Workers at DeepMind unionizing is possibly the most effective way to fight all the bad stuff AI is causing https://www.wired.com/story/google-deepmind-unionization-talks-are-off-to-a-rocky-start/

🆕 On the blog: How I built Calibre’s new dynamic hexagonal world map, in just 10kb.

https://calibreapp.com/blog/building-our-beloved-hex-map

What @pluralistic has been saying for months: Big Tech monopolies bring "earnings so gigantic they’ll be hard to grow big from here".

Now they are betting big on AI to keep the mirage of hyper growth, but "the companies making money from AI are the ones selling the tools … and their customers are having a hard time monetizing that equipment". https://fortune.com/2026/03/19/rob-arnott-stock-market-outlook-magnificant-7/

Disney has lost Roger Rabbit
by Cory Doctorow @pluralistic

«Copyright is a very weak tool for protecting creators’ interests, because copyright only gives us something to bargain *with,* without giving us any bargaining *power,* which means that copyright becomes something we bargain away.

«The middlemen who control access to our audiences have all the power, so when Congress gives creators new copyrights to bargain with, the Big Five (or Four, or Three, or Two, or One) just amend their standard, non-negotiable contract to require creators to sign those new rights over as a condition of doing business.
https://doctorow.medium.com/https-pluralistic-net-2025-11-18-im-not-bad-im-just-drawn-that-way-8cde6fd19a8a

The coffee machine only works with the coffee that comes with the SUBSCRIPTION, and has a big touch screen that displays ADVERTISEMENTS.

Said subscription doesn't include cleaning the machine or any other benefit. It's just a regular coffee machine that has been locked to only work with one brand of coffee.

Heard today in my office:

"Do you remember the password?"

Two colleagues were trying to make the COFFEE MACHINE work.

@pluralistic's enshittification is everywhere

It's not a railroad type bubble, it's a tulip type bubble

It was an interesting article until I read this but:

"The infrastructure investments may have long-term value, but are we already in bubble territory?"

The 400 billion in GPUs that will be obsolete in 2030 is "long-term value", like a railroad. https://www.wreflection.com/p/ai-dial-up-era

OpenAI, Nvidia Fuel $1 Trillion AI Market With Web of Circular Deals

Bloomberg: https://www.bloomberg.com/news/features/2025-10-07/openai-s-nvidia-amd-deals-boost-1-trillion-ai-boom-with-circular-deals

I'm also a big audiobook lover, and also hate that Audible hogs them all, so I backed @pluralistic's #Enshittification audiobook and added a signed copy as a treat 😌 https://mamot.fr/@pluralistic/115090070282146769

The Document Foundation @tdforg warns about Microsoft intentionally making the Open Document format too complex in order to extinguish the competition cc @pluralistic https://blog.documentfoundation.org/blog/2025/07/18/artificially-complex-xml-schema-as-lock-in-tool/

"Remember back in 2023 when hackers exposed (and fixed) malicious anti-repair software in Polish trains? Well, it turns out that the manufacturer, Newag, is at it again" cc @pluralistic https://www.ifixit.com/News/112008/polish-train-maker-is-suing-the-hackers-who-exposed-its-anti-repair-tricks

The problem with LLMs as an alternative to search engines is that, if it's a topic you don't know much about you won't be able to judge if it is lying; and if it's a topic you know much about, then asking the LLM is a loss of time.

Today I asked GPT about some technical stuff. It quoted some official manual, including a link.

I asked it to double-check it, which it did searching the internet, and confirmed the quotation was correct.

I opened the link and of course it wasn't, which I told it. GPT searched the web again, assuring me again that the quotation was verbatim.

I copy-pasted the page of the manual. At that point it surrendered and admitted it wasn't true.

Of course I was sure it was lying, but what if it's a topic I don't know much about? Assuming 10% of people is willing to challenge the answer, only 1% would have triple-checked it.

«As @Zitron points out: this industry [AI] is projecting $327b in spending this year, with $18b in revenue and zero profits. For comparison: smart watches are a $32b/year industry.»
@pluralistic https://pluralistic.net/2025/06/30/accounting-gaffs/#artificial-income

Me during a standup meeting

"How To Think About Scraping" by @pluralistic, every time I hear about me copyright laws to "fight AI" https://pluralistic.net/2023/09/17/how-to-think-about-scraping/

cc @mildsunrise @alvaro

Spain made it into the @protonprivacy Internet Censorship Observatory of February, thanks to LaLiga (professional football league) blocking thousands of sites. Congratulations 🇪🇸!

https://protonvpn.com/internet-censorship-observatory

This is important