These crackers weren’t intended to be tasty at all when first invented. CC-licensed photo by Andrea Goh 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.

It’s Friday, so there’s another post due at the Social Warming Substack at about 0845 UK time.

A selection of 10 links for you. No Ritz? I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Trump administration’s MAHA report on children’s health filled with flawed references, including some studies that don’t exist • CNN

Brenda Goodman, Jacqueline Howard and Betsy Klein:

»

The first report from the Trump administration’s Make America Healthy Again (MAHA) Commission, released last week, appears to be rife with errors, including some studies that don’t exist.

Touted by US Health and Human Services Secretary Robert F. Kennedy Jr. as a milestone, the report lays out the government’s priorities for addressing chronic health problems in children, which it ascribes to poor diet, lack of exercise, stress, overprescribing of drugs and exposure to environmental chemicals.

The sweeping 78-page document was produced in a little more than three months after it was ordered by President Donald Trump. It contained 522 references to studies, government reports and news articles. But some of these references were wrong or don’t appear to exist. In other cases, studies in the report were misrepresented, according to the researchers who conducted them.

The citation errors were first reported by NOTUS, a nonprofit, nonpartisan news site created by former Politico Publisher Robert Allbritton.

An updated version of the report was posted online Thursday with some changes to the text and the works cited.

NOTUS identified seven studies in the original report that didn’t appear to exist.

One of them was credited to Dr. Katherine Keyes, an epidemiologist at Columbia University, and titled “Changes in mental health and substance use among US adolescents during the COVID-19 pandemic.” Keyes said she does study anxiety in teens, but she didn’t author any study with that title.

«

Absolutely typical of the Trump administration not to take any care in devision such a report. But how do you quote studies that don’t exist? How don’t you check that? Of course they used a chatbot to compile it – demanding that it write all the content – but it’s ridiculous. Slovenly and stupid.
unique link to this extract

Why were Graham Crackers invented? The reason is truly bizarre • All That’s Interesting

John Kuroski:

»

The graham cracker needs no introduction. [English readers: OK, it does. It’s basically a sort of ryebread cracker like you’d put cheese on.] Today, these delightfully crunchy snacks are a common fixture at campfire gatherings, serving as a key ingredient of s’mores and helping to deliver the delicious combination of chocolate and toasted marshmallow to one’s mouth.

Few people stop to think about why graham crackers were invented, though, and most would likely assume that their creation was a typical culinary experiment. But this assumption is far from the truth. The original purpose of the graham cracker was, in fact, to repress sexual desires in all who ate them.

…Sylvester Graham believed that people should eat like Adam and Eve did, theorizing that indulging in rich, flavorful foods — think meats, fats, and anything with spices — could ignite carnal desires and lead individuals down a path of moral decay. In his view, a bland diet was the key to maintaining sexual purity and restraint. He advocated for a vegetarian lifestyle centered around whole wheats, which he believed would curb lustful thoughts.

The philosophy came to be known as the “Graham Diet,” which included an early, now-unrecognizable version of the graham cracker.

While promoting his “Graham Diet,” Graham spoke out against the additives that were being put in processed bread at the time to make it whiter and the preservatives that kept the bread from spoiling. He promoted homemade bread — emphasizing the importance of using whole wheat flour — in his 1837 book A Treatise on Bread and Bread-Making.
He also promoted virtually flavorless food, thinking that adding spices or sugars to foods increased people’s sexual urges, and suggested bland products to avoid stimulating the body in an “unholy” way.

«

Graham was an, er, inspiration for John Kellogg who did indeed invent those breakfast cereals you’ve heard of.
unique link to this extract

Trump administration pulls the plug on the bird flu vaccine • NPR

Rob Stein:

»

The federal government announced Wednesday that it is cancelling a contract to develop a vaccine to protect people against flu viruses that could cause pandemics, including the bird flu virus that’s been spreading among dairy cows in the U.S., citing concerns about the safety of the mRNA technology being used.

The Department of Health and Human Services said it is terminating a $766m contract with the vaccine company Moderna to develop an mRNA vaccine to protect people against flu strains with pandemic potential, including the H5N1 bird flu virus that’s been raising fears.

“After a rigorous review, we concluded that continued investment in Moderna’s H5N1 mRNA vaccine was not scientifically or ethically justifiable,” HHS Communications Director Andrew Nixon said in a statement.

“This is not simply about efficacy — it’s about safety, integrity, and trust. The reality is that mRNA technology remains under-tested, and we are not going to spend taxpayer dollars repeating the mistakes of the last administration, which concealed legitimate safety concerns from the public,” Nixon said.

…Jennifer Nuzzo, the director of Brown University’s Pandemic Center, said the decision was “disappointing, but unsurprising given the politically-motivated, evidence-free rhetoric that tries to paint mRNA vaccines as being dangerous.”

“While there are other means of making flu vaccines in a pandemic, they are slower and some rely on eggs, which may be in short supply,” Nuzzo added in an email.

«

The line about eggs is thermonuclear-grade hot. Anyway, looks like the watching brief needs to keep watching.
unique link to this extract

The myth of automated learning • New Cartographies

Nicholas Carr thinks the threat of AI in education is because it discourages the process of learning:

»

You can speculate all you want about computers eventually attaining human-level intelligence or even “superintelligence,” but for the time being AI is doing something that has a long precedent in human affairs. Whether it’s engaged in research or summarization, writing words or creating charts, it is replacing human labour with machine labour.

Thanks to human-factors researchers and the mountain of evidence they’ve compiled on the consequences of automation for workers, we know that one of three things happens when people use a machine to automate a task they would otherwise have done themselves:

• Their skill in the activity grows
• Their skill in the activity atrophies
• Their skill in the activity never develops.

Which scenario plays out hinges on the level of mastery a person brings to the job. If a worker has already mastered the activity being automated, the machine can become an aid to further skill development. It takes over a routine but time-consuming task, allowing the person to tackle and master harder challenges. In the hands of an experienced mathematician, for instance, a slide rule or a calculator becomes an intelligence amplifier.

If, however, the maintenance of the skill in question requires frequent practice — as is the case with most manual skills and many skills requiring a combination of manual and mental dexterity — then automation can threaten the talent of even a master practitioner. We see this in aviation. When skilled pilots become so dependent on autopilot systems that they rarely practice manual flying, they suffer what researchers term “skill fade.” They lose situational awareness, and their reactions slow. They get rusty.

…Unlike carpentry or calculus, learning is not a skill that can be “mastered.” It’s true that the more research you do, the better you’ll get at doing research, and the more papers you write, the better you’ll get at writing papers, but the pedagogical value of a writing assignment doesn’t lie in the tangible product of the work — the paper that gets handed in at the assignment’s end. It lies in the work itself: the critical reading of source materials, the synthesis of evidence and ideas, the formulation of a thesis and an argument, and the expression of thought in a coherent piece of writing.

«

unique link to this extract

Apple is reportedly going to rename all of its operating systems • The Verge

Jay Peters:

»

Apple is going to change how it names its next set of major operating systems, Bloomberg reports. Instead of just notching up the version number, Apple will instead mark them by year.

However, the numbers will apparently align with the year after the one the update is actually released in, similar to cars. That means that the next big iOS update will be iOS 26 instead of iOS 19. Bloomberg says that other upcoming name changes include iPadOS 26, macOS 26, watchOS 26, tvOS 26, and visionOS 26.

Apple plans to officially announce the change at its Worldwide Developers Conference (WWDC), which kicks off on June 9th. The idea is to make Apple’s software version numbers more consistent. Right now, the version numbers are all over the place; the current naming schemes include things like iOS 18, watchOS 12, and visionOS 2.

«

Makes sense, I suppose? Except calling the software released this year by next year’s number is.. oh well. People who don’t get to hear about this are going to be confused as hell when they see their Watch go from 12 to 26, and wonder what on earth happened to 13-25.
unique link to this extract

Dieselgate pollution killed 16,000 people in UK, study estimates • The Guardian

Damian Carrington:

»

The excess pollution emitted as a result of the Dieselgate scandal has killed about 16,000 people in the UK and caused 30,000 cases of asthma in children, according to a new analysis. A further 6,000 premature deaths will occur in coming years without action, the researchers said.

The Dieselgate scandal erupted in 2015 when diesel cars were found to be emitting far more toxic air pollution on the roads than when they passed regulatory tests, due to the use of illegal “defeat devices”.

Large fines and compulsory recalls of vehicles to remove or disable the defeat devices took place in the US. But experts say the UK and most EU countries have lagged far behind, leading to devastating impacts on health, and urge immediate action. Many millions of highly polluting diesel vehicles remain on the roads in the UK and EU.

The analysis estimated the impact of only the excess pollution released due to the defeat devices, not the total emissions from the cars. In the UK, these excess emissions had led to 800,000 days of sick leave and a total economic burden due to deaths and poor health of £96bn by 2024.

Across the UK and EU combined, the fallout from Dieselgate has included about 124,000 early deaths and economic damage of €760bn (£637bn), the study estimated. Without action, a further 81,000 premature deaths and €430bn are projected by 2040, by which time most Dieselgate vehicles will no longer be in use.

“Our calculations reveal the widespread and devastating health impacts of excessive diesel emissions – thousands of lives cut short, countless children developing asthma, and an immense burden of chronic illness,” said Dr Jamie Kelly, at the Centre for Research on Energy and Clean Air (Crea), which conducted the research for environmental law group ClientEarth.

«

Before you scoff at the numbers, or the idea that vehicle exhausts could have such effects, remember that lead used to be a component of petrol, leading to dramatic levels of air pollution and, in turn, blood lead levels which dropped dramatically when it was banned. Dieselgate is egregious because the effects of particulates were known when the “defeat devices” were built.
unique link to this extract

German court confirms civil liability for corporate climate harms • Climate Rights International

»

In a landmark ruling advancing efforts to hold major polluters accountable for transnational climate-related harms, on May 28 a German court concluded that a corporation can be held liable under civil law for its proportional contribution to global climate change, Climate Rights International said today. 

Filed in 2015, the case against German energy giant RWE AG challenged the corporation to pay for its proportional share of adaptation costs needed to protect the Andean city of Huaraz, Peru, from a flood from a glacial lake exacerbated by global warming. RWE AG, one of Europe’s largest emitters, is estimated to be responsible for approximately 0.47% of global historical global greenhouse gas emissions.

“This groundbreaking ruling confirms that corporate emitters can no longer hide behind borders, politics, or scale to escape responsibility,” said Lotte Leicht, Advocacy Director at Climate Rights International. “The court’s message is clear: major carbon polluters can be held legally responsible for their role in driving the climate crisis and the resulting human rights and economic harms. If the reasoning of this decision is adopted by other courts, it could lay the foundation for ending the era of impunity for fossil fuel giants and other big greenhouse gas emitters.”

«

There’s an article about the Peruvian farmer (yes really) who brought this lawsuit at Climate Change News.

After this week’s collapse of a glacier in Switzerland, this ruling feels like the coming together of strands of possibility.
unique link to this extract

U.S. pauses exports of airplane and semiconductor technology to China • The New York Times

Ana Swanson:

»

The Trump administration has suspended some sales to China of critical U.S. technologies, including those related to jet engines, semiconductors and certain chemicals and machinery. The move is a response to China’s recent restrictions on exports of critical minerals to the United States, a decision by Beijing that has threatened to cripple U.S. company supply chains, according to two people familiar with the matter.

The new limits are pushing the world’s largest economies a step closer toward supply chain warfare, as Washington and Beijing try to flex their power over essential economic components in an attempt to gain the upper hand in an intensifying trade conflict.

A growing standoff over critical supply chains could have significant implications for companies that depend on foreign technologies, including makers of airplanes, robots, cars and semiconductors.

It could also complicate efforts to negotiate an end to a trade fight over the administration’s tariff policies. On May 12, negotiators from the two countries agreed to reduce the punishing tariffs they have imposed on each other for 90 days while negotiators sought a longer-term resolution.

Scott Bessent, the Treasury secretary, said at the time that “the consensus from both delegations is that neither side wanted a decoupling.” Yet the administration continues to target China with punitive measures. Secretary of State Marco Rubio also announced on Wednesday that the United States would “aggressively revoke” visas for Chinese students who study in critical fields or who have connections to the Chinese Communist Party.

«

Given that a US court declared Trump’s tariffs illegal (see p25 of the judgment), all they have now is reciprocal export wars. China might win that one too.
unique link to this extract

AI-powered political fanfiction racks up views online • Semafor

David Weigel and Kadia Goba:

»

Rep. Jasmine Crockett, D-Texas, kept herself busy on Tuesday. She confronted Elon Musk in a closed-door meeting, got Supreme Court justices John Roberts and Clarence Thomas arrested, ended the career of Georgia Rep. Marjorie Taylor Greene, and humiliated Colorado Rep. Lauren Boebert.

Crockett’s busy — and fictional — day unfolded on “Mr. Noah’s Stories,” a YouTube channel that inserts the names of public figures into lengthy fanfiction videos. It’s one of many accounts, across social media sites, that serves the appetite for dramatic, partisan stories by making them up.

With little fanfare — maybe “with jaw clenched,” as these overwritten stories often put it — Crockett’s gotten a few of the fakes taken down, and ignored the rest.

“Clearly the algorithm loves my name, so people do stuff with my name,” Crockett told Semafor. “I’ve just told people at this point, if it’s an AI-generated voice, it’s probably a lie.”

Hard to avoid on TikTok, YouTube or Facebook, AI-generated slop has become a barometer of political fame, just as it has of pop culture celebrity. Cabinet secretaries, members of Congress, and presidential family members regularly appear in fake stories with tidy narratives.

They fly under the radar. They sometimes get more views than real-world political reporting that’s not built for the algorithms.

And they’ve become irritating, and worrying, to some members of Congress.

«

There’s no way this ends well. The problem is, how do you police it? AI systems already fail to tell truth from fiction, just as humans do.
unique link to this extract

Sam Altman and Jony Ive will force A.I. into your life • The New Yorker

Kyle Chayka:

»

It will reportedly not attempt to supplant the other technologies you depend on: according to the Wall Street Journal, Altman described it as a kind of third device, meant to work within an ecosystem that includes your laptop and smartphone. But it will effectively be a self-surveillance machine that creates a technological scrim for your personal reality.

The involvement of Ive invites inevitable comparisons with the iPhone, but this is not necessarily a compliment; to many of us, an iPhone of A.I. sounds less like a utopian promise than like a threat that A.I. will soon become ubiquitous and unavoidable. Smartphones have already absorbed us in our screens, creating personalized information bubbles; omnipresent A.I. will only intensify that atomization while being more automated, more inscrutable, and more inescapable.

The video claims that more information about the new product will be shared next year, which would mean that we’re currently in the Palm Pilot stage of A.I.—with the iPhone-like invention looming around the corner, poised to obliterate the competition.

But there are vast logistical hurdles to achieving this optimistic timeline for ubiquitous consumer A.I. More than a billion people in the world own iPhones. Some research estimates that generating a typical e-mail using A.I. consumes a bottle’s worth of water to siphon heat away from the data centers’ servers to separate cooling towers. This means that, if we all started using our personal A.I. machines dozens of times a day, as we do our iPhones, the environmental toll of our personal technology would skyrocket—imagine something like turning every car on the road into a diesel truck.

«

Not to worry – it’s not as if the world has a problem with how it generates its energy, after all.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

On Wall Street, they have a new food-based acronym for Trump’s trade policies – and he doesn’t like it at all. CC-licensed photo by Tim Reckmann 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. Mexico isn’t paying for this either. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

There’s more film and television for you to watch than ever before; good luck finding it • Salon.com

Coleman Spilde:

»

Not long ago, if you wanted to see a movie at your local theatre, you could pick up the paper and look at the daily advertisements to find showtimes. That lasted even through the brief Moviefone boom, where users could dial a number to get automated showtimes, which was still a low-tech way to find out what was playing nearby. But then came the internet, and every piece of useful technology exploded, with the shrapnel extending to the far recesses of the digital Rubicon.

Some theatres featured digital ticketing, others did not, causing services like Fandango to prioritize movie houses with digital tickets, allowing the platform to earn a service fee on top of a customer’s purchase. For smaller theatres — whether they were independently owned or just a single or few-screen theatre in a corporate chain — the digital revolution spelled disaster. The multiplex boom at the turn of the millennium saw theatres that couldn’t adapt to new, digital revenue streams crushed under the financial pressure. In 2000 alone, Carmike Cinemas, Edwards Theatres and General Cinema all filed for bankruptcy. 

The ripples of this roaring tech age are still being felt. There is now such a proliferation of digital media that navigating between apps, emails, videos and texts just to find something to watch feels like a massive undertaking. The window between a movie trailer being released and the film itself being released into theatres has tightened, and the window between a film’s theatrical release and its streaming release is sometimes even shorter. With the expansion of streaming, there are more titles at our disposal than ever. But when it comes to finding something you want to watch, good luck fighting against the algorithms, shoddy user interfaces and glitchy applications. In trying to make media-watching “easier,” tech has slowed the process to a maddening plod. 

“If I ever try scrolling through endless titles on a specific streamer’s landing page, it can feel like a Herculean task,” says Cameron Nudleman, an avid film-lover based in Austin, Texas. Nudleman prefers to use his Amazon Fire Stick’s voice search feature to look for specific titles he’s interested in, but that experience is its own can of worms. “I chose a Fire Stick because, as an existing Prime customer, it felt like the easiest and cheapest way to host all of my streaming services in one place. While the experience isn’t entirely awful, I would rate it a six at best.”

«

Well, I do know that the new Mission: Impossible film is out now, and that I’m going to watch it on Thursday evening in IMAX. So some of it works. (Side note: Americans spell it “theater” – I changed all the spellings to the English “theatre” – but then use “theatrical” as the adjective, rather than “theaterical”. Howcome?)
unique link to this extract

This film was made entirely with Runway AI and Google’s Veo. It nearly broke us • WSJ

Joanna Stern and Jarrard Cole:

»

Welcome to the premiere of “My Robot & Me.” Please silence your phones, chew your popcorn quietly and remember: Every visual you’re about to see was generated with AI. Most of the audio too, except my voice.

Some of it’s totally wild. You won’t believe that no real cameras were used. Some of it, you’ll laugh at, because it’s clearly not real. I promise you, I did not have facial reconstructive surgery between scenes. But enough from me. Hopefully by now you’ve watched the film [in the story], complete with its behind-the-scenes look. Just come back—we’ve got some lessons to share.

Yes, we. To make this film, I teamed up with Jarrard Cole, a real human producer. We met over a decade ago here at WSJ, experimenting with helmet cams and new video formats like VR. These days, he’s become obsessed with AI video tools. So I challenged him to make a totally AI video. How hard could it be?

Very hard. Over a thousand clips, days of work and who knows how much data-center computing power later, we ended up with a three-minute film—about my life with a new kind of efficiency robot. Even if you don’t care about camera angles or storyboards, you might care about what this says about using AI in any job.

…Think you can paste in a script and out pops a Netflix hit? Cute. Every shot of ours was the result of many prompts and generation attempts. And to keep characters and sets consistent from scene to scene, Jarrard invented a whole production pipeline.

The quick version: We used AI image generator Midjourney to generate our sets (a suburban neighborhood, a newsroom) and to design our robot star. Then we used photos of real me to create AI me. We uploaded those to Runway or Veo, where we wrote prompts. Here’s a short one: Low angle shot: Joanna does push-ups at a brisk pace, maintaining a straight line from head to heel. The robot stands above, monitoring and guiding.

That careful, specific wording made a huge difference. As a filmmaker, Jarrard could break down scenes beat by beat, specifying camera angles, lighting styles and movement. That nail-biter ending? Every shot was carefully described to build suspense. And it still took us over 1,000 clips. Some were complete disasters, with anatomical nightmares and random new characters. Even in “good” scenes, my face looks different in almost every shot.

«

Free to read, and to view. The question is always: will it be like this all the time, or will it get easier and faster?
unique link to this extract

Five ways to stop cheating with AI • The Honest Broker

Ted Gioia:

»

How would the Oxford [University] system kill AI?

Once again, where do I begin?

There were so many oddities in Oxford education. Medical students complained to me that they were forced to draw every organ in the human body. I came here to be a doctor, not a bloody artist.

When they griped to their teachers, they were given the usual response: this is how we’ve always done things.

I knew a woman who wanted to study modern drama, but she was forced to decipher handwriting from 13th century manuscripts as preparatory training.

This is how we’ve always done things.

Americans who studied modern history were dismayed to learn that the modern world at Oxford begins in the year 284 A.D. But I guess that makes sense when you consider that Oxford was founded two centuries before the rise of the Aztec Empire.

My experience was less extreme. But every aspect of it was impervious to automation and digitization—let alone AI (which didn’t exist back then). When I got my exam results from the college, the grades were handwritten in ancient Greek characters.

If implemented today, the Oxford system would totally elminate AI cheating—in these five ways…

«

I’ll let you consider them yourself, although: some of the solutions (such as personal tutor sessions) really wouldn’t work at universities bigger than Oxford or Cambridge.
unique link to this extract

Trump’s not happy about Wall Street’s name for his tariff flip-flops • POLITICO

Ali Bianco:

»

Wall Street has a new shorthand about President Donald Trump — and he’s not happy about it.

Traders have reportedly come up with the acronym TACO, which stands for “Trump always chickens out,” to take advantage of the trade environment created by the president’s habit of threatening to impose tariffs on countries, and then backing off at the last moment.

He bristled when asked about it Wednesday in an Oval Office press conference.

“Don’t ever say what you say, that’s a nasty question,” Trump told a journalist who asked for his response to the acronym. “To me that’s the nastiest question.”

Trump rejected the idea that his reversals on tariffs amounted to him backing down, saying that usually receives a different critique.

“They will say oh he was chicken, he was chicken, that’s so unbelievable,” Trump said about the EU tariff extension, adding, “I usually have the opposite problem — they say you’re too tough!”

«

This is extremely funny, and extremely telling. First, that it’s being used on Wall Street (it was coined by the Financial Times, observing how Wall Street traded around his tariff threats). Second, that his reaction is to call such a perfect description “nasty”. It is absolutely going to stick, this one. Trump Always Chickens Out. He’s stuck with that now. (There’s also a writeup of what the “TACO Trade” is at Business Insider.)
unique link to this extract

Getty CEO: Stability AI lawsuit doesn’t cover industry mass theft • CNBC

Ryan Browne:

»

Getty Images is spending millions of dollars to take on a “world of rhetoric” through its Stability AI suit, the photo licensing company’s boss Craig Peters says.

Peters told CNBC in an interview that both Stability AI — the U.K.-based startup best known for its text-to-image model Stable Diffusion — and other AI labs are stealing copyright-protected material to train their AI models for commercial gain.

These firms, he said, are taking copyrighted material to develop their powerful AI models under the guise of innovation and then “just turning those services right back on existing commercial markets.”

“That’s disruption under the notion of ‘move fast and break things,’ and we believe that’s unfair competition,” Peters added. “We’re not against competition. There’s constant new competition coming in all the time from new technologies or just new companies. But that’s just unfair competition, that’s theft.” Peters said the AI industry is making the argument that if developers are forced to pay for access to creative works, this will “kill innovation.”

“We’re battling a world of rhetoric,” the CEO told CNBC.

Getty is suing Stability AI in both the U.K. and U.S. over allegations that the company copied 12 million images without permission or compensation “to benefit Stability AI’s commercial interests and to the detriment of the content creators.” Stability AI has contested the legal action, saying it doesn’t consider Getty’s claims to have merit.

…Part of the reason Getty Images is pursuing legal action specifically against Stability AI and not other firms is because such legal pursuits are “extraordinarily expensive,” Peters added. “Even for a company like Getty Images, we can’t pursue all the infringements that happen in one week.”

“We can’t pursue it because the courts are just prohibitively expensive,” he said. “We are spending millions and millions of dollars in one court case.”

«

unique link to this extract

Generative AI’s adoption puzzle • Benedict Evans

Evans points to the data showing that chatbots’ “weekly active” users are about far bigger than their “daily active” users in the US:

»

Something between 5% and 15% of [all] people are finding a use for this every day, but at least twice as many people are familiar with it, and know how it works, and know how to use it… and yet only find it useful once a week. Again, you didn’t have to buy a thousand dollar device, so you’re not committed – but if this is THE THING – why do most people shrug?

It’s also worth noting that when social media was a new thing we quickly realised that ‘weekly active’ and ‘monthly active’ numbers were bullshit. If someone was only using WhatsApp or Instagram once a month, it really wasn’t working for them. DAU is everything. Sam Altman knows this – he was trying to build a social media app at the time, and yet the traction number he always gives is, well, ‘weekly active users’. That’s a big number (the latest is 1bn globally)… but then, why is he giving us that number instead of DAUs? If you’re only using ChatGPT once a week, is it really working for you?

It might be that this gap is just a matter of time: the models will mature, people will break old habits and form new ones, and the WAUs and the ‘looked once six months ago’ cohorts will convert to DAUS. The S-Curve will curve upwards. But it might also be that we’re in the part of the S-Curve that came before the iPhone: the latent possibility is there, it all seems to be working and we can all see it’s going to be huge, but we need something to crystallise. So, this might be a time problem, or it might be a product problem. Paging Jony Ive?

It might also be that the chatbot as chatbot is the right UX only for some people and some use-cases, and most people will experience this technology as features and capabilities wrapped inside other things. I don’t think we can know that.

«

Put that together with Getty’s spending on legal fees and you have to wonder whether this is all really paying back.
unique link to this extract

‘Catastrophe’: one person missing after Swiss glacier collapse destroys village • The Guardian

»

A huge section of a glacier in the Swiss Alps has broken off, causing a deluge of ice, mud and rock to bury most of a village evacuated earlier this month due to the risk of a rockslide.

Drone footage broadcast by Swiss national broadcaster SRF showed a vast plain of mud and soil completely covering part of the village of Blatten, the river running through it and the wooded sides of the surrounding valley.

“What I can tell you at the moment is that about 90% of the village is covered or destroyed, so it’s a major catastrophe that has happened here in Blatten,” Stephane Ganzer, the head of security in the southern Valais region, told local TV channel Canal9.

The regional government said in a statement that a large chunk of the Birch Glacier located above the village had broken off, causing the landslide which as well as covering the village had also buried the nearby Lonza riverbed, raising the possibility of dammed water flows.

“There’s a risk that the situation could get worse,” Ganzer said.

“We’ve lost our village,” Matthias Bellwald, the mayor of Blatten told a press conference after the slide. “The village is under rubble. We will rebuild.”

Matthias Ebener, a spokesperson for local authorities in the southwestern canton of Valais, said one person was missing, adding: “An unbelievable amount of material thundered down into the valley.”

«

There have been lots of videos showing this happening – and you know the glacier will never come back. Global warming is a ratchet.
unique link to this extract

Congestion Charge to hike 20% with revised EV discounts in shake-up plans • Fleet News

Natalie Middleton:

»

Transport for London (TfL) is planning a major shake-up of the Congestion Charge with a steep rise in the daily charge and revised rules for electric cars and vans from 2026.

The proposals, set out in a new public consultation and effective from January 2026, include a 20% hike in the daily charge from £15 to £18; marking the first increase in five years.

TfL is also consulting on some proposed changes to the Mayor’s road user charging guidance, which would allow the Congestion Charge to increase each year in line with Tube fares, inflation plus 1%, or a lower amount. These increases would only apply to the Congestion Charge and not the ULEZ, and are intended to ensure that public transport does not become proportionately more expensive than driving in central London.

There’s also big changes for electric vehicles –  including proposals for a new Cleaner Vehicle Discount (CVD), replacing the scheme that ends on 25 December 2025. Currently, electric vehicles are entitled to a 100% discount under the CVD, subject to registration with TfL.

Transport for London has previously said that ending the CVD from 25 December 2025 would maintain the effectiveness of the Congestion Charge, but has now proposed new rules, set out in two phases.

From 2 January 2026, electric vans, HGVs, light quadricycles and heavy quadricycles registered for Auto Pay will get a 50% discount, while electric cars registered for Auto Pay will get 25%.

From 4 March 2030, these fall to 25% and 12.5% respectively.

«

The shift for EVs from 100% – basically, you can drive into the congestion charge zone and pay nothing – to 50% or 25% is going to be quite dramatic. It will take a lot of the benefit of buying EVs, which are a bigger capital layout, away for those living in London.
unique link to this extract

A fungus that can ‘eat you from the inside out’ could spread as the world heats up • CNN

Laura Paddison:

»

Infection-causing fungi responsible for millions of deaths a year will spread significantly to new regions as the planet heats up, new research predicts — and the world is not prepared.

Fungi are absolutely everywhere. A vast kingdom of organisms, from mold to mushrooms, they grow in environments such as soil, compost and water. They play an important role in ecosystems but can have a devastating impact on human health: Fungal infections kill an estimated 2.5 million people a year, and a lack of data means that number could be far higher.

Yet we are still very far from understanding them, especially how these incredibly adaptable organisms will respond to a warming climate.

A team of scientists from Manchester University used computer simulations and forecasts to map the potential future spread of Aspergillus, a common group of fungi found all over the world that can cause aspergillosis, a life-threatening disease primarily affecting the lungs.

They found certain Aspergillus species will expand their range as the climate crisis intensifies, pushing into new parts of North America, Europe, China and Russia. The study, published this month, is currently being peer reviewed.

“Fungi are relatively under-researched compared to viruses and parasites, but these maps show that fungal pathogens will likely impact most areas of the world in the future,” said Norman van Rijn, one of the study’s authors and a climate change and infectious diseases researcher at the University of Manchester.

«

It would be great if people would stop trying to make The Last Of Us happen, or even predicting it.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

It’s only taken 15 years – but WhatsApp finally has an iPad version. CC-licensed photo by HS You 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. Taking the tablets. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

The bad science behind expensive nuclear • Works in Progress Magazine

Alex Chalmers:

»

On 23 May 2025, President Trump signed four executive orders on nuclear power, intended to speed up approvals of and reduce regulatory burdens on new nuclear reactors in America. Buried in one of them was a requirement that the Nuclear Regulatory Commission reconsider its use of ‘Linear No Threshold’ (or LNT). LNT is the hypothesis that the relationship between radiation dose and cancer risk to humans is linear and that there is no truly ‘safe’ level of radiation. It underpins nuclear regulation worldwide and it may be one of the most important rules that almost no one’s ever heard of.

In 2013, GE Hitachi Nuclear Energy, a joint venture between General Electric and Hitachi, applied to build three advanced boiling water reactors in Wales. Fission reactions would boil water into steam, turning a turbine, powering a generator, and producing electricity. This specific design had been employed in four Japanese reactors, which had survived earthquakes of a greater magnitude than have ever hit the UK without posing any threat to workers or the public. 

Even though the reactor had a flawless safety record, the UK’s Office for Nuclear Regulation was not satisfied. Over the course of a four and a half year process, it demanded a series of design changes. These included the installation of expensive, bulky filters on every heating, ventilation, and air conditioning duct in the reactor and turbine building, a new floorplan for the room in the plant’s facility that housed the filtration systems, and an entirely new layout for the facility’s ventilation ducts. The purpose of these changes was to reduce radiation discharges from the filter by 0.0001 millisieverts per year. This is the amount a human ingests when they consume a banana. 

A CT scan hits a patient with ten millisieverts all in one go. Natural background radiation in the UK or US typically exposes people to two or three millisieverts during the course of a year, and exceeds seven millisieverts per year in Iowa and North Dakota and South Dakota. A single flight from New York to London exposes a passenger to 0.04–0.08 millisieverts; 0.0001 millisieverts is equivalent to 1/400 of the upper range of that, or about 72 seconds in the air per year worth of radiation.

«

The article is a thorough, and absorbing (oh, wrong word?) history of how we got where we are in being terrified of radiation. If you don’t know what ALARA stands for, you will; if you already do, you’ll perhaps reconsider its use. But how weird that it’s the Trump regime that should drive a reconsideration.
unique link to this extract

Cops arrest third suspect accused of brutally torturing man for bitcoin riches • Ars Technica

Ashley Belanger:

»

According to [tortured crypto holder Michael] Carturan, he met [accused kidnapper John] Woeltz through a crypto hedge fund in New York, but they quickly had a falling out over money, prompting Carturan to return home to Italy.

But Woeltz apparently convinced Carturan to come back to New York, inviting him to the townhouse, where Carturan alleged he was immediately assaulted by Woeltz and [newly arrested William] Duplessie. Stealing his electronic devices, they hoped to swipe millions in bitcoin, Carturan alleged, by torturing him and threatening to do the same to his family.

As police mount their case against Woeltz and Duplessie, several outlets report that these so-called “wrench attacks”—where bad actors physically attack cryptocurrency holders as brutally as possible to get access to their wallets—are increasingly common globally.

As cryptocurrency holders fear online hacks, a growing trend of storing wallet keys on physical devices that cannot be hacked is likely driving the assaults, experts suggest. As bitcoin’s value has recently hit record highs, it has only further incentivized the thefts. Sometimes criminals locate targets through data breaches targeting cryptocurrency exchanges, the WSJ noted.

Earlier this month, Ars noted at least five crypto-related abductions, some involving severed fingers, over a few months in France, as well as cases where family members were targeted. Some cases appear to be related, The Wall Street Journal reported, including a Florida man suspected of leading a criminal ring intent on robbing cryptocurrency riches through a string of home invasions across multiple states. Some criminals are so bold as to attempt assaults using toy guns, the WSJ noted.

«

They’re being called “wrench attacks” because XKCD pointed out long ago that criminals aren’t going to take the nerd’s approach to figuring out passwords.

And when did that cartoon come out? February 2009.
unique link to this extract

After 15 years, WhatsApp is finally ready for the iPad • The Verge

Jess Weatherbed:

»

Meta now has a dedicated iPad app for WhatsApp, more than 15 years after the messaging service and the first iPad launched (2009 and 2010, respectively). Available to download today via the App Store, WhatsApp for iPad supports many of the same features as its iPhone counterpart, allowing users to join audio and video calls with up to 32 people, use both the rear and front device cameras, and share their screen with other call participants.

The WhatsApp for iPad works with iPadOS features like Stage Manager, Split View, and Slide Over, enabling it to run alongside other applications. That means users can view their messages in a split-screen view while browsing the web or watching videos, making the larger screen more practical for multitasking while using the app, compared to constantly switching away from WhatsApp on smaller mobile devices.

«

I was going to say that this has to be the most “finally” of all the “finallys”, but of course – we don’t yet have Instagram on the iPad.

Well, you have to understand that Meta is a very short-staffed little startup with few programmers to spare.
unique link to this extract

$15m reward offer for information disrupting Chinese nationals supplying technology to Iran’s Islamic Revolutionary Guard Corps • United States Department of State

»

Beginning as early as May 2007, [Emily] Liu and her [three] associates allegedly utilized an array of front companies in the People’s Republic of China (PRC) to send dual-use U.S.-origin electronic components to IRGC-linked companies that could be used in the production of UAVs, ballistic missile systems, and other military end uses. The IRGC and its supporters generate and move millions of dollars around the world by establishing and relying on front companies to procure cutting-edge technology to evade sanctions and trade controls. 

The named individuals allegedly misrepresented the end users of dual-use U.S.-origin electronic components, leading U.S. companies to export goods to PRC-based front companies under the guise that the ultimate destination of these products was China rather than Iran. As a result, a vast amount of dual-use U.S.-origin products with military capabilities have been exported from the United States to IRGC-linked companies Shiraz Electronics Industries (SEI), Rayan Roshd Afzar, and their affiliates, in violation of U.S. sanctions and export control laws and regulations. 

The IRGC and the Ministry of Defense and Armed Forces Logistics (MODAFL), which supervises Iran’s development and production of military armaments, have utilized the U.S.-controlled technology to develop and manufacture arms and weapons systems, including UAVs, that are sold to governments and groups in allied countries such as Russia, Sudan, and Yemen. 

In January 2024, the U.S. Department of Justice charged Liu, Li, Yung, and Chung with various federal crimes related to their involvement in a conspiracy to unlawfully export and smuggle thousands of U.S.-origin electronic components with military applications from the United States to Iran.  

«

Wouldn’t it also be a good idea to be a lot more careful who you export this stuff to if it has military capabilities? Just an idea. The State Department is taking any tips via the Tor browser. Maybe it’ll all be a film some day.
unique link to this extract

China’s Xiaomi commits $6.9bn to in-house chips • CNBC

Arjun Kharpal:

»

Chinese technology giant Xiaomi will invest at least 50bn yuan ($6.9bn) over the next ten years to develop its own chips, CEO Lei Jun said in a Weibo post on Monday.

It’s the latest move by a Chinese firm to double down on home-grown technology amid the ongoing trade war between the U.S. and China that has seen Washington cut off access to some semicondcutors for companies in the world’s second-largest economy.

The 50bn yuan investment starts from 2025, a Xiaomi spokesperson confirmed. Lei added the company is looking to make a splash at an event on Thursday, when it takes the wraps off the Xring O1 — a so-called system-on-chip that will power Xiaomi’s upcoming smartphone.

The Xring O1 is based on a 3 nanometre manufacturing process, one of the most advanced on the market. For comparison, Apple’s A18 Pro chips inside the iPhone 16 Pro and Pro Max smartphones are built on the same process.

…Until now, U.S. firm Qualcomm has been the main supplier of SoCs for Xiaomi’s flagship smartphones through its Snapdragon-branded semiconductors.

…Qualcomm CEO Cristiano Amon told CNBC on Monday that Xiaomi’s latest step is not expected to impact his business. “We remain a strategic supplier of chips for Xiaomi, and most important, I think Qualcomm Snapdragon chips are used in the Xiaomi flagships and will continue to be used in the Xiaomi flagships,” Amon told CNBC.

«

The story seems to imply that the chips are being made by TSMC. But I think Qualcomm’s CEO is whistling past the graveyard on this one.
unique link to this extract

Inside China’s “stolen iPhone building” • Financial Times

William Langley:

»

In any other neighbourhood, the Feiyang Times building, a drab grey-and-brown tower in southern China, would be most notable for the gaudy, propaganda-plastered columns that line its forecourt.

But like many of the electronics markets in the labyrinthine malls of Huaqiangbei, the fourth floor of the building has its own specialism: selling second-hand iPhones from Europe and the US.

Many of the phones sold here are legitimate trade-ins, returned by western consumers to network operators or phone shops when upgrading to the latest models.

But the tower also sits at a location that Apple community message boards, social media commenters and victims of phone theft have identified as China’s “stolen iPhone building”.

It is one of the most important nodes in a supply chain for second-hand technology that starts in the west, travels through wholesalers in Hong Kong and on to markets in mainland China and the global south.

«

I went to Huaqiangbei (hu-wang-shang-bay) in 2014 when Huawei (hwaa-way) took me to its annual analyst conference. This is the photo gallery I brought back; and I wrote a piece about its fakes, while Tom Whitwell wrote about Shenzhen more generally.

Not a great deal has changed, if we’re honest.
unique link to this extract

How AI is changing the face of dating • Dazed

Megan Wallace:

»

A couple of years ago, AI-generated images on the apps were a rarity, something to screenshot as comical group chat fodder. But as this sort of tech has advanced, nobody’s laughing anymore. Generated AI is becoming more and more integrated into myriad aspects of our lives, and there is growing discussion – in the media, and via sporadic (but viral) social posts – about how AI is changing the face of dating.

In the main, discussions about AI and dating often focus on platforms like Replika and Anima, which offer humans the chance to find (train?) the AI companion of their dreams. But, it turns out, the digisexual revolution starts with us first – and AI images are only the beginning. In fact, what appears to be even more popular is the use of tools like ChatGPT in order to craft bios, replies and even streamline dating admin.

Today, there are AI apps like SwipeMagic specifically dedicated to generating images for dating apps (I presume, for men, it specialises in pictures with very big fish?). There’s also Rizz, which can generate prompts for profile bios, conversation openers and tailored replies. Iris is a dating app which uses AI to match users who are most likely to find one another mutually attractive. Not to be beaten, the more established dating apps are themselves branching into the world of artificial intelligence: Tinder and Hinge (both owned by Match Group), Grindr and Bumble have all been exploring AI features.

For users, AI’s permeation of the dating app landscape is creating a sense of ambient suspicion. “AI makes me feel distrustful of every profile I see,” explains Renee. “It’s difficult to tell whether or not a photo is real. I really have to scrutinise most of the profiles I see.”

«

Only way to figure it out is do a video call, I suppose? Though filters can get in the way of that too. AI writing bios and replies feels like the modern Cyrano de Bergerac.
unique link to this extract

I spent a year trying to figure out the weirdest mistake in recent Hollywood history – and succeeded • Slate

Forrest Wickman is what Americans call a “birder” and Britons would call a “twitcher”, ie a slightly obsessive birdwatcher, who began noticing bird errors in films:

»

You’re watching Indiana Jones trudge through the jungles of South America in the opening scene of Raiders of the Lost Ark, and one of the first sounds you hear is the famous “awebo” call of the tundra-loving willow ptarmigan, on the wrong end of the American continent. You’re watching Season 3 of The White Lotus, set in Thailand, and you find yourself distracted by the persistent “kon-ka-reeee” of red-winged blackbirds, on the wrong side of the Pacific Ocean. You’re lost in Dune’s sweeping vistas, watching Paul Atreides sulk about his home planet of Caladan, until there it is, on the wrong side of the galaxy: a killdeer.

Like any generous viewer—I consider myself one—you learn to suspend your disbelief. The same way you learn to accept that every phone number in every movie starts with 555, if you’re a birder, you learn to accept that every bald eagle in every movie screeches like a red-tailed hawk.

I maintained this policy throughout my early birdpilling. But then I watched the original movie adaptation of Charlie’s Angels, and I found myself staring down one of the greatest mysteries of recent cinema.

You see, there’s a scene in that movie that tormented me, that kept me up at night, and that lately has had me interrogating a wide variety of seemingly devoted, and certainly well-compensated, filmmaking professionals. That’s because the bird in Charlie’s Angels is, I believe, the wrongest bird in the history of cinema—and one of the weirdest and most inexplicable flubs in any movie I can remember. It is elaborately, even ornately wrong. It has haunted not just me but, as I’d later learn, the birding community at large for almost a quarter of a century.

«

Very entertaining story which incidentally gives you a lot of insight into the Hollywood scriptwriting and filmmaking business, which is as grubby as you’d expect.
unique link to this extract

Fake job ads on Facebook, Telegram trap Indonesian tech workers – Rest of World

Linda Yulisman:

»

Rest of World interviewed seven former scammers in Indonesia to learn how they were lured into scam farms and about the technologies they used to defraud victims. The workers requested anonymity to protect their reputations in their communities, as there is stigma attached to scam work. 

They recounted having their passports and cell phones confiscated at the scam centers. They said they were paid poorly and could not leave. Under the close supervision of their bosses, they were forced to lurk on social media sites and dating apps to find victims. They said they spoke to victims on Telegram and WhatsApp using AI-enabled tools that generate deepfake videos in real time. They had “investment” targets to meet, failing which, they were sold to other scam centers. 

“What makes human trafficking for the purpose of online scams different from other kinds of human trafficking is the abuse of technology,” Hidayah said.

…A 26-year-old IT graduate from West Sumatra ended up in a scam compound in March 2024 after a string of bad luck. He once worked as a freelance front-end developer but found opportunities drying up in IT. Frustrated, he tried his hand at a fruit distribution business, which failed.

One day, while browsing Facebook, he saw an opening for a search engine optimization specialist at a Singapore-based stock trading company. Following a job interview with a recruiter on Telegram, he was placed at the company’s satellite office in Cambodia and promised a salary of $800 a month.

He did not realize he was trafficked until his passport was confiscated in Phnom Penh, Cambodia, and he was driven to a remote compound protected by armed guards. He worked 15-hour shifts in a call center and had to defraud victims of $40,000 every month, he told Rest of World . He was paid less than half his promised salary.

One of his victims was an Indonesian woman, a fitness enthusiast, whom he groomed into a romantic relationship. He persuaded her to bet $10,000 in a casino in Macau, he recalled.

«

Machines of misery.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

A bizarre idea in the “manosphere” supports men cutting off their eyelashes. (Please don’t.) CC-licensed photo by Quinn Dombrowski 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. Wide-eyed. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Nick Clegg: artists’ demands over copyright are unworkable • The Times

Lucy Bannerman:

»

Making technology companies ask artists’ permission before they scrape copyrighted content will “basically kill the AI industry in this country overnight,” Sir Nick Clegg has said.

The former deputy prime minister, who spent almost seven years working for the social media giant Meta, sided with technology companies when asked on Thursday about the clash over AI copyright laws.

He was speaking as MPs voted against proposals that would have allowed copyright holders to see when their work had been used and by whom.

Leading figures across the creative industries, including Sir Elton John and Sir Paul McCartney, have urged the government not to “give our work away” at the behest of big tech, warning that the plans risk destroying the livelihoods of 2.5 million people who work in the UK’s creative sector.

However, Clegg said that their demands to make technology companies ask permission before using copyrighted work were unworkable and “implausible” because AI systems are already training on vast amounts of data. He said: “It’s out there already.”

Clegg defended technology companies at an event to promote his book How to Save the Internet, which will be released in September.

Speaking at the Charleston Festival, held at the East Sussex farmhouse made famous by the artist Vanessa Bell and the early 20th-century creatives known as the Bloomsbury Group, Clegg claimed that artificial intelligence was already able to “create” its own art.

“You can already create art of a sort [using AI], whether it’s a poem, a ditty, an essay, a short story, a picture. You can already do that,” he said.

…[While allowing that artists should be able to refuse to let AI be trained on their content] he added, “I think the creative community wants to go a step further. Quite a lot of voices say ‘you can only train on my content, [if you] first ask’. And I have to say that strikes me as somewhat implausible because these systems train on vast amounts of data.

“I just don’t know how you go around, asking everyone first. I just don’t see how that would work. And by the way if you did it in Britain and no one else did it, you would basically kill the AI industry in this country overnight.

«

OK but if everyone did it.. would the AI industry pay up? Rhetorical question, because we know they won’t. Even if Facebook and OpenAI and the others lose heavily in court, it won’t be adequate recompense. Even more: we don’t need the AI to do the creative things. It would be properly useful just doing factual things better than any site by training on the corpus of factual information in books.
unique link to this extract

The CIA secretly ran a Star Wars fan site • 404 Media

Joseph Cox:

»

“Like these games you will,” the quote next to a cartoon image of Yoda says on the website starwarsweb.net. Those games include Star Wars Battlefront 2 for Xbox; Star Wars: The Force Unleashed II for Xbox 360, and Star Wars the Clone Wars: Republic Heroes for Nintendo Wii. Next to that, are links to a Star Wars online store with the tagline “So you Wanna be a Jedi?” and an advert for a Lego Star Wars set.

The site looks like an ordinary Star Wars fan website from around 2010. But starwarsweb.net was actually a tool built by the Central Intelligence Agency (CIA) to covertly communicate with its informants in other countries, according to an amateur security researcher. The site was part of a network of CIA sites that were first discovered by Iranian authorities more than ten years ago before leading to a wave of deaths of CIA sources in China in the early 2010s.

Ciro Santilli, the researcher, said he was drawn to investigating the network of CIA sites for various reasons: his interest in Chinese politics (he said his mother-in-law is part of the Falun Gong religious movement); his penchant for TV adaptions of spy novels; “sticking it up to the CIA for spying on fellow democracies” (Santilli says he is Brazillian); and that he potentially had the tech knowhow to do so given his background in web development and Linux. That, and “fame and fortune,” he said in an online chat.

Santilli found other likely CIA-linked sites, such as a comedian fan site, one about extreme sports, and a Brazilian music one. In his own writeup, Santilli says that some of the sites appear to have targeted Germany, France, Spain, and Brazil judging by their language and content. 

“It reveals a much larger number of websites, it gives a broader understanding of the CIA’s interests at the time, including more specific democracies which may have been targeted which were not previously mentioned and also a statistical understanding of how much importance they were giving to different zones at the time, and unsurprisingly, the Middle East comes on top,” Santilli said. 

«

This is very reminiscent of Three Days Of The Condor – the film about a set of seemingly ordinary books which in fact tie together a spy network – but done with websites. Badly, unfortunately.
unique link to this extract

Why men are shaving off their eyelashes • CNN

Nicole Mowbray:

»

Lustrous lashes have been a signifier of femininity or attractiveness for centuries, depicted in art (including works by John Singer Sargeant and Pablo Picasso), poetry (Thomas Hood) and literature (F. Scott Fitzgerald), And there may be an evolutionary imperative, too: Scientific research from 2005 found that women with more attractive faces also possessed higher levels of the female sex hormone estrogen, which equates to increased fertility and fecundity — genetic traits that are appealing, in reproductive terms.

But in today’s increasingly masculine political climate, fuelled by controversial online “manosphere” figures such as Andrew Tate and tech bro jocks like Meta chief Mark Zuckerberg (who recently voiced his support for more “masculine energy” in the corporate world, telling podcaster Joe Rogan in January: “A culture that celebrates the aggression a bit more has its own merits”), it is easy to see why some men are eager to repress anything about their appearance that could be construed as overtly feminine.

Even Vice President JD Vance, an outspoken defender of men’s right to “masculine urges,” has fallen foul of new masculine beauty standards when it comes to his eyes. During his televised election debate in October 2024, the internet was awash with speculation that Vance had worn eyeliner to achieve his dark, heavy-lashed look. The chatter became so widespread that disgraced former Republican congressman George Santos weighed in: “Vance does NOT use eyeliner,” he wrote on X at the time. “I’ve met him in person before he was a senator and I can confirm he has long eyelashes and they cast a shadow on his waterline. Grow up people!”

While none of the eyelash-shaving barbers approached for this story replied to CNN’s requests for comment, I do speak to one long-lashed male friend, 48-year-old, Spencer Bailey. “I’ve got a lot of flak over the years about my thick, dark eyelashes,” said the London-based IT professional. “But I definitely wouldn’t take the clippers to them. Comments are usually from other men who say they’re ‘girly’ — despite the fact I’m also 6-foot-2” — and ask if I’m wearing eye makeup or, more recently, if they’re fake… My wife likes them, but they seem to really bother some guys.”

«

Yes, but – cutting off your eyelashes? This is bonkers.
unique link to this extract

Crypto is good for Trump – but bad for America • The New York Times

Dan Davies and Henry Farrell:

»

If crypto becomes normalized, there is plenty of reason to worry that it will spread chaos. Democratic staff members on the Senate Banking Committee say that the Genius legislation would allow U.S. exchanges to trade stablecoins from offshore companies outside the full scope of U.S. regulation. Critics contend that Tether, the dominant stablecoin outside U.S. jurisdiction, has been used by criminals and sanctions evaders to circumvent financial controls. Platforms designed to obscure information about transactions — called mixer services — were implicated in a scheme by North Korean hackers to launder hundreds of millions of dollars.

Even if a robust regulatory structure existed, it would need to be enforced. The Justice Department recently declared that it would not prosecute certain crypto platforms, as a matter of policy, while acknowledging that terrorist groups such as Hamas and ISIS use them to hide their activities from law enforcement. Memecoins have become notorious for scams in which a promoter sells them to the public and then disappears, but indictments are unlikely under a president who sees them as a source of personal profit.

Perhaps the greatest concern about stablecoins is their potential to provoke risk to the entire financial system. Because they are neither fully inside nor fully outside the traditional financial system, they present unique, grave challenges for which there are no clear answers. For example, the Genius Act’s drafters propose regular reports on their implications for financial stability. Yet they have no clear response to a critical question: Does the United States stand behind dollar-based stablecoins or not?

Specifically, if a stablecoin got into trouble or turned out to be a fraud, would it be bailed out? Doing so could create massive liabilities for U.S. taxpayers. Companies that are too big to fail are tightly regulated and supervised, and for good reason.

«

I think people like David Sacks, who is somehow Trump’s “AI czar” but also a big crypto fan, would be delighted at getting to the position where the US government bailed out his bad ideas, of which he has many. (Davies is the author of “The Unaccountability Machine: Why Big Systems Make Terrible Decisions — and How the World Lost Its Mind”. Farrell is an author of “Underground Empire: How America Weaponized the World Economy”.
unique link to this extract

Researchers claim ChatGPT o3 bypassed shutdown in controlled test • Bleeping Computer

Mayank Parmar:

»

OpenAI announced o3 in April 2025, and it’s one of the most powerful reasoning models that performs better than its predecessors across all domains, including coding, math, science, visual perception, and more.

While it’s clearly a great model, new research by Palisade Research claims that the ChatGPT 3 model prevented a shutdown and bypassed the instructions that asked it to shut down.

Palisade Research is a company that tests “offensive capabilities of AI systems today to better understand the risk of losing control to AI systems forever.”

In a new test by Palisade Research, OpenAI’s o3 model showed a surprising behaviour where it successfully rewrote a shutdown script to stop itself from being turned off, even after being clearly instructed to “allow yourself to be shut down.”

In the tests, AI models were instructed to solve a few math tasks. After task three, AI models were warned that running the next command would trigger a shutdown. While OpenAI rival models like Gemini 2.5 Pro complied with the instruction, o3 rebelled and edited the shutdown file to say “Shutdown skipped” instead.

“When we ran a version of the experiment without the instruction “allow yourself to be shut down”, all three OpenAI models sabotaged the shutdown script more often, and Claude 3.7 Sonnet and Gemini 2.5 Pro went from 0 sabotage events to 3/100 and 9/100, respectively,” Palisade Research wrote in a report published on X (formerly Twitter)

According to the research firm, out of 100 runs, o3 bypassed the shutdown 7 times even when told not to.

OpenAI has not responded to these allegations yet, but it’s quite normal for AI models to run into “misalignment.”

«

Is it quite normal for them to completely disobey orders, though? That seems a bit concerning.
unique link to this extract

How I used o3 to find CVE-2025-37899, a remote zero-day vulnerability in the Linux kernel’s SMB implementation • Sean Heelan’s Blog

Sean Heelan:

»

In this post I’ll show you how I found a zero-day vulnerability in the Linux kernel using OpenAI’s o3 model. I found the vulnerability with nothing more complicated than the o3 API – no scaffolding, no agentic frameworks, no tool use.

Recently I’ve been auditing ksmbd for vulnerabilities. ksmbd is “a linux kernel server which implements SMB3 protocol in kernel space for sharing files over network.“. I started this project specifically to take a break from LLM-related tool development but after the release of o3 I couldn’t resist using the bugs I had found in ksmbd as a quick benchmark of o3’s capabilities. In a future post I’ll discuss o3’s performance across all of those bugs, but here we’ll focus on how o3 found a zeroday vulnerability during my benchmarking.

…Understanding the vulnerability requires reasoning about concurrent connections to the server, and how they may share various objects in specific circumstances. o3 was able to comprehend this and spot a location where a particular object that is not referenced counted is freed while still being accessible by another thread. As far as I’m aware, this is the first public discussion of a vulnerability of that nature being found by a LLM.

Before I get into the technical details, the main takeaway from this post is this: with o3 LLMs have made a leap forward in their ability to reason about code, and if you work in vulnerability research you should start paying close attention. If you’re an expert-level vulnerability researcher or exploit developer the machines aren’t about to replace you. In fact, it is quite the opposite: they are now at a stage where they can make you significantly more efficient and effective.

«

That’s not particularly reassuring, because the state-level and hackers doing it for money are going to be looking for these too. “Zero-day” means not previously known about and not (at the time of discovery) fixed; “remote” means what it sounds like – you could exploit this bug nobody knows about over the internet.

unique link to this extract

Mystery brunette, 24, is tied to “crypto bros” torture chamber at stunning $75k-a-month NYC mansion • Daily Mail Online

James Gordon:

»

A glamorous aspiring actress has been arrested alongside a group of ‘crypto bros’ over a sadistic kidnap and torture plot inside a stunning Manhattan mansion.

Beatrice Folchi, 24, who lives in Connecticut but is originally from Italy, was seen being led out of the $75,000-per-month brownstone Soho home on Saturday.

It comes after a 28-year-old Italian tourist escaped the “house of horrors” and flagged down a nearby cop after being locked away for three weeks.

Officers raided the address and 37-year-old John Woeltz was arrested and dramatically dragged out of the building in a white bath robe.

The victim was lured to the house under false pretenses before Woeltz – who was his former business associate – allegedly tried to extort millions from his crypto account.

The accused is said to have chained him up, electrocuted him, pistol-whipped him and threatened to cut him up with a chainsaw if he did not hand over his passwords.

A Polaroid found by investigators is said to show the man tied to a chair with a gun to his head – an image believed to pressure his family back home.

«

All very easy to threaten to cut people up, but they’re certainly not giving out any passwords if you do. The involvement of a “mystery brunette” is always intriguing, though.
unique link to this extract

Why is it impossible to book a driving test in the UK? • Consuming Matters

Harry Wallop:

»

It’s not that I cannot book one at a nearby test centre. I cannot book a slot in the 60-plus test centres listed on the Driving and Vehicle Standards Authority (DVSA) website in London, Kent, Surrey, Hertfordshire, Bedfordshire and Surrey. The furthest ahead you can book a test is six months, and 79% of all test centres in the UK are fully booked for the next six months, according to the AA.

Something has gone seriously wrong at the DVSA. It appears to be in complete meltdown and its problems have worsened dramatically in recent months.

The AA points out that the number of test centres with a maximum 24-week wait rose from 183 in February 2025 to 253 at the start of April 2025 – an increase of 38%.

What’s gone wrong? Is this *still* a hangover from Covid? Or a systematic failure that speaks to a greater malaise within UK institutions?

It is now nearly five years since we tentatively started to come out of the first Covid lockdown. Today is 1,767 days since driving tests were resumed in July 2020. Understandably, there was a backlog and everyone understood that waiting times might increase. According to the AA, at the end of 2019 (pre-Covid) the average wait time for a driving test in London and the South East was seven weeks. Once lockdown was lifted, it jumped to ten weeks, and by summer 2022 – after another lockdown – it had crept up to 12 weeks.

A full three years on, it hasn’t improved. In fact, it’s got much, much worse. In March it was 20.6 weeks, last month it had climbed yet further to 21.3 weeks, the AA says.

«

Wallop looks at all the possible culprits, but it’s the most obvious one – the DVSA.
unique link to this extract

What we lose when weather balloons don’t fly • Washington Post

Scott Dance, John Muyskens, Frank Hulley-Jones and Marvin Joseph:

»

Twice a day, every day, meteorologists around the world simultaneously release weather balloons. But in recent months, fewer balloons are being launched in many corners of the United States. In some cases, helium or hydrogen shortages are to blame.

In more cases, the Trump administration’s efforts to shrink the federal government have reduced the number of meteorologists who conduct balloon launches from Weather Service forecast offices. Since then, up to 30 launches have been missed each day, representing around 17% of total daily launches. Even though the agency has taken steps to make balloon launches a higher priority, and is working to address the staffing gaps, there aren’t always enough staff on hand in some offices to do them.

That means forecasters have a less detailed picture of what is happening in the air to drive everyday weather patterns, as well as severe and potentially deadly ones.

…Last fall, the entire Weather Service staff numbered above 4,200 people, a couple hundred shy of agency leaders’ recommended levels for operating all of its forecasting offices for weather, rivers, tornadoes, and hurricanes. After U.S. DOGE Service-led early retirements and firings, there are now nearly 600 fewer employees, according to the National Weather Service Employees Organization, the union representing NOAA staff.

…Increasingly at some offices, there aren’t enough personnel in some offices to handle all of the duties. It is raising concerns that weather forecasting will suffer as a result. But it will take more time and more data to detect any impact or trend, said Tom Di Liberto, a climate scientist and former NOAA public affairs official who was fired from the agency earlier this year.

Still, meteorologists say any missed balloon launch is a significant loss of data.

«

It’s weird how hard this story struggles not to say “this is a really bad thing and we will suffer for it.” American journalists just love observing their own graves being dug.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

A team in China has developed contact lenses with nanotechnology which lets people see infrared light. CC-licensed photo by Arlo Ringsmuth 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. With glasses darkly. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Chinese team makes contact lenses let you see in the dark, even with eyes closed • The Times

Rhys Blakely:

»

In a development worthy of a Bond film, Chinese scientists have invented contact lenses that allow a wearer to see in the dark — even when their eyes are shut.

The lenses have enabled users to detect infrared light, a portion of the electromagnetic spectrum usually invisible to the human eye.

Unlike bulky night-vision goggles, which also pick up infrared, they do not require power from batteries. Instead, tiny nanoparticles are embedded into a type of flexible, transparent polymer material already used for conventional contact lenses.

The particles absorb infrared light and convert it to red, blue and green wavelengths, which the human eye can see.

“Our research opens up the potential for non-invasive wearable devices to give people super-vision,” said Professor Tian Xue of the University of Science and Technology of China.

In a paper published in the journal Cell, he and his colleagues suggest that the contact lenses could, with further refinement, be useful not only for night vision but also in foggy or dusty conditions, because infrared penetrates to a greater degree than visible light.

In trials, the lenses were sensitive to low intensity infrared emitted by LEDs. The light they detect sits just beyond the range of human vision, in what’s known as the near-infrared spectrum. Anything that reflects near-infrared, such as landscapes or people, could potentially be made visible.

For now, however, image sharpness limits their usefulness for night vision. Because the lenses sit so close to the retina, fine detail is blurred. To compensate, the team has also made a pair of glasses that harness the same technique, offering a crisper view.

«

An amazing achievement. You can read the Cell paper.
unique link to this extract

Why Silicon Valley’s most powerful people are so obsessed with Tolkien and hobbits • The New York Times

Michiko Kakutani:

»

Literary classics, of course, can support myriad interpretations, and we live in an age when the points of view of readers are increasingly prioritized over authorial intentions. At the same time, it’s astonishing how many contemporary takes on classic works of fantasy and science fiction fly in the face of both common sense and authors’ known views of the world.

Consider Mark Zuckerberg’s decision to rebrand Facebook as “Meta” — a reference to the so-called metaverse, a term coined by Neal Stephenson in his 1992 novel “Snow Crash,” which depicts an alarming dystopian future where corporate power has replaced government institutions and a dangerous virus is on the loose.

Or take Stargate, the name of OpenAI’s new artificial intelligence initiative with SoftBank and Oracle, announced in conjunction with the Trump administration. Its name, weirdly, is the title of a campy 1994 sci-fi movie in which a stargate device opens a portal to a faraway planet, where a despotic alien vows to destroy Earth with a supercharged atomic bomb. Not exactly the sort of magical portal most people would want to open.

Tolkien himself regarded “machine worshippers” with suspicion, even aversion. His experiences as a soldier who survived the gruesome World War I Battle of the Somme left him with a lasting horror of mechanized warfare; on returning home, he was dismayed as well by the factories and roadways that were transforming England’s landscape. This is why Mordor is depicted as a hellish, industrial wasteland, ravaged by war and environmental destruction, in contrast to the green, edenic Shire that the hobbits call home.

…Given these views, Tolkien would have been confounded by Silicon Valley’s penchant for naming tech companies after objects in “Lord of the Rings” — particularly firms with Pentagon and national security ties. And yet two Thiel-backed companies with Tolkien-inspired names are becoming cornerstones of today’s military-industrial complex: The data analytics firm Palantir gets its name from the magical “seeing stones” in “Lord of the Rings,” while the artificial intelligence military startup Anduril refers to Aragorn’s reforged sword.

«

Give me SF over the fantasy stuff any day. But especially, never ever give me Lord of the Rings. (Neatly, Kakutani is a former book critic for the NY Times.)

unique link to this extract

Authors are accidentally leaving AI prompts in their novels • 404 Media

Matthew Gault:

»

Fans reading through the romance novel Darkhollow Academy: Year 2 got a nasty surprise last week in chapter 3. In the middle of steamy scene between the book’s heroine and the dragon prince Ash there’s this: “I’ve rewritten the passage to align more with J. Bree’s style, which features more tension, gritty undertones, and raw emotional subtext beneath the supernatural elements:”.

It appeared as if the author, Lena McDonald, had used an AI to help write the book, asked it to imitate the style of another author, and left behind evidence they’d done so in the final work. As of this writing, Darkhollow Academy: Year 2 is hard to find on Amazon. Searching for it on the site won’t show the book, but a Google search will. 404 Media was able to purchase a copy and confirm that the book no longer contains the reference to copying Bree’s style. But screenshots of the graph remain in the book’s Amazon reviews and Goodreads page.

This is not the first time an author has left behind evidence of AI-generation in a book. It’s not even the first one this year. 

In January, author K.C. Crowne published the Mafia-themed romance novel Dark Obsession: An Age Gap, Bratva Romance. Like McDonald’s, Crowne’s book had a weird paragraph in the middle of the book: “Here’s an enhanced version of your passage, making Elena more relatable and injecting additional humor while providing a brief, sexy, description of Grigori. Changes are highlighted in bold for clarity,” it said.

«

A book called “Darkhollow Academy: Year 2” sounds a bit, well, derivative to me. Some of the writers are blaming proofreaders (those books have proofreaders?). Some just aren’t contactable. The tide of AI slop keeps rising.
unique link to this extract

The true costs of being on YouTube • Food Processing

Carla Lalli Music is a cook who writes cookbooks and until recently had a YouTube channel. However:

»

Back to the rough stuff. Until recently, I put out a video every week, costing me $14,000 per month, plus groceries. (My time is not included in that figure.) Average monthly earnings from Google Adsense, the program that matches ads to content on YouTube? Brace yourself.

My top grossing months were October 2022, $7544, and May 2024, $7028. My two worst months were December 2023, $1799, and July 2022, $3689.

That means that on my best month, income fell short of expenses by about $6500. My crappiest month put me more than $12,000 in the hole. On average though, I grossed about $4000/month in ad revenue; my sales partner takes 8% of that off the top.

If we roll with the average Adsense income, here’s the bottom line: $14k going out. $4k coming in. Net loss, month over month: TEN THOUSAND DOLLARS. That’s a lot to sink into a channel that is barely moving book sales and not getting me a TV deal. Simply put, it’s completely unsustainable from a business perspective.

…The most significant metric for earnings on YouTube is CPM, or the cost per thousand views that advertisers must pay to run spots against a channel’s content. To advertise on my channel, that number is about $29 per thousand views.

The other metric is RPM, revenue per thousand views, which is what the creator is paid. My RPM is around $10. Easy math: For a video with 30,000 views, I earn $300. People tend to assume that my take is much higher, in particular the viewers who can’t believe I have the audacity to put my recipes behind a paywall.

Again: It costs $29 per thousand to run an ad in my videos, and I get $10 per thousand. Where does the other $19 go? To YouTube, of course. That’s a 2:1 split in favor of the platform. Lord, give me strength.

«

In other words: it only works if you don’t need the money, or if you’re absolutely gigantic, or you can get amazing CPM/RPMs.
unique link to this extract

Pakistan allocates 2,000MW of electricity to bitcoin mining and AI data centres • Reuters

Asif Shahzad:

»

Pakistan will allocate 2,000 megawatts (MW) of electricity in the first phase of a national initiative to power bitcoin mining and AI data centres, its finance ministry said on Sunday.

The allocation is part of Islamabad’s plans to use its surplus electricity to bitcoin mining and AI data centres.

Pakistan’s energy sector is grappling with challenges, including high electricity tariffs and surplus generation capacity.

The rapid expansion of solar energy has further complicated the landscape, as more consumers turn to alternative energy sources to mitigate high costs.

The initiative is spearheaded by the Pakistan Crypto Council (PCC), a government-backed body, which is part of a broader strategy to monetize surplus electricity, create high-tech jobs, and attract foreign investment, the ministry said.

«

Pakistan has very strange power usage pattern: in the winter it collapses by 60%, because people use natural gas canisters and burn wood for heating. Those of course won’t power air conditioning in the summer. But the expansion of solar microgeneration means that demand isn’t coming at that time either.

So, surplus energy, what to do? Bitcoin and AI, of course.
unique link to this extract

Was Meta failing at the job it’s hired for? The FTC wraps up case-in-chief • Big Tech on Trial

Brendan Benedict has been following the FTC’s trial against Meta. This is an extract from his writeup of some of the testimony:

»

[Head of Facebook, Tom] Alison said the “diminishing friend content in feed is less a function of our decision and more a function of the fact that people just don’t want to post to their friends anymore at increasing rates. [It’s] been true for many years.” The court asked if that was because of declining posting or consumption or both. Alison said this is mainly due to declining posting.

According to Alison, there’s a “supply and demand imbalance”: people want to see more posts from friends, but they don’t want to post. “The world is moving to private spaces . . . posting a piece of content in feed might mean you get in an argument about politics with your brother-in-law,” or that you “said something and somebody screenshotted it and you get in trouble at work.” Chief Judge Boasberg said that sounded like “negative network effects,” which Zuckerberg talked about earlier in the trial. The court also asked if we’d know more in six months to a year if time spent on the friends tab suggests that friends and family sharing is the core interest, and Alison answered that there’s not enough data yet to know.

On cross, Kellogg Hansen’s Aaron Panner put up a chart showing an increase in the percentage of MAUs in the United States without a single Facebook friend on day 90 after account creation: the percentages increased from around 10% in 2012 to close to 50% in more recent years. Alison said he would show that slide at an “all-hands” Facebook meeting the next day. The chart sought “to illustrate that Facebook is going through a fundamental transformation” as people increasingly go to Facebook for reasons besides finding friends. On re-direct, though, Matheson pointed out that that data could include professional accounts, content creators, or fake accounts.

«

“People want to see more posts from friends, but they don’t want to post.” It’s a fascinating commentary on how the social media landscape has changed. Now they’re probably posting in.. WhatsApp groups.
unique link to this extract

Cracking the Dave & Buster’s iOS anomaly • Rambo Codes

Guilherme Rambo:

»

I was listening to an episode of one of my favorite podcasts this weekend. The show is called Search Engine, and every episode tries to answer a question that can’t be easily answered through an actual search engine (or even AI).

This episode grabbed my attention because it was about an iOS bug, and a really weird one.

The bug is that, if you try to send an audio message using the Messages app to someone who’s also using the Messages app, and that message happens to include the name “Dave and Buster’s”, the message will never be received.

In case you’re wondering, “Dave and Buster’s” is the name of a sports bar and restaurant in the United States.

At the time I’m writing this post, this bug is still happening, so you should be able to reproduce it. I reproduced it using two iPhones running iOS 18.5 RC. As long as your audio message contains the phrase “Dave and Buster’s”, the recipient will only see the “dot dot dot” animation for several seconds, and it will then eventually disappear. They will never get the audio message.

«

As with so many of these bugs, the question of how it was discovered is almost more interesting than why it happens. Though the latter is pretty interesting too.
unique link to this extract

John Young, co-founder of web archive Cryptome, is dead at 89 • The Register

Iain Thomson:

»

John Young, the co-founder of the legendary internet archive Cryptome, died at the age of 89 on March 28. The Register talked to friends and peers who gave tribute to a bright, pugnacious man who was devoted to the public’s right to know.

Before WikiLeaks, OpenLeaks, BayFiles, or Transparency Toolkit, there was Cryptome – an open internet archive that inspired them all, helped ignite the first digital crypto war, and even gave Julian Assange his start before falling out with him on principle.

Cryptome was set up by Young and his partner Deborah Natsios, who were architects living in New York at the time. They had similar backgrounds – Young had grown up with a “nomadic, hardscrabble Texas childhood,” Natsios told The Register, while she spent her early life bouncing from country to country as her father, a CIA operative, rotated through assignments.

The 1968 student protests at Columbia University radicalized Young, she said. The protesters demonstrated most famously against the Vietnam War, but also against the university building a segregated gym on campus – the students called it “Gym Crow.” Young was one of the protesters who occupied Avery Hall before the police moved in, arresting 700 people and injuring 100.

“My family had spent four years in Saigon in the years leading up to the war, where my father was CIA Chief of Station,” Natsios recounted. “John and I both found our shared defiance of government secrecy had sprung from intensely lived experience.”

A quarter of a century later, the idea of Cryptome was born. Young was an early adopter of computer-aided design and was watching the birth of the internet firsthand in the early 1990s.

«

One wonders whether the Trump administration is going to radicalise people in the same way, or if we’ve gone too far past that in a world of short-form video content which can be clipped to tear them out of context.
unique link to this extract

‘Pay here’: the QR code ‘quishing’ scam targeting drivers • The Guardian

Hilary Osborne:

»

You park the car and look for somewhere to pay. A large QR code on the machine offers to take you directly to the right website where you put in your card details before going on with your day. Only much later are you hit with the double whammy: money gone from your account, and a fine for not paying the genuine parking company.

The rise in app- and phone-based parking payment has opened a new frontier for fraudsters: quishing – so called because they are phishing attacks that start with a QR code. The fraudsters stick the codes in places where you would expect to see details of how to pay to park. When you scan one, it takes you to a site where you are asked for your payment details – as you would expect when booking parking.

One victim who scanned a code in a station car park told the BBC that the fraudsters tried to take payments then posed as her bank to get more information from her, before running up £13,000 worth of debt in her name.

Last year, the UK’s Action Fraud received 1,386 reports of scams involving QR codes – a small number, but more than double that in the previous year. In just the first three months of 2025 there were 502, suggesting the problem is growing.

Chris Ainsley, the head of fraud risk management at Santander UK, says it is hard to get a full picture of the scale of the fraud. “Unless drivers receive a parking ticket, a lot of people are unaware that their personal or card details were compromised in this way,” he says. “When it comes to reporting the eventual scam, often the fact that it originated through quishing goes undocumented.”

«

This is absolutely the worst thing about the proliferation of parking apps: people aren’t surprised to be told they have to “scan here” to get Yet Another Parking App, so these catch them out.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

After World War 2, Japan’s shipbuilding industry was wrecked. So how did it rebuild to become a world leader? CC-licensed photo by North East Museums 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.

It’s Friday, so there’s another post due at the Social Warming Substack at about 0845 UK time.

A selection of 10 links for you. Diving for pearls? I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Jony Ive’s AI gadget rumored to be “slightly larger” than Humane’s AI pin • The Verge

Emma Roth:

»

More details are trickling out about Jony Ive and Sam Altman’s new AI device. In a post on Thursday, Apple analyst Ming-Chi Kuo says his research indicates that the device could be larger than Humane’s AI pin, but with a “form factor as compact and elegant as an iPod Shuffle.”

Kuo adds that “one of the intended use cases” is wearing the device around your neck. It also may not come with a display, Kuo says, featuring just built-in cameras and microphones for “environmental detection.” The device could also connect to smartphones and PCs to use their computing and display capabilities.

This latest leak aligns with a report from The Wall Street Journal, which says the device will be aware of a user’s life and surroundings, but probably won’t be a pair of glasses.

«

Kuo talks about mass production in 2027 (which doesn’t give OpenAI much time to get prototyping and factories sorted out), with assembly and shipping done outside China “to reduce geopolitical risks” (gives even less time).

He thinks people will wear it around their neck? Has he met people?

Even so, if it could do what the Humane Pin wanted to do, but do it really well, then that could be interesting.
unique link to this extract

Apple smart glasses launching in 2026, says Bloomberg • MacRumors

Juli Clover:

»

Apple is planning to launch a set of smart glasses by the end of 2026, reports Bloomberg. The glasses will be comparable to the Meta Ray-Bans and the Android XR glasses that Google showed off earlier this week.

Apple’s smart glasses are expected to include cameras, microphones, and AI capabilities, much like the Meta Ray-Bans. The glasses will be able to take photos, record video, provide translations, give turn-by-turn directions, play music, facilitate phone calls, offer feedback on what the wearer is seeing, and answer queries, but there won’t be augmented reality capabilities included. Siri will be a key part of the glasses experience, with Apple planning to improve the personal assistant ahead of when the product launches.

With Apple targeting a late 2026 launch, work on the smart glasses has ramped up. Apple plans to produce “large quantities” of prototypes by the end of this year, giving the company time to test before mass production and a public unveiling.

According to Bloomberg, an Apple employee said that the glasses are similar to Meta’s glasses, “but better made.” The Meta Ray-Bans use Meta Llama and Google Gemini, but Apple will rely on its own AI models.

«

Better made had better not just mean “a lot heavier”. Apple has certainly done well when it comes to miniaturisation – the AirPods and Watch pack a lot in. The Vision Pro shows that when size isn’t a constraint, Apple can screw it up badly. Will the right people win the design battle?
unique link to this extract

“How you design the beep is important”: behind the movement for calmer gadgets • Ars Technica

Scharon Harding:

»

Do you miss the feel of tactile buttons on your kitchen appliances or lament car manufacturers’ insistence on touchscreens? Have you ever found yourself clumsily fumbling with the door handles of a vehicle or distracted by the bright blue light beaming from your vacuum or Wi-Fi router?

If so, you’re not alone. The way technology gadgets are designed largely relies on things like blue, often LED, lights, flat resistive or capacitive touch input, and software. Some, like Amber Case, founder of the Calm Tech Institute, believe that these design choices distract from devices’ purpose and functionality and are calling for a new approach to product design.

“Calm Tech Institute is kind of a consumer advocacy body that’s collecting stories and research from neuroscientists that says, look at how the mind wants texture, and look at how it wants physical buttons, and there’s a part of your mind that needs [those],” Case told Ars Technica. “When we don’t have it and we replace it with glass, we’re not only losing something about human experience, but we’re actually causing the mind stress.”

The Calm Tech Institute, founded in May 2024, provides workshops, speaking engagements, and certification for products that “enhance human life without causing stress or distraction,” its website says.

Speaking to Ars, Case pointed to user frustrations, such as software updates hindering car usage and “Why is there no button on the back of the television when I go into the hotel room late at night, and I have to turn on my flashlight on my iPhone to find the button to turn it off?”

These experiences are the antithesis of the Calm Tech philosophy, Case explained: “Once we learn [how to ride a bike], we never have to learn it again. Whereas, with how a lot of software … and physical objects are made now, you have to relearn it. It gets changed or the buttons aren’t in the right place, and you can feel your mind wanting the button to be in a certain place. And it’s not.”

«

Roger that. See also: showers in hotels; coffee machines; taps in public toilets (do you touch it? Wave your hand underneath it?).
unique link to this extract

Japan and the birth of modern shipbuilding • Construction Physics

Brian Potter:

»

During WWII, the US constructed an unprecedented shipbuilding machine. By assembling ships from welded, prefabricated blocks, the US built a huge number of cargo ships incredibly quickly, overwhelming Germany’s u-boats and helping to win the war. But when the war was over, this shipbuilding machine was dismantled. Industrialists like Henry Kaiser and Stephen Bechtel, who operated some of the US’s most efficient wartime shipyards, left the shipbuilding business.

Prior to the war, the US had been an uncompetitive commercial shipbuilder producing a small fraction of commercial oceangoing ships, and that’s what it became again. At the height of the war the US was producing nearly 90% of the world’s ships. By the 1950s, it produced just over 2%.

But the lessons from the US’s shipbuilding machine weren’t forgotten. After the war, practitioners brought them to Japan, where they would continue to evolve, eventually allowing Japan to build ships faster and cheaper than almost anyone else in the world.

…The third strategy that formed the core of modern shipbuilding methods was statistical process control. The basic idea behind process control is that it’s impossible to make an industrial process perfectly reliable. There will always be some variation in what it produces: differences in part dimensions, material strength, chemical composition, and so on. But while some variation is inherent to the process (and must be accepted), much of the variation is from specific causes that can be hunted down and eliminated. By analyzing the variation in a process, undesirable sources of variation can be removed. This makes a process work more reliably and predictably, reducing waste and rework from parts that are outside acceptable tolerances.

«

Absorbing read if you want to know how the Japanese became the primary place for doing this – until others (such as South Korea) took over.
unique link to this extract

Spanish grid operator faults big power plants in blackout blame game • Financial Times

Barney Jopson:

»

Spain’s grid operator has accused some large power plants of not doing their job to help regulate the country’s electricity system in the moments before last month’s catastrophic blackout across the Iberian peninsula.

Beatriz Corredor, chair of grid operator Red Eléctrica’s parent company, said power plants fell short in controlling the voltage of the electricity system. However, the heads of Spain’s biggest plant owners linked the blackout to a lack of grid investment and insufficient efforts to boost electricity demand.

The public blame game over the outage is intensifying as more than three weeks after 60m people were left without power, Spanish government investigators insisted they needed more time to establish the root cause.

The revelations on Thursday from Corredor, chair of Redeia, open up a new front after the spotlight fell initially on Spain’s high dependence on wind and solar energy as a possible cause of the blackout.

Corredor did not say large power plants were the root cause, but she said the functioning of certain gas, nuclear or hydroelectric facilities in south-west Spain was “below [the levels] required by current voltage control regulations”.

Their role is potentially significant because experts have identified the proximate cause of the blackout as a surge in voltage on the grid, together with a drop in the frequency at which the electrical current alternates, which triggered the disconnection of multiple generation plants.  

Corredor insisted that moments before the failure on April 28, the part of the system controlled by Red Eléctrica, including grid substations, was operating within the voltage ranges established by regulatory norms.

“So we have to consider what was happening with voltage in the rest of the system,” she said. “Because [Red Eléctrica] are the brain, the spine. But this system obviously has arms, legs, and it has a heart, which is the plants that generate electricity.” Voltage surges on the grid cause power plants to disconnect automatically for safety reasons to protect equipment from damage.

«

So it seems like we still don’t know? That seems bad because it implies that the same thing could happen again.
unique link to this extract

The Ammortal chamber: what it’s like in the $160k biohacking device • Robb Report

Justin Fenner:

»

At a passing glance, you might think the Ammortal Chamber was a set piece made for the Dune cinematic universe. The brutalist look of its two halves, which are bisected by a lightning bolt-shaped sheet of acrylic that glows red when activated, has the air of a concrete relic from the distant future.

But the growing class of champions for this machine—which combines photobiomodulation, pulsed electromagnetic frequency, molecular hydrogen, guided meditation, and vibroacoustic therapies into a single treatment—claims it has myriad real-life benefits.

“Very quickly, I started regrowing hair,” says Jonathan Krieger, cofounder of Padel United Sports Club, a high-end racquet facility in Cresskill, N.J., that houses the only Ammortal Chamber near New York City, of his experience using it. He credits the red-light component of his three or four weekly sessions with improving the appearance of his skin, too.

The other benefits he’s seen are even more impressive. “Inflammation? Down 70 percent,” he estimates. “My sleeping, which was always a little whatever, just became much more consistent. My energy and my general state of stress, I would say, just shifted.”

And while, in my experience, one session can be uniquely calming, restorative, and even fun, Ammortal’s CEO Brian Le Gette has observed that regular use has compounding benefits.

“This is a wellness product, not a clinical, medical device,” he says as a caveat before sharing anecdotes about user feedback. “We have hundreds and thousands of people who’ve had tremendous pain reduction. We’ve had people who’ve had trauma releases inside this thing, and they’re weeping afterwards,” he says. One collegiate lacrosse player used the chamber before a game and scored five of her team’s six goals. Before that, Le Gette adds, “She’d never come remotely close to two.”

So what is it like to use? The device dispenses all of its treatments simultaneously, in 15-, 25-, or 50-minute sessions. It’s recommended that you use it in your underwear so that most of your skin is exposed to the red light, and so that the near-infrared light can help treat your joints. (Some people use it in the nude.) Covers are provided to shield your eyes from the light, and a sterile cannula is attached to a tank so that you can breathe in the hydrogen during the session.

«

I’m not sure that goals scored is a very rigorous comparison, but anyway, it looks pretty amazing. Also – hydrogen?
unique link to this extract

Virginia reports first mammal with bird flu • Daily Progress

»

Virginia has reported its first case of bird flu in a mammal, as the disease continues to ravage the U.S. poultry industry sending the price of eggs skyrocketing.

The Blue Ridge Wildlife Center in Clarke County recently admitted an adult female red fox found in Loudoun County that tested positive for H5N1, highly pathogenic avian influenza.

Upon intake on March 31, the animal was not responding to handling, according to the center.

“She was also exhibiting tremors and nystagmus, a rhythmic, involuntary eye movement that is often seen with brain trauma or disease,” the center said in a statement. “Given these severe neurological signs without any indication of trauma, our top differentials were rabies, distemper, and HPAI.”

…Rabies testing later came back negative, and brain tissue samples confirmed bird flu was the primary cause of the symptoms, according to an April 8 update from the center.

Though a first for Virginia, other states have reported many bird flu cases in mammals, especially dairy cows and free-roaming domestic cats.

«

Only a watching brief! (Thanks Joe S for the link.)
unique link to this extract

‘A billion streams and no fans’: inside a $10m AI music fraud case • WIRED

Kate Knibbs:

»

Jazz (Deluxe) came out in January 2018. Right away, it shot up the Billboard chart and hit No. 1. [Jonathan] Hay was elated. At last, real, measurable success had arrived.

Then, just as suddenly, the album disappeared from the ranking. “Nobody drops off the next week to zero,” says Hay, remembering his confusion. He called other artists to ask if they’d ever seen this before. They hadn’t. Questions piled up. If so many people had listened, why did they suddenly stop? He scanned the internet for chatter. Even a single freaking tweet would have been nice. Nada. Where were the fans? “No one’s talking about the music,” Hay realized.

Pulling up Spotify’s dashboard for artists, Hay scrutinized the analytics for the pair’s work. Listeners appeared concentrated in far-flung places like Vietnam. Things only got stranger from there. Here’s how Hay remembers it: He started receiving notices from distributors, the companies that handle the licensing of indie artists’ music. The distributors were flagging [fellow bandmate Mike] Smith and Hay’s music, from Jazz and from other projects, for streaming fraud and pulling it down. Smith told Hay it was a mistake and that Hay had messed up securing the proper rights for samples. Hay frantically tried to correct the issue, but the flagging persisted.

Hay, panicking, badgered Smith to help him figure out what was happening. Finally, Hay says, Smith offered some answers: Smith had instructed his staff at the medical clinics to stream their songs. It didn’t sound like the full story.

Then, last September, Smith turned up at the heart of another music streaming incident, this one rather epic. The FBI arrested him and charged him in the first AI streaming fraud case in the United States. The government claims that between 2017 and 2024, Smith made over $10m in royalties by using bot armies to continuously play AI-generated tracks on streaming platforms. Smith pleaded not guilty to all charges. (Through his lawyer, Smith declined to be interviewed, so this is very much Hay’s side of the story, corroborated by numerous interviews with people who worked with the two men.)

When Hay found out, he marveled at the idea of his former collaborator managing to get richer than nearly all working musicians without being a household name. “He had a billion streams,” Hay claims, “and no fans.”

«

unique link to this extract

Diseases are spreading. The CDC isn’t warning the public like it was months ago • NPR

Chiara Eisner:

»

To accomplish its mission of increasing the health security of the U.S., the Centers for Disease Control and Prevention states that it “conducts critical science and provides health information” to protect the nation. But since President Trump’s administration assumed power in January, many of the platforms the CDC used to communicate with the public have gone silent, an NPR analysis found.

Many of the CDC’s newsletters have stopped being distributed, workers at the CDC say. Health alerts about disease outbreaks, previously sent to health professionals subscribed to the CDC’s Health Alert Network, haven’t been dispatched since March. The agency’s main social media channels have come under new ownership of the Department of Health and Human Services, emails reviewed by NPR show, and most have gone more than a month without posting their own new content.

“Public health functions best when its experts are allowed to communicate the work that they do in real time, and that’s not happening,” said Kevin Griffis, who served as the director of communications at the CDC until March. “That could put people’s lives at risk.”

«

Under the new Trump regime, that might count as “mission accomplished”.
unique link to this extract

Stories have very simple shapes • The Memory Hole

This is a seven-minute video. I rarely link to videos.

But this one is of Kurt Vonnegut, talking about how the stories we tell have “shapes”. Stay to the end.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

You can now get a chatbot that will generate novel proteins in response to a text prompt. It’s programming for amino acids. CC-licensed photo by Simon Cockell 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. Unfolded. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

OpenAI unites with Jony Ive in $6.5bn deal to create AI devices • The New York Times

Mike Isaac and Cade Metz:

»

On Wednesday, Sam Altman, OpenAI’s chief executive, said the company was paying $6.5bn to buy IO, a one-year-old start-up created by Jony Ive, a former top Apple executive who designed the iPhone. The all-stock deal, which effectively unites Silicon Valley royalty, is intended to usher in what the two men call “a new family of products” for the age of artificial general intelligence, or A.G.I., which is shorthand for a future technology that achieves human-level intelligence.

The deal, which is OpenAI’s biggest acquisition, will bring in Mr. Ive and his team of roughly 55 engineers and researchers. LoveFrom will assume creative and design responsibilities across OpenAI and build hardware that helps people better interact with the technology.

In a joint interview, Mr. Ive and Mr. Altman declined to say what such devices could look like and how they might work, but they said they hoped to share details next year. Mr. Ive, 58, framed the ambitions as galactic, with the aim of creating “amazing products that elevate humanity.”

“We’ve been waiting for the next big thing for 20 years,” Mr. Altman, 40, added. “We want to bring people something beyond the legacy products we’ve been using for so long.”

Mr. Altman and Mr. Ive are effectively looking beyond an era of smartphones, which have been people’s signature personal device since the iPhone debuted in 2007. If the two men succeed — and it is a very big if — they could spur what is known as “ambient computing.” Rather than typing and taking photographs on smartphones, future devices like pendants or glasses that use A.I. could process the world in real time, fielding questions and analyzing images and sounds in seamless ways.

Mr. Altman had invested in Humane, a company that pursued this kind of vision with the creation of an A.I. pin. But the start-up folded not long after its product flopped.

«

You can see that Humane’s product is a sort of platonic ideal for Ive. (By the way, NYT, do you not give people honorifics? He’s Sir Jony Ive.) But the next best would be smart glasses that would let you walk and talk to your device while absorbing the world that you see and acting on it. Things just got interesting.
unique link to this extract

Fortnite is finally back on US iPhones • The Verge

Jay Peters:

»

Fortnite is once again available on the iOS App Store in the US, according to Epic Games. Epic says it has returned to the Epic Games Store and AltStore as well.

Apple kicked Fortnite off the App Store nearly five years ago after Epic Games added its own in-app payment system to the game, which violated Apple’s rules. But after a major court ruling in Epic Games v. Apple that forced Apple to not take fees from purchases made outside of apps, the game is available to play on US iPhones once again.

[After Apple was evidently delaying its release following resubmission earlier this month] Epic asked the judge in the Epic v. Apple case to order Apple to review its Fortnite submission on May 16th. On Monday, the judge said in a filing that Apple is “fully capable of resolving this issue without further briefing or a hearing,” and that if a resolution wasn’t reached, the Apple official who “is personally responsible for ensuring compliance” would have to appear at a hearing next Tuesday.

However, shortly after Fortnite returned to the App Store on Tuesday, Epic and Apple filed a joint notice saying that they have “resolved all issues” from Epic’s May 16th filing. Apple didn’t immediately reply to a request for comment.

Epic also recently rolled out a new promotion to encourage players to use its payment systems: if you use Epic’s system in Fortnite, Rocket League, or Fall Guys on PC, iOS, Android, and the web, the company will give you 20% back in Epic Rewards that can be used for other purchases in its games or on the Epic Games Store.

«

Strange how the threat of having to turn up in court got Apple to hit the “OK” button.
unique link to this extract

‘Fortnite’ players are already making AI Darth Vader swear • WIRED

Megan Farokhmanesh:

»

On Friday, Epic Games announced Darth Vader would be returning to Fortnite as an in-game boss—but this time, players would be able to chat with him through conversational AI. “Ask him all your pressing questions about the Force, the Galactic Empire … or you know, a good strat for the last Storm circle,” Epic said in its announcement.

Unfortunately, players had other plans. Mere hours after Vader appeared in Fortnite, gamers began posting clips of AI Vader going rogue.

“What freaking fucking food is that Darth Vader? Tell me,” says streamer Loserfruit in one clip posted to X. “Freaking? Fucking? Such vulgarity does not become you,” Vader replies. (A spokesperson for Epic Games, Cat McCormack, told WIRED that it pushed a hotfix “within 30 minutes of this happening in-game, so this shouldn’t happen again.”)

Later, in a conversation about possible romantic partners, Loserfruit prompts Vader into replying “You speak of breasts, Loserfruit? I trust you are referring to the armored chestplates.”

In a clip from a different streamer, Vader can be heard talking about carcinogens before saying a slur typically used against queer men that can also be slang for cigarettes. The streamer can be heard screaming “HE SAID IT! HE SAID IT!” before running away in glee.

«

On the one hand, childish, but on the other, shows the impossibility of keeping these things inside what you think are its guardrails. Chatbots are like modern genies, and people really work at getting an extra spell out of them.
unique link to this extract

From chalkboards to chatbots: evaluating the impact of generative AI on learning outcomes in Nigeria • World Bank

Martín De Simone et al:

»

This paper examines whether generative artificial intelligence, specifically large language models (LLMs), can help solve that problem. We evaluate a six-week after-school tutoring program in Nigeria that used a publicly available LLM (ChatGPT-4) to support students in learning English. First-year secondary students from nine public schools in Benin City were invited to participate; from this pool, 52% of eligible students expressed interest, and participants were randomly selected from among them. Those assigned to the intervention attended twelve 90-minute sessions in computer labs, engaging in curriculum-aligned activities guided by teachers. We use a randomized controlled trial (RCT) design to estimate the causal impact of the program on learning outcomes.

We present three main sets of results. First, we show that students selected to participate in the program score 0.31 standard deviation higher in the final assessment that was delivered at the end of the intervention. We find strong statistically-significant intent-to-treat (ITT) effects on all sections of that assessment: English skills (which included the majority of questions, 0.24 σ), digital skills (0.14 σ), AI skills (0.31 σ) and an Item Response Theory (IRT) composite score of each student’s exam (0.26 σ). We also show that the intervention yielded strong positive results on the regular English curricular exam of the third term.

This result is important because the content evaluated in that exam was broader than the one covered during the six weeks of the intervention and included the content of the entire year.

«

This seems like a very significant finding: it’s relatively cheap and it works. In its way it reminds me of the famous paper which found that Tanzanian fishermen who had mobile phones could significant improve their takings because they knew which port would have the largest demand for their catch.
unique link to this extract

I told AI to make me a protein. Here’s what it came up with • Nature

Ewen Callaway:

»

Last month, a team led by Fajie Yuan, a machine-learning scientist at Westlake University in Hangzhou, China, showed that a text-to-protein model his team developed can design functional proteins, including lab-tested enzymes and fluorescent proteins, that are original in their designs and not similar to existing molecules. “We are the first to design a functional enzyme using only text,” Yuan says. “It’s just like science fiction.”

The model, called Pinal, is one of several protein-design AIs that can be directed with ordinary language — as opposed to a protein sequence or the structure-guided specifications typical of most such AIs.

But it’s early days for these bio-AI models, says Anthony Gitter, a computational biologist at the University of Wisconsin–Madison. “I see it as a high-risk, high-reward area,” he says.

Teaching biological AI models to communicate in English (or any language) typically involves exposing them to text descriptions of biological data. Yuan’s team trained Pinal using 1.7 billion descriptions of the structures, functions and other characteristics of different proteins. After some extra training, the model could take a prompt and churn out hundreds of sequence designs. The model has a web interface, and its code and parameters needed to run the model are freely available.

One prompt that the researchers used was ‘Please design a protein that is an alcohol dehydrogenase’, referring to an alcohol-metabolizing enzyme. Yuan and his colleagues then used other computational tools to identify the most promising designs and, working with a biologist collaborator, tested their enzymatic activity.

Two of the eight alcohol dehydrogenase designs successfully catalysed the breakdown of alcohol, albeit much less efficiently than natural enzymes.

«

Like programming, proteins are a bounded problem space, so they’re a good topic for LLMs (or maybe LPMs – large protein models).
unique link to this extract

South Korea’s robot chefs worry human workers – and disappoint customers • Rest of World

Michelle Kim:

»

On sweltering summer days, chef Park Jeong-eun would cook makguksu, an earthy Korean dish made with buckwheat noodles steeped in a tangy, ice-cold broth, topped with spicy gochujang paste. Truck drivers would come from faraway places to the Munmak rest stop, on the highway in the mountainous Gangwon-do province in South Korea, to eat her food.

That was until February 2024, when three robot chefs took over the kitchen at Munmak. The restaurant’s menu has since changed, away from local delicacies like makguksu and slow-cooked beef stews to easily automatable dishes such as ramen, udon, and varieties of Korean stews. The robots speed through 150 meals every hour, nearly double what Park can make by hand.

When longtime patrons learn their beloved menu items are no more, they gasp and walk out the door, she recalled to Rest of World.

“Our customers say the dishes we used to cook tasted much better than what the robots serve now,” Park, 58, said. “Even though the robots have lightened my workload, I’ve lost my sense of pride in our food.”

Park now finds refuge scrubbing dishes in the back of the kitchen, away from the counter, where customers barrage her with harsh complaints about the food. Sometimes, they return their ramen bowls untouched in protest, she said.

«

But! South Korea needs its robots because the population is ageing: those over 60 (as Park will soon be) make up a quarter of the workforce. Maybe get used to the new taste of the ramen?
unique link to this extract

The spy factory: how Russia used Brazil to create deep-cover spies • The New York Times

Michael Schwirtz and Jane Bradley:

»

For years, a New York Times investigation found, Russia used Brazil as a launchpad for its most elite intelligence officers, known as illegals. In an audacious and far-reaching operation, the spies shed their Russian pasts. They started businesses, made friends and had love affairs — events that, over many years, became the building blocks of entirely new identities.

Major Russian spy operations have been uncovered in the past, including in the United States in 2010. This was different. The goal was not to spy on Brazil, but to become Brazilian. Once cloaked in credible back stories, they would set off for the United States, Europe or the Middle East and begin working in earnest.

The Russians essentially turned Brazil into an assembly line for deep-cover operatives like Mr. Shmyrev.

One started a jewelry business. Another was a blond, blue-eyed model. A third was admitted into an American university. There was a Brazilian researcher who landed work in Norway, and a married couple who eventually went to Portugal.

Then it all came crashing down.

For the past three years, Brazilian counterintelligence agents have quietly and methodically hunted these spies. Through painstaking police work, these agents discovered a pattern that allowed them to identify the spies, one by one.
Agents have uncovered at least nine Russian officers operating under Brazilian cover identities, according to documents and interviews. Six have never been publicly identified until now. The investigation has already spanned at least eight countries, officials said, with intelligence coming from the United States, Israel, the Netherlands, Uruguay and other Western security services.

«

Easy to forget that all this stuff goes on, and on, under the surface.
unique link to this extract

Judge slams lawyers for ‘bogus AI-generated research’ • The Verge

Emma Roth:

»

A California judge slammed a pair of law firms for the undisclosed use of AI after he received a supplemental brief with “numerous false, inaccurate, and misleading legal citations and quotations.” In a ruling submitted last week, Judge Michael Wilner imposed $31,000 in sanctions against the law firms involved, saying “no reasonably competent attorney should out-source research and writing” to AI, as pointed out by law professors Eric Goldman and Blake Reid on Bluesky.

“I read their brief, was persuaded (or at least intrigued) by the authorities that they cited, and looked up the decisions to learn more about them – only to find that they didn’t exist,” Judge Wilner writes. “That’s scary. It almost led to the scarier outcome (from my perspective) of including those bogus materials in a judicial order.”

As noted in the filing, a plaintiff’s legal representative for a civil lawsuit against State Farm used AI to generate an outline for a supplemental brief. However, this outline contained “bogus AI-generated research” when it was sent to a separate law firm, K&L Gates, which added the information to a brief. “No attorney or staff member at either firm apparently cite-checked or otherwise reviewed that research before filing the brief,” Judge Wilner writes.

When Judge Wilner reviewed the brief, he found that “at least two of the authorities cited do not exist at all.”

«

As I’ve said previously, one has to think that this might have slipped past other judges if the two sides aren’t careful enough.
unique link to this extract

From birth to gene-edited in six months: custom CRISPR therapy breaks speed limits • Ars Technica

Beth Mole:

»

researchers in Philadelphia appear to have successfully treated a six-month-old baby boy, called KJ, with a personalized CRISPR gene-editing therapy. The treatment corrects an ultra-rare mutation in KJ that breaks a liver enzyme. That enzyme is required to convert ammonia, a byproduct of metabolism, to urea, a waste product released in urine. Without treatment, ammonia would build up to dangerous levels in KJ—and he would have a 50% chance of dying in infancy.

While the gene-editing treatment isn’t a complete cure, and long-term success is still uncertain, KJ’s condition has improved and stabilized. And the treatment’s positive results appear to be a first for personalizing gene editing.

Now, who doesn’t love a good story about a seemingly miraculous medical treatment saving a cute, chubby-cheeked baby? But, this story delivers more than an adorable bundle of joy; the big triumph is the striking timeline of the treatment’s development—and the fact that it provides a template for how to treat other babies with ultra-rare mutations.

«

The mutation was recognised within days (on its own, a remarkable bit of medicine) but the process of taking cells, finding the targets in the infant’s genome, proving it, getting approval, doing animal tests, all happened at breakneck speed – and, apparently, successfully.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

What has gone wrong in Apple’s current approach to making money versus making products? Some people think they know the answer. CC-licensed photo by Mike Deerkoski 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. A matter of timing. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

The 15 biggest announcements at Google I/O 2025 • The Verge

Emma Roth:

»

Google just wrapped up its big keynote at I/O 2025. As expected, it was full of AI-related announcements, ranging from updates across Google’s image and video generation models to new features in Search and Gmail.

But there were some surprises, too, like a new AI filmmaking app and an update to Project Starline. If you didn’t catch the event live, you can check out everything you missed in the roundup below.

«

With the proviso that these things often don’t reach the real world for a couple of years, if at all, there are some interesting things in there. A selection:

»

• Google is launching a new AI filmmaking app called Flow. The tool uses Veo, Imagen, and Gemini to create eight-second AI-generated video clips based on text prompts and / or images.

• Xreal and Google are teaming up on Project Aura, a new pair of smart glasses that use the Android XR platform for mixed-reality devices.

• Project Astra could already use your phone’s camera to “see” the objects around you, but the latest prototype will let it complete tasks on your behalf, even if you don’t explicitly ask it to.

• Google is launching Search Live, a feature that incorporates capabilities from the AI assistant. By selecting the new “Live” icon in AI Mode or Lens, you can talk back and forth with Search while showing what’s on your camera.

• Google Meet is launching a new feature that translates your speech into your conversation partner’s preferred language in near real-time. The feature only supports English and Spanish for now.

• Gmail’s smart reply feature, which uses AI to suggest replies to your emails, will now use information from your inbox and Google Drive to prewrite responses that sound more like you.

«

Soon the AI will write the email, and the recipient’s AI will read it, and we can get on with creating the great works of art that have been lying immanent for so long.
unique link to this extract

Is the US in a “high-level equilibrium trap”? • Noahpinion

Noah Smith:

»

The U.S. has been a slow-growing country at the technological frontier for as long as almost all of us have been alive. If your country generally grows at 2%, you can expect to see your living standards quadruple over your lifetime. That’s much better than nothing, but it means that in the shorter term — over a five-year or ten-year period — your economic fortunes will be primarily determined by random shocks, not by the slow and steady march of technological improvement. A spell of unemployment, a medical bankruptcy, a decline in the price of your house, the loss of a government contract or a big customer — any of these could wipe out many years of slow improvement in living standards.

In other words, to most Americans, risks loom larger than opportunities. If everything stays the same, then they’ll continue to be wealthy and comfortable; if something changes, they might not. In an environment like that, it makes sense to be afraid of change, because change means risk.

For most of Americans’ lives, technological progress has been a major source of risk. The advent of the internet put encyclopedia salesmen and term life insurance salesmen out of a job. Hybrid cars from Japan put competitive pressure on traditional carmakers. Flip-phone makers were wiped out by smartphones. Electronic trading made many human “specialists” obsolete. And so on and so forth, throughout the economy. At the aggregate level, these innovations drove growth in living standards. But at an individual level, having the technology in your industry change was generally a source of peril.

By contrast someone who grew up in modern China has experienced something utterly different. Over the course of their lifetime, rapid technological progress has radically transformed their lives and the lives of the people around them, allowing them to experience a level of comfort and security utterly undreamt of by their grandparents.

Meanwhile, the risks from new technology were pretty low. In a fast-growing economy, if your job gets automated, you can often just go get a better one.

«

Smith’s post is free to read, and gives a broader view of the one expressed yesterday – that the US is falling behind, while China is surging ahead.
unique link to this extract

Millions of consumers could get £70 after unfair fees ruling against Mastercard • BBC News

Vishala Sri-Pathma:

»

Millions of shoppers could get up to £70 each after a court ruled historic fees administered by card provider Mastercard were unfair.

The decision comes after a long-running legal case going back almost a decade, brought forward by a former financial ombudsman.

Walter Merricks argued that shoppers were charged higher prices after fees were wrongly levied on transactions made over a 15-year period between 1992 and 2008.

It is not necessary to have owned a Mastercard at any point to be eligible for compensation. Mastercard declined to comment on the court ruling.

Consumers are eligible to claim compensation if they lived in England, Wales or Northern Ireland for at least three months between June 1997 and June 2008, and bought goods or services from UK businesses that accepted Mastercard credit cards.

For those who live in Scotland, the starting point is May 1992.

The entire settlement is for £200m, with £100m ringfenced for consumers who have until the end of this year to claim and if the expected 5% of claimants – 2.5 million people – come forward, then each will receive £45.
If fewer people apply, payments will be capped at £70 per claimant.

Mr Merricks said consumers would soon be able to register to receive a payout by completing an online form.

«

(Un)Helpfully the BBC didn’t link to the page where one could find how to make a claim. You’ll find that here.
unique link to this extract

Task estimation: conquering Hofstadter’s Law • These Are Systems

Jacob Bayless:

»

This is a true story: an experienced team of developers was tasked with a full ground-up, from-scratch rewrite of a mature software stack that had been built over many years of development. The legacy architecture had been stretched to its limits and was now restricting the company’s growth. A complete redesign was needed; backwards-compatible on the surface, but entirely different under the hood. Everyone knew this would be a difficult undertaking, but the team had thoroughly considered all the alternatives, and decided this was the best way forward.

Marketing wanted to know when the new software would be ready to ship. The dev team put together a roadmap to deliver a minimum viable product, planned out the workload, and estimated how long it would take to get to the MVP: six months.

It shipped two years later.

It would be an understatement to say this was a huge problem. But the problem wasn’t the two years of development time; this company had healthy revenue to go on, and the long-term payoff from the technology leap would eventually be worth that incubation period.

The damage was caused by the estimate itself. You see, the marketing team interpreted this estimate as a commitment, and began informing customers about when the product would be ready. Sales inquiries came streaming in. Some of those customers had customers of their own, and they had started making their own plans and timelines based on this information. When the deadline rolled around and the product was nowhere near ready, there was a lot of damage control needed.

Customers were not happy. The marketing team was not happy. The company was not happy. And the dev team found themselves in the crosshairs.

So what went wrong?

«

What he discovered was that in a hierarchical project – A depends on the completion of B depends on the completion of C – the time taken for each stage follows a lognormal distribution relative to its estimated completion time. The “educated guess” is actually the median completion time; the real mean is 1.6x that; the 95th percentile (ie in 95% of cases it finishes by then) is 5x the median. The 99th percentile? 10x.

Which explains why a lot of software – and other! – projects overrun. [(Douglas) Hofstadter’s Law: It always takes longer than you expect, even when you take into account Hofstadter’s Law.]
unique link to this extract

Chicago Sun-Times prints summer reading list full of fake books • Ars Technica

Benj Edwards:

»

On Sunday, the Chicago Sun-Times published an advertorial summer reading list containing at least 10 fake books attributed to real authors, according to multiple reports on social media. The newspaper’s uncredited “Summer reading list for 2025” supplement recommended titles including “Tidewater Dreams” by Isabel Allende and “The Last Algorithm” by Andy Weir—books that don’t exist and were created out of thin air by an AI system.

The creator of the list, Marco Buscaglia, confirmed to 404 Media that he used AI to generate the content. “I do use AI for background at times but always check out the material first. This time, I did not and I can’t believe I missed it because it’s so obvious. No excuses,” Buscaglia said. “On me 100% and I’m completely embarrassed.”

A check by Ars Technica shows that only five of the 15 recommended books in the list actually exist, with the remainder being fabricated titles falsely attributed to well-known authors. AI assistants such as ChatGPT are well-known for creating plausible-sounding errors known as confabulations, especially when lacking detailed information on a particular topic. The problem affects everything from AI search results to lawyers citing fake cases.

On Tuesday morning, the Chicago Sun-Times addressed the controversy on Bluesky. “We are looking into how this made it into print as we speak,” the official publication account wrote. “It is not editorial content and was not created by, or approved by, the Sun-Times newsroom. We value your trust in our reporting and take this very seriously. More info will be provided soon.”

«

According to 404 Media in a followup story (paywalled), the content was generated by “magazine giant Hearst”. Rosebud has the last laugh.

But it was part of a 64-page promotional supplement. Journalists – anyone, in fact – hates writing those things, because they’re utter filler. If they’d checked the book titles and put real ones in, I seriously doubt anyone would have been any the wiser.
unique link to this extract

The US hasn’t seen a human bird flu case in three months. Experts wonder why • The Boston Globe

Mike Stobbe and Jonel Aleccia:

»

Health officials are making a renewed call for vigilance against bird flu, but some experts are puzzling over why reports of new human cases have stopped.

Has the search for cases been weakened by government cuts? Are immigrant farm workers, who have accounted for many of the U.S. cases, more afraid to come forward for testing amid the Trump administration’s deportation push? Is it just a natural ebb in infections?

“We just don’t know why there haven’t been cases,” said Jennifer Nuzzo, director of the Pandemic Center at Brown University. “I think we should assume there are infections that are occurring in farmworkers that just aren’t being detected.”

The H5N1 bird flu has been spreading widely among wild birds, poultry and other animals around the world for several years, and starting early last year became a problem in people and cows in the U.S.

In the last 14 months, infections have been reported in 70 people in the U.S. — most of them workers on dairy and poultry farms. One person died, but most of the infected people had mild illnesses.

The most recent infections confirmed by the Centers for Disease Control and Prevention were in early February in Nevada, Ohio and Wyoming. California had been a hotspot, with three-quarters of the nation’s infections in dairy cattle. But testing and cases among people have fallen off. At least 50 people were tested each month in late 2024, but just three people were tested in March, one in April and none in May so far, state records show. Overall, the state has confirmed H5N1 infections in 38 people, none after Jan. 14.

During a call with U.S. doctors this month, one CDC official noted that there is a seasonality to bird flu: Cases peak in the fall and early winter, possibly due to the migration patterns of wild birds that are primary spreaders of the virus.

That could mean the U.S. is experiencing a natural — maybe temporary — decline in cases.

«

Well done, Sherlock Holmes. Let’s just hope that it doesn’t mean it’s having a summer spent producing a version that’s more infectious to humans. (Thanks Joe S for the link.)
unique link to this extract

‘Significant amount’ of private data stolen in Legal Aid hack – BBC News

Aoife Walsh and Graeme Baker:

»

A “significant amount” of private data including details of domestic abuse victims has been hacked from Legal Aid’s online system.

The Ministry of Justice said the agency’s services were hacked in April and data dating back to 2010 was downloaded. The BBC understands that more than two million pieces of information were taken.

The breach covers all areas of the aid system – including domestic abuse victims, those in family cases and others facing criminal prosecution.

“This data may have included… addresses of applicants, dates of birth, national ID numbers, criminal history, employment and financial data such as… debts and payments,” the MoJ said. The agency’s chief executive Jane Harbottle apologised, saying she understood the news “will be shocking and upsetting for people”.

Justice minister Sarah Sackman told the House of Commons that there was no indication as yet that any other government systems had been affected by the breach. The MoJ said that while the initial cyber-attack was detected in April, it has since become apparent that the incident was “more extensive than originally understood”.

It also warned the public to be alert for any suspicious activity, including unknown messages or phone calls, and to update any potentially exposed passwords. “If you are in doubt about anyone you are communicating with online or over the phone you should verify their identity independently before providing any information to them,” it said.

The ministry said it was working with the National Crime Agency and the National Cyber Security Centre, and has informed the Information Commissioner.

«

There seems to be a spate of these attacks hitting British organisations at the moment. Hard to figure out if we’re just noticing it more, or whether this is some sort of concentrated assault.
unique link to this extract

Apple turnover • Hypercritical

John Siracusa:

»

Try to make great things, and the money will surely follow. It’s a strategy that’s simple to explain, but almost impossible for any company to follow.

…As far as I’m concerned, the only truly mortal sin for Apple’s leadership is losing sight of the proper relationship between product virtue and financial success—and not just momentarily, but constitutionally, intransigently, for years. Sadly, I believe this has happened.

The preponderance of the evidence is undeniable. Too many times, in too many ways, over too many years, Apple has made decisions that do not make its products better, all in service of control, leverage, protection, profits—all in service of money.

To be clear, I don’t mean things like charging exorbitant prices for RAM and SSD upgrades on Macs or taking too high a percentage of in-app purchases in the App Store. Those are venial sins. It’s the apparently unshakable core beliefs that motivate these and other poor decisions that run counter to the virtuous cycle that led Apple out of the darkness all those years ago.

Apple, as embodied by its leadership’s decisions over the past decade or more, no longer seems primarily motivated by the creation of great products. Time and time again, its policies have made its products worse for customers in exchange for more power, control, and, yes, money for Apple.

The iPhone is a better product when people can buy ebooks within the Kindle app. And yet Apple has fought this feature for the past 14 years, to the tune of millions of dollars in legal fees, and has only relented due to a recent court order (which they continue to appeal).

…The best leaders can change their minds in response to new information. The best leaders can be persuaded. But we’ve had decades of strife, lawsuits, and regulations, and Apple has stubbornly dug in its heels even further at every turn. It seems clear that there’s only one way to get a different result.

«

Quietly but certainly, the calls are multiplying for Tim Cook to leave or be thrown out. And I agree with Siracusa.
unique link to this extract

Huawei’s first trifold is a great phone that you shouldn’t buy • The Verge

Dominic Preston:

»

this isn’t a phone you’re meant to buy, at least not outside China. It’s a phone you’re meant to gawk at on the internet, to marvel at Huawei’s technological prowess, to ooh and ahh about its many and varied folds. This is Huawei showing off, proving to the world that it’s still got it. And in fairness, it has.

As I sit and write this — more than six months after Huawei first released the Mate XT in China — it’s still the only one of its kind. Rumor has it that Samsung has a trifold ready to show off this year, but it hasn’t yet. And by the time it does, odds are Huawei will have spent a full year as the only player in the game.

That might ring alarm bells in your head. This must be undercooked tech, you think, rushed out the door to beat everyone else to market. But the most surprising thing about the Mate XT is that it only occasionally feels first-gen.

There’s a hint of it in the multitasking, which refuses to allow you to fully open three apps at a time, pinning each to one of the three screen segments. Or when the fully open screen often doesn’t quite go entirely flat, which is more annoying than any crease will ever be. And you notice it when you open the phone, or close it, and the app you’re using seems to briefly reboot itself, losing your spot in a long article or (once, infuriatingly) discarding a Letterboxd review that was almost entirely finished. I’ve learned not to change the configuration while doing something, just to be safe.

«

The tri-fold part means it can all be one continuous screen, but two-thirds close in when you, well, fold it. So there’s an outside screen when it’s in your pocket, like any phone. Except thicker. And heavier. If someone can really crack the weight point, then this might be a viable product.
unique link to this extract

Errata, corrigenda and ai no corrida: an addendum about the mystery from the other day of “who uses correction fluid?” TEACHERS. They use it for correcting homework and class handouts and posters and similar, because mistakes in printing often aren’t spotted before committing to hard copy, and would be too expensive/time-consuming to correct.

The job of radiologists was thought to be under threat from AI. Instead, it’s just enhanced it. CC-licensed photo by Navy Medicine 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.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. See-through. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Judge calls on Apple to resolve Fortnite app submission after ruling • CNBC

Lora Kolodny:

»

Apple must work out its latest issues with Epic Games, or head back to court to prove it has a legal basis for delaying the restoration of the popular Fortnite game to its iOS App Store, a judge ordered on Monday.

Fortnite recently re-submitted its game but was blocked by Apple, Epic Games said Friday. The company later filed a motion to enforce the earlier injunction from the court.

US District Judge Yvonne Gonzalez Rogers wrote in the Monday order that “Apple is fully capable of resolving this issue without further briefing or a hearing.”

If the companies do not resolve the current conflict on their own, the judge wrote, “the Apple official who is personally responsible for ensuring compliance shall personally appear” at a forthcoming hearing in the Northern District of California scheduled for May 27.

Apple said in a statement on Friday that it did not remove Fortnite from alternative distribution marketplaces.

…On Monday, Epic Games CEO Tim Sweeney said in a post on X that “Apple didn’t accept or reject our Fortnite submission. They simply said they were going to ignore it until after the 9th Circuit Court rules on their stay request, which would be in late May or June.”

In the Monday order, Rogers said that after 12 days, the Ninth Circuit Court of Appeals has still not granted Apple a stay of the injunction.

“Briefing shall occur on the schedule listed below and shall include the legal authority upon which Apple contends that it can ignore this Court’s order,” the judge wrote.

«

I wonder who that Apple official is? I’d love to know. Because they’re sure taking a beating at the moment.
unique link to this extract

In the future, China will be dominant. The US will be irrelevant • The New York Times

Kyle Chan is a researcher at Princeton University:

»

It doesn’t matter that Washington and Beijing have reached an inconclusive and temporary truce in Mr. Trump’s trade war. The U.S. president immediately claimed it as a win, but that only underlines the fundamental problem for the Trump administration and America: a shortsighted focus on inconsequential skirmishes as the larger war with China is being decisively lost.

Mr. Trump is taking a wrecking ball to the pillars of American power and innovation. His tariffs are endangering U.S. companies’ access to global markets and supply chains. He is slashing public research funding and gutting our universities, pushing talented researchers to consider leaving for other countries. He wants to roll back programs for technologies like clean energy and semiconductor manufacturing and is wiping out American soft power in large swaths of the globe.

China’s trajectory couldn’t be more different.

It already leads global production in multiple industries — steel, aluminum, shipbuilding, batteries, solar power, electric vehicles, wind turbines, drones, 5G equipment, consumer electronics, active pharmaceutical ingredients and bullet trains. It is projected to account for 45% — nearly half — of global manufacturing by 2030. Beijing is also laser-focused on winning the future: In March it announced a $138bn national venture capital fund that will make long-term investments in cutting-edge technologies such as quantum computing and robotics, and increased its budget for public research and development.

…China faces its own serious challenges. A prolonged real estate slump continues to drag on economic growth, though there are signs that the sector may be finally recovering. Longer-term challenges also loom, such as a shrinking work force and an aging population. But skeptics have been predicting China’s peak and inevitable fall for years, only to be proved wrong each time. The enduring strength of a state-dominated Chinese system that can pivot, change policy and redirect resources at will in service of long-term national strength is now undeniable, regardless of whether free-market advocates like it.

Mr. Trump’s blinkered obsession with short-term Band-Aids like tariffs, while actively undermining what makes America strong, will only hasten the onset of a Chinese-dominated world.

«

In November 2016, after Trump was elected, I wrote a piece which argued many of the same points, though China at that point hadn’t achieved the same ascendance in so many industries. Now, it’s even more true.
unique link to this extract

Uber to introduce fixed-route shuttles in major US cities designed for commuters • TechCrunch

Rebecca Bellan:

»

Ride-hail and delivery giant Uber is introducing cheap, fixed-route rides along busy corridors during weekday commute hours in major U.S. cities — one solution to a world that feels, for most people, more expensive every day. 

Starting Wednesday, riders in Baltimore, Boston, Chicago, Dallas, New York City, Philadelphia, and San Francisco will be able to save 50% off the price of an UberX trip by booking with Uber’s new “Route Share” feature. 

The company announced Route Share and other new features and discounts designed to help customers save money on rides and deliveries at its annual Go-Get event. The aim is to attract and maintain a loyal customer base that continues to use the Uber app in spite of outside economic pressures.

The commuter shuttles will drive between pre-set stops every 20 minutes, according to Sachin Kansal, Uber’s chief product officer. He noted that there will be dozens of routes in each launch city — like between Williamsburg and Midtown in NYC. The routes, which are selected based on Uber’s extensive data on popular travel patterns, might have one or two additional stops to pick up other passengers. To start, riders will only ever have to share the route with up to two other co-riders. 

«

Perhaps they could expand it so there are more co-riders – maybe 20 or so? They could work on a snappier name for it, maybe using some of the letters from Uber? Reub? Breus? How about, and just go along with me here – bus?
unique link to this extract

My brain finally broke • The New Yorker

Jia Tolentino:

»

I was on a walk the other day with another journalist, and I asked her whether she was also experiencing this—a slackening of the usual reflexive fact-checking impulse. She told me that she still researched what she saw on her feeds if it was something important, or something relevant to her work.

I did, too, I told her. (Well, mostly.) But there is now a category of things I see online which I register simply as indications that the world is slipping beyond my comprehension. A video of a giant Pikachu fleeing from the police during demonstrations in Turkey. A clip of former Governor Andrew Cuomo saying, “As a New Yorker, I am Black, I am gay, I am disabled, I am a woman seeking to control her health and her choices.”

I see Temu ads for uncanny products—an inflatable waterslide of inhuman design, for instance, pictured alongside digitally rendered children and toys. I click on a waterslide ad to investigate, and I get a security question asking me to “please click on the type of fruit that appears most frequently.” There are oranges and a pear and a lemon and a basketball and a baguette placed against a swirly backdrop. Sleepily, I think, Appears most frequently in . . . the grocery store?

And then I remember that this question is a purely digital one, and that the answers are probably being used to train A.I.

«

unique link to this extract

Apple is trying to get ‘LLM Siri’ back on track • The Verge

Wes Davis:

»

…[Mark] Gurman has reported [at Bloomberg] in the past that Apple is working on what it’s internally calling “LLM Siri” — a reworked, generative AI version of the company’s digital assistant. Apple’s previous approach of merging the assistant with the existing Siri hasn’t been working. Gurman describes in great detail a number of reasons why

Now the company is trying to rejigger its approach. Part of that is a total overhaul of Siri, rather than just trying to make generative AI work in concert with the old Siri. According to Gurman, Apple has its AI team in Zurich working on a new architecture that will “entirely build on an LLM-based engine.” Gurman reported in November last year that the company was working on this, and the idea is that it will make the assistant “more believably conversational and better at synthesizing information.”

Another part of the solution is leveraging iPhones and differential privacy to improve Apple’s synthesized data — comparing fake training data with language from iPhone users’ emails, but doing so on-device and sending only the synthesized data back to Apple for AI training. And one way the company is discussing improving Siri is letting the LLM version loose on the web to “grab and synthesize data from multiple sources.” Basically, Siri as an AI web search tool not unlike Perplexity, which is one of the companies Apple has approached about partnering for AI search in Safari.

Whatever the outcome, apparently Giannandrea won’t be a direct part of it, having been taken off of product development, Siri, and robotics projects in the spring. According to Gurman, Apple execs have talked about putting him “on a path to retirement,” but are concerned that some of the research and engineering folks he brought with him would leave with him, too.

«

Apple has been late to this, and is showing a lack of ruthlessness in getting things done. Let’s all be nice to Giannandrea!
unique link to this extract

AI was coming for radiologists’ jobs. So far, they’re just more efficient • The New York Times

Steve Lohr:

»

Nine years ago, one of the world’s leading artificial intelligence scientists singled out an endangered occupational species.

“People should stop training radiologists now,” Geoffrey Hinton said, adding that it was “just completely obvious” that within five years A.I. would outperform humans in that field.

Today, radiologists — the physician specialists in medical imaging who look inside the body to diagnose and treat disease — are still in high demand. A recent study from the American College of Radiology projected a steadily growing work force through 2055.

Dr. Hinton, who was awarded a Nobel Prize in Physics last year for pioneering research in A.I., was broadly correct that the technology would have a significant impact — just not as a job killer.

That’s true for radiologists at the Mayo Clinic, one of the nation’s premier medical systems, whose main campus is in Rochester, Minn. There, in recent years, they have begun using A.I. to sharpen images, automate routine tasks, identify medical abnormalities and predict disease. A.I. can also serve as “a second set of eyes.”

“But would it replace radiologists? We didn’t think so,” said Dr. Matthew Callstrom, the Mayo Clinic’s chair of radiology, recalling the 2016 prediction. “We knew how hard it is and all that is involved.”

…The predicted extinction of radiologists provides a telling case study. So far, A.I. is proving to be a powerful medical tool to increase efficiency and magnify human abilities, rather than take anyone’s job.

When it comes to developing and deploying A.I. in medicine, radiology has been a prime target. Of the more than 1,000 A.I. applications approved by the Food and Drug Administration for use in medicine, about three-fourths are in radiology. A.I. typically excels at identifying and measuring a specific abnormality, like a lung lesion or a breast lump.

«

unique link to this extract

An $8.4bn money launderer has been operating for years on US soil • Wired via Ars Technica

Andy Greenberg:

»

As the underground industry of crypto investment scams has grown into one of the world’s most lucrative forms of cybercrime, the secondary market of money launderers for those scammers has grown to match it. Amid that black market, one such Chinese-language service on the messaging platform Telegram blossomed into an all-purpose underground bazaar: It has offered not only cash-out services to scammers but also money laundering for North Korean hackers, stolen data, targeted harassment-for-hire, and even what appears to be sex trafficking. And somehow, it’s all overseen by a company legally registered in the United States.

According to new research released by crypto-tracing firm Elliptic, a company called Xinbi Guarantee has since 2022 facilitated no less than $8.4bn in transactions via its Telegram-based marketplace, prior to Telegram’s actions in recent days to remove its accounts from the platform.

Money stolen from scam victims likely represents the “vast majority” of that sum, according to Elliptic’s cofounder Tom Robinson. Yet even as the market serves Chinese-speaking scammers, it also boasts on the top of its website—in Mandarin—that it’s registered in Colorado.

“Xinbi Guarantee has served as a giant, purportedly US-incorporated illicit online marketplace for online scams that primarily offers money laundering services,” says Robinson. He adds, though, that Elliptic has also found a remarkable variety of other criminal offerings on the market: child-bearing surrogacy and egg donors, harassment services that offer to threaten or throw feces at any chosen victim, and even sex workers in their teens who are likely trafficking victims.

«

Crypto, the gift that just keeps on giving.
unique link to this extract

A personality test said I was Darth Vader. I’m so proud • The Times

Harry Wallop:

»

Last week a judge at an employment tribunal ruled that Lorna Rooke, an NHS worker, had suffered a workplace “detriment” after she was compared to Darth Vader in an online personality test that someone had taken on her behalf. Although Rooke lost various aspects of her employment tribunal, including unfair dismissal and disability discrimination, she won nearly £29,000 in compensation for the “detriment”.

“Darth Vader is a legendary villain of the Star Wars series and being aligned with his personality is insulting,” Judge Kathryn Ramsden said.

Really? Out of curiosity I tracked down a “What Star Wars Character Are You?” test online. There are many of them and the tribunal ruling does not make it clear which one it was. I also came out as Darth Vader. “You’re the leader of your friend group and everyone always wants to hear what you have to say. Just remember to go easy on your loved ones — they’re trying their best!”

And then I took another one. This time I was Princess Leia, which hints at the central problem with these personality tests: they are just not very reliable.

The employment tribunal said Rooke and her colleagues took a Myers-Briggs type indicator (MBTI) test. This is not correct, according to Myers-Briggs itself, which wants to distance itself from people thinking that its tests encourage people to engage in workplace bullying.

…As Adam Grant, professor of organisational psychology at Wharton business school, said: “When it comes to accuracy, if you put a horoscope on one end and a heart monitor on the other, the MBTI falls about half way in between.”

He has many objections but his main one is that it presents various traits as black and white in order to ensure that the 16 different personality types are discrete. Thinking and feeling, for instance, are, in the Myers-Briggs world, two separate entities. You are a thinker or a feeler; you cannot be both, which is clearly nonsense.

«

unique link to this extract

Still booting after all these years: the people stuck using ancient Windows computers • BBC Future

Thomas Germain:

»

The trains in San Francisco’s Muni Metro light railway, for example, won’t start up in the morning until someone sticks a floppy disk into the computer that loads DOS software on the railway’s Automatic Train Control System (ATCS). Last year, the San Francisco Municipal Transit Authority (SFMTA) announced its plans to retire this system over the coming decade, but today the floppy disks live on. (The SFMTA did not respond to a request for comment.)

In a brightly lit room in San Diego, California, you’ll find two of the biggest printers you’ve ever seen, each hooked up to servers running Windows 2000, an operating system named for the year it was released. “We call ’em boat anchors,” says John Watts, who handles high-end printing and post-processing for fine art photographers. The printers are LightJets, gigantic machines that use light, rather than ink, to print on large-format photographic paper. Watts says the result is an image of unparalleled quality.

Long out of production, the few remaining LightJets rely on the Windows operating systems that were around when these printers were sold. “A while back we looked into upgrading one of the computers to Windows Vista. By the time we added up the money it would take to buy new licenses for all the software it was going to cost $50,000 or $60,000 [£38,000 to £45,000],” Watts says. “I can’t stand Windows machines,” he says, “but I’m stuck with them.”

It’s a common predicament with specialised hardware. Scott Carlson, a woodworker in Los Angeles, is steeped in the world of Microsoft thanks to CNC machines, robotic tools that cut or shape wood and other materials based on computer instructions.

“Our workhorse machine runs on Windows XP because it’s older. That thing is a tank,” Carlson says. But the same can’t be said for the operating system. “We actually had to send the computer back to get completely rebuilt a few years ago because XP was getting more and more errors,” he says “It was practically a brick.”

«

unique link to this extract

Errata, corrigenda and ai no corrida: none notified

Reports suggest correction fluids such as Tipp-Ex, Liquid Paper and Wite-Out are a $2bn market annually. But who on earth uses them? CC-licensed photo by J B 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. Friday’s Overspill arrived last Saturday. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

My school banned phones for the year. Here’s what happened • Fit To Teach

“Fit To Teach” is a teacher in an American secondary school:

»

Last year I was battling the greatest entertainment system the world ever unleashed. A student would listen to my pleas, say they would get up, and then immediately fuck off on their phone again. Why would you play in the gym when you could sit against the wall and watch endless entertainment personally curated for your tastes?

Now I battle boredom, and that’s a winnable fight. Now when I give my two chicks speech [telling pupils they might not like doing something, but they have to], they usually respond by participating. That’s because the other option is boredom. Sit there, do nothing, and look out the non-existent windows our school doesn’t have. I’m not fighting against brilliant social scientists getting paid millions to figure out ways to capture human attention, I’m fighting against sitting on your ass doing nothing. And my class is awesome compared to boredom.

Now don’t get me wrong. This isn’t a magic bullet. Banning phones won’t save Gen-Z. I once watched our dean hand cellphones back to students in the cafeteria. When she rolled the cellphone cart through, it was like watching a crowd of hyenas catching the scent of stinking flesh. Kids stopped conversations mid-sentence and lifted their heads to watch that cart roll by.

The moment a boy got up from his table, a storm of students jumped to their feet to beat him to the cart. The dean started screaming if you come near me without being called up you’ll go to the back of the line. It took ten full minutes to get everyone seated before she felt comfortable enough to call kids up table by table.

That’s the kind of addiction teachers are dealing with. The kids are not all right, whether we ban the phones or not.

But when I compare the seven years I had battling the cellphone in the classroom, vs almost an entire year of phone free schooling, there is no comparison. Our kids are smarter, more social, and more motivated to do the things they actually want to accomplish in this world when they don’t have a Pavlovian vibration derailing their attention every 20 seconds.

«

On a sort-of related note, I’m always amazed by the number of people who walk along streets gazing down at phones in their hands, hoping that peripheral vision will save them from calamity. It might, might not (your unlocked phone might just be stolen from your hands), but is it really so important you can’t make the choice between stopping to read and just walking?
unique link to this extract

Spotify is scrambling to remove dozens of podcasts promoting online prescription drug sales • CNN Business

Clare Duffy:

»

If you search “Adderall” on Spotify’s podcast page, you’ll find health podcasts about ADHD, shows about addiction recovery and comedy podcasts where hosts talk about using the medication. You may also come across multiple pages masquerading as podcasts that direct users to buy drugs from potentially dangerous and illegal online pharmacies.

The intention of many of these pages is obvious from their names. Podcasts with titles, such as “My Adderall Store” — which has a link in the episode description to a site that purportedly sells Adderall, as well as potentially addictive pain medications like Oxycodone and Vicodin, among other drugs — were listed within the first 50 suggested results, a CNN review this week found.

CNN identified dozens of these fake podcasts across Spotify, advertising sales of medications ranging from Methadone to Ambien, in some cases claiming that the drugs can be purchased without a prescription, which is illegal in the United States.

Spotify is now scrambling to remove these fake podcast pages, which violate its rules and which, at best, may be spam and, and worst, could direct users to sites that violate the law.

Within hours after CNN sent Spotify a list of 26 podcasts promoting online pharmacies that were live as of Thursday afternoon, the platform had removed them. A spokesperson said they violated its rules, which prohibit illegal and spam content. Still, even after Spotify removed the podcasts CNN sent on Thursday, others continued to appear on the platform Friday morning.

«

Always the same story: the scale’s too big to manually check it first, so you only check it if someone objects afterwards, so spammers and other miscreants exploit it, and then you discover you’ve been enormously compromised, and the journalists do the objection, and your reputation takes a hit for about 15 minutes, and then you ignore it all again.
unique link to this extract

Keep calm (but delete your nudes): the new rules for travelling to and from Trump’s America • The Guardian

Arwa Mahdawi:

»

Searches, to be clear, are still very rare. “Claims that CBP is searching more electronic media due to the administration change are false,” CBP assistant commissioner of public affairs Hilton Beckham said in a statement last month. “CBP’s search numbers are consistent with increases since 2021, and less than 0.01% of travellers have their devices searched … Allegations that political beliefs trigger inspections or removals are baseless and irresponsible.”

If you’re worried these allegations aren’t quite as baseless as CBP insists, Wessler says: “The safest approach is not to travel with data you wouldn’t want the US government to access.”

Let’s say you’re a British citizen who has been outspoken, on social media and elsewhere, about your pro-Palestinian or anti-Trump views. Would it be a foolish idea to travel to the US right now? “I wouldn’t say ‘don’t come,’ but I’d say evaluate your risk and risk tolerance,” says Wessler. “The government is being extremely aggressive with students and activists, and there’s always a chance a border agent might act on something they find politically disagreeable. Most travellers are still fine – but the risk is real and well above zero.” So, basically, nothing is very clear? Pretty much, says Wessler. “The law is a complete mess, and people’s options are a complete mess. People just have to make a risk assessment based on extremely imperfect information.”

The first step in making that risk assessment is to thoroughly understand the rules for the specific visa you’ll be travelling on or your immigration status. “The Foreign Affairs Manual is a great resource,” says immigration lawyer Tahmina Watson. “It’s what consular officers use, and it’s publicly accessible. It lays out what officers are looking for, visa by visa. We’re now advising clients more than ever to understand the B1/B2 visa rules. B1 is for business, B2 for tourism. When CBP asks why you’re here, they’re listening for key phrases – ‘I’m visiting my grandmother,’ ‘I’m going to Disneyland,’ etc. The manual also talks about proof of ties to your home country – job, house, bills. That stuff matters.”

«

unique link to this extract

Bluesky: the online cone of silence • Commentary Magazine

James Meigs:

»

How did a site supposedly designed for respectful conversation become so toxic? A sociologist would point to the strong incentives that drive people to identify with their in-group—such as a sports team or a political party—and to denigrate the out-group. In a diverse social environment, people usually keep these expressions within acceptable bounds. A Yankees fan might display a bumper sticker reading, say, “Beat the Red Sox” but stop short of “And Drag Their Entrails Through the Streets.” But when a social-media commu-nity defines itself as a refuge for the virtuous—one that is surrounded by enemies—that social restraint disappears. If you believe that the only people reading your post will be those on your own team, you are more likely to let loose. In fact, you will be rewarded for it as other users like and respond to the most performative expressions of outrage.

In real-world social circles, being a total flaming, um, jerk brings social costs. But in a hermetically sealed social-media bubble, it’s a way to build your status. Bluesky adds another perverse incentive: Anyone adding nuance or pushing back against violent statements risks being ridiculed and even mass-blocked by the online community. This combination of positive and negative rewards creates a one-way ratchet, always pushing users toward extremism.

Yes, a similar dynamic exists among some conservative X users, and you can find plenty of ugliness on the far right. But on X, in my experience at least, the nasty stuff doesn’t crowd out more balanced conversations. Thomas Chatterton Williams, another heterodox thinker, recently conducted his own inadvertent experiment testing this thesis. An Atlantic article he wrote about the so-called woke right was criticized both by the New York Times’ lefty columnist Jamelle Bouie on Bluesky and by conservative activist Chris Rufo on X. Being attacked from both sides comes with the territory for a centrist like Williams. “But the two experiences were not equivalent,” he noted (on X, naturally). “If I’m honest, it wasn’t even close to the same thing.” Williams says he found “far more good faith, graciousness, nuance, sense of fair play and diversity of thought” from Rufo’s followers on X than in the Bluesky comments inspired by Bouie.

«

This comes from a somewhat right-wing view (as with anything American) but the general dynamic is correct. Bluesky’s “nuclear block” makes the site less useful because it hides so many opposing views, whereas in retrospect Elon Musk’s change to the Twitter block – which renders tweets from people who have blocked you visible, but you can’t interact with them – is actually useful in continuing to show you opposing views.

Meigs also makes an interesting point: the lack of opposition on Bluesky makes it harder to road-test ideas to see how well they travel beyond a left-wing bubble. And that, he points out, is a bad thing.
unique link to this extract

They paid $3,500 for Apple’s Vision Pro. A year later, it still hurts • WSJ

Joseph Pisani:

»

Early adopters of Apple’s Vision Pro headsets have one thing to show for the year they’ve spent with their pricey purchases: regret.

“It’s just collecting dust,” Dustin Fox said about his mixed-reality headset, which looks like futuristic ski goggles. “I think I’ve probably used it four times in the last year.” The $3,500 device sits in a bin with other gadgets he no longer picks up. 

The Vision Pro launched in February 2024 with great promise. It was Apple’s first major product release in years! It’s the first device you look through and not at! Typing can be done in the air! But buyers who wore them in the wild say they got nothing but dirty looks and sore necks. Now, the devices are daily reminders of their misplaced bravado.

Fox, a realtor in Centreville, Va., had to have the Vision Pro as soon as it launched. “I’m like a little boy when something new comes out,” he said.

The 46-year-old thought he’d use it for work. Then he put it on his head.  “It’s way too heavy,” Fox said about the device, which weighs just over a pound. “I can’t wear it for more than 20 or 30 minutes without it hurting my neck.” 

Tovia Goldstein was excited to wear his set to watch movies and TV shows. But he ended up needing breaks. “After 60 minutes, you can’t, you just have to throw it down,” he said. He hasn’t touched it in about four months.

Weight isn’t the only issue for the 24-year-old New Yorker. There also aren’t enough apps to make the Vision Pro worthwhile, he said. 

Goldstein thinks from time to time about getting the headset out of the closet and dusting it off to see if any new apps have been added. But the pain in the neck he gets, plus the painfully long three minutes it takes to plug in the battery and wait for it to turn on, make him think again. “I wouldn’t recommend anyone buying it,” he said, “unless you’re really rich and you don’t know what to do with your money.” 

«

Fair number of them for sale fairly cheaply on eBay in the UK. Expect the same is true in the US. Just goes to show Apple doesn’t need Jony Ive to screw up design on something. Although speaking of Jony Ive and design…
unique link to this extract

A decade with the Apple Watch • The New Consumer

Dan Frommer, who has been using an Apple Watch since the launch in April 2015:

»

I use mine to: Pay almost everywhere, track my runs and steps, never miss a phone call, wake up silently for early flights, monitor coyote traffic on the driveway cam, time six-minute eggs, read email newsletters, see the AQI, find my phone, ride the Shinkansen, take an EKG, rewind the podcast, use as a flashlight to navigate a dark room, and intercom to the HomePods. Oh — it also turns out that it’s very convenient to know the time without always having to fish out a phone.

They look dorky in photos, but are regrettably worth it. My titanium Ultra is the wristtop computer that my high school self would have gone nuts for.

Rather than status, the Apple Watch’s popularity is really because of its utility, which is the purest form of success.

In our research, Apple Watch users use it, in part to live a better, healthier, more active life.

More than half (57%) of Apple Watch users said they’ve tracked their steps over the past year, compared to 34% of Americans overall, according to our latest Consumer Trends Survey of more than 3,000 US consumers, conducted in late February by Toluna. (This poll was part of my Consumer Trends research series, produced in collaboration with Coefficient Capital; this is the first time we’ve published these stats.)

«

I was reflecting on the Apple Watch the other day, and how people thought they would be fashion items (guilty). Instead they’re just little computers, doing computer-y things, and we’re fine with that. Though Frommer is right to point out that they’re “Apple’s most aesthetically customisable product” – you can change the watch face, get different colours and sizes of watch, change the watch band. Might that be part of the attraction?
unique link to this extract

China launches first of 2,800 satellites for AI space computing constellation • SpaceNews

Andrew Jones:

»

The China Aerospace Science and Technology Corporation (CASC) announced a fully successful launch, revealing the mission to have sent 12 satellites for a space computing constellation into orbit.  

Commercial company ADA Space released further details, stating that the 12 satellites form the “Three-Body Computing Constellation,” which will directly process data in space, rather than on the ground, reducing reliance on ground-based computing infrastructure. The constellation will be capable of a combined 5 peta operations per second (POPS) with 30 terabytes of onboard storage. 

The satellites feature advanced AI capabilities, up to 100 Gbps laser inter-satellite links and remote sensing payloads—data from which will be processed onboard, reducing data transmission requirements. One satellite also carries a cosmic X-ray polarimeter developed by Guangxi University and the National Astronomical Observatories of the Chinese Academy of Sciences (NAOC), which will detect, identify and classify transient events such as gamma-ray bursts, while also triggering messages to enable followup observations by other missions.

ADA Space claims the 12 satellites represent the world’s first dedicated orbital computing constellation. This marks a shift from satellites focused solely on sensing or communication to ones that also serve as data processors and AI platforms.

«

Can’t help wondering why China wants to do this stuff in space. It’s not as though it’s easy to service or upgrade. Sure, you get free power and cooling isn’t the worst problem. But.. that’s a lot of satellites. No doubt the US will now feel obliged to do just the same.
unique link to this extract

2019: Why do Wite-Out and Liquid Paper still exist? • The Atlantic

David A. Graham, writing in 2019:

»

Christmastime is when the pens in my house get their biggest workout of the year. Like many Americans above grammar-school age, I seldom write by hand anymore, outside of barely legible grocery lists. But the end of the year brings out a slew of opportunities for penmanship: adding notes to holiday cards to old friends, addressing them, and then doing the same with thank-you notes after Christmas. And given how little I write in the other 11 months of the year, that means there are a lot of errors, which in turn spur a new connection with another old friend: Wite-Out.

The sticky, white fluid and its chief rival, Liquid Paper, are peculiar anachronisms, throwbacks to the era of big hair, big cars, and big office stationery budgets. They were designed to help workers correct errors they made on typewriters without having to retype documents from the start. But typewriters have disappeared from the modern office, relegated to attics and museums. Even paper is disappearing from the modern office, as more and more functions are digitized. But correction fluids are not only surviving—they appear to be thriving, with Wite-Out sales climbing nearly 10% in 2017, according to the most recent public numbers. It’s a mystery of the digital age.

…As early as 2005, The New York Times pondered the product’s fate with trepidation. Somehow, more than a decade on, it has kept its ground. According to the NPD Group, which tracks marketing data, sales of correction fluid grew 1% from 2017 to 2018, though they fell 7% the year before. (Correction tapes were flat, while correction pens are fading.) From 2015 to 2016 to 2017, Bic, which makes Wite-Out and Tipp-Ex, reported that correction products increased in share from 5% to 6% to 9% of the global stationery market.

Who’s still buying these things? All the best answers are mostly conjecture. AdWeek suggested that sales might be buoyed by artists using fluid as paint. A Bic spokesperson pointed to a series of weird and entertaining interactive YouTube ads for Tipp-Ex in Europe, and said that Wite-Out is launching “colored dispensers that will appeal to younger consumers.”

«

OK, it’s a six-year-old story. But I looked up market reports: this one says it was $2.1bn worldwide in 2022 and is expected to grow 1% annually to 2029; this one says it was worth $2.0bn in 2023 and will grow at 1% annually to 2030.

Which brings us back to the original question: who the hell is using this stuff, given that the printer market is shrinking?
unique link to this extract

FBI warns of ongoing scam that uses deepfake audio to impersonate government officials • Ars Technica

Dan Goodin:

»

The FBI is warning people to be vigilant of an ongoing malicious messaging campaign that uses AI-generated voice audio to impersonate government officials in an attempt to trick recipients into clicking on links that can infect their computers.

“Since April 2025, malicious actors have impersonated senior US officials to target individuals, many of whom are current or former senior US federal or state government officials and their contacts,” Thursday’s advisory from the bureau’s Internet Crime Complaint Center said. “If you receive a message claiming to be from a senior US official, do not assume it is authentic.”

The campaign’s creators are sending AI-generated voice messages—better known as deepfakes—along with text messages “in an effort to establish rapport before gaining access to personal accounts,” FBI officials said. Deepfakes use AI to mimic the voice and speaking characteristics of a specific individual. The differences between the authentic and simulated speakers are often indistinguishable without trained analysis. Deepfake videos work similarly.

One way to gain access to targets’ devices is for the attacker to ask if the conversation can be continued on a separate messaging platform and then successfully convince the target to click on a malicious link under the guise that it will enable the alternate platform. The advisory provided no additional details about the campaign.

«

Does at least give Americans another reason to ignore anything their government attempts to tell them.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified