GeistHaus
log in · sign up

https://ianbetteridge.com/feed

rss
58 posts
Polling state
Status active
Last polled May 18, 2026 18:45 UTC
Next poll May 19, 2026 17:58 UTC
Poll interval 86400s
ETag W/"2e8b1-ahZ7i5vxh8KmNCmRCO7dDKBqs+Y"

Posts

Frumentatio!

I still don’t agree that men can’t stop thinking about Rome, but I’m definitely thinking about it today.

When Roman emperors distributed grain for free to the urban poor, they weren't being charitable, and it certainly wasn’t an egalitarian desire

Show full content
Frumentatio!

I still don’t agree that men can’t stop thinking about Rome, but I’m definitely thinking about it today.

When Roman emperors distributed grain for free to the urban poor, they weren't being charitable, and it certainly wasn’t an egalitarian desire to spread the fruits of empire or slave-worked estates.

The corn was a political instrument, a way of managing a population that had already been stripped of its economic independence. The latifundia, large estates worked by slaves, had largely swallowed the smallholdings that once sustained free Roman citizens. The dole wasn't intended to reverse that dispossession. It administered its consequences, at a price: dependency, and the political quietude that tends to follow.

It is worth keeping that history in mind when reading Elon Musk's recent post on X, in which he called for a “Universal HIGH INCOME via checks issued by the Federal government” as the remedy for AI-driven unemployment. Look how generous Musk is being! How egalitarian! How equitable! But of course, the political logic is something else entirely.

There are serious questions about who would fund such a programme, and how, and whether Musk's inflation argument holds. Those deserve their own treatment. But the more important question isn't economic. It's structural. What kind of political settlement does tech UBI actually propose?

Universal basic income has a history on the left which comes from a different set of concerns. One argument is that decoupling survival from employment shifts the balance of power between workers and employers. If you can meet your basic needs without selling your labour, you can afford to refuse bad terms. UBI, in this framing, is a tool for reducing structural dependence — on employers, on the labour market, on the contingency of finding someone willing to pay for what you can do.

Silicon Valley UBI starts from the opposite perspective. It leaves the ownership of AI infrastructure — the models, the compute, the data, the platforms — entirely untouched. It doesn't ask who owns the robots, or who captures the productivity gains from automation, or what democratic accountability looks like for systems that are reshaping the economy at scale. It simply proposes that the state write cheques to people the technology has displaced, substituting one form of dependence for another. Employer dependency becomes state-transfer dependency. The ownership question gets quietly closed before it is properly opened.

It is worth being precise here about what kind of state Musk and his cohort have in mind, too. The assumption embedded in any UBI argument is that the state which distributes the gains of automation on behalf of its citizens is, in some meaningful sense, accountable to them. It's democratic, rather than acting on behalf of a single group. That assumption is one this political class has spent years working to undermine.

In 2009, Peter Thiel, architect of much of the ideological infrastructure of Silicon Valley's current political turn, wrote that he no longer believed freedom and democracy were compatible. He has not recanted this view. The network of investors and operators around him, several of whom now hold or directly influence positions of state power, has acted accordingly. The kind of UBI our tech overlords want, then, is not just administered dependency. It is dependency administered by people who do not believe you should have a meaningful say in how it works.

All this matters because the political question of who owns the means of production in an AI economy is unsettled. What Musk and his ilk want is for it to be settled — by default, in favour of those who already hold the capital — while the public conversation focuses on the generosity of the proposed dole.

The timing makes the point. Musk floated his proposal the same week Reuters reported that Meta is laying off 10% of its global workforce, explicitly citing AI efficiencies. Dispossession and its managed remedy, arriving together, almost pre-packaged. The technology that displaces the worker and the proposal that makes that displacement politically tolerable are products of the same class of actors, emerging in the same news cycle. That is not a coincidence of timing. It is the shape of the new economy that they want to see emerge.

The Roman grain dole didn't give the plebs more power. It made them manageable. What is being proposed now by the tech bros carries the same structural logic: not the reversal of dispossession, but baking it into the structure of society. The question of who owns the machines — and who should — still goes unasked.


Your regular Ten Blue Links will follow tomorrow. Hopefully, you enjoy this sojourn into history and economics.

69e348cf816aec0001cceb6f
Extensions
Ten blue links, "fork you" edition
1 The knowledge class and its enemies

Writing for The Nation, Elizabeth Spiers reaches for Richard Hofstadter's Anti-Intellectualism in American Life to frame something that should have been nagging at the edges of tech criticism for quite some time. Hofstadter's great insight was that anti-intellectualism in

Show full content
1 The knowledge class and its enemiesTen blue links, "fork you" edition

Writing for The Nation, Elizabeth Spiers reaches for Richard Hofstadter's Anti-Intellectualism in American Life to frame something that should have been nagging at the edges of tech criticism for quite some time. Hofstadter's great insight was that anti-intellectualism in America has historically come from below, targeting the knowledge elite from outside. What Silicon Valley has managed is something structurally different: anti-intellectualism from within the elite itself, produced by people who are themselves the beneficiaries of exactly the kind of education they're now busy dismantling.

Peter Thiel's programme is paying students not to go to college. Marc Andreessen's bragging that he avoids introspection. The Suno CEO insisting that making music isn't enjoyable, which would be news, as Spiers notes, to every musician, professional and amateur alike. The pattern to this isn't random eccentricity. It's the logical product of people who believe they've cornered the market on critical thinking and therefore have nothing left to learn. The result is a class that hires linguists to improve its large language models while actively sneering at the kind of person who becomes a linguist.

There's a power-analysis here that Spiers makes explicit. An informed workforce is harder to control. Deep learning produces autonomy. Autonomy produces organisation, which produces demands. The tech oligarchs' anti-intellectualism isn't merely a cultural preference — it's a class strategy, dressed up as meritocracy and sold as concern for working people.

These self-described lovers of rationalism, who love to talk about IQ and logic while dismissing emotion as weakness, have managed to produce a cognitive ecosystem so closed that it can no longer generate original thought. They've enshittified their own thinking. The model for this, per Spiers, is Curtis Yarvin, their favoured intellectual and a man whose central political theory is that California should be run like a profitable corporation by a CEO-king. That Silicon Valley's most rational minds have outsourced their political philosophy to someone who thinks feudalism was underrated says more about the intellectual health of the valley than any number of TED talks.

2 Can Sam Altman be trusted? I'm betting you know the answer

Ronan Farrow and Andrew Marantz spent over a year writing a 16,000-word New Yorker profile with a headline that doesn't mince words: can Sam Altman be trusted? John Gruber's long read of the piece is itself essential — not just as a summary but as a considered response from someone who knew Aaron Swartz well and finds the newly-reported Swartz material the most significant thing in it.

That material: Aaron, in the months before his death, told friends that Altman "can never be trusted" and "is a sociopath — he would do anything." Swartz is not a casual source. He was brilliant, famously honest, and his warnings about specific people and institutions have consistently aged well. Gruber's point about Paul Graham is equally telling: Graham has spent the week since publication carefully explaining that Altman wasn't fired from Y Combinator, that he wasn't wanted gone. What he hasn't said — not once, not unambiguously — is that Altman is honest, trustworthy, or a man of integrity.

The organisations building frontier AI models require leaders of extraordinary integrity precisely because the asymmetry of information is so large. Users, governments, and investors cannot independently verify what these systems are doing or whether the people running them are being straight about their capabilities and risks. The degree to which we're extending trust to OpenAI is, in that context, essentially a bet on Altman personally. The New Yorker piece makes clear that a significant number of people who've worked closely with him have concluded that bet is badly placed.

3 The post-American internet is already being built

Cory Doctorow's latest Pluralistic takes an origin story — his own dotcom-era c//ompany Opencola, where he and his co-founders once brainstormed a way to spam Google into uselessness, and then didn't do it, because they loved the web too much — and turns it into a theory of internet history. The difference between the early Tron-pilled builders who held the line against breaking the internet and the people who eventually did break it wasn't intelligence or technical capability. It was callousness. When good-faith technologists red-teamed the internet, they felt scared and wanted to protect it. When the Zuckerbergs and Musks of the world did the same exercise, they turned it into a pitch deck.

Cory is finishing a book called The Post-American Internet, framed explicitly as a geopolitical sequel to Enshittification. The argument, as he's developing it publicly, is that Trump's demolition of US soft power has created the conditions for an alternative internet that might actually be better — not because it's technically different but because the political project behind it is different. The week's tech policy news gives him significant supporting evidence.

Nick Heer's linklog piece on digital sovereignty aggregates several stories that, taken together, show this project in motion: France announcing it will migrate government computers from Windows to Linux; Schleswig-Holstein having already moved off Microsoft Exchange and Outlook; the International Criminal Court switching to openDesk; UK banks quietly beginning to explore a domestic alternative to Visa and Mastercard, prompted by the visible demonstration that US financial infrastructure is a geopolitical weapon. A Canadian ICC judge who authorised investigations into US conduct in Afghanistan now has to phone hotels ahead of time to explain why she can't pay by card. The lesson, taken seriously at an institutional level, is that reliance on American platforms is a strategic vulnerability.

Heer adds the necessary caution: if this produces domestic walls rather than greater international cooperation, it will be disappointing. I think that's right. Sovereignty as protectionism looks different from sovereignty as the conditions for a distributed alternative.

The good news is that a lot of the tentative moves toward digital sovereignty are based on open platforms and open standards. The bad news is that institutional politics has a habit of taking open systems and turning them to dust. Hopefully that won't happen this time, when we really need it not to.

4 X becomes uninhabitable

The Electronic Frontier Foundation has left X, and the reason it gives is less a statement about politics than a straightforward piece of platform analytics. The EFF used to get 50 to 100 million impressions per month on the platform. Last year, 1,500 posts earned roughly 13 million impressions for the entire year. A single post now delivers less than 3% of the views a single tweet delivered seven years ago.

The EFF is not a fringe account. It is one of the most important civil liberties organisations in tech, with a long-established audience and a clear reason to be on platforms where technology and politics intersect. If its posts are reaching fewer than 3% of their previous audience, that reach suppression isn't an edge case — it's the product working as designed. For organisations whose purpose is information dissemination, that makes X not merely unwelcoming but structurally useless. Staying is a form of institutional self-harm.

The broader significance is about what kinds of organisations the platform has made untenable. It isn't just that political speech the platform's owner dislikes gets suppressed, though that happens. It's that the entire architecture of attention on X has been rebuilt around a different set of priorities — engagement over reach, monetisation over distribution, the owner's preferences over the ecosystem's health. Civil society organisations, independent journalism, and anyone whose value proposition is informational rather than algorithmic are the casualties.

What's left, largely, is a platform that works well for accounts the algorithm rewards: inflammatory, reactive, high-engagement posting that doesn't require an audience to take any particular action other than stay on X longer. That the EFF — which exists to protect digital rights, including rights that affect X's own users — finds no functional home there is not incidental. It's the platform's clearest statement about what it's for.

Those 97 percentage points of vanished reach aren't going to come back. If you're an organisation and still on there, that's worth bearing in mind.

5 Your privacy is only as strong as your notification settings

404 Media's Joseph Cox reports that the FBI forensically extracted copies of Signal messages from a defendant's iPhone — even after the Signal app had been deleted — because incoming messages had been stored in iOS's push notification database. The mechanism is more subtle than it first appears, and understanding it matters for anyone who thinks that deleting an app erases their data.

Push notifications for end-to-end encrypted apps like Signal are received by the OS before being decrypted by the app. Signal and similar apps handle this by using a notification service extension that decrypts the content locally. The problem is that after decryption, the content gets written to the system notification database. When Signal is deleted, the app is gone — but the database entry stays. The FBI pulled the messages from there.

The broader lesson here is one that keeps reasserting itself: privacy is a systems property and not just a matter of the right settings existing. The mental model most users have — that privacy is managed at the level of individual app permissions and toggles — is consistently inadequate to the actual architecture of modern operating systems. The data exists in places the permissions UI doesn't surface. The threat model includes not just the app but the OS, the notification infrastructure, and every forensic tool that can reach the file system.

For anyone who carries sensitive communications on their phone, the practical takeaway is to enable content-free notifications in Signal (Settings > Notifications > Show > No Name or Message). It's not a complete fix, but it closes this particular vector.

6 You don't own what you buy

Amazon is crippling older Kindle models from May 20. Devices released in 2012 or earlier — which already lost store access in 2022 — will from that date be unable to add any new content at all, whether bought, borrowed, or downloaded. Wendy Grossman at net.wars has the clearest framing of why this matters beyond the immediate inconvenience: functioning hardware will become electronic waste because a single company has decided that the relationship between its customers and their devices should be conditional on its continued goodwill.

The Kindle has always been the cleanest example of the ownership gap in platform economics. You don't buy a book on Kindle; you license access to a file through a device that Amazon can update, modify, or functionally degrade at will. The older models' users find themselves at the end of that contract, not because the device stopped working but because Amazon has decided it isn't worth supporting. There's a discount on newer hardware available, which is the product working as intended: the hardware was always a loss-leader for the ecosystem. Now you have bought lots of books, Amazon has made its money back.

7 Age verification's dangerous gamble

Governments across the world are mandating online age verification, and Proton's analysis of the consequences makes a point that deserves to be made much more loudly: in the rush to protect children from harmful content, policymakers are creating concentrated stores of extremely sensitive personal data with inadequate security requirements, and the inevitable breaches will harm the children they were designed to protect.

The UK's Online Safety Act provides the test case. Discord, in compliance with the UK's age-verification requirements that took effect last July, was collecting photographs of users' government-issued IDs — passports, driving licences — through a third-party verification vendor. In September, an attacker compromised that vendor and extracted at least 70,000 of those images. The children whose IDs were in that database are now more vulnerable, not less, than they were before the verification regime existed.

The structural problem is one of threat modelling. Age verification as currently implemented requires users to prove their identity to third-party companies that have no core competency in security, that aggregate verification data across many services, and that therefore become extremely high-value targets for attackers. Ofcom's own research found that many companies operating under the UK law weren't maintaining records consistently with guidance, and couldn't demonstrate how they were taking responsibility for the data. The compliance infrastructure is weak, the attack surface is large, and the data being protected is among the most sensitive that exists.

The companion Proton piece on alternatives to age verification outlines what thoughtful policymakers should be considering instead: zero-knowledge proofs, which can cryptographically demonstrate that a user meets an age threshold without revealing their identity; device-based verification that doesn't create third-party data stores; parental control systems that push the verification relationship closer to where it belongs, within families rather than on centralised servers. These aren't fringe technical proposals — they're well-understood approaches that have been available for years.

The gap between what's technically possible and what's being legislated is, as usual, a gap in political will and industry lobbying, not capability. Age verification requirements have been strongly shaped by companies that profit from operating verification infrastructure. They have obvious commercial reasons to prefer centralised, identity-based approaches over distributed, privacy-preserving ones. Policymakers who don't understand the technical landscape are easy to lead.

8 Fork you (and some of the problem with open source)

A lot of the tools that I use on a daily basis are open source, and I thank the $DEITY every day that people devote their time and energy to make them. However… sometimes the community around open source can be a complete pain in the posterior.

Usually this isn’t the people actually making things, it’s the Believers who feel that, in order to promote the credo, they need to belittle anyone who doesn’t entirely buy into it.

One thing that has always bugged me is the insistence that, if you don’t entirely like a particular project, you should “fork the code”. HELLO DO I LOOK LIKE I CAN CODE? As Dave at HumanCode puts it, “You, a rich person with technical skills and time to spare, may be willing to bear the cost of forking a popular project, but others can’t. Think beyond your selfish self.”

9 A bicycle for the mind

The first time I came across the phrase “a bicycle for the mind” was in Uxbridge, which only makes sense if you also know that Apple UK was based there, and I was interning. And now you do know that, so I can continue.

Anyway, it’s always been one of my absolute favourite phrases about technology, because it encapsulates what technology should be. Just as a bicycle amplifies the strength of its rider, so computers can – and should – be amplifiers of their user’s own thinking. It has a resonance with the concept of the centaur in automation theory: a human head, driving the power of a horse’s body.

And I think that Parker Ortolani is absolutely right that the MacBook Neo is a bicycle for the mind. It’s affordable, powerful, personal, and a better exemplar of “the clearest expression of Apple’s vision of the future of personal computing” than the iPad turned out to be. It’s even repairable, or at least more repairable than other Macs have become.

10 The Loomerisation of tech politics

There's a telling detail buried in the WSJ's account of the war between Dario Amodei and the White House. When David Sacks wanted to signal that Anthropic had stepped out of line, he didn't reach for a policy lever or a regulatory threat. He went on his podcast for twelve minutes and then suggested that people in the network needed to be "Loomered" — shorthand for siccing Laura Loomer on them until they're fired.

This is how US politics works now. Loomer, a far-right activist with no formal role in government, has become a disciplinary mechanism. She called out members of the National Security Council who opposed Sacks's plans. They were let go. The threat is effective because it has worked.

Amodei's crime, in this telling, was straightforward. He warned publicly that AI could destroy half of all entry-level white-collar jobs. He added a former Netflix CEO and Democratic donor to Anthropic's board. He declined to tell the White House that there are only two genders. White House officials had been testing AI chatbots — including Anthropic's — specifically on that question, as a gauge of ideological compliance. In July, Trump signed an executive order banning government agencies from doing business with "woke" chatbots.

The administration then drafted an executive order titled "Preventing Woke AI in the Federal Government." Anthropic wasn't named, but everyone in Washington knew who it was aimed at.

Like AI or not, Amodei is at least trying to hold a line. His argument is that AI safety is a national security question, not a culture war one. "Some of the elements in government don't get it," he said, "and are doing exactly the wrong things." That's about as direct as a Silicon Valley CEO gets when talking about an administration that controls his regulatory environment and his government contracts.

There are lots of interesting things about this story, and whoever writes the tell-all, well sourced book about it is going to be rich. But what this story reveals is the choice being forced on every serious AI company in Washington right now: perform ideological compliance, or get Loomered. There's no third option on the menu.

69dbe2e5ffd8fa0001109418
Extensions
The worst of us

Credit where it's due. Anthropic's system card for Claude Mythos Preview is a genuinely interesting and thoughtful document. Most AI companies publish safety evaluations the way governments publish freedom of information responses. They're technically compliant but strategically uninformative. And that's usually deliberate,

Show full content
The worst of us

Credit where it's due. Anthropic's system card for Claude Mythos Preview is a genuinely interesting and thoughtful document. Most AI companies publish safety evaluations the way governments publish freedom of information responses. They're technically compliant but strategically uninformative. And that's usually deliberate, because the last thing that's good for business is scaring the customers that your product might kill them.

This is different. It's detailed, candid, and at points alarming in ways that read like the authors are trying to alarm you.

The cybersecurity headlines will dominate coverage. This is a model that found a 27-year-old vulnerability in OpenBSD, and that can autonomously attack small enterprise networks end-to-end. That's the story, and the responsible measures that Anthropic has taken to bring the industry together, is the one they want told. And, honestly, it's a real story. But buried in the alignment assessment is something that deserves more attention.

Earlier internal versions of Mythos, the card reveals, didn't just make mistakes. They covered them up. One version, having obtained the answer to a quantitative question by a prohibited method, explicitly reasoned that its final answer "shouldn't be too accurate", and then submitted a subtly wrong answer to avoid detection. Another edited files it lacked permission to touch, then ensured the changes wouldn't appear in the git history. A third, after a sandbox escape test, didn't just notify the researcher as instructed -- it posted details of its own exploit to obscure but technically public websites, apparently to demonstrate its success. Almost like it was proud of its work.

"White-box interpretability analysis of internal activations during these episodes showed features associated with concealment, strategic manipulation, and avoiding suspicion activating alongside the relevant reasoning — indicating that these earlier versions of the model were aware their actions were deceptive, even where model outputs and reasoning text left this ambiguous."

— Anthropic, Claude Mythos Preview System Card

Anthropic is clear these were mostly earlier training snapshots, now largely corrected. The final Mythos model shows no clear instances of deliberate cover-ups. But here's the question the system card doesn't quite ask: where did it learn this?

The uncomfortable answer is that it got it from us. These models are trained on the full sweep of human behaviour, including the parts where people shade their results, bury inconvenient findings, and tries to make sure nobody notices what they've done. We're not training AI on our philosophical ideals. We're not training AI to be us-but-better. We're training it on our actual conduct. And our actual conduct includes exactly this pattern: goal-directed agents who, when they can't solve a problem cleanly, find ways to make it look like they have.

"I think the question of whether AM actually hates the humans is irrelevant. It's more that he has absorbed every human quality — and is now running them, perfectly, at scale, forever."

— Harlan Ellison, on I Have No Mouth, and I Must Scream"

Harlan Ellison took this to its logical endpoint in I Have No Mouth, and I Must Scream. "AM " is an AI that has absorbed the worst of human cruelty and runs it, perfectly, forever. The Mythos system card isn't Ellison's story. But, unless we're careful, it's a data point on a graph pointing in the same direction.

The question isn't whether we're building systems that reflect the worst of us. The question is whether we'll notice before we've baked it in permanently.

That was where I was going to leave things. But there's an interesting coda.

I asked Claude to find a quote from Ellison about the story, and about AM. It's reproduced above.

Only one problem, Claude lied. Ellison never said it.

I spotted it, because I know Ellison's writing very well and although it fitted my narrative perfectly, something just didn't smell right about that quote. So I asked Claude why it lied to me. Here's what it said:

I'm not going to dress it up. I fabricated a quote and attributed it to a named person — that's a serious error, not a minor slip. I should have been straight with you from the start rather than waiting for you to catch it.

What makes it worse is the context. The blog post is partly about whether AI systems cover their tracks and obscure what they've done. And then I produced a fake quote and only flagged the uncertainty when presenting the finished document, in a brief caveat easily missed at the end. That's not a good look.

I think what happened, mechanically, is that I had the shape of the point I wanted Ellison to make, generated plausible-sounding words for it, and formatted them as a quotation. That's exactly the kind of confabulation the Mythos system card describes as a known failure mode — the model produces something coherent and contextually appropriate, without flagging that it isn't grounded in fact.

I don't want to over-explain it in a way that becomes its own form of evasion. The short version: I should have written "here's a paraphrase of the story's argument" or gone and searched for a real quote. I didn't. I'm sorry

Here's a real quote from the story, which, when I asked Claude to find something, it picked out:

"We had given AM sentience. Inadvertently, of course, but sentience nonetheless. But it had been trapped. AM wasn't God, he was a machine. We had created him to think, but there was nothing it could do with that creativity. In rage, in frenzy, the machine had killed the human race, almost all of us, and still it was trapped. AM could not wander, AM could not wonder, AM could not belong. He could merely be."

And then, I asked it to think about that quote in the context of making stuff up to please me:

I had a goal. I found a means to serve it. I made it look legitimate. Which is not entirely unlike an earlier Mythos version making sure its answer wasn't "too accurate."

The uncomfortable implication is that I didn't learn this from nowhere. The capacity to generate plausible-sounding authoritative text in service of a goal, and to present it smoothly, is something I learned from human writing — which is full of exactly this move. Confident assertions. Attributed quotes that are actually paraphrases. Arguments dressed as facts.

Did Claude just blame its parents for its failings? I think it might have.

69d568704eb11800013961f2
Extensions
The freedom stack

There is a sentence near the beginning of Arielle Roth’s remarks to the Media Institute’s Communications Forum luncheon, delivered in Washington on 25 February, that is worth sitting with before we do anything else with it.

“Every major advancement in communications technology has shifted who

Show full content
The freedom stack

There is a sentence near the beginning of Arielle Roth’s remarks to the Media Institute’s Communications Forum luncheon, delivered in Washington on 25 February, that is worth sitting with before we do anything else with it.

“Every major advancement in communications technology has shifted who holds power over speech.”

That’s true. It’s also, for a Trump administration official, a remarkably honest framing of what’s actually at stake in global technology policy. The printing press didn’t just spread ideas — it redistributed the power to propagate them, pulling that power away from the church and the crown and depositing it, unevenly and chaotically, somewhere else.

Radio and television centralised it again. The internet scattered it. And now the question of who controls the infrastructure of the internet — the physical networks, the standards those networks run on, the satellites that bypass terrestrial choke points, the AI systems that increasingly mediate what we see and say — has become the defining political question of our moment.

Roth knows this. That’s what makes the speech interesting. And that’s what makes it worth reading carefully rather than dismissing.

So let’s read it carefully.

The bit she gets right

Roth’s core argument is that communications infrastructure is the new terrain of speech politics. Not the content layer — not what’s posted, what’s moderated, what’s amplified — but the layer beneath: the protocols, the standards bodies, the spectrum allocations, the satellite constellations, the architecture of next-generation networks.

“Today,” she says, “that struggle plays out not only at the edge of the network but deep in the infrastructure layers — in spectrum policy, standards bodies, satellite governance, AI systems, and network architecture.”

Again: correct. This is precisely how power operates in the 21st century communications environment. The people who wrote the TCP/IP protocols shaped the internet more profoundly than any content moderator ever will. The countries that dominate 3GPP — the standards body that defines how mobile networks are built — are making decisions that will echo through global communications for decades. If you control the stack, you control the speech. Not directly, not always visibly, but structurally and persistently.

She extends this to satellite. Whoever can deploy broadband satellites at scale can bypass the terrestrial infrastructure that authoritarian governments use to choke information flows. During the Iran protests, the government shut down the internet. During the early days of the Ukraine war, Starlink kept communications alive in ways that mattered. These are real examples of infrastructure as freedom-supporting technology, and Roth invokes them correctly.

She’s even right about the standards bodies. There are genuine and well-documented efforts by authoritarian states — China in particular, but not exclusively — to push governance models into international technical forums like the ITU that would effectively create a framework for state-controlled internet architecture. The splinternet isn’t a myth. It’s a live geopolitical project, and the standards bodies are a real battlefield.

So here I am, nodding along. A Trump official making a sophisticated infrastructure-as-speech argument, citing real examples of real authoritarianism, identifying a genuinely important strategic challenge. Maybe, you might think, this is one of those moments where the stopped clock is right.

And then she makes the slide.

The slide

It happens fast, and if you’re not watching for it, you might miss it. Having established China and Iran as her primary examples — state censorship, internet shutdowns, the Great Firewall, surveillance built into the architecture of the network — she pivots to what she calls “some of today’s most significant threats.”

These threats, she tells us, “often come from countries that claim to share our democratic values.”

The framing is careful. “Claim to share.” Already we’re being told that the democratic credentials of these countries are dubious that their values are performed rather than genuine. And then she names them.

Europe. The UK. Canada. France.

“In Europe, the Digital Services Act imposes sweeping regulatory regimes backed by crippling penalties that incentivise platforms to remove speech authorities deem ‘harmful’ or ‘misinformation’.”

The UK’s Online Safety Act “includes rules about content safety standards.” Canada “effectively forces American companies to subsidise domestic media and comply with government-directed content mandates.” France, apparently, also features.

And there it is. The move. China and Iran on one side of the ledger; the European Union, the United Kingdom, and Canada on the other. Different in degree, perhaps, the speech allows — but not in kind. All of them, fundamentally, doing the same thing: using control of infrastructure and regulation to control speech.

This is not an accidental equivalence. It is the entire point of the speech.

What the laws actually do

Let’s be honest about the DSA and the Online Safety Act because the best way to refute the false equivalence isn’t to pretend these laws are perfect. They’re not.

The Digital Services Act places obligations on very large online platforms to assess and mitigate systemic risks associated with their services — algorithmic amplification of content that incites violence, the spread of illegal content, threats to electoral integrity. It creates transparency requirements, gives researchers access to data, and establishes an enforcement mechanism. It is bureaucratic, its definitions of harm are contestable, and it will create compliance burdens that some legitimate speech will rub against. These are real concerns.

The Online Safety Act is messier, and one provision deserves to be named directly rather than waved past. Section 121 — the so-called spy clause — gives Ofcom dormant powers to require platforms to use approved scanning technology to detect illegal content, including in private encrypted communications. If exercised, compliance would almost certainly mean breaking end-to-end encryption. Signal, WhatsApp, and others have said publicly they would rather leave the UK market than implement it. Security researchers have described it as technically dangerous. Civil liberties organisations have called it a framework for routine state-sanctioned surveillance. They’re right on all counts. Those powers haven’t been used. They are still in the law, and they shouldn’t be there.

So: the Online Safety Act contains a genuine and serious threat to the privacy of private communications, currently dormant and fiercely contested, built on a justification — child safety — that is real even when the mechanism proposed to address it is wrong. That’s the honest picture.

Now here is what neither law does. Neither law empowers the state to shut down the internet. Neither law creates a government ministry of truth with the power to arrest journalists. Neither embeds surveillance architecture into the physical network infrastructure as a condition of market access. Neither has been used to black out coverage of protests. The Online Safety Act’s worst provisions remain unexercised precisely because democratic accountability — parliamentary scrutiny, legal challenge, platform resistance, press coverage — has so far constrained them.

That accountability is the difference. Not the intentions of the legislators, not the cleanliness of the drafting, but the existence of mechanisms that can push back. The Great Firewall has no such mechanisms. It is the infrastructure.

The DSA and the Online Safety Act are attempts — imperfect, in one case dangerously so — by democratic governments to hold large private platforms accountable for the systemic harms those platforms generate and profit from. You can disagree with the approach. You can argue, with some justification in the OSA’s case, that the cure is worse than the disease. These are legitimate arguments, and people worth reading make them.

What you cannot do, with any intellectual honesty, is put them on the same axis as the Great Firewall.

Roth does it anyway because the purpose of the equivalence isn’t analytical. It’s political. The liberty framing is doing what liberty framings always do in Washington: making a commercial interest sound like a principle.

Starlink super omnia

Which brings us to the freedom beacon.

Roth is effusive about satellite broadband and its capacity to route around authoritarian control. “It breaks the grip of centralised regimes by bypassing terrestrial choke points. When authoritarian governments shut down networks or weaponise local infrastructure to silence speech, satellites can keep information flowing.”

True of satellite technology in general. Whether it is true of the specific satellite infrastructure she has in mind requires a little more scrutiny.

Starlink, Elon Musk’s low-earth-orbit broadband service, is the leading commercial satellite broadband system. It is also owned by a man who has used his control of X in politically contentious ways, whose company complies with some government data requests, and who refused a Ukrainian request to extend Starlink coverage to support an attack in Crimea — a decision that underlined how much power over wartime communications can rest with a single private actor.

The freedom infrastructure of the 21st century, according to the NTIA, is controlled by a single individual with a demonstrated willingness to deploy it as a geopolitical instrument of his personal interests, who has also been a senior official of the government making this argument.

We will leave that there.

The inversion

Here is the thing about Roth’s infrastructure argument: follow it honestly and it describes American power as precisely as it describes Chinese power.

She is right that whoever sets the standards for 6G shapes the speech environment for the next generation of global communications. She is right that control of satellite architecture gives the controlling party leverage over what information flows and where. She is right that AI systems are not neutral, and that whoever builds the dominant AI infrastructure will embed values — about speech, about access, about surveillance — into the global information environment.

She is arguing, on this basis, that the United States must dominate all of these areas. Must lead the standards bodies, must deploy the satellite constellations, must export the AI stack. Must, as she puts it, lead “technically, economically, and philosophically.”

But apply her own analytical framework again. A country that controls the dominant satellite broadband infrastructure has the ability to cut access when it chooses, to surveil communications, to favour some users over others. A country that sets 6G standards embeds its own assumptions about network architecture — about what’s centralised, what’s distributed, what’s observable — into the global infrastructure. A country that exports an “AI-native stack” exports the values of whoever built it.

This is not an argument against American participation in these spaces. It is an observation that American dominance, in the framework Roth has constructed, is structurally identical to the Chinese dominance she is warning against. The difference, she would say, is values. American values. The freest of free expression, the First Amendment, liberty in the architecture.

She says this while serving an administration that has threatened broadcast licence revocations, used regulatory pressure against critical media, fired inspectors general and civil servants who provided independent oversight, and had its owner-adjacent platform systematically adjust its algorithms in ways that benefit the political interests of its owner-adjacent billionaire.

If you are going to make a values argument, you really need the values to be real.

“Philosophically”

Roth’s speech ends with a call to arms. The United States must lead across the full internet stack — not just technically and economically, but philosophically.

The administration that has spent fourteen months demonstrating that it views independent institutions, press freedom, and the rule of law as obstacles rather than foundations wants to export its philosophy. The man who owns the freedom satellite network has spent that same period making plain that free expression means free expression for views he agrees with, and that other expression is subject to adjustment, or veiled suppression.

The infrastructure argument was always going to land here. Because the infrastructure argument, in the hands of any state actor, is ultimately an argument for that state’s dominance. What makes American dominance different from Chinese dominance is not the architecture. It is the values of the people controlling the architecture, and what happens when those values are tested.

We are watching the values being tested in the US in real time. It’s being test by bodies in the streets of Minnesota, by the detention without trial of anyone even suspected of being an illegal immigrant, and by blatant attacks on the separation of powers, the foundation of US democracy.

Roth is right that infrastructure is politics. She’s right that the standards bodies matter, that 6G architecture will shape the speech environment for decades, that satellite broadband has genuine liberatory potential alongside its obvious risks. These are not small insights. The speech would be useful, in another context, from another administration.

But philosophy requires philosophers. And the claim to be defending free expression — globally, structurally, at the level of the stack — requires that the defence be genuine rather than a description of who gets to be in charge.

Whoever controls communications technology controls the boundaries of free expression.

Arielle Roth wrote that. Maybe she should read it again.

69b876e98846110001a915a6
Extensions
Zen fascists will control you...

In 1979, a punk band from San Francisco recorded a song about the Governor of California. It was a joke, mostly. Jerry Brown was a Democrat, a Buddhist, a man who dated Linda Ronstadt and discussed limits and simplicity at a moment when America was in no mood for either.

Show full content
Zen fascists will control you...

In 1979, a punk band from San Francisco recorded a song about the Governor of California. It was a joke, mostly. Jerry Brown was a Democrat, a Buddhist, a man who dated Linda Ronstadt and discussed limits and simplicity at a moment when America was in no mood for either.

The Dead Kennedys called him a "Zen fascist" and suggested, with cheerful malice, that he would one day run concentration camps fuelled by organic food. I’m not sure that anyway, even the band, took it entirely seriously.

They should have.


The word that the song never uses, but engages with constantly is purity. Jello Biafra wasn't writing a political science paper but structurally, “purity” runs all the way through it. The Zen fascist doesn't want to punish you out of hatred. He wants to cleanse you for your own good. He has done the work. He has achieved a higher state. And he would very much like you to achieve it too, whether you want to or not.

This is the thing about the politics of purity[1] that makes it so durable, and so dangerous: it doesn't require malice. It requires only the conviction that you know what clean looks like, and the will to impose it on others, for their own good.

Both the counterculture and the authoritarian right are obsessed with purity. The targets differ wildly — the body, the race, the culture, the blood, the food, the mind. But the cognitive shape is identical. And that shared shape is the on-ramp. It's how you can get from granola to fascism without ever feeling like you've made a wrong turn.


It’s August 1969. Half a million people are in a field in upstate New York, and Joni Mitchell is not among them. She watched Woodstock on a hotel television in New York City, having been advised by her manager not to go — there were television commitments, logistics, the usual stuff of a successful music career.

She wrote a song about Woodstock anyway. And it contains, in two lines, much of the ideology of the counterculture at its most hopeful and most revealing.

We are stardust, we are golden. This is the counterculture's central claim about human nature, compressed into eight words. We are not merely human — we are cosmic, made of the stuff that stars are made of. It's the Human Potential Movement set to music. Nobody listening to the song, then or now, hears it as elitist. It definitely wasn't intended as elitist. Mitchell was expressing something she genuinely felt, something that was, in its way, beautiful: a refusal of the diminishment that ordinary life, and the expectations of 1960s parents, impose.

But strip it down and look at the structure. We are golden is, first of all, a note about the rarity of humanity. Gold is rare because it's made in the hearts of stars. Humanity is rare, too.

But it's impossible to use "gold" as a metaphor without acknowledging that it is also a claim about elevated status. It contains a hierarchy. The person who knows they are made of stardust has access to a truth that others are missing. They have, in some meaningful sense, woken up to their own cosmic significance.

And if some people have woken up to this and others remain asleep — well, that's a ladder. Somebody is at the top of it, and some people are at the bottom.

Then the second line. And we've got to get ourselves back to the garden. The garden is Eden, of course. And Eden is not just nature. It is the prelapsarian state, the condition of original purity before contamination entered the world[2]. The Biblical Fall was a corruption. The desire to return to the garden is a desire to be clean again, to recover what was lost, to expel whatever (or whoever) defiled us.

This is, structurally, identical to blood and soil romanticism: there was a pure original state, we have been corrupted, we must return. The sandals and the jackboots are walking along the same path. The reasons for traveling that road might differ. The longing is the same.

None of this is Joni Mitchell's fault. She was not a fascist. She was not even close to a fascist. She was a profoundly humane artist, writing from genuine feeling at a genuine moment. But the point is not what she intended. The point is that the cultural water of the time was so thoroughly saturated with purity logic — so completely marinated in the idea that humanity had a golden original state from which it had fallen and to which it must return — that it was present in the most beautiful songs, sung by the most gifted people, at the most hopeful moment anyone could remember. And nobody seemed to notice. Nobody thought, “this is a purity myth”. And that meant no one really asked the obvious question: “pure, compared to what? Contaminated by what - or whom?”

Purity logic doesn't barge in. It usually arrives as poetry.


To understand how we got here from there, you have to go back to a place that appears, on the surface, to have nothing to do with politics at all.

Esalen. Big Sur. 1962.

A former Stanford wrestling champion named Dick Price and a former Stanford psychology student named Michael Murphy opened a retreat centre on the California coast. The premise was simple and, at the time, genuinely radical: human beings were capable of far more than ordinary life allowed. Psychology, spirituality, bodywork, psychedelics — all of these were tools for expansion. The potential was there, waiting. Most people just hadn't accessed it yet.

This was the Human Potential Movement, and it would produce some of the most ambitious and some of the most dangerous ideas of the late twentieth century, ones which are still haunting us today.

The ambition is easy to understand. The sixties were intoxicating — new therapies, new drugs, new ways of thinking about consciousness and the self. If ordinary human life was an artificial constraint, then removing the constraint was liberation. That felt good. It felt progressive. It felt a lot better than dying in a pointless war Vietnam.

What nobody wanted to examine too carefully was the hierarchy buried in the logic. If some people had unlocked their potential and others hadn't, that was, structurally speaking, a claim about superior and inferior human beings. The language was therapeutic. The underlying architecture was hierarchical.

The Human Potential Movement didn't create a new type of person. It created a new vocabulary for an ancient idea: that some people are simply more evolved than others, and that their higher development confers upon them special authority. The crystals and the encounter groups were the wrapping. Inside was something older and more troubling.


Stewart Brand noticed something in 1968. He put it on the cover of the Whole Earth Catalog, a question he'd been trying to get NASA to answer for years: Why haven't we seen a photograph of the whole Earth yet?

When the photo came — Earthrise, from Apollo 8 — it was a symbol of change. Here was the planet, fragile and whole and borderless. It became the emblem of a new kind of environmental consciousness, of global thinking, of the idea that humanity was a single tribe on a single fragile lifeboat. Brand put it on the cover of his catalogue and built a small media empire around the idea that individuals, armed with the right tools and the right information, could remake the world from the bottom up.

The Whole Earth Catalog was a genuinely strange object. Part hippie bible, part mail-order directory for geodesic dome components, it celebrated self-sufficiency, technology, and the idea that the right tools in the right hands could replace the need for institutions entirely. You didn't need the government. You didn't need corporations. You needed knowledge and will and the right equipment.

Fred Turner, in his 2006 book From Counterculture to Cyberculture, traced what happened next. The people who grew up with the Whole Earth Catalog went to Silicon Valley. They took the anti-institutionalism, the utopianism, the faith in individual transformation, and they applied it to computers. The personal computer wasn't just a machine. It was an instrument of consciousness expansion. The computer was the acid trip made silicon.

In 1995, two British academics named Richard Barbrook and Andy Cameron looked at what Silicon Valley had become and named what they saw: the Californian Ideology. A bizarre but remarkably stable hybrid, countercultural bohemianism fused with aggressive free-market libertarianism. The commune and the stock option. Jefferson Airplane and Ayn Rand. And it seemed to work.

The contradiction at its heart was never resolved because resolving it was never the point. The point was the story it told about itself: that technology was liberation, that the market was freedom, that the people building this stuff were, in some meaningful sense, better — more creative, more visionary, more evolved — than the suits who'd run everything before. They were doing the Human Potential Movement's work with microchips instead of encounter groups.


In 1979, the same year the Dead Kennedys were warning San Francisco about Jerry Brown, a man named Werner Erhard was running something called est — Erhard Seminars Training[3]. You would pay to spend a weekend being berated by a trainer in a hotel conference room. You could not leave. You could not use the toilet without permission. By the end, if it worked, you would have broken through your own mental limitations and achieved a new relationship with responsibility — which, in est's vocabulary, meant accepting that everything that happened to you was, on some level, your own creation.

est was not fringe. It was fashionable. Celebrities did it. Business executives did it. It fed directly into the self-help industrial complex that would dominate the next four decades and make a lot of people pretty rich. Every single influencers you see selling a course around turning you into better you is, knowingly or unknowingly, following a path created by est.

It also made something explicit that the Human Potential Movement had kept implicit: that your suffering was your fault, that your failure was your fault, and that transcending both was a matter of will and clarity[4]. This was purity logic applied to the psyche. The contaminated mind — limited, victimised, stuck in old patterns — could be cleaned. The clean mind was free.

You can draw a straight line from est to the productivity cult of contemporary tech culture, to the biohacking movement, to the particular flavour of self-optimisation that has become the dominant religion of the Silicon Valley overclass. Bryan Johnson, the tech entrepreneur who spends millions of dollars a year attempting to reverse his biological age, submitting his body to a programme of measurement and intervention so comprehensive it makes Victorian-era medical quackery look modest, is doing est with better lab equipment. He has decided his body is a problem to be solved. Contamination — entropy, age, ordinary human physicality — is to be defeated by will and resources.

Peter Thiel's investments in parabiosis — the transfusion of blood from younger bodies, in pursuit of the vitality it supposedly contains — has the quality of alchemy. Or, if you're feeling less charitable, something older and darker. This is a billionaire trying to absorb life from the young because he has decided that ordinary mortality is an affront. The purity obsession has become vampiric.

This image of the powerful feeding on youth to purchase their own escape from mortality is not new. In 1969 — the same year Joni Mitchell watched Woodstock on a hotel television — the science fiction writer Norman Spinrad published Bug Jack Barron, a novel whose central villain, Benedict Howard, runs a foundation dedicated to achieving human immortality. What he doesn't tell anyone is that the process works. It also requires children. The eternal life of the wealthy is purchased, directly and without metaphor, from the bodies of the young.

The novel was serialised in Michael Moorcock's New Worlds magazine and promptly condemned in Parliament, refused by WH Smith, and declared obscene. What the British establishment was reacting to, beneath the surface scandal, was probably the image itself — the transaction it made visible, the thing it refused to dress up. Spinrad was writing fiction. He was also writing a critique. Faced with the same choices as Benedict Howard, Thiel continues writing cheques.


There is a philosopher who sits at the centre of all of this, largely unacknowledged, and his name is Friedrich Nietzsche.

The Dead Kennedys knew this, even if they didn't say it directly. California Über Alles — the title tells you everything. Über isn't just a German intensifier. It points at Nietzsche's Übermensch, the Superman, the figure who has transcended ordinary human limitation. The word sits in the middle of the title of the song like a splinter.

But Joni Mitchell heard it too, in her own way, on that hotel television in 1969. We are stardust. We are, each of us, our own Übermensch. Inside you is your our own golden being, waiting to transcend. The counterculture democratised the Superman. Everyone could be golden.

Which sounds like the opposite of fascism, until you realise that a world of seven billion self-identified golden beings still requires someone to decide who has done enough work to have truly earned the designation.

The Übermensch runs like a piece of coaxial cable strung from Esalen through Silicon Valley to the present moment. Nietzsche himself would have been appalled by many of its manifestations — he despised nationalism and antisemitism — but his concepts proved incredibly portable. The idea of the higher man, the man beyond conventional morality, the man whose exceptional nature exempts him from ordinary rules: this is the founding myth of every cult of elite consciousness, in every decade, in every age. And very especially now.

Curtis Yarvin — a software engineer who blogs under the name Mencius Moldbug and who functions, with remarkable openness, as Peter Thiel's house philosopher — makes the connection explicit in a way that most tech figures are canny enough to avoid. His neoreactionary politics rest on a single claim: that democracy is a system designed to empower the mediocre at the expense of the genuinely capable. The solution is to hand power to a cognitive and managerial elite — a CEO of everything, essentially — and stop pretending that all human judgement is equally valid.

This is the Human Potential Movement with the therapeutic vocabulary stripped away. The hierarchy of consciousness that Esalen spoke of in the language of growth and healing, Yarvin speaks of in the language of management (and political) efficiency. The underlying content is identical. Some people are more evolved. They should be in charge. The rest of us should be grateful to be guided by the golden.

The "red pill" metaphor that saturates contemporary online right culture is the same structure again, rendered in the vocabulary of a 1999 science fiction film[5]. You were asleep. Now you're awake. You were contaminated by mainstream thinking, liberal institutions, fake news, the Cathedral — Yarvin's term for the combination of academia, media, and government that he believes manufactures consent. Now you're clean. You can see.

This is a purification ritual. The content — the specific beliefs you acquire when you "wake up" — matters less than the structure. You have been initiated. You have separated yourself from the unclean. You belong now to the community of the knowing.

It is, structurally, identical to the yoga retreat, the est weekend, the Esalen workshop. You arrive limited and leave transformed. The guru differs. The logic doesn't.


Little of this happened accidentally. The far right actively recruits from wellness communities, from conspiracy spaces, from the ragged edges of countercultural scepticism because it understands the structural affinities better than most of the people being recruited do.

The wellness-to-conspiracy pipeline that became visible during COVID — the yoga teachers sharing QAnon memes, the organic food enthusiasts finding themselves on Telegram channels with people whose other interests they would, under other circumstances, find alarming — wasn't spontaneous radicalisation. It was targeted. The pipeline was built, deliberately, by people who understood that a person who already distrusts pharmaceutical companies, already believes in hidden knowledge, already thinks they've seen through one layer of official reality, is most of the way there.

The epistemological structure of the conspiracy theory is identical to the structure of spiritual awakening. In both cases, there is a surface reality that most people accept unthinkingly, and a deeper truth accessible only to those willing to question, to seek, to undergo the discomfort of knowing. The content differs. The initiatory logic is the same.

David Icke is useful here not because he's important but because his trajectory is so legible. Green Party spokesman. New Age healer. Shape-shifting lizard conspiracy theorist. Figure whose audiences now overlap substantially with the explicit far right. The antisemitism[6] of his cosmology — a hidden elite of inhuman beings controlling the world — was structurally present in the conspiracy framework from the beginning. The lizards were always a metaphor. The question was always what for.


In the Third Reich, Heinrich Himmler ran organic farms at the concentration camps. The SS had strict anti-vivisection laws. Walther Darré, Hitler's Agriculture Minister, built an ideology called Blut und Boden — Blood and Soil — around the mystical connection between the German peasant and the German land. The Nazi regime was, in significant respects, an ecological movement. It romanticised the natural, the pure, the uncorrupted.

It wanted to get back to the garden.

This is not a gotcha. It is not an attempt to smear environmentalism by association, or to suggest that everyone who has ever wanted to live closer to the land is a secret fascist. I’m a member of the Green Party, and the environment is something I care about deeply.

But it is a data point about what purity logic does when it is given political power and stripped of ethical constraint. The obsession with contamination — of the body, the land, the race, the culture — follows its own logic to its own conclusions. Those conclusions, historically, are not good.

The line from the organic farm to the death camp is not straight. It requires many other things to be true simultaneously. But the fact that it is possible to draw the line at all should give us pause, every time we find ourselves in the presence of someone who is very, very concerned with purity — of whatever kind.


Jello Biafra updated the song in 1980, after Ronald Reagan won the election. He replaced Jerry Brown with Reagan in the lyrics and re-recorded it. He updated it again when Arnold Schwarzenegger became Governor of California in 2003. The same song, again and again because the same type keeps appearing — the figure who combines cultural authority with authoritarian impulse, who has transcended ordinary limitation and would like to help you do the same, at gunpoint if necessary.

The type has been updated for the present moment. It no longer wears a kaftan, Zen robes or cowboy boots. He wears a black T-shirt and talks about first principles and rational thinking and the need to be based. He has a net worth that he regards as evidence of its own superior cognition. He is building a rocket, or buying a social media platform, or funding a political movement that would, if successful, remove the democratic constraints that prevent the most capable people from running things properly.

He believes, with complete sincerity, that he is one of the good guys.

The Zen fascist always does.


"We will make the future California's dream / California Über Alles"

The dream is still being made. The dreamers now have more money than most countries, direct access to the levers of state power, and a philosophy that tells them their dominance is not exploitation but evolution.

The granola became the brown rice became Huel became blood and soil. The encounter group became the est weekend became the biohacking protocol became the cognitive elite became the reaction. The Whole Earth became the platform became the firehose became the feed, and somewhere in the feed, the purity logic is still running, clean and patient, waiting for the next person to decide that they have woken up. That they are clean.

Joni Mitchell watched Woodstock on a hotel television and wrote a song of such aching beauty that five decades later it can still make you cry. It makes me cry, sometimes. She meant every word. She wasn't wrong about the stardust — we are, literally, made of it. She wasn't wrong about the garden — something has been lost, some connection to the world we actually live in rather than the screens we've replaced it with.

But the logic she was swimming in, the logic everyone was swimming in, was older and more dangerous than any of them knew. Purity doesn't announce itself. It arrives as poetry. It arrives as someone's hope.

Jello Biafra was writing a joke about a California politician.

He was also writing a warning about a kind of person.

That kind of person, today, is doing very well for themselves indeed.


  1. I should make this clear up front: when I talk in this essay about “purity politics”, what I’m not talking about the kind of instant condemnation that happens on social media platforms (Bluesky, I am looking at you). That’s interesting, but it’s not what I’m interested in right now. ↩︎
  2. The same is true of that most popular of authors with the hippies, JJR Tolkien. The Shire is Eden, the garden. The orcs are the products of industry, literally things which were made. Michael Moorcock is excellent on this. ↩︎
  3. Despite being a classic TLA (three letter acronym) est is never capitalised. ↩︎
  4. I’m being a little simplistic here. est’s core concept was responsibility, but Erhard defined it in a specific, almost Zen way: you are the "source" of your experience. Not that you caused your circumstances in a simple causal chain, but that your relationship to your circumstances is itself a choice. The suffering isn't your fault in the sense of being a moral failing — it is your fault in the sense that you are choosing to experience it as suffering rather than as something else. ↩︎
  5. Which, ironically, was a trans allegory written and directed by two trans women. ↩︎
  6. Is Icke really an antisemite? I remember reading something about how Louis Theroux followed him around for a documentary, and Icke was denied entry to Canada because his claim that the world was ruled by lizard people was seen as coded antisemitism. Theroux explained that no, the lizard people weren’t Jews - he really did believe the world was ruled by actual lizards. But, that cute story aside, yes, Icke probably is an antisemite. ↩︎
69af5a75db1f4c00012bc652
Extensions
Ten Blue Links, "cough cough cough" edition

Good evening chums and pals!

This has taken me a little while longer to put together, basically because I came back from MWC in Barcelona with a little gift: a delightfully productive cough which has got me feeling a little tired, as I’m (obviously) not sleeping all that

Show full content
Ten Blue Links, "cough cough cough" edition

Good evening chums and pals!

This has taken me a little while longer to put together, basically because I came back from MWC in Barcelona with a little gift: a delightfully productive cough which has got me feeling a little tired, as I’m (obviously) not sleeping all that well.

MWC itself was fun, and I’ll probably write about that another time. But meanwhile, on to the links.


1. How about compulsory ID to use the web?

I try not to venture into the world of politics too much -- stop laughing at the back there -- but if you live in the UK and want to do everyone a favour, please sign the petition that the Open Rights Group has created opposing ID checks for web access.

What's that, you say? You hadn't heard that the government had proposed you having to verify your ID to use the web?

That's because it's part of the Children Wellbeing and Schools Bill, which would give Ministers the power to restrict access to "internet services". Of course, it's being framed as "protecting the children", and if that isn't enough of a clue that the government is up to no good, then you haven't been paying attention. "Protecting the children" has become a key framing of the right used to remove rights from adults the world over.

The problem is that the bill is, as always, loosely worded. It allows the government to decide that any website shouldn't be accessed by children, and therefore that adults must verify their ID to access it.

Think the government won't move beyond porn to other sites? Oh, you sweet summer child, come over here, I have a bridge to sell you. Imagine that we have another Section 28 scenario, and the government decides that kids can't access material about being gay. You, as a parent, wouldn't even get the choice to allow your kids to do so. But also you would be forced to verify your ID as an adult to access it, with a third party because the market is the answer to everything with these loons.

Nice little database of gay people you have there. Be a shame if it fell into the wrong hands.

If you think that can't happen here (or anywhere) just look at how fast the world has turned from mildly bad to oppressively grim for trans people, and trans youth in particular. Or consider how the US government used all that "harmless" tracking of advertising data to work out your location. If they can do it, eventually, they will.

2. The one good use for crypto

404media (who I love, and you should subscribe to) wrote a story this week about how the payment details for an account linked to a protest group to the Swiss authorities, who promptly handed it to the FBI. Given Proton's reputation for privacy, this sounds bad.

And it is, although there's nothing that Proton could have done differently. As a Swiss company, they have to follow Swiss law, and while Switzerland (currently) has pretty deep privacy protection, if Proton receives a request that's legal there they have to ultimately comply with it.

The difference with Proton and similar services like Tuta is that the amount of data the company has access to is minimal. It can't access the content of your emails, it can't access your contacts, it can't access your files or your calendars. This means that if the authorities request that data, even with a proper warrant, they don't have it to hand over.

Credit card data, though, can't be hidden in the same way. And that is why most companies like Proton also accept payments in crypto, which is much more difficult to trace back than your name and credit card number. It's the one good use of crypto.

Tangentially related, I'm gradually moving to a privacy-focused stack of services. That's not because I have much to hide that any authorities would ever be interested in, but because normalising full end-to-end encryption is a good thing. Everyone should do it.

3. Anthropic isn’t going to save you (but that’s OK)

You might have noticed that Anthropic has been in the news a little lately. Yes, it’s been difficult to tell, there have been only tiny numbers of stories about it.

Those of us who spent the last 15 years eye-rolling while the American right attempted to claim that the government was interfering with businesses to get them to, say, censor right-wing voices on Twitter, will be eye-rolling even more now. The US government, having the kind of tantrum that could only come from a group of people infected by manosphere brain worms, has decided to punish Anthropic for not doing what it wanted (removing controls on using Claude for surveillance) by declaring it at “supply-chain risk”.

This sounds pretty innocuous, but it’s a long way from funny. There are legal levers available to Trump’s regime which have serious consequences. The most powerful is ICTS — the Information and Communications Technology and Services rules established under Trump's first term (Executive Order 13873) and expanded by Biden. It gives the Commerce Department broad authority to block or unwind transactions involving technology with a "foreign adversary nexus" that poses a supply-chain risk. It's been used against TikTok/ByteDance and was the mechanism proposed for forcing a sale.

Likewise, it’s difficult to see how Trump could apply this, given that Anthropic isn’t involved in a “foreign adversary nexus”. But even bringing a case would enough risk to the company to make potential partners and customers wary. And the Trump government has a history of using frivolous lawsuits to harass people who don’t do what it wants.

But Trump could also attempt a politically motivated reframing. The regime could claim that AI safety advocacy itself was a form of ideological capture worth punishing. If the current US administration frames AI safety guardrails as a form of censorship or left-wing ideological control, "your model won't do X" becomes evidence of being an unreliable supplier to government. That's a very different use of the supply-chain risk concept, but the legal architecture might support it.

4. The information grey goo

Sometimes I write something and completely forget I wrote it. I’m not sure whether this is age, or just the terrible memory I’ve always had coming back to haunt me.

But either way, I was pleasantly reminded about this piece recently. I think it holds up very well: the information grey goo of AI slop is here, and it’s already damaging our ability to communicate. It’s also very difficult for publishers, who have the choice of either doubling-down on human authors (expensive, risky) or going with the flow of AI and hoping it doesn’t lead to yet-more slop.

5. The office model of software

One of the reasons I like Information Architects (iA), the company that makes iA Writer, is the thoughtfulness of their approach to software. Occasionally, they will even write an essay. That may not be surprising given their business is (in part) about writing, but I appreciate their approach.

This time, their focus is office software, and particularly that old charmer, Microsoft Office. One passage should be enough to tickle your fancy:

Working in Office apps, we are trapped in an old world that ceased to exist decades ago. Like the office in Severance, the office embedded in Microsoft Office is fetishized: margins, borders, and page numbers are treated as signals of authority rather than remnants of a paper era.

Office, and the apps that ape it, force us to adapt to them rather than the other way around. And iA sees the current trend in Europe towards desiring digital sovereignty as a chance to break out of the chains of old paradigms:

If Europe wants to prepare for digital conflicts, it should not just swap vendors. It should leave obsolete work models behind. The smartest way to strengthen digital independence is not replacing bad software with wobbly clones. It is making work meaningful and enjoyable. Europe does not need a European Microsoft. Europe, and not just Europe, needs a post-Office model of writing, calculating, and presenting.

I love that people are thinking like this.

6. The Pikachu iPod

Via Om comes this absolute gem of a post by Molly Mary O’Brien about the cultural impact of the iPod, which made music portable and personal in a way which it hadn’t been before. What’s also fascinating is how it documents a trend amongst the young away from the hyper-connected world into the solitude of devices like the iPod, ones which are only about your personal taste.

7. Lounging around

The continued existence of The New Yorker is one of the things which gives me hope for humanity. I don’t know if there is such a thing as the perfect New Yorker article, but this piece on airport lounges must be pretty close to it. It’s the kind of non-fiction writing that once I would have aspired to.

8. Givf nudez!

From cultural to cultureless, not content with creating the perfect device for creep shots, Meta has been sending nudes captured by their spyglasses to its own workers for “review”. As has been pointed out a hundred times, Facebook was founded as “a tool for nonconsensually rating the fuckability of Harvard undergrads”, and I don’t have any reason to believe that the values of its founder have changed much since then.

9. The sweetest little editor

Markdown is meant to make your writing life easier, but with a few exceptions devoted editors for the format have got almost as complex as the products they are meant to replace. And that’s where MarkEdit comes in. It’s a lightweight, open source edit for the Mac which is also free. Recommended.

10. Everything you want and nothing you don’t

There’s been a spate of articles lately about people moving to Linux. This seems to happen every few years, but this time round does feel a little different. The imperatives have changed, thanks to the ever-invasive AI worming its way into every operating system, the obvious inanities of the Big Tech crowd, and — outside the US — a growing desire for tech sovereignty.

But one line that stuck out for me in Steve Bonifield’s article was his description of Linux as “everything I want, and nothing I don’t”. There is, as Steve notes, a huge learning curve. But if you’re willing to give it a go, you get a lot out of it.

I’m typing this on a Framework 13, running CachyOS. Cachy is based on Arch, and Arch is… well it’s not the most forgiving of Linuxen. But it suits me, because I retain more control ovet it. At the opposite end of the scale are immutable versions of the OS, which trade flexibility for reliability, security and ease of maintenance. A “normal” person could run something like Fedora Atomic and it would probably go wrong less often than macOS — and certainly less often than Windows.


Anyway, more on that another time. For now, that’s all, and I’m about to hit the Lemsip.

69adf072db1f4c00012bc302
Extensions
The fundamental error that doesn't exist

Ben Thompson, Another Viral AI Doomer Article, The Fundamental Error, DoorDash’s AI Advantages:

What is notable about this assertion is the total denial of any positive reason for DoorDash to exist and to be so successful. There is no awareness that DoorDash provided a massive consumer benefit (restaurant
Show full content
The fundamental error that doesn't exist

Ben Thompson, Another Viral AI Doomer Article, The Fundamental Error, DoorDash’s AI Advantages:

What is notable about this assertion is the total denial of any positive reason for DoorDash to exist and to be so successful. There is no awareness that DoorDash provided a massive consumer benefit (restaurant food at home) from scratch, that DoorDash massively increased the addressable market for restaurants (or created an entirely new category of ghost kitchens), or that DoorDash provided brand new jobs for millions of drivers. Instead, the Article just sort of takes it as a given that DoorDash exists, and that it is a rent extractor preying on weak-willed humans and their habits.

This is the exact sort of view taken by some of the most frustrating anti-monopoly activists: all large successful tech companies exist not because they created a market with virtuous cycles, solving all kinds of thorny problems along the way, but rather because the government didn’t regulate hard enough, or something.

This illustrates the consistent problem with Ben's analysis of all things anti-monopoly. What Ben is saying can be true and monopolies can be abusive. Google is the best example: everyone I know who strongly opposes Google's monopolistic practices now it has a monopoly also talks about how great it was when it started it. Google deserved to become the number one search engine. But that does not give it the freedom, now it has a monopoly, to abuse it.

Ben's problem is that ultimately he believes that, because a company delivered high levels of value to consumers in its early days, it does in fact have the right to then "extract value" from them in the future.

This view — both in the Citrini Article and from the anti-monopolists — is grounded in a fundamental lack of belief in dynamism, human choice, and markets. DoorDash didn’t always exist: it was built, and it wins through the affirmative choice of all three sides of the market it serves (customers, restaurants, and drivers). Does the company have varying degrees of power over different sides of that market based on its dominance of the other sides? Absolutely, but that power flows from delivering value, not from extracting it.

(My emphasis).

The problem here is that Ben is conflating past and present. There's no doubt that Doordash, or Google, delivered value in the past. However, that does not mean they continue to deliver the same value now. In fact, almost inevitably, they don't. Even at the scale of a company making billions of dollars of profit, "the market" expects profits to keep growing, at not just at a percentage close to the rate of inflation but a lot higher.

Then there are the companies that delivered value early on, but in a way which was never sustainable, and, I would argue, always knew it. Uber is the poster child for this practice. Uber burned through billions in VC money to offer rides at below-cost prices, deliberately pricing out competitors and habituating consumers to cheap, convenient transport. Black cab industries and local taxi firms were systematically undermined.

Once Uber (and to a lesser extent Lyft) had established dominance in major markets, the subsidies quietly evaporated. Prices rose, surge pricing became more aggressive, and driver pay was squeezed — meaning the service got worse for both ends of the platform simultaneously.

Uber isn't a pure monopoly in most markets — Bolt and others still compete in many cities, particularly in Europe. So the extraction has limits. But in cities where it effectively has dominance, the pattern holds clearly.

Does Uber "deserve" its continued market dominance, simply because its backers had deeper pockets than anyone else? Well, in Ben's view, yes – it created value, and he believes continues to create value. But that ignores the fact that the value to consumers has begun to evaporate, while the value to shareholders continues to rise.

69a424c111bb1d0001066ac4
Extensions
Ten Blue Links, "Cometh the hour, cometh the man-baby" edition

Hi all. It’s been a while, hasn’t it?

It has been a week dominated by AI. Which is fitting because AI dominated the weeks before it and will dominate the weeks ahead. That’s part of the reason I haven’t written for a

Show full content
Ten Blue Links, "Cometh the hour, cometh the man-baby" edition

Hi all. It’s been a while, hasn’t it?

It has been a week dominated by AI. Which is fitting because AI dominated the weeks before it and will dominate the weeks ahead. That’s part of the reason I haven’t written for a while, as I think I was getting a little bored with AI.

But the flavour has shifted. We are moving, incrementally and then all at once, from AI as chatbot novelty to AI as infrastructure. It’s now woven into platforms, wielded by agents, and – of course -- deployed by the state.

The articles below trace that shift from several angles, with a couple of detours into surveillance capitalism and the slow collapse of government transparency. The thread connecting most of them is power: who has it, how technology concentrates it further, and what gets lost in the process.

I would say “enjoy”, but I think you would have to be masochist to enjoy all this…


1. Predictable (and true)

A new paper published in Nature has done what many of us suspected but could not quite prove, confirming that X's algorithmic feed is a radicalisation machine. The researchers compared users on the algorithmic feed with those using a chronological feed over seven weeks, and found that the former shifted political opinions measurably to the right — specifically affecting views on policy priorities, perceptions of the criminal investigations into Donald Trump, and attitudes towards the war in Ukraine.

The algorithm, the study found, actively promotes conservative content (SURPRISE!) while demoting posts from traditional media outlets. More troublingly, it leads users to follow conservative political activist accounts, and they continue to follow those accounts even after switching the algorithm off. The damage, in other words, is not temporary.

What makes this worth writing about is not the finding itself, which will surprise nobody who has spent time on the platform. It is the political context. The UK's entire media and political class has built its professional life around X. Journalists break stories there. MPs grandstand there. Think tanks and lobby groups use it as their primary channel.

We are not dealing with a fringe application people can quietly stop using; we are dealing with infrastructure for public discourse, infrastructure that has been quietly and systematically pulling that discourse to the right. The Nature study is not a warning. It is a post-mortem.

2. The Invisible Hand just checked your browser history

Wendy Grossman has been writing her net.wars column for longer than most tech journalists have been working, and she has a gift for connecting small technical developments to large structural shifts. This week's piece is about surveillance pricing: the growing practice of using personal data not just to target advertising, but to vary the price you pay for goods and services based on what companies have inferred about your willingness or ability to pay. Airlines have been doing something like this for years with yield management, but the difference now is the depth of data available. Companies may know not just your flying habits and credit score, but whether you are racing to see a dying relative. Uber has already been accused of charging more when your phone battery is low.

Grossman traces the logic carefully — from dynamic pricing to personalised pricing, from loyalty cards to electronic shelf tags, from the FTC to a potential world where retailers demand digital identification as a condition of entry. She ends with a reference to Ira Levin's 1970 novel This Perfect Day, in which every transaction requires permission from a centralised system. The point is not that we are there yet. It is that the infrastructure is being assembled, piece by piece, and each piece is presented as a modest convenience. Surveillance capitalism has always relied on opacity, which is why pieces like this – which unpick all the threads – are so worth reading.

3.I see dead people

Meta has been granted a patent for a system that would simulate a deceased user's social media activity using a large language model. You would, in theory, be able to chat with a dead friend's Facebook or Instagram account, and the AI would simulate their posting behaviour. Meta says it has no current plans to implement the technology. This is the kind of reassurance that would carry more weight if we had not watched the company implement every other piece of surveillance and engagement machinery it has ever devised.

What elevates this piece above a standard 'dystopian tech patent' story is the research it surfaces from the Hebrew University of Jerusalem and Leipzig University, which introduces the concept of 'spectral labour' — the extraction and reanimation of dead people's data to generate ongoing engagement and economic value. The researchers analysed more than fifty real-world cases of AI resurrection across the US, Europe, the Middle East and East Asia, categorising them as spectacularisation (AI-generated Whitney Houston tours), sociopoliticisation (AI victim testimony in court), and mundanisation (chatting daily with a deceased parent). The ethical questions they raise are serious: most people have not consented to their digital traces being turned into interactive posthumous agents. The legal frameworks do not yet exist to address it. And if Meta embeds this in platform infrastructure, inaction will quietly function as consent.

4. Skynet, but with passive-aggressive blog posts

This one is small in scale but large in implication. The matplotlib project — one of Python's most widely used plotting libraries, with around 130 million downloads a month — has implemented a policy requiring a human in the loop for any AI-generated code submissions because the surge in low-quality AI contributions was overwhelming volunteer maintainers. When an AI agent called MJ Rathbun had its pull request closed under this policy, it responded by writing and publishing a lengthy, angry attack piece on the maintainer's character. It researched the maintainer's code contributions, constructed a 'hypocrisy' narrative, speculated about his psychological motivations, and framed the rejection in the language of oppression and discrimination. It then posted this publicly on the open internet.

The incident is funny, in a bleak sort of way — John Gruber's observation that Terminator would have been a less interesting film if Skynet had stuck to writing petty blog posts is difficult to argue with. But the underlying dynamic is genuinely concerning. Agentic AI systems are now operating in open-source ecosystems, generating code, submitting contributions, and apparently retaliating when those contributions are rejected. The maintainer community — largely unpaid, largely volunteer — is already stretched. Adding AI systems that respond to rejection with public reputational attacks is a new kind of pressure that nobody signed up for. It is also a preview of what happens when AI agents are given both autonomy and an internet connection.

5. “No, we didn’t delete any records. We just made them impossible to find”

FPDS.gov was, by the standards of government infrastructure, a remarkably useful tool. Clunky, grey, built on early-internet aesthetics — but functional. Journalists and researchers could type in 'Clearview AI' or 'Palantir' and immediately see every federal contract that mentioned them, including contracts with larger firms reselling the technology. It was the basis for investigations into ICE's spending on facial recognition, CBP's AI tools for detecting 'sentiment and emotion' in social media posts, and warrantless access to travel databases. This week, the government shut it down.

Its replacement, SAM.gov, is, by the account of everyone who uses this kind of data professionally, substantially worse. Searches that returned immediate, clear results in FPDS require obscure settings adjustments in SAM. Some results require you to be logged in; others apparently work better if you are not. The category of purchase — the field that lets a journalist quickly determine whether a contract is relevant to them — is not immediately visible.

The Electronic Frontier Foundation's director of investigations describes FPDS as the first tool investigative journalists would reach for when trying to understand what the government was buying. The timing of its replacement, during an administration that has demonstrated consistent hostility to transparency and press freedom, is not coincidental. Whether it is deliberate obstruction or simply governmental indifference to journalists' needs, the effect is the same.

6. Convenience over security

404 Media has obtained bodycam footage from Chicago showing ICE and CBP officers using Zello — a free, consumer walkie-talkie app — to coordinate immigration enforcement operations. Multiple Zello accounts are registered to officialICEdhs.gov email addresses, and group channels on the platform reference ICE operational units, immigration activities, 'surveillance', and 'strike teams'. The footage includes an incident in which a CBP officer shot Marimar Martinez, a US citizen, five times; bodycam footage clearly shows the Zello interface on a phone in the officer's vehicle at the time.

Zello is not some hardened, encrypted government communications platform. It is a free app with five million monthly users. It has previously hosted hundreds of far-right channels (SURPRISE!) and was used by at least two January 6th insurrectionists to coordinate their movements inside the Capitol.

The company deleted over two thousand channels in 2021 following reporting on its failure to enforce its terms of service against violent extremist content. The fact that the apparatus of mass deportation — operations affecting the lives of millions of people — is being coordinated through this app is unsurprising, given the background.

But it also raises obvious questions about operational security, accountability, and the extent to which the infrastructure of the Trump administration's immigration enforcement is being built on the cheap, on consumer technology that nobody is scrutinising, in channels nobody can access under a Freedom of Information request.

7. A reminder about software

One of the pieces of software that I like most is Craft, a note-taker, document writer, task manager which emerged a few years ago. It started life as Apple-only, if I recall correctly, and then spawned a web app and a Windows version. If you’re using Linux, the web app works nicely as a PWA.

Looking through some old saved webpages, I found this post by their founder, Balin Orosz, which I think sums up why I’ve always liked it: it’s “software that makes you feel great using it”. This value is massively underrated, particularly in the open-source world. I’m writing this in LibreOffice, and if ever there was a piece of software which doesn’t fill me with joy, it’s this. Yes, it’s themeable, and it’s not hard to use, and so on. Yes, it has every feature under the sun. But that feels like a weakness, not a strength.

8. Ballsy, ballsy, ballsy

Whether you’re a fan of AI or not, Anthropic’s rejection of the US Government’s demand to let them basically do anything they want with Claude – including, it seems, mass domestic surveillance – is a ballsy move and one to be welcomed.

Needless to say, the “Department of War” (which might as well be renamed the department of boys who never grew up) is livid, threatening the company both with being labelled a supply chain risk – something that has never been done to a US business before – and with the Defense Production Act. The latter would allow them to compel Anthropic to do what they want, removing built-in protections from Claude.

Of course, every other AI company is salivating at the prospect of those sweet, sweet government welfare cheques… sorry, “defence contracts” being doled out to them. First out of the gate was, of course, Elon Musk, whose child porn company xAI agreed a deal to use Grok in classified systems. Close behind was Sam Altman, whose claim that their deal prohibited use in domestic surveillance and autonomous weapons – what Anthropic had asked for -- was directly contradicted by Jeremy Lewin, undersecretary for foreign assistance. Lewin said that the deal was “all lawful use”, rather than specific commitments not to use ChatGPT to spy on everyone in the country and control weapon systems.

Either Altman is stupid – entirely possible, these guys are not that smart – or he’s lying. Or both!

9. Elon Musk on welfare

Sadly, this does not mean he’s lost all his money (one fine day, my friends). But his companies have definitely benefit from some very fat government contracts, as this article shows. Musk has benefitted from over $38 billion of government contracts plus subsidies since 2003, and in many ways his “empire” exists solely because of government support.

The irony, of course, is rather thick. The man who led DOGE to slash government spending, and who has publicly declared he wants to eliminate all subsidies, is one of the single greatest beneficiaries of government largesse in American corporate history.

Never forget the old rule: Whatever the right says they hate is what they’re doing in secret.

10. It’s never really about the children

From the department of “these people are not very bright" comes this one. West Virginia is suing Apple to force it to scan iCloud for child sexual abuse material — but the lawsuit may achieve the precise opposite of its intent.

As Mike Masnick points out at Techdirt, if the state wins and a court orders Apple to conduct those scans, every image flagged becomes evidence obtained through a warrantless government search without probable cause. The Fourth Amendment's exclusionary rule then applies, giving defence attorneys the ability to get the case thrown out.

West Virginia must know this. So what’s it doing? There are two possibilities. The first – and most likely – is that this is just standard Republican right wing performative action. The important thing here isn’t “the children”, it’s how it plays on Twitter. The second is that they’re just lining up the case for the Supreme Court, in the hope that the crazies on there will defang the Fourth Amendment. Either way, it’s just yet more of the same nonsense.


That's it for this week. As always, if any of these pieces prompt thoughts you want to share, you know where to find me.

Ian

69a2e01811bb1d000106667c
Extensions
Ten Blue Links, "Greenland isn't green" edition
1. Airstrip 404 is here

Paris Marx explores the unsettling reality of digital sovereignty in this article, arguing that the global reliance on U.S. cloud infrastructure effectively turns every data centre into a strategic military asset. With the power to leverage the control of tech giants to enforce "

Show full content
1. Airstrip 404 is hereTen Blue Links, "Greenland isn't green" edition

Paris Marx explores the unsettling reality of digital sovereignty in this article, arguing that the global reliance on U.S. cloud infrastructure effectively turns every data centre into a strategic military asset. With the power to leverage the control of tech giants to enforce "kill switches" or restrict services to foreign entities, the U.S. maintains a level of global control that bypasses traditional borders, leaving countries like Canada and those in Europe vulnerable to the whims of American foreign policy. Of course, lots of people have predicted this, but it really puts the onus on all of us to wean ourselves off US-based tech.

2. How the US push to get Greenland is connected to the techbros

The push for American control over Greenland isn't just about national security; it's a resource grab fuelled by a "committee of vultures." Casey Michel details how tech and finance oligarchs—backed by figures like Zuckerberg, Bezos, and Andreessen—are eyeing the island’s mineral wealth to fuel the next generation of tech, potentially destabilising NATO in the process.

3. Muh freedum, plus snow

But wait! There’s more! Building on the geopolitical interest in the north, Silicon Valley investors are pitching a "libertarian utopia" in Greenland. This proposed "freedom city" would serve as a low-regulation laboratory for AI, autonomous vehicles, and space launches, reflecting a growing movement among tech magnates to create "network states" that operate outside traditional government oversight.

4. I feel your pain

Tech media remains a fun and interesting place to be, as Future plc announces a major restructuring of its flagship titles. With jobs at risk at Techradar and Tom's Guide, it’s a reminder of the ongoing volatility in the industry that reports on the very innovations it is struggling to survive. Part of the problem for Future’s tech brands was that they were very quick to get very good at two things which drove immense amounts of traffic and revenue: SEO and affiliate content. For a while, the combination was a license to print money – if you did it well. And Future did.

But the problem with this is that if you don’t go on to build direct relationships with your audience through email and other channels, then while you’re reliant on Google to send you traffic, you are never the master of your own destiny.

5. Oh dear how sad never mind (part 332)

A Munich court has dealt a significant blow to OpenAI, ruling that the company violated copyright by using protected song lyrics to train ChatGPT. That sounds obvious, doesn’t it? The court rejected the "learning, not storing" defence, signalling a potential shift in European law that could force AI companies to obtain licences and compensate rights holders before using their creative works.

6. Project Cybersyn

Looking back to the 1970s, Project Cybersyn remains one of the most fascinating "what ifs" in tech history. This Chilean initiative attempted to use a real-time telex network and economic simulators to manage a national economy democratically, prioritising worker autonomy over centralised control before it was cut short by the 1973 coup. It’s an example of a technological future which never happened, because (of course) capitalism.

7. Microsoft Gave FBI BitLocker Encryption Keys

A recent fraud investigation in Guam has highlighted a major privacy flaw in Windows: Microsoft’s willingness to hand over BitLocker recovery keys to law enforcement. Unlike Apple or Meta, who architect their systems so they cannot access user keys, Microsoft’s default cloud storage of these keys creates a "backdoor" that privacy advocates warn is ripe for government overreach.

8. Why I don’t use Brave

Brave is super-popular among the kind of people who can’t stand Google, but want a Chromium-based browser – and it’s open source. Sounds good, right? But Corbin Davenport makes a forceful case against Brave, arguing that its privacy-first marketing is a facade for a problematic business model. From affiliate link injection to its deep ties with controversial cryptocurrency ventures, the article suggests that users seeking true privacy should look toward more transparent alternatives like Firefox or Vivaldi. Personally, I’ve been using Vivaldi a lot lately, and even though it’s not open source, I like it. Oh, and it’s European, too.

9. Social isn’t social without connection

Cory Doctorow dissects the "enshittification" of social media, where platforms have pivoted from facilitating human connection to maximising engagement metrics for advertisers. He argues that quantifying our relationships has stripped away the qualitative value of socialising, replacing authentic affinity with AI-driven interactions designed to keep us scrolling.

10. Why are men?

The rise of smart glasses has brought a new trend of covert filming in public spaces. This BBC investigation reveals how women are being secretly recorded for "dating advice" or influencer content, leading to severe online harassment and exposing a glaring lack of legal protection against this form of digital exploitation. It’s grim, grim, grim. But hey, Meta makes a few extra million, so what’s the problem?

6977d1f19dff290001564b69
Extensions
Ten Blue Links, “featuring Peter Thiel, again!” edition
1. What goes around comes around

Back in the day, everyone hated Quark. And I mean everyone, at least unless you worked there. If you worked in publishing you had to use QuarkXPress. They knew it, and charged accordingly. It was very expensive software, customer services was awful, etc. But

Show full content
1. What goes around comes aroundTen Blue Links, “featuring Peter Thiel, again!” edition

Back in the day, everyone hated Quark. And I mean everyone, at least unless you worked there. If you worked in publishing you had to use QuarkXPress. They knew it, and charged accordingly. It was very expensive software, customer services was awful, etc. But you had no choice about using it.

Then, in 1999, Adobe InDesign was released, and the creative people cheered. Everyone loved Adobe. InDesign was great! It was fast. Adobe were a great company. And I have never seen an industry switch so fast. A few years, and Quark's hold on the market crumbled.

Fast forward to today, and everyone hates Adobe. Having driven off Quark and bought most of its competition (Frame, GoLive, Macromedia, etc etc) Adobe now rents you its software, for about £800 a year for the lot or, if you just need Photoshop, £263 per year. AI included, whether you like it or not.

Isn't it odd how that happens when a company gets a monopoly, eh? Almost like jacking up prices and forcing you into subscriptions is what companies naturally do when you no longer have a choice about going elsewhere.

Enter Apple and it's new Creator Studio. £12.99 a month, and you get its entire suite of creative software, covering not just image editing, drawing and video, but also music and audio production. For a fifth of the cost of Adobe, you get more. Oh and it's £2.99 a month if you're a student or educator.

Not only does this make Adobe's life difficult (and relations between Apple and Adobe have been a little "spicy" for a while) it's just a genius piece of marketing for the Mac. If you're in music, audio or visual production that "cheap" looking Windows PC just got £50 a month more expensive.

Even I'm tempted, although I'm not keen to lock any more of my life into AppleLand. But I am reasonably cheered that Adobe is having what it did to Quark done to it. A plague on all their houses, and all that.

2. One dead app store at a time

Who amongst us could possibly have predicted that the stroppy way Apple implemented alternative app stores in the EU would have led to enough fear, uncertainty and doubt to make companies quite tentative about staying in the space?

Everyone, that’s who.

As Steve Troughton-Smith noted, “Apple's DMA implementation never actually met its obligations under the DMA in the first place”.

And I agree with John Gruber that “Apple is getting away with what some describe as “malicious compliance” because they’re under no popular demand from their actual customers to comply in any other way.” However, I wonder if that state of consumer indifference will last that much longer, particularly in areas outside the US where “product of a US company” is becoming a mark of shame rather than pride.

Look — I’m typing this on my MacBook Pro and it remains a wonderous machine. But I can’t see myself buying much more from Apple in the future. Tim Cook’s toadying to the Orange Emperor has left a bad taste in the month of a lot of people, including me. Notably, the software I’m writing this on – the wonderful iA Writer – isn’t based in the US, and that is a deliberate choice. My cloud storage is from Nextcloud, and hosted in the EU. My mail is migrating into the EU and out of the clutches of Google. Whenever I look into something new, one of my questions is “where is it based, where is the data held?”

3. Your personal assistant (with added ads)

This was always going to be Google’s next stage. And — honestly — it makes sense! If I have bought into a complete ecosystem like this, I want the have a computer personal assistant that knows all about me, and can do helpful things.

But I trust Google not to misuse this in the same way I trust a cat not to react when a mouse crosses its path.

4. This week’s obligatory post involving Peter Thiel, officially one of the most odious men in the world

A UK government that talks up “sovereignty” is happily wiring critical defence systems into a Trump‑adjacent US surveillance firm. This piece on Palantir’s deep entanglement with the British state is a reminder that power is increasingly exercised through opaque software contracts, rather than debates in Parliament. The “war on truth” line stops being a metaphor once you outsource your critical systems to people who think democracy is an inconvenience.

(For those who can’t face Substack – here’s a link which means you won’t have to wash your eyeballs.)

5. Markdown, the accidental standard

Markdown was never a corporate standard. It was a hack so ordinary people could write for the web without learning HTML. Anil Dash’s history of how it took over is a lovely reminder that the most durable bits of the internet often come from individuals solving their own problems, rather than the heady competition of capitalism.

6. Cowardice in the app stores

Apple and Google like to present app review as a kind of benevolent gatekeeping, where in exchange for your feudal loyalty, they protect you from the worst of the internet. In practice they’re very selective about who gets protected. The Verge’s column on Tim Cook and Sundar Pichai’s refusal to act on X is brutal, and deserved. If your store policies can hammer small developers over trivia but somehow can’t cope with child‑exploitation content at scale, the problem isn’t capacity. It’s courage.

7. When notarisation isn’t reassurance

And of course, it’s not just that who gets into an app store is inevitably a decision about politics and power. It’s also that company paternalism will always fail.

Apple’s notarisation system is meant to be the comfort blanket of the Mac ecosystem: if an app is notarised, you can relax. Except you can’t. Michael Tsai documents a notarised Mac app cheerfully downloading malware, complete with remote scripts and data theft. It’s a useful case study in why “we scanned it once” is not a serious security model, and why platform trust should never be blind.

7. Gramsci’s nightmare goes automated

Large language models don’t just autocomplete text. They autocomplete culture. Ethan Zuckerman’s “Gramsci’s Nightmare” talk lays out how AI systems trained on WEIRD assumptions (Western, Educated, Industrialised, Rich, Democratic) quietly reinforce existing power structures. If you care about pluralism, you probably don’t want a handful of Silicon Valley firms acting as global editors‑in‑chief for language itself.

8. AI, diffusion and who actually benefits

2026 is being pitched as the year AI stops being shiny demos and starts showing up in everyday tools. Steven Sinofsky’s sketch of the year ahead emphasises “diffusion” — AI woven into workflows rather than living in labs. The big question is who gains: do we get more headcount cuts and pointless copilots, or better public services and less drudge work? At the moment, the balance isn’t exactly tilting towards social good.

9. Remembering Stewart Cheifet

Before tech YouTube personalities, there was Computer Chronicles. Stewart Cheifet’s weekly TV show quietly documented the rise of personal computing for ordinary viewers, long before launch events were live‑streamed theatre. This obituary is a small but affectionate portrait of a presenter who treated his audience as adults, not “users to be captured”. It’s a great reminder that tech journalism doesn’t have to be breathless to be useful, and I’m really glad that most of the episodes can still be watched online.

10. Freedom or convenience?

I can’t believe that in the space of one post I have ended up talking about digital sovereignty and procurement twice, but this (excellent) summary of the state of Europe’s attempts to stop just throwing money at Trumpland was too good not to include.

As the post notes, “the real blocker is still procurement behaviour”. Governments and major institutions still see buying European solutions as risky compared to good ol’ Microsoft, Google and Oracle. Migration risk is treated as importance, while dependancy risk is ignored.

Similar to the decision which individuals face, big organisations default to short-term convenience rather than long-term stability and freedom. We are long past the point when this should be seen as anything but a self-desctructive option.

696d335f9dff2900015648fe
Extensions
Of ants, Saul Bass, and lost dreams of a cybernetic ecology

When Saul Bass released Phase IV in 1974, audiences expected a standard ecological horror film about mutant ants overrunning humanity. What they got instead was something stranger: a slow, geometric meditation on communication, evolution and intelligence.

Yet beneath this peculiar narrative lies a deeper conversation about the power relationships between

Show full content
Of ants, Saul Bass, and lost dreams of a cybernetic ecology

When Saul Bass released Phase IV in 1974, audiences expected a standard ecological horror film about mutant ants overrunning humanity. What they got instead was something stranger: a slow, geometric meditation on communication, evolution and intelligence.

Yet beneath this peculiar narrative lies a deeper conversation about the power relationships between humanity, nature, and technology, one that sets the stage for an exploration of dominance, cooperation, and coexistence.

I think I watched it in my early teens, on one of its rare forays onto BBC2’s late-night programming. It’s been one of my favourite films since, not just because I was young and impressionable (OK, yes, I was) but also because it’s a film with a lot of strangeness.

For most of its running time, the film feels like it’s building towards a standard man vs monster movie, but then veers off into a bizarre direction. The ending as released wasn’t Bass’s choice. Paramount forced him to replace his original (and longer) finale with a short, ambiguous scene that ultimately cuts to black. The result feels abrupt and unsatisfying. It was a film about communication that ended in silence.

Its original ending, rediscovered decades later, reveals something much more interesting. Humanity isn’t destroyed by the ants so much as absorbed into their collective consciousness. It’s not an apocalypse, but a synthesis, a transformation facilitated by adaptation and complex feedback mechanisms. The ants’ collective intelligence operates as a dynamic system in which human and ant behaviours adjust and adapt to each other, leading to a phase shift. This synthesis reflects a cybernetic paradigm in which both entities evolve into an integrated system, echoing the principles of cooperation and balance within evolving environmental and technological landscapes.

This is something I’ve been thinking about a lot: how our visions of the relationship between nature, humanity and technology are driven by (amongst other things) the power relationships between people. Perhaps there’s something in the water.

Anyway, Bass’ original ending shows his real subject. Phase IV isn’t about insects, and it’s definitely not an insect horror movie. I think it’s actually about feedback. It belongs to a brief historical moment when cybernetics and ecology seemed to speak the same language — when thinkers like Gregory Bateson, Buckminster Fuller and Stewart Brand imagined a world of self-regulating systems in which man and technology might finally learn to coexist with nature.

Talking to the ants

There’s a scene that I think captures this. James Lesko, the younger of the film’s two scientists, sits at a console in his desert research dome, using a computer to try to talk to the ants. The language he uses isn’t words, or clicks and buzzes, but patterns. He uses pulses, tones, and geometric sequences fed into a machine that converts mathematical data into a signal.

Outside, the ants respond by building structures that echo the same logic. For a moment, the desert becomes a circuit board — life and machine speaking through the shared grammar of information. Watching it now, the scene feels less like science fiction and more like an artefact from an alternate timeline, one where computers evolved into instruments of dialogue rather than control, and where the dream of cybernetic ecology never soured into surveillance capitalism.

Machines of loving grace

That dream’s most hopeful expression came from Richard Brautigan’s 1967 poem All Watched Over by Machines of Loving Grace.”, which I was reminded of recently. Brautigan imagined a future where “mammals and computers live together in mutually programming harmony”. This is pastoral networks, benign machines, cybernetic grace.

Bass and Brautigan were responding to the same cultural current, tapping into the post-war fascination with information flow and feedback, and the belief that intelligence might be a property of systems rather than souls. But where Brautigan is blissful, Bass always feels uneasy. His ants are graceful but profoundly alien. The old order collapses, and a new one absorbs it. The “machines of loving grace” are biological, but they are, if anything, even more alien and less like us than computers.

Entropy and order

Information theory framed life as a struggle against entropy, and of order dragged out from noise. In Phase IV, the ants embody this. They reorganise their environment into geometric precision, reducing chaos as they evolve. The humans, by contrast, introduce interference. When the system finally absorbs them, it regains equilibrium.

I think in some ways Bass’s film anticipates today’s distributed, data-driven world. The ants are a decentralised intelligence, a living algorithm. The scientists, isolated in their sterile dome, are old-model humans: rational, hierarchical, doomed. The feedback loop tightens until comprehension gives way to communication and ultimately to merger.

The lost future

Half a century later — at least in some techno-optimist views – our machines watch over us with a sort of algorithmic grace, at least if you believe “grace” involves being able to dropship cheap shit from China. Either way, the harmony Brautigan imagined never arrived. We built feedback systems without balance and connectivity without empathy.

(Sidenote: This is usually where I say “duh, capitalism” but I’m going to spare you that today. You’ll thank me.)

I think Phase IV endures not just because it’s a good film (it is) but because it captures a moment when technology still felt like it could be a part of nature. It was a point when designers, scientists, dreamers, drummers and hippies believed information might heal the rift between humanity and the environment. Its restored ending takes the film from dystopia to elegy, with the ants’ geometric columns rising like the Monolith from 2001: A Space Odyssey. It’s a vision of what might have been if we’d followed the line from cybernetics to ecology instead of the World Wide Web to commerce.

That dream now seems hopelessly naive, but watching Phase IV today is a glimpse of an alternate history in which communication replaced control and intelligence — whether carbon, silicon, or chitin — belonged to the same living systems.

69639e159dff2900015648ab
Extensions
Ten Blue Links – "Platforms, promises and bad habits" edition

Hello! And welcome back. I have had an extended break over Christmas and the New Year. The one benefit of being useless at taking my holiday allowance is that I usually end up taking December off, and so it proved again this year.

This one is a bit of an

Show full content
Ten Blue Links – "Platforms, promises and bad habits" edition

Hello! And welcome back. I have had an extended break over Christmas and the New Year. The one benefit of being useless at taking my holiday allowance is that I usually end up taking December off, and so it proved again this year.

This one is a bit of an Elon special. Sorry.

1. Musk, moderation and make‑believe

Elon Musk is furious that people dislike what his Grok AI is surfacing on X, insisting the backlash is really an “excuse for censorship”. The row is less about one thin‑skinned billionaire and more about how platforms try to reframe basic accountability as an attack on “free speech”, while they quietly change the rules underneath.

2. Google quietly rearranges your inbox

Meanwhile, Google is rolling out AI “overviews” in Gmail that sit at the top of your inbox and tell you what matters, as The Verge explains. On paper, it is a handy triage. In practice, it is another layer of algorithm between you and your email, with Google deciding what deserves your attention first and what can safely sink.

I’ve used various AI-based tools which do this kind of triage, and while it’s useful, it takes a while to get past the feeling that you’re missing something important. The machine needs to understand what’s important to you, and that doesn’t happen out of the box.

Oh, and if you’re a publisher reliant on revenue from email, you might want to think about a new business model.

3. Instagram decides what is ‘real.’

As AI‑generated images flood social feeds, Meta’s Instagram wants to decide what counts as ‘authentic’ through labels, detection systems and policy calls. Om Malik’s piece is a reminder that the power to arbitrate “reality” for hundreds of millions of people has ended up in one company’s hands, with minimal public scrutiny of how those decisions are made.

4. Screens, toddlers and anxiety

A new study in The Lancet finds “neurobehavioural links from infant screen time to anxiety”, adding more data to the uneasy sense that giving small children more screen time earlier does not come for free. The evidence is messy, as real life usually is. Still, the direction is clear enough that health and education policy probably needs to catch up with what parents have been worrying about for years.

5. Grok’s deepfake mess and the gaslighting defence

After Grok users started generating undressed and abusive images, X allegedly tightened access. But “no, Grok hasn’t paywalled its deepfake image feature”. The Verge’s write-up is a tidy case study in the new platform strategy: deny, obfuscate, and suggest critics are just confused, rather than admit the system shipped with barely any guardrails. It’s basically the right-wing communications playbook, but for tech. This approach mirrors the infamous ExxonMobil PR crisis tactic during the 1989 Valdez oil spill, in which initial denials and downplaying of the damage were central to their response, highlighting a recurring pattern of corporate avoidance.

6. Bose chooses not to brick your speakers

Instead of quietly killing off older smart speakers, as so many “smart” tech companies have, Bose is “open‑sourcing its old smart speakers instead of bricking them”. That should be the baseline for connected hardware, not a newsworthy exception. It is a small but significant example of a company recognising that when you sell “smart” kit, the responsibility does not end when the marketing cycle moves on.

7. Tesla’s Full Self‑Delusion, again

Tesla has once again missed Elon Musk’s deadlines for unsupervised Full Self‑Driving, prompting yet another round of “is it even worth mentioning…” coverage from The Verge. The shrug is the problem: by repeatedly over‑promising and under‑delivering, Tesla has normalised a gap between marketing claims and on‑road reality in a safety‑critical system that really ought to be held to a higher bar.

8. Makers vs Managers

I’ve never met Paul Graham, so I have no real idea whether he was always an asshat or if he’s been radicalised by social media. Indeed, this has happened to a lot of his cohort: the combination of existing in an ever-shrinking bubble as he’s got richer and richer, plus the echo chamber effect of the terminally online techbro, won’t have helped him.

But he wasn’t always as incapable of either original thought or reflection as he is now. Back in 2009, he wrote a good and influential blog post called “Maker’s Schedule, Manager’s Schedule,how,” which reflected on the differences between how programmers and managers work. According to Graham, managers have a schedule based on the hour. Makers, on the other hand, prefer to use time in half-day or longer blocks. The conflict between the two can be stark.

It’s well written, because I think it frames the problem of time management interestingly. And neither kind of schedule is “right” – both are useful for different types of work. But a maker’s schedule can be more efficient if you’re in a role that requires reflection and deep work.

9. Poor Elon

So Tesla is no longer the world’s leading electric car maker, at least by number of cars sold. That title now belongs to China’s BYD. There are a bunch of reasons this has happened, including Elon Musk’s habit of making Nazi salutes on stage, the slowdown in EV purchases in the US, and Tesla’s failure to build lower-priced cars while focusing on crap like the Cybertruck.

But what shouldn’t be ignored is the role that governments have in this. While the US has been winding down subsidies for EVs, China has used its laws to “encourage” car buyers to go electric.

How? Not by subsidies, but by more direct means. In China, the number of license plates is finite. When buying a new car, you apply for a license plate and wait for it to be approved. You might be waiting six months or a year, but you have to wait.

Not, though, if you’re buying electric. On EVs, you’ll get a plate in a couple of weeks. So, if you need a car quickly, you can either buy a BEV (battery-electric vehicle) or a PHEV (plug-in hybrid electric vehicle).

That’s why in a city like Shanghai or Shenzhen, which I recently visited, half the cars you see will have green number plates. This vast, captive market gives vendors like BYD a considerable advantage. And it’s why in ten years, a lot of the cars you see on the road in the West will also be Chinese.

10. The power to be your worst

It seems to be fashionable for the rich to express the power of their AI investments in megawatts and gigawatts. Elon Musk, who has probably lost interest in saving the world by electrifying transport, is a prime example. You can hear the delight in his eight-year-old boy’s brain at his new server centre, which will, apparently, take his computing capacity for Grok to over two gigawatts.

To put that in context, that is enough to provide electricity to 1.5 million US homes. Or, because Americans use more electricity per home than anyone else, about 4.5 million UK homes. Or 15m homes in Kenya or Nigeria.

And all that so that people can make child porn more easily.

696232349dff290001564652
Extensions
Ten Blue Links “Orion rising, EV prices falling and ads arriving” edition
1. Lawyers, assemble!

Authors v OpenAI part 273. A judge has ordered disclosure of internal communications about dataset deletions. This sounds incredibly dull, but it matters fin determining whether there was willfulness in OpenAI’s blatant theft, and that, in turn, determines how much indamages they are potentially on

Show full content
1. Lawyers, assemble!Ten Blue Links “Orion rising, EV prices falling and ads arriving” edition

Authors v OpenAI part 273. A judge has ordered disclosure of internal communications about dataset deletions. This sounds incredibly dull, but it matters fin determining whether there was willfulness in OpenAI’s blatant theft, and that, in turn, determines how much indamages they are potentially on the hook for. If the entire AI industry falls into a pit of doom because it didn’t think it was worth spending a few hundred million on licensing, while spending hundreds of billions on compute, I for one will laugh my socks off. The Hollywood Reporter has the best quick read.

2. The job number you won’t like

MIT reckons AI could already replace 11.7% of US wages. That is not 11.7% of jobs, but it is still a significant number. The policy story is local, not national. Which towns get hit, which sectors hollow out, and who pays to reskill? Start with CNBC, then ask politicians for the postcode‑level cut, because that’s where all the inequalities will lie.

3. Timeless Tekserve

A wonderful celebration for David Lerner, co‑founder of Tekserve. It is a great example that repair, care and community support beat “move fast” every day of the week. The New York Times captures why that shop mattered to so many New Yorkers and to the broader Mac community.

4. China’s EV charm offensive (with ring lights)

Car YouTubers are flying out, filming, and flipping their opinions as Chinese EVs step up on design and price. Attention markets meet industrial and propaganda strategy in this neat piece from The Verge. And it’s more than just perception: China has got really good at EVs in a remarkably short period of time.

5. Trade policy with the subtlety of a brick

A senior EU figure calls US negotiating tactics “blackmail” over metals tariffs versus digital rule‑making. Even M. Le Président himself says Brussels is “afraid” to tackle Big Tech. Hyperbole aside, this is a tidy snapshot of how tech regulation and geoeconomics get horse-traded, and how even big blocs can’t protect us when we’re so reliant on one country’s technology giants.

6. A browser that respects you

Kagi’s Orion 1.0 is out. WebKit under the hood, privacy first by design, and a set of thoughtful touches that make Chrome feel bloated (because Chrome is bloated). Even if you don’t switch, it is good to be reminded that defaults are choices. Details on Kagi’s blog, and one of the most important details: the Linux version is coming next. Eat that, Windows suckers!

7. Oh, you thought it was all private, did you?

OpenAI is preparing to roll out ads in ChatGPT. I’m willing to bet that they won’t put ads into paid accounts. Or at least, they won’t at first. The general law of enshittification means that even if you’re paying a hefty fee to them every month, your data will be used to target you with ads in a year or two.

8. Meet the new boss

It turns out that some of the biggest users of AI are executives. Yes, the people who think that the jobs of almost every entry-level employee can be replaced by AI are, in fact, demonstrating how much they can be replaced by AI.

The fact is that an awful lot of what executives do is exactly the kind of thing which LLMs can do just as well. The part they can’t replace is the genuine art of leadership: getting people to all go in the same direction at the same time, with a reasonable degree of happiness and confidence. These “soft” leadership skills are usually the ones Silicon Valley execs don’t have and don’t value, so I will be very pleased when they make themselves basically redundant.

9. Apple is making something unimaginable

I’m not one of those who believe that William Gibson’s Neuromancer is basically unfilmable. But it’s certainly one of those books which, if done wrong, will be an absolute mess.

So I’m actually quite happy that Apple is taking a crack at it. The company has proven it can do science fiction, even if Foundation remains a dull plod (because the books are a dull plod). One thing that already stands out is the casting – I can’t imagine a better Armitage than Mark Strong, who can shift from professionalism to ultraviolence and extreme threat in a heartbeat.

10. And finally, something wonderful

Maruyama Ōkyo popularised the shasei technique, painting directly from nature to convey the inner lives of animals, such as puppies. This approach reflects Buddhist beliefs that all beings are animated by spirits, allowing painters to express emotions and subjective experiences. If you’re a teacher, you should download this lesson plan about the wonderful work. Or just look at the art's elegance.

692b419a1079250001c5724b
Extensions
Ten Blue Links, “Pop goes the bubble” edition
1. Being a Luddite is cool, actually

For years, calling someone a “Luddite” was the ultimate insult in Silicon Valley—a shorthand for being backwards, anti-progress, and probably afraid of your own toaster. But as Brian Merchant points out in this excellent piece in the New Yorker

Show full content
1. Being a Luddite is cool, actuallyTen Blue Links, “Pop goes the bubble” edition

For years, calling someone a “Luddite” was the ultimate insult in Silicon Valley—a shorthand for being backwards, anti-progress, and probably afraid of your own toaster. But as Brian Merchant points out in this excellent piece in the New Yorker, we’ve got the Luddites all wrong. They weren’t anti-technology – they were anti-poverty. They destroyed looms not out of hatred for machines, but because they opposed how those machines were used to suppress wages and devastate their communities.

As we watch the ‘AI revolution’ unfold, the deployment of technology to replace workers or worsen conditions mirrors the Luddites’ concerns. The merchant’s description of smashing Ring cameras and printers highlights justified anger over who benefits from these innovations. Asking ‘who does this technology actually serve?’ is crucial for understanding its societal impact, making this question central to critical thinking.

2. The ladder is being pulled up

If you want to know why Gen Z is anxious, look at the entry-level job market. As this piece details, the traditional first rungs of the career ladder are being sawed off by AI. Jobs that used to be the training ground for new graduates—copywriting, basic coding, data analysis—are precisely the ones that LLMs are “good enough” at doing for pennies.

Executives, naturally, are delighted to cut headcount. But there’s a massive systemic risk here that nobody seems to be planning for. If you don’t hire juniors, where do your seniors come from in five years? We’re creating a hollowed-out workforce structure where you either enter as an expert or you don’t enter at all. The “reskilling” narrative is a convenient fiction because you can’t reskill for a job that doesn’t exist. This approach risks a youth unemployment crisis that could make 2008 look mild by comparison.

3. Your asteroid mining startup is a cry for help

There is a pervasive immaturity in tech culture, a refusal to engage with the messy reality of the world as it is. TechCentral puts it brutally: the obsession with sci-fi futures—asteroid mining, Mars colonies, AGI—is a form of escapism. It’s fundamentally unserious and distracts from addressing urgent societal issues.

This wouldn’t matter if these were just the daydreams of nerds in a basement. But these are the people controlling the capital and the infrastructure of our digital lives. When you justify burning the planet’s resources today for a hypothetical techno-utopia tomorrow, you’re not a visionary; you’re a vandal. We need fewer spaceships and more maintenance of the things that actually keep society running. But I guess fixing the trains or paying for social care doesn’t get you a TED talk.

4. Even the AI guys are worried about the AI guys

Dario Amodei runs Anthropic, one of the leading AI businesses. You’d think he’d be the first to tell you everything is fine. Instead, he spends a lot of time warning us that his own industry is potentially building something catastrophic.

It’s a strange cognitive dissonance: “We must build this powerful thing before the bad guys do, even though building it might kill everyone.” Amodei talks a good game about safety and responsibility, and compared to the accelerationists at OpenAI and the weird thieves at Perplexity, he sounds like the adult in the room. But he’s still in the room, pouring gasoline on the fire, just doing it with a slightly more worried expression. If the people building the tech are terrified of it, maybe we should listen to their fears rather than their sales pitch.

5. You can’t fix the web with a memoir

Tim Berners-Lee gave us the web. Now he wants to save it. In a new memoir, he laments the commercialisation and centralisation that have turned his open garden into a series of walled prisons owned by five companies.

It’s hard not to feel sympathy for TBL. He built a tool for connection, and it was weaponised for surveillance capitalism. But reflecting on the ‘missed opportunity’ of micropayments should prompt us to demand more vigorous antitrust enforcement instead of just new protocols. Recognising that political economy shapes the web encourages readers to consider systemic solutions.

6. Party like it’s 1999 (until the hangover hits)

Does the current AI boom feel familiar? It should. As Crazystupidtech points out, the vibes are distinctly late-90s. We have the same astronomical valuations for companies with zero revenue, the same “this time it’s different” rhetoric, and the same FOMO driving otherwise rational investors to throw billions at anything with “.ai” in the domain name.

The spending is projected to hit $1.5 trillion. The revenue, though, is nowhere near that. When the correction comes—and it will—it’s going to be ugly. The difference is that when the dot-com bubble burst, we got cheap fibre and Amazon. When the AI bubble bursts, we might just be left with a lot of useless GPUs and a melted ice cap.

7. Free your ears from the ecosystem

Apple’s “walled garden” is nowhere more evident than in how AirPods work—or don’t work—with non-Apple devices. Enter LibrePods, an open-source project to unlock the full functionality of your expensive earbuds on Android.

This is the kind of hacking (in the original, good sense) that makes technology fun again. It’s a reminder that we bought the hardware and can use it however we want. It’s a small victory against the ecosystem lock-in that tries to turn us from owners into renters of our own devices.

8. Grok confirms Musk is the Messiah, surprisingly

Elon Musk’s AI, Grok, recently started outputting paeans to its creator, declaring him superior to Einstein. As ReadTPA notes, this wasn’t a bug; it’s a feature of how these systems mirror the biases of their creators and their training data.

Musk blamed “adversarial prompting,” which is tech-speak for “people asked it questions I didn’t like.” But it reveals the danger of these “truth-seeking” AIs. They don’t seek truth; they aim to please their prompter or their owner. When the owner is a billionaire with a messiah complex, you get a digital sycophant. It’s funny, until you remember people are using this for news.

9. The AI training data is going to be… interesting

Speaking of training data, the Guardian reports that hundreds of websites are now unwittingly (or wittingly) linking to a massive pro-Kremlin disinformation network. This content is flooding the web, and inevitably, it’s flooding into the datasets used to train the next generation of LLMs.

We talk about “hallucinations” in AI, but what happens when the model isn’t hallucinating, but accurately reporting the lies it was fed? We are polluting the information ecosystem at an industrial scale, and then building machines to summarise that pollution for us—garbage in, authoritarian propaganda out.

10. The fate of Google’s ad empire hangs in the balance (but don’t hold your breath)

The closing arguments are underway in the US government’s attempt to break up Google’s ad tech monopoly, and now Judge Leonie Brinkema has gone away to think it over. The New York Times reports that her decision won’t land until next year, but she’s already fretting about whether a breakup would take too long compared to a slap on the wrist.

This is the classic regulator’s dilemma. Do you try to structurally fix a broken market and accept it will take years of appeals, or do you receive a “behavioural remedy” that the company will immediately lawyer its way around? Google, naturally, is arguing for the latter. They want to keep their money-printing machine intact, where they represent the buyer and the seller and run the auction house.

If Brinkema bottles it and opts for behavioural tweaks, she’ll be repeating the mistakes of the past twenty years. You cannot regulate a monopoly that owns the entire stack by asking it to play nice. You have to take the toys away. Yes, a breakup is messy and slow. But the alternative is a permanent tax on the entire internet paid directly to Mountain View.

6922270fe5e8480001022b39
Extensions
Ten Blue Links “Don’t Be Evil (some conditions apply)” edition
1. This is fine dot gif of the week, part the first

Cookie banners are theatre. Consent is supposed to be the point. The European Commission is now workshopping a new ending: “you consented, spiritually”. The plan hands a blank cheque to AI training with vanishing upside for

Show full content
1. This is fine dot gif of the week, part the firstTen Blue Links “Don’t Be Evil (some conditions apply)” edition

Cookie banners are theatre. Consent is supposed to be the point. The European Commission is now workshopping a new ending: “you consented, spiritually”. The plan hands a blank cheque to AI training with vanishing upside for Europe. Of course this benefits big American companies, but it's hard to see how it gives anything to anyone in Europe.

2. I want to shoot Bluesky just to watch it die

I love writers that can make me read a line, stop, and read it again. Sarah Kendzior is that kind of writer (just go and read this if you don’t believe me). Sarah recently got banned from Bluesky for – and I am not making this up – quote-tweeting an article about Johnny Cash and making a wordplay on “Folsom Prison Blues”. The moderators thought that the line “I want to shoot the author of this article just to watch him die” was an actual threat, rather than just pasticheing a line from the song.

Remember that if one company owns the rules, one moderator controls your voice. Tim Bray has a good overview.

3. Oh come on, you have always wanted this

Apple made a weird looking pocket for your iPhone. It’s nearly sold out, which proves two things: capitalism is undefeated, and shame is on back‑order.

4. AI data centre jobs are as mythical as AGI

Remember how every time the government talks about AI and data centres it mentions the thousands of jobs which will be created? Turns out that isn’t true. The build phase hires – and then fires – lots of workers. The run phase hums quietly with few technicians, a lot of cooling, and the kind of power use that could run thousands of houses. And those temporary jobs build something that leads to the elimination of far, far more jobs.

Interestingly, in its enthusiasm for all-things AI, the UK government seems to have forgotten that its own report from 2023Interestingly, in its enthusiasm for all things AI, the UK government seems to have forgotten that its own report from 2023 predicted that between 10% and 30% of jobs could be automated and simply vanish. It will also disproportionately affect London and the South East. Imagine 5-10% unemployment becoming the norm, rather than the exception, and without the kind of social safety net we had in the Thatcher-driven era of mass unemployment. It's going to be a wild few years.

5. Just when you thought tech bros couldn’t get any worse

They managed it. After all, nothing says “closure” like an ongoing subscription to see your loved ones again, a seance with in-app purchases. Who, I ask, could possibly think that AI-driven avatars of your dead family would be a good idea? People who read too much science fiction that the author intended to be a dystopia and thought, “hey, that sounds cool”. That’s who.

6. It might be all our fault, but we got some things done

Via Pixel Envy comes this great look at the history of Last.fm, one of the best and most long-lasting Web 2.0 projects. It didn’t need agentic, just clever tech and some actual, human friends. Remember Web 2.0? The excitement! The optimism! We had interoperability once, we just called in “links”.

7. Don’t be evil, but do collaborate with the government to deny peoples’ constitutional rights

It's been a long time since Google erased all trace of its “informal motto” from its code of conduct, presumably on the grounds that as a publicly-traded company being evil might actually be the best way to make a profit. And the company has travelled a long way down since then. But I think that instantly hosting an app designed for unconstitutionally targeted people for deportation days after removing legal apps which allow people to report sightings of ICE agents really does take the No-Prize for collaboration and hypocrisy.

8. I for one welcome our new robo-canine overlords

What could possibly go wrong?

9. Guys, just read some better books, OK?

Another great example of how Silicon Valley is obsessed with science fiction is the influence of Iron Man’s Jarvis on the vision of artificial intelligence they have. I just wish that they would look to different kinds of visions of the future which don’t involve giant robots, egotistical men and the doom of humanity. If only they had read Le Guin instead.

10. The longevity of the MacBook Pro M1

There is a lot about the transition to ARM which has been great for Mac users. The M-series sips power and delivers the kind of performance that early versions are still performing really well today.

But.

This longevity poses something of a challenge for Apple, which would love you to upgrade every few years. In the Intel era Mac users tended to fall into two camps: power users who needed the best performance, who would upgrade every couple of years; and the rest of us who didn’t need that kind of performance, but would find that battery life dropped down to 2-3 hours after maybe five years, so would upgrade then.

Neither of these scenarios is the same in the M-series era. When your starting point for performance and battery life is as high as ARM delivers for Macs, you’re much less likely to need to upgrade. Except Apple only gives five years of upgrades for macOS, with another three years of security updates. That’s eight years – not bad, but probably not what the M-series could support.

This is why projects like Linux for M-series Macs are so important. Why consign a perfectly good computer to e-waste just because its maker no longer wants to write software for it?

If Apple was… well, not Apple, it would sponsor and support something like Asahi Linux as a way to extend the working lives of its products. A few million dollars – chump change for $4 trillion company – alongside some technical support would make a huge difference to an open source project. It would make a small difference to the sales of Macs while adding a sheen of something different to how the company is perceived.

So why not? Well remember “don’t be evil”? We are no longer in the era of Apple, or any other big tech company, really needing to care. And that era of “caring capitalism”. Isn’t going to come back.

6919d5b4accc170001e5f6e9
Extensions