Sam Altman wants an actual license to kill you. This bill: the "Artificial Intelligence Safety Act"😒 would give companies like OpenAI immunity for mass murder including nuclear, biological, and radiological attacks assisted by their bots. youtu.be/34bMvXOQfEM

Even chatbots made with regex do this if designed this way. Conversational AI is either about service or it’s about extraction, and you can either invest in the former or the latter.

Doing a shot for shot remake of Bill & Ted’s excellent adventure but it’s about Steely Dan.

I think "shut up you ugly fuck" was the correct response to this man

the world cup protection racket

the only good bubble billionaire is the guy who has a pizza place he runs at a loss that still books ska and goth nights and streams every show

This month marks 20 years since the International Chiptune Resistance World Tour kicked off at a converted underground Victorian public toilet in London, and would subsequently take @bit.shifter.net and I through 20 cities in 9 countries over the next 4 weeks. ⚠️ THREAD INCOMING ⚠️

The want control not efficacy

Writing a book called “All the Presidents’ Memes”.

Happy to criticize AI writing but college admissions essays are the worst possible dataset to investigate this. Schools have ruthlessly imposed homogeneity on those; the impact of AI is to let poor kids write the same bland essays that expensive consultants do. www.nytimes.com/2026/05/27/o...

Anthropic’s response and influence on the encyclical is most evident in the claim that “even its creators don’t understand how it works.” In June, I wrote about what that actually means & why Anthropic says it. (Written while I was in Rome, FWIW). www.techpolicy.press/the-black-bo...

Dow Jones retire bitch.

I cannot emphasise strongly enough that X's "community notes" are part of that website's over-arching fascist project. Yes sometimes it's a funny little top-comment style reply but the entire concept is pure silicon valley disinformation binfire

A very pleasant good morning to all my knickerbockers.

maybe you'd like to read an actual article about myra cheng and her research into AI and stanford break-up texts instead of some facebook screencaps?

“There are unsettling structures we’re finding” — but are they releasing the measurements? The actual evidence? Or just gesturing toward concern while continuing deployment?

Grimly amusing how Microsoft researchers showed, with substantial data, that frontier LLMs corrupt the hell out of documents in long delegated workflows, just so Microsoft could make them post, without any data in support, that the problem can be totally mitigated. www.microsoft.com/en-us/resear...

If I’m not evolved enough to understand it, then why did they have to steal my writing to train it?

The Running Man movie reboot is a valiant attempt to suppose what would happen if instead of the entire cast being on cocaine, what if the entire cast was on SSRIs.

Pope Leo quoting Tolkien in his encyclical on AI seems like a pointed reference for the tech oligarchs who have appropriated LOTR for their own ends.

people don’t know everyone can see them buying mothers days flowers from the plug

Went to an anti AI datacentre protest march today and a friend of mine got a picture that summed it all up

TFW you finally start coming for white collar jobs.

AI is more unpopular than ICE

The guy criticized working with a beloved children’s franchise helmed by a hateful bigot is branching out into other mediums of problematic authorship.

People forget that the bubble exists so the boy within is not killed by germs. Bubbles are good as long as you can see outside one.

“You know i’m something of a burlesque performer myself” - Me talking to an attractive person while under the mistaken impression this means “sings like Burl Ives”.

(not endorsing this beyond making a technical observation but) Local LLMs running on the laptop you're developing with now already run well enough, fast enough, accurately enough, that coding new features faces bigger blockers getting through the process of agreeing on merging if you have one.

i think some pro-AI people are also taken aback by what feels like the sudden implication that they’re not cool when they’ve been told the last couple of years how they’re on the cutting edge and anyone who didn’t get on board would end up looking like a loser when maybe it’s the other way around

AI boosters claim that LLMs are the future of software development. But that could not be further from the truth. LLMs are the apotheosis of a paradigm 40 years in the making: the tech industry assigns you problems and then sells you the solutions. But that world is now coming apart at the seams.

“You know i’m something of a burlesque performer myself” - Me talking to an attractive person while under the mistaken impression this means “sings like Burl Ives”.

The google map icons are out of pocket

The google map icons are out of pocket

are they hitting on me

“People using AI chatbots to generate their ideal faces are increasingly arriving at surgeons’ offices with briefs demanding flawless skin, sharply sculpted cheekbones, refined noses and near-perfect symmetry…”

sometimes I wonder if people in other fields have any clue what we go through as women in computational fields... (graph from: www.npr.org/sections/mon...)

Stop asking AI to do democracy for you.

Papa skimmin’

Now that ai winter is upon us can we discuss what has happened to the Entra logo

The first 20 seconds of AOC holding up a mason jar of filthy drinking water will become the image most Americans conjure up when they hear "data center." Tech is rekt!

Today is Ona Judge Day. While the United States gears up for its 250th celebration, which inevitably includes the likes of George Washington, Ona Judge made her escape on May 21, 1796. The Washington household attempted her capture several times, but she was never caught. Forever free

Tonight we congratulate our friend Stephen Colbert and his writers, staff and crew for eleven years of excellence. Watch Colbert’s finale tonight, and then never watch CBS again.

Wow it's a bloodbath for AI enthusiasts. These numbers are truly remarkable given the enormous amount of money poured by the most powerful individuals on Earth to endlessly hype up AI.

Anyway wasnt the entire business case for AI that training LLMs is super expensive but once you do it inference is dirt cheap so there are low marginal costs? But now every big AI provider has to do marginal charges for compute or they lose money? What am I missing?

new magfest class action suit coming

Imagine this is annoying for judges.

When is the government going to pay out reparations to people (me) whose birthday (jan 6th) has been permanently ruined by assholes (chuds).

As your friendly neighborhood AI consultant, I don’t usually get this much space to hold court on why AI projects fail: techstartupnetwork.co/meet-the-fou...