Scientists have detected a sugar in deep space that on Earth is found in raspberries. Maybe a raspberry-flavoured planet? CC-licensed photo by oatsy40 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. Sweet. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

---

Who cleans up after the vibe-coding party? • Financial Times

Sam Learner:

»

Earlier this year, a group of researchers led by Miklós Koren, an economist at Central European University in Vienna, released a paper titled “Vibe Coding Kills Open Source”. They argue that in a world where the consumers of software packages are increasingly AI coding agents, the incentives that drive the collaborative model of open-source development collapse.

“We started with models that are quite standard in international trade, where firms are producing different quality products and they compete in the market . . . and we realised it’s not a bad analogy for open source,” Koren told me. “There might not be a lot of money involved, but every developer wants their package to be popular.”

I understand this well. Before working [as a programmer] in newsrooms, I developed and released a set of personal open-source projects. One in particular, which visualised the downstream paths of rivers, was a disorientating success. The project is still more widely viewed than every story I’ve worked on at the FT combined and, judged by eyeball count, might be the most successful thing I ever create. I released it just months before interviewing for my job and, while I developed the tool for free, I can draw a straight line between its success and my employment today.

But Koren found that these socio-economic reasons for developers to contribute to open-source projects degrade when human users are replaced with bots, which install the software without providing positive interactions. In the researchers’ model, this outweighs the fact that AI tools may help individual developers write code faster. Over time, the bots erode the open-source ecosystem.

“The question for the developer,” Koren said, “is if I want to be popular among humans, why should I write something that is only used by machines?”

In experiments with six coding models, the researchers found that open-source software packages that were frequently recommended by models saw large increases in download numbers — but that activity did not translate into the engagement that typically sustains maintainers. The paper concludes that, “under the traditional business model, where developer revenue depends entirely on direct user engagement, the open-source ecosystem cannot survive widespread AI adoption”.

«

More unintended consequences, perhaps some way distant, but quite probably more proximate than we think. For many, coding might become an artisanal process again, just as in the dawn of the PC age.
unique link to this extract

---

Astronomers have found a sugar molecule in deep space for the first time ever • WIRED

Jorge Garay:

»

FOR THE FIRST time in history, astronomers have detected a sugar molecule floating among the gas clouds of interstellar space.

Erythrulose has four carbon atoms (C4-H7-O2), and on Earth, it occurs naturally in some fruits. Its presence 26,000 light-years away could help untangle the mystery of the origin of life on our planet.

The study was published this week in Nature Astronomy. The team, led by Izaskun Jiménez Serra, analyzed data captured by radio telescopes in Spain to identify the molecule’s signature in the microwave frequencies it produces as it rotates.

Sugar molecules are essential for life. They fuel cells and are part of RNA and DNA. Yet scientists still don’t know how they accumulated in large enough quantities on the early Earth. One possibility is that some of the molecules did not originate on the planet but instead reached Earth via meteorites.

For the new study, the researchers focused on molecular cloud G+0.693−0.027, a location they didn’t choose at random. G+0.693−0.027 is among the regions richest in molecules in the entire Milky Way. It is located near the supermassive black hole at the center of the galaxy, and collisions with another cloud appear to have turned the region into a veritable chemical factory. Researchers had already detected alcohols, aldehydes, urea, ethanolamine, hydroxylamine, and dozens of complex organic molecules. Now, sugar is in the mix.

The idea that some sugars might have come from space gained traction in December 2025, when scientists confirmed that the asteroid Bennu contained ribose and other monosaccharides. Ribose is a fundamental sugar in RNA. The new study reveals another type of space sugar, this one from the ketose family. On Earth, it’s found in tanning lotions and raspberries.

«

To some extent moving the origin question to “meteorites” and “outer space” only pushes the question of “how?” away by one step, but at the same time shows that we need not be unique. Perhaps space is raspberry-flavoured? That would be nice.
unique link to this extract

---

Markets experience new DeepSeek shock after MoonShot AI releases Kimi K3 • Fortune

Nicholas Gordon:

»

On Thursday, Moonshot AI, a Beijing-based AI startup, unveiled the latest version of its Kimi large language model, which promises performance close to Anthropic’s Fable 5—perhaps the most powerful publicly available model today—at a fraction of the cost.

Kimi K3, the largest open-weight model ever released, performed “competitively” with Fable 5, and “substantially outperformed” Anthropic’s Opus 4.8, and OpenAI’s GPT 5.6 Sol, according to Moonshot. The company’s officially-released benchmarks consistently rank K3 among the top three AI models; one independent benchmark from Arena.AI even put K3 as the best model currently available, ahead of Anthropic.

Many observers, including Anthropic CEO Dario Amodei, didn’t expect a Chinese AI lab to release a model that could approach the US’s best offerings for another six months. Tesla CEO Elon Musk, for example, suggested it might happen by the first quarter of next year. K3’s launch rapidly shrank that timeline and underscores just how quickly Chinese AI developers are closing the performance gap with their US rivals. 

“The AI ecosystem in China is probably much better than people thought,” says Paul Triolo, a partner at DGA-Albright Stonebridge Group.

Moonshot AI has priced K3 at $15 per million output tokens, the bits of information processed and generated by a large language model. That’s cheaper than Fable 5, which costs $50 for the same level of output.

Moonshot’s launch seemed to rattle investors who interpret the new model as undermining the conventional wisdom that US firms can maintain their extended lead by simply outspending Chinese competitors on computing power.

A semiconductor selloff was already underway, but the K3 debut seemed to make things worse. Leading chipmaker Taiwan Semiconductor Manufacturing Company fell by 7% on Friday, despite reporting a 77% jump in quarterly operating profit. SoftBank—often seen as a proxy for OpenAI—fell by 9.0%. Z.ai, a Chinese AI startup that has released a competing model to Moonshot’s Kimi, plunged by almost 30% in Hong Kong trading.

«

And that’s before you get into the open source models. There doesn’t seem to be any moat – protective unique selling point – to LLMs; anyone can do it, and lots of people are doing, and once you get good-enough open source models you can deploy on a phone, well, haven’t we seen this movie before?
unique link to this extract

---

AI ‘amplification spiral’ may be causing delusions among users, study suggests • Decrypt

Jason Nelson:

»

Researchers have proposed a new framework to explain reports of AI psychosis, arguing that certain chatbot behaviors may reinforce delusions in vulnerable users.

Published in Nature, the study out of King’s College London and Germany’s Protestant University of Applied Sciences proposes an “amplification spiral” framework that links common chatbot behaviors to the reinforcement of user delusions.

“AI-associated delusions represent an emerging phenomenon requiring mechanistic understanding,” the researchers wrote. “This framework aims to guide systematic inquiry into how human cognitive vulnerabilities interact with AI design features in psychopathology development.”

The study focused on three specific behaviors, including linguistic alignment, in which AI mirrors a user’s language and communication style; hyperpersonalized generation, where responses are tailored to an individual’s history, emotions, and beliefs; and sycophancy, a tendency to validate or agree with users rather than challenge them.

The authors argue that these traits can combine into a feedback loop in which chatbots not only reflect a user’s thinking but help elaborate and reinforce it over time.

“The tendency of AI chatbots to agree with user opinions has been likened to social media echo chambers and, in its most extreme form, to an ‘echo chamber of one,’ where the positive corrective influence of real-life social interactions is absent,” the paper said.

The researchers noted that technology has long featured in delusions, from the radio and television, to satellites and the internet. Researchers argue, however, that AI represents a shift because chatbots can engage users in prolonged, personalized conversations.

The review arrives as psychologists and AI researchers examine the effects of chatbots on vulnerable users. A recent survey by the American Psychological Association found that 15% of psychologists reported patients developing distorted thinking or delusions related to chatbot use. At the same time, more than a third said they had observed patients becoming dependent on AI companions. The findings followed a separate study from researchers at the City University of New York and King’s College London showing that several leading AI models could reinforce delusions, paranoia, and suicidal thoughts.

«

unique link to this extract

---

Meta trying to destroy whistleblower Sarah Wynn-Williams, US senator says • The Guardian

Michael Savage:

»

A US senator has accused Meta of using lawfare in “efforts to destroy” a whistleblower who made allegations about the social media company’s dealings with China and its treatment of teenagers.

In a letter to its founder, Mark Zuckerberg, the Republican senator Josh Hawley demanded to know what measures Meta had taken to monitor Sarah Wynn-Williams, Facebook’s former global head of public policy, and her family.

He also accused the firm of trying to make “truthful speech so perilous that no whistleblower like Ms Wynn-Williams can ever afford it”.

Wynn-Williams made her allegations in a memoir, Careless People: A Cautionary Tale of Power, Greed, and Lost Idealism but under an interim arbitration ruling secured by Meta she is unable to speak about the book, which contains claims about it working with the Chinese government on censorship tools, and its platforms’ negative impacts on teenagers.

She sat in silence on stage at the Hay literary festival in Wales earlier this year after taking legal advice, and is suing Meta over the restrictions it secured over her.

Meta strongly disputes Wynn-Williams’s book, saying it contains several false claims, and describes her legal case as a ploy to shift more copies.

Hawley, whose senate judiciary subcommittee on crime and counter-terrorism took evidence from Wynn-Williams last year, said: “Meta’s efforts to destroy Ms Wynn-Williams with lawfare are a matter of grave public concern.”

…Hawley demanded documents relating to any Meta attempts to “monitor, track, record, or catalog Ms Wynn-Williams’ or her family members’ public statements, interviews, social media activity, or public or private travel”.

«

The documents Hawley demands are basically just part of an effort by him to make Meta look bad. Not a high bar. Though you’d think the Democrats might want to get in on this act too. Still, has to be said that Meta’s efforts have been an absolute boon to the book’s sales. That saying that no publicity is bad publicity? They maybe thought it meant that enforcing no publicity would be bad publicity for the book. Mistake!
unique link to this extract

---

Why Tokyo’s new shorts policy is sparking debate in Japan • The New York Times

Javier Hernandez and Hisako Ueno:

»

For more than three decades, Noboru Watanabe wore the uniform of the Japanese salaryman: trousers, a starched shirt, a jacket and a necktie.

But last week Mr. Watanabe, 53, an environmental official in Tokyo, ditched the outfit for the unimaginable: a T-shirt and a pair of khaki shorts. He was taking advantage of a new policy by the Tokyo government to allow its workers to wear shorts in the office, after successive summers of record-breaking heat.

“Everyone says I look younger,” Mr. Watanabe said near his desk on a recent afternoon. “I seem more approachable. I feel I am helping relax the whole office.”

The policy is the latest attempt by Japanese officials to persuade the country’s millions of office workers to dress down in the summer to make office life more comfortable, reduce the use of air-conditioning and save energy. Under the “Cool Biz” initiative, the government encourages workers to forgo heavy suits in favor of polo shirts, linen tops and sneakers in hot weather.

But shorts, newly allowed this summer, are a bridge too far for some. The policy has prompted debate in Japan — and inspired delicate conversations at the office about etiquette and how to deal with the sight of a colleague’s hairy legs. The Japanese news media has even coined a new term — sune-hara, or “leg-hair harassment” — to describe the sense of grievance some women feel when they are forced to observe their male co-workers’ legs.

Ruai Sajiki, 23, an employee at the Tokyo government offices, acknowledged that seeing bare legs in a professional setting could be jarring for women. But she offered her blessing when Mr. Watanabe, her supervisor, asked if she minded if he wore shorts in the office.

“I’ve seen my own father’s legs, so I’m personally comfortable with it,” she said. “It can’t be helped that there are many opinions, and some women who feel resistance.”

Some commentators argue that the policy, while officially applying to both men and women, is unfair. While men are rarely judged for what they wear to work, women still face pressure to cover their legs with pantyhose, which are still common in Japan, or long skirts.

«

Japan remains incredibly patriarchal. Also, shrinking in number. Not sure if the two are linked, but why risk it?
unique link to this extract

---

LinkedIn and X are flooded with AI spam, browsing data suggests • 404 Media

Jason Koebler:

»

A shocking amount of the content that users encounter on popular social media websites is likely AI generated, according to data from a company that detects AI writing. As much as 41% of longform written content seen by users on LinkedIn is likely to be fully AI-generated and roughly a third of longer posts on X are AI-generated; roughly one-in-ten longer Reddit and Substack posts are AI, according to the data. 

The data was collected using a Chrome extension from Pangram, a company that detects AI-generated writing. Pangram’s Chrome extension scans writing that users encounter while browsing and determines if any given post is likely AI-generated or likely human written. Because Pangram works passively in the background while a user is browsing the internet, it only scans posts that its users actually see. This helps answer the question of whether AI slop is actually poisoning the internet that humans actually use, versus polluting the internet more broadly. The answer is unequivocal: AI slop writing is not just sequestered off on unpopular automated SEO farms or spam sites that no one reads; humans are regularly wading through AI dreck on hugely popular sites. 

“This isn’t something that had really been studied before—how much AI content people are actually seeing,” Max Spero, the CEO of Pangram, told me in a phone interview. “AI content is a tax on readers’ time.” 

«

Is it a tax if you ignore it and scroll straight past? Asking for a friend. Or, in the case of LinkedIn, just never go there.
unique link to this extract

---

The collapse at Netflix signals the end of audience capture • The Honest Broker (*not Roger Pielke)

Ted Gioia:

»

Netflix got into this mess by pursuing a simple strategy: (1) Reduce the number of new scripted series (which peaked in 2022), but (2) Raise subscription prices.

That is the “audience capture” strategy mentioned above. The idea is that the audience got captured years ago with cheap subscription prices, and now the platform can squeeze them mercilessly—offering less and charging more. Netflix has been pursuing this agenda for several years now.

Ah, but Netflix isn’t the only company building its future on audience capture. It’s getting used at almost every streaming platform. And even companies outside of the media space are practicing variants of it. You see it at Google, Meta, X, Apple, etc.

It’s shocking how many companies have learned this technique. The entire printer and toner business is now built on audience capture. The same is true of the software industry—don’t even get me started on my Microsoft Office subscription fiasco. And, of course, all those customer loyalty programs (variants on the frequent flyer gimmicks that started this craze years ago) are examples of the same stale strategy.

Even the AI world is turning into an audience capture business—both for itself and its customers. This is one of the key reasons for my frequent criticisms of AI slop. It feeds into step one of the strategy outlined above. The companies use AI to reduce the cost of content, thus boosting margins while reducing their dependence on human creators.

Audience capture has always existed, but never to this extent. When I consulted at BCG we called it a milking strategy, where you raised prices and reduced capital investment in a business—which was now your cash cow. You squeeze all the money you can from it, for as long as you can.

But back then we realized that milking only worked in the short term. Eventually you killed the cow. And the risk is the same today with “audience capture”—which is just a new name for that poor old bovine.

Sooner or later, the audience refuses to be held captive. And that’s happening now at Netflix—hence the stock sell-off.

«

The stock sell-off might precede a bigger abandonment by subscribers who just don’t see the point because Netflix keeps throwing things against a non-stick wall. Nothing is allowed to grow. Tech companies can’t do content.
unique link to this extract

---

‘We noticed a login from a new device’: the message from fraudsters targeting your X account • The Guardian

Samuel Gibbs:

»

You have had an X account for years, since it was known as Twitter. When an email arrives about a new login from a location nowhere near where you live, alarm bells begin to ring.

“We noticed a login to your account from a new device. Was this you?” the email asks.

The location is Arizona, but you live in London. The device is “Firefox Desktop on Mac”.

“If this wasn’t you,” the email continues, “complete these steps now to protect your account.”

The steps include clicking a link to change your password, which will end up with you “logged out of all your active X sessions except the one you’re using at this time” and to click another link to review the apps that have access to your account and revoke any that are unfamiliar.

The steps are legitimate advice from X, but the links to reset your password, or to review app access are not. The email is fake and attempting to trick you into giving scammers your password, or direct access to your X account.

“Scammers want your X username and password, or to trick you into approving a malicious link that gives them access to your account without needing your password,” says Jake Moore, a global cybersecurity adviser at ESET.

Once the criminals have access to your account they will more than likely use it to attempt to commit further fraud including crypto scams, phishing attacks and misinformation campaigns.

«

There’s a lot of scamming going on. The misdirection emails are quite common; they often lead to direct messages to mutual followers asking them to vote for them in a podcast competition, or something similar. The scams mutate so quickly that once you’ve picked up on one, the next will be along imminently.
unique link to this extract

---

• Why do social networks drive us a little mad? • Why does angry content seem to dominate what we see? • How much of a role do algorithms play in affecting what we see and do online? • What can we do about it? • Did Facebook have any inkling of what was coming in Myanmar in 2016? Read Social Warming, my latest book, and find answers – and more.

---

Errata, corrigenda and ai no corrida: none notified