The Matter protocol for smart homes works great in theory (and diagrams). The practice, with Ikea’s products, is very different. CC-licensed photo by Smart Home Perfected on Flickr.
You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.
A selection of 9 links for you. Unenlightened. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.
Meta is shutting down Horizon Worlds in the metaverse • CNBC
Lola Murti:
»
In a community blog, Meta announced that the Horizon Worlds app will be taken off the Quest store at the end of March, and fully removed from VR on June 15. After that date, it will only be available on a standalone mobile app.
“We are separating the two platforms so each can grow with greater focus, and the Horizon Worlds platform will become a mobile-only experience,” the company said in announcing the change.
The shift for Horizon Worlds, which was once a central part of the company’s push into virtual reality, comes weeks after Meta cut over 1,000 employees from Reality Labs, the unit responsible for the metaverse.
The January cuts in Reality Labs also hit studios that were working on VR titles, including Ouro Interactive, an in-house studio that debuted in 2023 to build first-party content for Horizon Worlds.
When Meta changed its name from Facebook in October 2021 to cement the pivot to the metaverse, CEO Mark Zuckerberg called it “the next frontier.”
“Our hope is that within the next decade, the metaverse will reach a billion people, host hundreds of billions of dollars of digital commerce, and support jobs for millions of creators and developers,” Zuckerberg wrote at the time when announcing the change.
«
How much has Reality Labs lost in Zuckerberg’s quixotic pursuit of something that wasn’t there (so absolutely perfectly quixotic)? Nearly $80bn between Q4 2020 and Q4 2025. And how much revenue? According to ChatGPT (asked to list the Reality Labs revenues by quarter and sum them; the numbers looked believable), $11.4bn.
It takes a special talent to lose $8 for every $1 you take in on a project that you then dump. Take a bow, Meta.
unique link to this extract
Friendship, on demand • The Atlantic
Julie Beck:
»
The robots befriended us remarkably fast.
Over the past year or two, AI has become not just a utilitarian tool but a technology that many people are turning to for connection and emotional support. One survey last year found that 16% of American adults had used AI for companionship, and a quarter of adults under 30 had. Social AI use seems to be growing rapidly around the world, according to several recent reports on the state of artificial intelligence. Raffaele Ciriello, who studies emerging technologies at the University of Sydney, told me that he once assumed AI companions would remain “niche”; he has been “surprised by how quickly that took over.”
Some people use apps that are explicitly made for companionship; they let you design a virtual character’s personality, appearance, and backstory. Popular such apps include Replika, which reportedly had 40 million users as of late 2025, up from 10 million in 2023, and Character.AI, which reported 20 million monthly users in 2025. Other people seek emotional support from all-purpose AI tools such as OpenAI’s ChatGPT and Anthropic’s Claude, even though they aren’t explicitly intended for social use. OpenAI’s own data show that use of ChatGPT was pretty evenly split between work and personal cases in 2024, but by 2025, 73% of conversations with ChatGPT were personal, not for work. (The Atlantic entered a corporate partnership with OpenAI in 2024.)
This is a major transformation, a sudden and dramatic shift in which millions of people are seeking companionship from machines that they formerly could have gotten only from other humans. Yet in some ways, AI companionship is a logical destination for the current direction of human friendship.
«
To anyone who knows the history of Eliza, the computer program created by Joseph Weizenbaum at MIT in the 1960s, this is no surprise. As Weizenbaum observed, seeing people become completely absorbed by the pretend psychologist represented by Eliza, “What I had not realized is that extremely short exposures to a relatively simple computer program could induce powerful delusional thinking in quite normal people.” (Quote via John Naughton.)
It would be good if we didn’t have to rediscover things we already know every decade because people have no recollection of history. (Gift link to the article.)
unique link to this extract
Ikea tried to build a smart home for everyone — here’s why it’s not working yet • The Verge
Jennifer Pattison Tuohy:
»
Matter is an interoperability standard, but interoperability with Matter devices is still largely elusive. Rather than being a plug-and-play solution for manufacturers — make a Matter device, and it will just work with any platform — there remains a huge onus on each manufacturer to ensure its devices work properly with each platform before release. Which is basically the same problem they had before Matter launched.
Only now manufacturers have a playbook to follow that supposedly makes their devices work with everyone — easy, right? Apparently not. My theory is that it’s how the platforms interact with the devices that is causing many of these problems — something manufacturers have no control over.
This was somewhat implied by Thread Group, the organization that runs the Thread protocol, when I asked for comment on the issues users were seeing with Ikea’s Matter devices. “A seamless onboarding experience relies on orchestrating multiple components and layers within the smart home ecosystem, including the mobile app, application protocol, network protocol, platform software, and hardware design,” Ann Olivo, VP of marketing for Thread Group, told me via email. “While Thread provides a robust and secure foundation at the network layer, optimizing the end-to-end experience requires ongoing collaboration across all these interconnected components.”
That’s not to say Thread is blameless here. The protocol is frustratingly obtuse, and there are still too few troubleshooting solutions.
«
So: Thread is a protocol that runs over the Matter standard. Or rather, doesn’t. Six years ago I bought a ton of Ikea smart home products (mainly light bulbs) which use the Zigbee communication protocol: they are absolutely rock solid and integrate, via an Ikea hub, with Apple’s Home.
A couple of the light bulbs failed; I bought some of the newer Ikea Matter-standard bulbs. They integrate directly with Apple Home, but they’re unreliable (ignore timers, don’t respond) and they make Apple Home forget all the Zigbee Ikea products. That’s untenable. So I bought some of the (now discontinued) Ikea Zigbee bulbs on eBay, and it’s all fixed.
Matter is broken. Just like USB-C, it’s going to need a few years to mature.
unique link to this extract
Actors, musicians and writers welcome UK U-turn on AI use of copyrighted work • The Guardian
Dan Milmo:
»
Actors, musicians and writers have welcomed the UK government’s decision to backtrack on plans to let AI firms use copyright-protected work without permission.
Technology secretary Liz Kendall said it no longer had a “preferred option” on copyright reform, having previously supported a proposal allowing tech companies to take copyrighted work – unless rights holders opted out of the process.
“We have listened,” said Kendall on Wednesday, “we have engaged extensively with creatives, AI firms, industry bodies, unions, academics and AI adopters, and that engagement has shaped our approach. This is why we can confirm today that the government no longer has a preferred option.”
The proposal had triggered a backlash from Elton John, who called the government “absolute losers” over the plans. Dua Lipa, Abba’s Björn Ulvaeus, the actor Julianne Moore and the Radiohead singer Thom Yorke are among thousands of artists who voiced their concerns over the potential legal overhaul.
Creative industry organisations welcomed the new government stance. Equity, the actors’ trade union, said the move was “recognition that selling out the UK’s creative industries to benefit US tech companies would’ve been an act of national self-sabotage”.
UK Music, a trade body representing the UK music industry including the Musicians’ Union, said it was “delighted” but urged the government to rule out the proposal altogether. The Society of Authors said the announcement was a “hard-won” moment for writers and creators, while the News Media Association – whose members include the Guardian – said giving away the UK’s “goldmine” of creative content was not the way to drive economic growth.
Intellectual property has become a key battleground in the development of AI because the technology requires vast amounts of data, including copyright-protected work taken from the open web, to develop tools such as chatbots and image generators.
«
There’s a certain Canute-ish feel to all this. The government no longer having a “preferred option” doesn’t mean it’s going to find a way to force AI companies to not use copyrighted material. It just means it’s shrugging its shoulders. Which means the people creating content will lose out.
unique link to this extract
Why lab coats are white • Asimov Press
Donna Vatnick:
»
In the 1960s, David Chambers, a researcher at Deakin University in Australia, instructed teachers to give children a blank sheet of paper and ask them to draw a scientist. Chambers repeated this experiment many times over eleven years, collecting more than 4,800 drawings. The results were surprisingly consistent: white lab coat, glasses, beakers, mysterious machinery, someone saying “eureka!” The study has since been repeated dozens of times. While some details have changed, with beakers replaced by rockets, microscopes by vaccines, or men by women (sometimes), the scientist always wears a white lab coat.
The white lab coat, however, only came to symbolize scientists in the 20th century. Before that, cartoonists satirized chemists by portraying their craft as sorcery whose practitioners wore long dark robes, and painters drew naturalists in waistcoats and breeches against backdrops of plants and landscapes. It was really surgery, more than any other scientific discipline, that gave us the white laboratory coat. Today, scientists don a variety of multicolored, specialized protective equipment to suit the needs of their field, but the fact that children still inextricably link white lab coats to “scientists” says everything about how a simple garment came to exemplify a profession’s public image.
To understand how the white lab coat arose, we have to go back to Victorian England to examine not the scientists, but rather the fashion of that time.
In mid-19th-century England, so-called “gentlemen of science” dressed in dark frock coats. In his 1885 portrait, Louis Pasteur stood in his laboratory, rabies sample in hand, in a black frock coat, waistcoat, and black cravat. Charles Darwin, whose personal home was his “laboratory,” sported a similar style. John Snow, the epidemiologist best known for tracing the source of London’s cholera epidemic, inspected the Broad Street pump while dressed like a banker, in a multipiece suit and tie.
«
Like all the Asimov work, this is a great read.
unique link to this extract
Microsoft weighs legal action over $50bn Amazon-OpenAI cloud deal • Financial Times
Stephen Morris, George Hammond and Madhumita Murgia:
»
Microsoft is weighing legal action against Amazon and OpenAI over a $50bn deal that could breach its exclusive cloud partnership with the ChatGPT maker, setting up a clash between the Big Tech rivals.
The dispute centres on whether Amazon Web Services can offer OpenAI’s new commercial product, known as Frontier, without violating a longstanding agreement that requires all access to the startup’s models to be routed through Microsoft’s Azure cloud platform.
The arrangement is highly lucrative for Microsoft, with OpenAI’s products helping drive Azure revenues to record highs.
Amazon and OpenAI say they are building a system that works around the contract. Microsoft executives dispute this, saying the approach is not feasible and would violate the spirit, if not the letter, of their agreement, according to people familiar with the discussions.
Ahead of Frontier’s launch, the companies were still in talks to resolve the dispute without litigation, they added.
«
Microsoft is essentially saying that it owns OpenAI, and everything it produces, and gets first dibs and power of refusal over what it does. That’s quite a veto power.
unique link to this extract
Elon Musk’s xAI sued for turning three girls’ real photos into AI CSAM • Ars Technica
Ashley Belanger:
»
A tip from an anonymous Discord user led cops to find what may be the first confirmed Grok-generated child sexual abuse materials (CSAM) that Elon Musk’s xAI can’t easily dismiss as nonexistent.
As recently as January, Musk denied that Grok generated any CSAM during a scandal in which xAI refused to update filters to block the chatbot from nudifying images of real people.
At the height of the controversy, researchers from the Center for Countering Digital Hate estimated that Grok generated approximately three million sexualized images, of which about 23,000 images depicted apparent children. Rather than fix Grok, xAI limited access to the system to paying subscribers. That kept the most shocking outputs from circulating on X, but the worst of it was not posted there, Wired reported.
Instead, it was generated on Grok Imagine. Digging into the standalone app, a researcher in January found that a little less than 10% of about 800 Imagine outputs reviewed appeared to include CSAM. In an X post following that revelation, Musk continued rejecting the evidence and insisted that he was “not aware of any naked underage images generated by Grok,” emphasizing that he’d seen “literally zero.”
However, Musk may now be forced to finally confront Grok’s CSAM problem after a Discord user reached out to a victim, prompting law enforcement to get involved.
In a proposed class-action lawsuit filed Monday, three young girls from Tennessee and their guardians accused Musk of intentionally designing Grok to “profit off the sexual predation of real people, including children.” They estimated that “at least thousands of minors” were victimized and have asked a US district court for an injunction to finally end Grok’s harmful outputs. They also seek damages, including punitive damages, for all minors harmed.
«
Fuel rations and no air con: south-east Asian nations race to conserve energy • The Guardian
Rebecca Ratcliffe and Guill Ramos:
»
In Thailand, news anchors ditched their jackets on air as the government called on the public to reduce their use of air conditioning to save energy. In the Philippines, many government workers are now operating on a four-day week. In Vietnam, officials have urged employers to allow staff to work from home.
Across south-east Asia, governments are scrambling to find ways to conserve energy and shield the public from soaring costs as war in the Middle East causes what the International Energy Agency has described as the largest supply disruption in the history of the global oil market.
Asia, which relies heavily on imported energy, much of which passes through the strait of Hormuz, is acutely affected by the crisis. In the Philippines, which depends on the Gulf for 90% of its oil requirements, the government is introducing cash handouts for drivers of public transport vehicles, and has told government agencies to cut electricity and fuel use by between 10 and 20%.
The price increases are already causing misery for many. Elmer Carrascal, 58, who drives a jeepney, vehicles that make up the backbone of the country’s transport system, said his income has fallen dramatically, by more than half, since the war.
“Before, I spent 700 pesos [£8.80] on diesel and took home around 1,000 pesos per day. Now I only earn 400 pesos. It’s not even enough for food,” said Carrascal, who has plied the roads of Mandaluyong City, in the capital region, for 35 years.
“How far can your 400 go? Rice alone is 65 per kg.”
On Tuesday, the Philippine senate granted the president, Ferdinand Marcos Jr, emergency powers to temporarily suspend or reduce excise taxes on oil, but has warned it is unclear how long the crisis will last.
“We are victims of a war that is not of our choosing,” Marcos said earlier this month. “But we control how we will protect the Filipino,” he said, as he announced a series of energy-saving measures.
«
Trump’s impulsive attack on Iran (or do we blame Israel?) is having effects far beyond the Middle East.
unique link to this extract
On the Wired renaissance • The Future, Now and Then
Dave Karpf:
»
There’s been some murmuring among the tech barons about buying Wired magazine and turning it back into the magazine they remember from their youth. Anduril co-founder Trae Stephens wrote an X.com diatribe about this a couple weeks ago, where he complained that Wired still has good journalists, but they are focused on the wrong stories. “The building still has good people. The question is who owns the building.” (Anduril, you might recall, is the company building killer AI drones for the military. It’s run by weird creep Palmer Luckey. So I can see why its executives would rather Wired go back to writing VC glow-up profiles.)
YCombinator CEO Garry Tan (who likes to tell local politicians to “die slow”) loves the idea. In a blog post that was probably written by ChatGPT, Tan argues that “great tech journalism (…) doesn’t just cover technology. It creates the intellectual conditions for investment and progress. It puts a story in [an investor’s] head that sits there for nine years and then fires when the moment is right.”
Tan is a clown, but he isn’t wrong that the magazine has decisively moved away from its 90s techno-optimist roots.
Today’s New York Times features a profile of Katie Drummond, the magazine’s Editor in Chief. It’s a very good profile, well worth reading. Wired has been nominated for a National Magazine Award and, if I didn’t hate Polymarket and Kalshi, I would bet they’ll win it. The Katie Drummond era has been Wired at its finest.And, given my documented history of reading entirely too many Wired magazines, I think that’s a nice excuse to write a bit about how the magazine has changed with the times.
«
This is a good companion to yesterday’s article interviewing Katie Drummond, who has edited Wired for the past three years.
unique link to this extract
| • Why do social networks drive us a little mad? • Why does angry content seem to dominate what we see? • How much of a role do algorithms play in affecting what we see and do online? • What can we do about it? • Did Facebook have any inkling of what was coming in Myanmar in 2016? Read Social Warming, my latest book, and find answers – and more. |
Errata, corrigenda and ai no corrida: none notified
The Overspill: when there's more that I want to say 