind says it has a security team to weed out bad actors, like those who may try to register under fake email addresses.) Still, she found it more helpful than not because the layoffs came as less of a shock after she spent time on the site. “I was trying to get prepared mentally,” she said.
Blind is capitalizing on the increased interest with new products. It plans to unveil a service called Blind AI, which will allow employers to simulate their workers’ reactions to certain changes, like a stricter in-office mandate. And it is close to releasing a feature to alert users that layoffs are imminent.
The approach, still in development, aims to produce processors that reach transistor densities comparable to a 1.4-nanometer process by 2031. That level of density is widely viewed as the next milestone for leading chipmakers such as Intel, TSMC, and Samsung Electronics, all of which rely on extreme ultraviolet lithography systems… Read Entire Article Source link
Monday morning, the Roman Catholic Church made its biggest foray yet into the discourse on artificial intelligence and the role it should play in human life as the technology develops.
In the first encyclical of his papacy, titled Magnifica humanitas (Latin for “magnificent humanity”), Pope Leo XIV argued that AI is not intrinsically immoral, but that its adoption needed to be slowed in order to build moral guardrails, to establish better social safety nets for those displaced by economic and labor disruptions, and to create democratic processes that will ensure the public remains in control of these developments, rather than a small subset of tech oligarchs. The document also contended that the “intelligence” in artificial intelligence was a misnomer: Intelligence is something only human persons possess, and technology will never be human.
The first encyclical, or official teaching letter, of Leo XIV’s papacy, dropped Monday.
It centers the uniqueness of humanity, the dignity of work, and the challenges that artificial intelligence poses to the world order and humans’ relationships with each other and God.
The Catholic Church has a long tradition of reasserting authority in the modern era, starting with the current pope’s namesake, Leo XIII, who confronted the rise of the Industrial Revolution and changing global economies.
There are deeper spiritual and material reasons the pope, and the church, are so concerned with AI now.
Encyclicals are official teaching documents of the Catholic Church: letters issued by popes to bishops after consultation with theologians, historians, and experts on pressing matters that affect humanity or the church, with the expectation that all people, faithful or secular, can learn from them and help shape their consciences and lives. Magnifica humanitas is Leo’s first encyclical since becoming pope last year, and its release now underscores the focus the new pope is putting on AI and technology. Notably, Leo also used the occasion to make a historic formal apology for the Church’s previous defense and justification of slavery — a reminder that the Catholic Church has not always been on the right side of social ills.
Though popes are traditionally not present during the release of these documents, which first began in the 18th century, Leo XIV was in attendance at its presentation, and delivered his own comments — something Vatican observers indicated reflected his desire to make sure the Church’s stance was properly understood. The Chicago-born pontiff spoke in English and was joined by AI experts and industry leaders, including Anthropic co-founder Chris Olah, who consulted on the document.
So how should the public process and think about this new document? It’s helpful to first understand the context in which the Church is speaking up about this at all.
Advertisement
Why the Catholic Church cares so much about AI’s development
Magnifica humanitas is dropping more than a week after Pope Leo XIV actually signed it on May 15. The timing matters.
That day marked 135 years since the release of Rerum Novarum, the seminal work of Pope Leo XIII, the current pope’s namesake, who was leading the church during the late stages of the Industrial Revolution. As they are today, the faithful, and the clergy, were facing a rapidly changing world. And the Church, the world’s leading moral authority at the time, had yet to establish its place in it.
That 19th-century document made philosophical arguments about the relationship between labor and capital, warning about the perils of communism. But it also redefined the church’s relationship to the modern world, with the papacy reasserting itself as both a source of power and a moral authority in an era of rapid change. The encyclical set a template for how a 2,000-year-old institution could still remain relevant in a modern age.
Advertisement
In presenting this encyclical, Pope Leo XIV made this parallel clear. He sees the rise of artificial intelligence as the defining global challenge of the day, and of his pontificate: “Like the earlier Leo, I feel entrusted to look upon another huge transformation with eyes of faith, with lucidity of reason, with openness to mystery and with cries of the poor and the earth resounding in my heart,” the pope said while presenting his encyclical.
The newer encyclical builds off his predecessor’s tradition, and the various arguments popes have made about the importance of preserving the dignity of the human person and valuing modern technology only so long as it benefits everyone, not just its creators or the rich.
In 2015, Pope Francis, for example, wrote about “the technocratic paradigm” that has taken root in modern capitalism: the sense that technological progress is unstoppable, that it will demand unlimited concessions from nature and from people, and that the world had no choice but to submit to change.
“Leo is concerned that we don’t just submit to inevitability on questions of AI, but ask critical questions and push back in ways that are necessary before it’s too late to push back, before damage is done that can’t easily be undone,” Dan Rober, an associate professor of Catholic Studies at Sacred Heart University, told me.
Advertisement
That role of questioning, pausing, and coming to consensus has defined the Catholic Church’s leadership and operation for the last two papacies: the notion of synodality, or teaching and making decisions based on consensus. Before AI and the technologists who have created it become the sole determinants of how politics, the economy, and society operates, the Church is asserting itself as a counterweight — even as it includes some of those leaders in the process.
“Pope Leo is trying to clearly walk in those footsteps, and I think he’s very concerned, as are a lot of people, about the possible implications, particularly for job markets and for people’s lifestyles being sustainable day to day with the rise of AI systems that may render a certain significant amount of jobs able to be automated very rapidly,” Rober said.
This kind of reflection has become standard procedure for the Vatican. Since at least the turn of the century, the Church has found itself increasingly weighing in on the crises of the day, albeit often a bit too late.
As the Catholic writer Christopher Hale has noted, “Francis took up the climate fight with Laudato Si in 2015, after decades of scientific consensus had been ignored. Benedict XVI took up the global economic order with Caritas in Veritate in 2009, after the financial system had already collapsed. Both documents arrived in the long shadow of the crises they addressed.”
Advertisement
In Magnifica humanitas, Pope Leo XIV may be seeking to intervene early in the development or takeover of a new technology this time, and show that the Church wants to both work with Silicon Valley and assert itself as a powerful defender of modern values, as it has done in its defense of the liberal international order and aspects of humanism, like human autonomy and reason.
In the background, there’s also a more sci-fi element: the notion that AI could end up coming between the Church and the people — serving as a filter or layer between regular people and God, and perhaps even usurping the role of the Church itself. The Catholic Church, famously, is concerned with the proper interpretation of scripture, Biblical truths, ethics, and God. Bloody wars waged and hard-fought reformations turned on this central question of who and how one can commune with God. Now, AI enters as another middleman.
“That’s closely related to the question of people using AI as a therapist,” Rober told me. “You could see a way in which AI becomes its own kind of religion, and certainly the way a lot of the Silicon Valley founders talk about it, it does have religious overtones to it. You listen to the Google founders talk about the singularity, and that sounds a lot like religion.”
It’s in this context that this document, and its specific teachings, lands.
Advertisement
What the Church is teaching about AI
Magnifica Humanitas is not the Vatican’s first examination of the role of AI in modern life.
Just last year, in the twilight of Pope Francis’s pontificate, the Vatican released a teaching note, Antiqua et nova, that laid the groundwork for Leo’s encyclical. That 2025 document established that the Church is not opposed to the development of AI: technological progress and scientific discoveries are part of the natural way that humans are meant to honor God, his creation of humanity in his image, and the natural outpouring of God’s gift of reason and rationality.
But it also established a distinction between human intelligence and machines that analyze data and perform processes. It insisted that artificial intelligence, like all technology, should serve humanity, not the other way around. And it emphasized the risk that new technology poses to the ability to, right to, and dignity of work, especially for the least well off in society.
Advertisement
In the encyclical, Leo uses the biblical parable of the Tower of Babel — a warning about human hubris — to make this case: “We must, then, avoid the ‘Babel syndrome,’ namely the idolatry of profit that sacrifices the weak,” he writes. “The risk of dehumanization — of building a future that excludes God and reduces the other to a means — is an ancient and ever-new temptation that today takes on a technical guise.”
This builds off of a long tradition of focusing on not just the dignity of work and workers, but also more recent concerns that modern capitalism facilitates a “throwaway culture” that views people and things as, at best, cogs in the service of a greater machine.
“He wants to talk about the idea that our humanity is meaningful in and of itself and that work is part of that, even if AI systems are able to allow for more leisure and even if something like universal basic income were to be made available, people need to find to have work of some kind to have meaning in their lives,” Rober said.
The encyclical’s teachings can be broken up into three broad categories: regulations on how AI is developed and how individuals adopt it, the responses required to handle the economic effects of AI, and limits on AI’s usage in war.
Advertisement
The practical recommendations and concerns Leo outlines include:
The need for a “more active” democratic process for people to decide how AI will develop, “that is capable of slowing things down when everything is accelerating, and of protecting the opportunities for communities still to be able to participate and ask questions.”
Regulation of how companies collect and use personal data, which “should not be treated as something to be sold off or entrusted to a select few.”
Better education of adults, teachers, and young people for using AI in their daily lives, specifically to avoid sexual exploitation, blackmail, grooming, and disinformation.
Environmental regulations, given AI infrastructure’s impact on the natural world.
A duty for governments to protect access to, and the dignity of, work, to provide job training and professional help to workers affected by AI disruptions, and to redistribute the wealth and value created by AI to those it displaces.
Flexibility from labor unions and organizations to “be open to new types of employment and the corresponding needs of workers, in order to represent and defend them.”
New rules of war and accountability for AI usage in combat, given that “just war” theory is being made obsolete by the growing automation of warfare. “When a decision to strike becomes automated or opaque, the risk of abdicating responsibility increases,” the encyclical says. “For this reason, the chain of responsibility must be identifiable and verifiable; those who design, train, authorize and employ technology must be held accountable for their decisions.”
A new international compact on how AI should be used to avoid “the technological arms race and ensure robust protection for civilians and the infrastructures necessary for their survival.”
From an eagle-eye view, the document is fairly wonky and detailed: concerned with very practical matters and specific recommendations that could have come from academia, or a secular background — underscoring just how much Leo may hope it can provide guidance for leaders and individuals, as opposed to remaining siloed to the intellectual class. So as this technology continues to develop, the pope and the church want to help shape it. They want the faithful to be reminded that whatever AI offers is not reality, not personhood, and not God. It is a tool that should not dominate or determine the lives of its users. And it should not replace the role of the Church in teaching morality and ethics.
For the greater secular world, the Church wants to remind the public that they should have a say in how AI shapes their world; they should not allow business and tech leaders to define the terms of existence through their machines; and that they have a powerful ally in the Roman Catholic Church in the effort to preserve human dignity in the face of unprecedented technological change.
I’ve been covering deals since 2013, and I still think Memorial Day sales are worth checking out. There are so many made-up shopping holidays these days that it can be difficult to ascertain when items are actually cheaper than usual. Now is one of those times. To help you out, I’ve scoured through our buying guides full of hand-tested items to find gadgets and gizmos that are discounted—and worth buying. If you’re in the market for something we’ve tested, keep checking in through May 26 to see if you can snag a deal. We’ll update this post again between now and then.
Updated May 25: I’ve added 5 new deals, removed expired discounts, and checked for accuracy throughout.
WIRED Featured Deals:
Advertisement
Birdfy Lite Smart Bird Feeder (Lifetime AI, No Solar) for $120 ($110 off)
Netvue
Birdfy Lite Smart Bird Feeder
A few different combinations of our favorite smart bird feeder are on sale. You can choose a version with or without solar charging. There’s also the option of buying a model that comes with lifetime AI bird identification; otherwise, if you decide you want it later, it’ll cost you $5 per month, but it isn’t required for enjoyment. This bird feeder is equipped with a 1080p camera that can capture decent footage of whoever comes to snack. It has a wide field of vision as well, and the feeder is easy to clean and refill.
Last week, the US government announced $2 billion in investments in quantum computing companies, allocating $100 million each to a range of startups in exchange for equity in the companies. Those could be make-or-break investments for many companies that are likely years away from a product that could see widespread use. But a member of the US Congress is now arguing that those deals are illegal, as Congress did not allocate the money for this purpose—instead, it was meant to support public research in semiconductors.
But the biggest chunk of money would go to a company that likely wouldn’t exist if it weren’t for the government’s backing. Anderon will be set up with a billion dollars each from IBM and the government and will inherit personnel and IP from IBM. It will serve as a foundry for fabricating quantum processing units and will contract its services out to IBM and any other company that wants access to cutting-edge hardware.
Is any of this legal?
Zoe Lofgren (D–Calif.), the ranking member of the House Science, Space, and Technology Committee, made it clear that she is not happy with how the government is using its money to support this technology.
“This announcement is illegal and troubling on so many levels,” Lofgren said one day after the announcement, pointing out that the money being used for the deal comes from the CHIPS and Science Act, which was passed during the Biden administration and was allocated “specifically for microelectronics R&D, with a focus on semiconductor technology.”
Advertisement
That technology overlaps only partially, at best, with what’s used in quantum processors. In addition, Lofgren says the money was allocated to foster public/private research partnerships, which these deals most decidedly are not. Finally, she noted that the largest sum of money will go to IBM, and she suggested that a former IBM executive (Dario Gil, current Under Secretary for Science at the Department of Energy) was involved in the negotiations that led to this deal.
None of this, she noted, means that quantum processing technology is a bad investment or that any of these companies are unworthy of support. She just argues that doing so would require Congress to allocate the money to do so.
Google I/O 2026 kicked off this week, and to no one’s surprise, it was almost completely about AI.
To start with, Google is rolling out its “intelligent, AI-powered Search box” globally. Instead of just autocompleting your text, it uses AI to anticipate your intent and help you formulate questions. You will be able to use images, video files and entire Chrome tabs as direct search inputs. Its AI Mode, now powered by the new Gemini 3.5 Flash, will still live alongside it, for your follow-up questions/corrections.
Google also introduced Gemini Spark. Running in the cloud, Spark is a digital assistant that can autonomously monitor credit card statements for hidden subscriptions, track updates from your kid’s school emails, or pull notes together into a Google Doc. It can even interact with third-party apps like OpenTable and Instacart to complete tasks—though it promises to ask for your confirmation before making any final purchases or sending emails. Google may have figured out a way to take the fun out of shopping.
Advertisement
There were glimpses of future hardware: Google and Samsung teased a collaboration with Gentle Monster and Warby Parker and offered the first look at two models of Android XR smart glasses. The glasses will allow you to chat with Gemini, get real-time audio translation in the speaker’s voice, translate real-world text in your line of sight and snap photos on the go.
Naturally, with all these new features, Google adjusted its AI subscription tiers, adding a new $ 100-per-month mid-range (hah!) tier. The AI Ultra Plan will offer five times higher usage limits than the standard $20 Pro plan, as well as priority access to Google’s Antigravity coding tool and 20TB of cloud storage.
Meanwhile, down from its original $250 price tag, Google’s top-tier Ultra plan features 20 times higher usage limits and exclusive access to Project Genie — Google’s experimental research preview that lets you build interactive 3D worlds using real-world Google Street View imagery.
We dig further into a few more of the biggest announcements below.
Advertisement
— Mat Smith
Xreal’s Project Aura smartglasses are a maximalist take on Android XR
Karissa Bell for Engadget
Instead of subtle, normal-looking glasses with some smart features, Xreal’s smartglasses aim for an immersive AR experience, with a focus on entertainment. That approach is very much the same as the company’s Android XR-powered Project Aura, but based on Karissa Bell’s time with these new smartglasses, they may offer more than just another wearable screen.
Compared with the low-key (arguably lower-spec) audio-only smartglasses from Warby Parker and Gentle Monsters, Xreal’s Project Aura packs in three cameras, a 70-degree field of view and hand-gesture navigation. The main gesture is a pinching motion that will feel pretty familiar to anyone who has used other AR setups, such as Apple’s Vision Pro.
Advertisement
Speaking of which, Project Aura comes with a tethered puck, just like Apple’s wearable display. It comes with a trackpad and a fingerprint sensor, although neither worked on the demo model we tested. The company hasn’t revealed pricing details yet (expect those alongside the formal launch later this year), but they’re likely to exceed Xreal’s One Pro glasses, which cost $650. Continue reading.
Google’s Gemini Omni can generate ‘anything from any input,’ starting with video
Gemini Omni is Google’s new gen-AI model that can “create anything from any input”. Gemini Omni Flash is rolling out now to the Gemini app, Google Flow and YouTube Shorts. It’s apparently the next evolution of Nano Banana and, presumably, its current video generator, Veo 3.1. It lets you “combine images, audio, video and text as input and generate high-quality videos grounded in Gemini’s real-world knowledge.”
Advertisement
Apparently, Omni also better understands physical forces such as gravity, kinetic energy and fluid dynamics, making scenes more realistic.
Spotify is adding more AI gunk for podcasts and audiobooks
Announced on its investor day, Spotify is expanding a feature that enables users to generate “personal podcasts.” It announced this tool last month, with the option to use AI agents such as OpenClaw and Claude Code to whip up synthetic audio. The company says it’s adding a feature that lets users generate personal podcasts directly within Spotify. It notes that after you enter a prompt, it will generate audio that draws on factors such as your Spotify taste profile and world knowledge. You can also feed in text, PDFs and links to give the tool more context for what you’d like to hear about. Spotify says eligible Premium users in the US will get access to personal podcasts next month. The Roman Empire, Final Fantasy VII Remake and Liverpool FC, please. In under 30 seconds. Go. Continue reading.
Meta’s latest attempt to copy its rivals is a Reddit clone
Meta
Meta has launched a new app called Forum. Social media consultant and analyst Matt Navarra spotted Forum in the App Store, where it is described as “a dedicated space for the conversations that matter most to you,” created specifically for Facebook Groups. As Navarra notes, Meta seems to be pushing the idea that the app can help users get “real answers” from “real people,” making it sound very similar to Reddit, offering an escape from the atrocious genAI search results we’ve all been suffering from recently. (Did we mention Google I/O was this week?) Continue reading.
You must be logged in to post a comment Login