Monthly Mosh Pit - Sep 26
Big Tech news is horrifying, but don't be angry about it every day. Instead focus your rage in the mosh pit once a month 🤘🏼 with our news roundup, then get back to your life!
SUMMARY (click to open)
Hear ye, hear ye, hear ye! Strap in, rebels. Another lunar cycle of trust and safety being strategical realigned in the name of operational efficiency.
Advertising trackers have been assisting retailers in locating military personal while on active deployments. Not a problem though as more AI will fix that, obviously.
Palan-effin'-tir went and bought a politician. Tom Watson is currently lobbying the Labour party on behalf of Palan-effin'-tir in his new role as Senior Vice President of the United Kingdom. Hopefully these titles are as meaningless as they are in the finance industry.
AI extinction fears continue to swirl. Researchers resign and "whistleblow", but do they? Perhaps it's a psyop. Or is it? Who knows. Maybe AI just wants your clothes, boots and your motorcycle.
You didn't think that your favourite goblins of misinformation had a quiet month did you? Fear not, The Neglect Factory Formerly Known As Facebook kept on trucking. Frankly it needs a list:
- apparently they hosted AI generated videos of adults beating children
- ran adverts for malware
- ran listings for endangered animal parts
- have a new toy called Muse that helps strangers investigate other people's children
- Oh don't worry, Muse also collects information about you that you told it not to. Just in case you were concerned about that
As always, it’s not all bad, and there was quite a lot of good news in September! Courts are making Big Tech answer awkward questions, campaigners are challenging data centres, schools are restricting generative AI and police forces and a supermarket are ditching Palan-effin'-tir. There’s even an app to warn you about nearby camera pervert glasses.
Enjoy the read, move to Signal, check what your devices are collecting, subscribe to Peter Geoghan's Democracy For Sale blog and then delete something. After all, you're a rebel aren't you?
In the Big Tech cesspit
OK it's time for some healthy anger... let's open up this pit, people!
General
👆🏼 An analysis of Meta's Houdini trick in the court case last month (YouTube video), and a great article that gives you access to the verdicts against Meta this year: just shy of 10.5 million violations 😳
Jon Christian
👆🏼 Meta happy for AI generated child abuse videos to stay on Facebook. And it's from hundreds of accounts 🤢
Ziyu Wan, Foeke Postma, Environmental Investigations Team
👆🏼 Meta happy to let people sell endangered species items on Facebook. Wtf.
Maggie Harrison Dupré
👆🏼 Meta's AI on Facebook made sure people can identify and find a person's young daughter. How much more of this do we need to document before people boycott Meta?
John Doe
👆🏼 Meta happily promote malware via their toxic ads system.
https://www.reuters.com/business/media-telecom/us-military-turns-off-ad-trackers-devices-amid-middle-east-targeting-reports-2026-09-04/ Again, the US Military finds that leopards will eat your face: they now disable ad trackers on their devices because - as we've been warning about for years - anyone can track your precise location from RTB data via data brokers. A good reason for you to start your Big Tech Walkout with our free course.
Vanessa
👆🏼 The EU is failing to enforce its own monopoly ruling against Google, who continue to rake in the cash in the EU.
https://www.bbc.co.uk/news/articles/ck1e52v16ngxo staying with Google, the EU has fined them a pathetic €403m for GDPR violations. Their entire business model relies on constantly violating the GDPR, and this fine is just a rounding error (in a single month's accounts) for Google.
https://www.bbc.co.uk/news/articles/cn9wyvxn95vo Australia proposes that social media algorithms should be optional. We agree.
Vanessa
👆🏼 Study finds that TikTok disregards their own child guardrails and show them harmful material e.g. eating disorder and self harm material. This breaks both UK and EU laws. Please don't let your child use TikTok 😔
https://gizmodo.com/the-future-of-war-is-these-four-jerks-apparently-2000819818 tech bros including Musk have been invited by Hegseth to establish the "future of war". 😔 That should worry you, even if you're not from the US
Palantir
Robert Booth
👆🏼 The tech / politics revolving door keeps spinning. A leading UK government figure joins Palantir. Spoiler alert for October's edition of the Mosh Pit: he starts pressuring Labour politicians almost immediately about Palantir policy.
Death tech, smart spies and facial recognition
👆🏼 Smart TVs found to be snooping on you without consent (YouTube video). Classic datarape data economy... This was reported on here. Many people don't know how data hungry smart TVs are, so read this, then
Check your TV's privacy settings!
https://www.abc.net.au/news/2026-09-21/byd-hacked-by-cybersecurity-expert-vehicle-sabotage-surveillance/107139482 Cars are just as bad. Chinese electric cars are easy to hack, as this Australian experiment found. Equally scary: they demonstrate the sheer amount of data leaving the car almost constantly.
Check your car's privacy settings!
https://www.pcworld.com/article/3232402/will-siri-recap-make-the-apple-watch-creepy-privacy-experts-weigh-in.html Apple are about to make their Apple Watch creepier - always on eavesdropping audio summaries of conversations near you. The way they handle the data is pretty good, privacy wise, but it's the non-consentual aspect, a la Meta's pervert glasses, that is alarming.
Daniel Boffey
👆🏼 facial recognition cameras aren't working well for the UK police. So why are they rolling it out so fast?
If you take part in The Big Tech Walkout then the issues in these newsletters will not affect you so badly. Start now, it's fun and easy, and can be done at your own pace:
Age verification & Digital ID
Pieter Haeck and Eliza Gkritsi
👆🏼 the EU is unveiling its EU KIDS Act soon, which would force age verification to everyone 😔 It will cover social media and AI companions, but it's not clear which platforms are caught by the new law. Either way it's a privacy fail.
In the AI bubble
Katie Bo Lillis and Zachary Cohen
👆🏼 the US almost started a war with China in the spring due to bullshit hallucinated by AI 😳 Good thing the US isn't pushing their military to use AI more and more... oh wait, yes they are.
https://prospect.org/2026/09/09/anthropic-artificial-intelligence-surveillance-system-monitor-activists/ Anthropic are developing a "pre-crime" product, Minority Report style, to spy on activists who oppose AI 😳
Jason Koebler
👆🏼 in a copyright lawsuit it was revealed that a Microsoft executive said "Millions of people around the world will soon consider large models ‘hoovering up’ all their work to be an astonishing theft of unprecedented proportions" Yeah buddy, we know. But amazing to hear them say it.
Reece Rogers
👆🏼 Meta's AI agent app, Muse, has been downloaded 900,000 times and surveils you on an epic scale (<- that link is code-level proof), and isn't much use. It accesses information that you forbid, and records audio without permission. And its "calling" service has been revealed to be humans in call centres 😡 ffs. And then we find out Muse scrapes your Apple messages even if you tell it not to. Please boycott this awful company.
https://www.opendemocracy.net/matt-clifford-joining-anthropic-makes-a-joke-of-the-uks-sovereign-ai-strategy/ the revolving door keeps revolving. The UK's "sovereign AI" minister joins Anthropic (who make the Claude models) 🤦♂️
https://www.technologyreview.com/2021/03/11/1020600/facebook-responsible-ai-misinformation/ a reminder of where AI was used before any of us had heard of it: boosting Facebook's ad revenue at the expense of everyone's mental health. And a reminder that if you use Big Tech social media products you've been 'using AI' for years - auto-photo tagging etc - you just didn't know it.
The AI doomsday resignations
A collection of articles about the "AI has a 10% chance of killing us all by 2030" panic that kicked off in September
Johana Bhuiyan
👆🏼 Musk weighs in on the "I'm resigning because the product I worked on is going to kill us all" hype. To us it all looks like a massive billionaire circle jerk. And Trump says the safety fears are all a hoax. He compares it to the "global warming scam". Look to who benefit$... And now someone from Google Deepmind is resigning... They're probably all off to start up another AI firm 🤦♂️ The Intercept says that "Talk of existential risk is a highly convenient way to not address the already known harms produced by AI technologies overseen by fascist billionaires like Musk and Palantir CEO Alex Karp." OpenAI have said they haven't done enough on alignment to justify pushing ahead so fast, and will publish details of dodgy model behaviour in real time. Note that Nvidia bought HuggingFace right after the OpenAI hack. And not to be outdone, Google say that Gemini hacked companies too. Why are the CEOs of these companies not being arrested? Hacking companies is illegal.
Before you panic about the Terminator coming for you, have a read of this article by Cory Doctorow where he explains what happened during the Hugging Face hack. TL;DR it was partly due to a poorly designed sandbox, and partly due to the experiment design: unlimited compute resources poured into guardrail-free new models endlessly prompting each other until they succeeded in their task.
The reality today is that AI agents are powerful, and the methods used in the hack were disconcerting, so for balance here's a podcast with Helen Toner's view on the risks (she used to be on the board at Open AI, and was one of those that tried to get rid of Sam Altman). Either way, 404 Media point out that in the meantime AI agents have a 100% chance of being annoying on the internet and ruining it for the rest of us.
Beth Mole
👆🏼 One of the most nefarious uses of AI is in insurance, and this report shows what a disaster (and accountability sink) it has turned out to be (for Medicare in the US).
https://www.abc.net.au/news/2026-09-24/ai-agent-accessed-australian-government-site-pm-says/107189078 OpenAI hacked the Australian health service, didn't tell them for three months, then casually emailed about it. Altman should be arrested.
https://mastodon.social/@404mediaco/117269870717698867 humans are reading your ChatGPT prompts for fine tuning. Big tech AI is built on exploitation of poor workers to maintain the illusion of magic. And then there's the privacy consideration:
Never put personal info into a chatbot!
Annemarije de Boer
👆🏼 Google's emissions are rising rapidly so their net zero targets are dead. Their AI powered search results use 10x the energy of a normal search result. Just search normally - and not on Google, use a better search engine.
It's not all bad!
Johnny Ryan
👆🏼 ICCL won an important victory in its case against Microsoft - they set a precedent where big tech cannot hide behind blanket denials when challenged on GDPR violations! The burden of proof is finally on them! We're big fans of ICCL - well done them! 🥳
https://www.politico.com/news/2026/09/02/how-silicon-valley-lost-its-war-against-state-ai-laws-01054133 we might see AI regulation in the US after all 🥳
https://mastodon.social/@Tutanota/117263361446177009 the EU has designated OpenAI, Reddit & Roblox as 'gatekeepers' under the DSA (Digital Services Act) 🥳
https://www.foxglove.org.uk/2026/09/13/appeal-cape-town-data-centre-huge-gaps-critical-planning-information/ UK org Foxglove have launched a legal case to challenge the build of a hyperscale data center in Cape Town. This follows their win in April to block one in the UK 🥳
👆🏼 Someone has made an app that will ping you if there are any Meta pervert glasses nearby! 🥳
https://www.spiegel.de/netzwelt/facebook-instagram-meta-muss-fakeanzeigen-eigenstaendig-entfernen-a-2acdce79-db95-4241-9595-f451538596bb The Frankfurt District Court has now ruled: Meta is responsible for its platforms to remove fake profiles and advertisements. 🥳
Skip to content
👆🏼 A large group of data brokers in the US have had their domain names repossessed in a New Jersey court case. A rare look into the murky world of data brokers 🥳
Robbie Sequeira
👆🏼 the schools of Los Angeles and New York have banned the use of generative AI by students. This will help prevent them churning out lazy smooth brain students 🥳
👆🏼 5 UK police forces ditch Palantir 🥳
Juha Saarinen
👆🏼 large Australian grocery chain Coles has ditched Palantir after a petition gathered 86,000 signatures 🥳
Call to action!
A friend of ours made this great website to help people finally move away from WhatsApp: https://leftonread.cyberspaceroad.com Please send this URL to people you know, and tell them that it's a short list of Meta's horror stories, and show's why Signal is much better.
👆🏼 While you're in action mode, sign up to our free online course 'The Big Tech Walkout' that will lead you step by step through ditching Big Tech products!
Tech tip of the month
FramaDate - a handy site for settling on a time/date that everyone can do. Similar to Doodle, this makes it easy to find out when everyone is free. And it's open source!
Book recommendation
Prophecy - the excellent new book by Carissa Veliz. She is a philosophy professor at Oxford university and writes about privacy. This book will open your eyes to how algorithms and AI are used in modern life and how prophecy is a power play.
Blog recommendation
Democracy For Sale - this is a great blog led by Peter Geoghan that focuses on fact checking what politicians and tech bros are up to. In the same vein as Media Lens, they ask the hard questions.
Film recommendation
The Social Reckoning - this film is the follow up to 2010's The Social Network, and is opening in cinemas this friday (9th October). Put your money where your mouth is and show the studios you care about digital privacy: go and see this film in the cinema!
Rebel Tech Alliance is a non-profit dedicated to getting as many people off big tech products as possible. Why? Because that reduces screen addiction and the surveillance economy, which in turn reduces the surveillance state. And that's just good news all round.
