Code embedded in Apple’s latest iOS release suggests it will launch a new “Invites” app soon. CC-licensed photo by Matt Biddulph 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. Uninvited. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

---

Meta warns that it will fire leakers in leaked memo • The Verge

Alex Heath:

»

Moments after Meta CEO Mark Zuckerberg’s all-hands comments to employees were widely leaked, a company executive warned in an internal memo that leakers will be fired.

“We take leaks seriously and will take action,” Meta’s chief information security officer, Guy Rosen, said in an internal memo I’ve seen. “When information is stolen or leaked, there are repercussions beyond the immediate security impact. Our teams become demoralized and we all waste time that is better spent working on our products and toward our goals and mission.”

Rosen goes on to say that Meta “will take appropriate action, including termination” if it identifies leakers and that “we recently terminated relationships with employees who leaked confidential company information inappropriately and exfiltrated sensitive documents.”

During today’s all-hands meeting, Zuckerberg told employees he would no longer be as transparent due to leaks. “We try to be really open and then everything I say leaks,” he said. “It sucks.”

«

As John Gruber points out,

»

It’s not fear of getting fired that keeps employees at most companies from leaking. It’s that they find themselves aligned with the company’s mission. They feel like part of a team that they want to see succeed, and they naturally adopt an attitude of being a team player. Team players don’t leak the playbook because they don’t like the coach’s play-calling or how much playing time they’re getting. I’ve never gotten the sense that that sort of attitude exists at Meta.

«

Can’t say better than that.
unique link to this extract

---

Could the US government fix the journal cartel problem? • Just Emil Kirkegaard Things

Emil Kirkegaard:

»

The USA publishes the most high quality science of all countries, though this is mainly due to the large population size, and not because American scientists are particularly productive. So given its dominant role, USA could try to do something about the issue, just as Elon Musk did for internet free speech by purchasing Twitter.

Perhaps the first idea you have is that open access for federally funded research should be mandatory. It sounds good. The public is sponsoring the research, so it is absurd they can’t read it. The journals found a nice way to game this system too. Open access fees. If you want to publish in a high ranked journal (e.g. Nature), and you want the paper to be readable by anyone, you can choose to pay a fee for this. How much is the fee? Well, whatever Nature says it is (right now it’s $12,290!). How could you decline, after having just gotten lucky enough to get accepted in one of the ‘best’ journals in the world? Who pays the fee? The tax payer of course (taken out of the research funding). So this solves only half the issue as the oligopoly still has a way to milk endless money but at least everybody can read the science.

My idea for solving this is that federal research funding comes with more stipulations to combat this oligopoly:

1: The research must be open access (from day 1). Many universities and research agencies across the world already have such mandates.
2: The publication fee must not exceed X USD, where X is, say, 100. Importantly, the rest cannot be paid by third parties. Otherwise, the universities would just pay this as a cost of doing business (they also want to publish in ‘top’ journals because university rankings depend in part on these).
3: The research materials must be public as well, including the data, questionnaires, computer code and whatever else is needed to evaluate the work. This is to make sure the public gets the most science for the money. Other scientists can reuse materials for other research. It also helps discover and prevent fraud because fraud is often proven when the data are analyzed by third parties.

The second stipulation removes the ability of the publishes to set arbitrarily high prices.

«

Very possibly the US government could fix the journal cartel problem. I’m going to go out on a limb though and suggest that the idea won’t even cross the collective mind of those in charge because they have absolutely no interest in science.
unique link to this extract

---

G/O Media is publishing AI slop again • Aftermath

Riley MacLeod:

»

several staffers at G/O Media (which previously employed all of us here at Aftermath) pointed out [at the end of January] the proliferation of AI-generated articles on news site Quartz. Written in a style that would get a high school “introduction to writing” student a B for effort, they… I can’t even think of a good way to sum them up, the whole thing just sucks.

As of publishing, the “Quartz Intelligence Newsroom” has written 22 articles today, running the gamut from earnings reports to Reddit communities banning Twitter posts to the Sackler settlement to, delightfully, a couple articles about how much AI sucks. Quartz has been running AI-generated articles for months, but prior to yesterday, they appear to have been limited to summaries of earnings reports rather than news articles. Boilerplate at the bottom of these articles notes that “This is the first phase of an experimental new version of reporting. While we strive for accuracy and timeliness, due to the experimental nature of this technology we cannot guarantee that we’ll always be successful in that regard.”

The articles, to their credit, do cite where the AI is gathering its information from. But even this is surface level: the Reddit article, for instance, cites Yahoo and the New York Post, but the Yahoo post is actually a repub from the Daytona Beach News-Journal, and the Post article cites NBC News as its own source. I cannot imagine how this game of telephone could go wrong, especially when letting a robot write news about contentious public figures and rapidly changing events.

G/O previously experimented with AI-generated articles back in 2023, most memorably producing a chronological list of Star Wars entries that wasn’t chronological, and a bunch of garbage on The AV Club and Deadspin. None of these sites are owned by G/O anymore; Quartz is one of three sites that are still standing, as well as commerce site The Inventory. Of those remaining sites, Kotaku saw layoffs back in November, and The Root recently made the news when, following the death of a writer, the site’s deputy editor asked staffers to write more to compensate.   

All of which makes Quartz’s use of AI just more proof–not that you need it–of how little G/O cares about the people who work there, and how little it thinks of its audience’s intelligence.

«

The audience’s intelligence doesn’t matter, though. The purpose of slop is to get indexed, so people doing a search click on it, and by the time they’ve figured out that the article is no use (if they do), the adverts have been shown to them. Job done.
unique link to this extract

---

ChatGPT vs. Claude vs. DeepSeek: the battle to be my AI work assistant • WSJ

Joanna Stern:

»

I keep waiting for my team to buy me a “WORLD’S BEST BOSS” mug. Then I remember they’re bots. Workplace brown-nosing isn’t one of their many skills.

The two AI co-workers on my org chart are OpenAI’s ChatGPT and Anthropic’s Claude. Over the past few months, they’ve taken on some of my work…so I can do even more work. And now I am auditioning a third assistant, DeepSeek.

They’re not just rewriting emails or summarizing meetings. These guys are building spreadsheets, prepping research, creating calendars and, yes, even ordering flowers for my wife.

I pay $20 a month for Claude and ChatGPT. Why both? Because we’re living in Turbulent AI Times where one week’s best AI assistant is the next week’s also-ran. Case in point: DeepSeek’s recent surprise debut. Fortunately, that’s free. I’ve also tested Google’s Gemini, Meta AI and Microsoft Copilot but, to paraphrase the great Shania Twain, they don’t impress me as much.

Choosing the best AI assistant for your work isn’t only about these ever smarter models, but also the tools and features that help you get things done. You will judge an AI not about how well it can do your job, but how many tasks you can offload to it.

“Every job is a bundle of tasks,” says Erik Brynjolfsson, a Stanford University economist and the founder of the AI-at-work consulting company Workhelix. “When you analyze jobs at that level, you can really make headway as to whether technology can help.”

What tasks you can outsource to these assistants depend on your job, your workflow and, most importantly, the AI’s capabilities. Yep, it’s a lot like hiring—you want the candidate with the right skills.

«

This is free to read. It helps that she’s writing a book about AI, so there’s a certain incentive to use these tools.
unique link to this extract

---

Deutsche Bank has published deck of 25 memes about DeepSeek • FT Alphaville

Bryce Elder, quoting the DB introduction to its deck explaining “AI in 2025: 25 themes in 25 memes”:

»

If a picture is worth 1,000 words, this chartbook should save you from reading 25,000 of them.

That counts for something in a week when so many millions of words have been written about the surprise arrival of China’s DeepSeek AI model.

AI has come of age in the era of the meme – and it turns out memes are one of the best ways of explaining where it is going.

«

These are actually pretty good!
unique link to this extract

---

Research Roundup: 7 cool science stories we almost missed • Ars Technica

Jennifer Ouellette:

»

It’s a regrettable reality that there is never time to cover all the interesting scientific stories each month. In the past, we’ve featured year-end roundups of cool science stories we missed. This year, we’re experimenting with a monthly collection. January’s list includes papers on using lasers to reveal Peruvian mummy tattoos; the physics of wobbly spears and darts; how a black hole changes over time; and quantum “cat states” for error correction in quantum computers, among other fascinating research.

«

These are fun.
unique link to this extract

---

Why we’ve suspended some Partner Program accounts this week • The Medium Blog

Scott Lamb:

»

Perhaps recently you’ve logged in to Medium only to encounter a sea of responses like these on a story: “Nice”, “Follow me please 🙏”, Good working ❤️❤️❤️”

Obviously, responses like that aren’t what we set out to make happen with Medium. It’s not why I get out of bed in the morning, or why anyone on our small but mighty team puts in the time they do. I hate seeing thoughtless comments like this on my writing here, frankly, or on your writing, or on anyone’s writing. We deserve better.

But these responses are especially damaging when they are organized in order to misuse our Partner Program and take earnings from other writers. And they’re only one form of behavior trying to extract money through deceptive content — we’ve also seen a massive recent uptick in low-quality, AI-generated posts behind the paywall, and coordinated activity like fake accounts created by a single person in order to engage with paywalled posts to generate earnings, and more.

We’ve heard your feedback and we see it ourselves, and we don’t like it. This isn’t the way to a better internet. It’s worth pointing out this isn’t limited to Medium; platforms everywhere are struggling with these challenges.

What is non-genuine engagement? We do not allow the following behaviors:

…• Using AI-generated content to earn money for stories and responses in the Partner Program

«

Inevitable. (And, by the way, remember Medium? Certainly lost out in a BIG way to Substack.)
unique link to this extract

---

Musk’s junta establishes him as head of government • Doomsday Scenario

Garrett M. Graff decided that “the American media would be more clear-eyed about the rise and return of Donald Trump if it was happening overseas in a foreign country, where we’re used to foreign correspondents writing with more incisive authority”:

»

WASHINGTON, D.C. — What started Thursday as a political purge of the internal security services accelerated Friday into a full-blown coup, as elite technical units aligned with media oligarch Elon Musk moved to seize key systems at the national treasury, block outside access to federal personnel records, and take offline governmental communication networks.

With rapidity that has stunned even longtime political observers, forces loyal to Musk’s junta have established him as the all-but undisputed unelected head of government in just a matter of days, unwinding the longtime democracy’s constitutional system and its proud nearly 250-year-old tradition of the rule of law. Having secured themselves in key ministries and in a building adjacent to the presidential office complex, Musk’s forces have begun issuing directives to civil service workers and forcing the resignation of officials deemed insufficiently loyal, like the head of the country’s aviation authority.

…Over the last two weeks, loyalist presidential factions and Musk-backed teams have launched sweeping, illegal Stalin-esque purges of the national police forces and prosecutors, as well as offices known as inspectors-general, who are typically responsible for investigating government corruption. While official numbers of the unprecedented ousters were kept secret, rumors swirled in the capital that the scores of career officials affected by the initial purges could rise into the thousands as political commissars continued to assess the backgrounds of members of the police forces.

The mentally declining and aging head of state, who has long embraced conspiracist thinking, spent much of the week railing in bizarre public remarks against the country’s oppressed racial and ethnic minorities, whom he blamed without evidence for causing a deadly plane crash across the river from the presidential mansion. Unfounded racist attacks on those minorities have been a key foundation of Trump’s unpredicted rise to political power from a career as a real estate magnate and reality TV host and date back to his first announcement that he would seek the presidency in 2015, when he railed against “rapists” being sent into the country from its southern neighbor.

«

unique link to this extract

---

iOS 18.3 hints at new ‘Invites’ app from Apple to manage events • 9to5 Mac

Filipe Espósito:

»

After analyzing the code, we believe that the app is designed to help users organize meetings and in-person events. Although Apple’s Calendar app can already be used for this purpose, the new Invites app will likely have some additional features.

Code suggests that the Invites app will integrate with iCloud and will even have a web version on iCloud.com. The new app also integrates with a new iOS 18 daemon called GroupKit, which manages database models for groups of people. This daemon has been present since the first release of iOS 18.0 and hasn’t been used by any Apple apps so far.

Essentially, the app will show you a list of the people invited to that event and who has already confirmed their attendance. It’s unclear whether Invites will actually be a stand-alone app or whether Apple has plans to integrate it with other parts of the system (such as a mini iMessage app). Presumably, the app will have a more fun interface than what the Calendar app currently provides for inviting someone to an event.

Apple never said anything about this app at WWDC 2024 when iOS 18 was announced, so there’s a chance that the company is just experimenting with the idea and may end up scrapping it or delaying it for a future version of iOS.

«

Unless this is cross-platform – both Android and Windows – what’s the point? Perhaps Apple will be satisfied just to get the US teen market organising its weekend parties.
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