A slop track was fraudulently added to the Volcano Choir list. ‘Ironically, the news comes just days after Spotify officially admitted it was being overrun by AI “slop.” The platform announced new policies to protect artists against “spam, impersonation, and deception.”’

“In one case, a woman who had been treating her schizophrenia with medications for years became convinced by ChatGPT that the diagnosis was a lie. She soon went off her prescription and spiraled into a delusional episode, which arguably wouldn’t have happened without the chatbot.”

“Surprisingly, we find that when developers use AI tools, they take 19 percent longer than without--AI makes them slower...hallucinations forced the developers to spend extra time cleaning up and reviewing code. The tools also performed worse when dealing with large and complex code repositories..."

Facebook will snitch on your friends.

45 billion in revenue, 2 trillion required seems like it's actually the 1.955 trillion dollar question

"By 2030, anticipated demand for AI services would require $2 trillion in annual revenues, leaving a shortfall of $800 billion globally to meet that demand." C/o @physticuffs.bsky.social

AI slop strikes again. www.wsj.com/politics/nat...

Mark Zuckerberg is introducing a feed that's just 100% AI slop. No pesky creativity from real people to get in the way. Eat up little piggies

Premium newsletter: Based on my estimates and analysis, OpenAI needs one trillion dollars in the next four years to build 17GW of data centers and other commitments, with at least $500 billion needed for company operations. There is not enough capital to do this. www.wheresyoured.at/openai-onetr...

There is not enough taxpayer money to actually fund this unless we spend the equivalent of the US military budget on OpenAI, and only OpenAI. By all means get the worry but this is entirely different to anything you have ever seen

I cannot express how wild this quote is. This is a top tech analyst at a well-regarded firm saying that there is not enough capital to actually pay for all the stuff that NVIDIA and OpenAI has promised. It is so important that everybody realizes that *there is not enough money to do this*

"AI machines — in quite a literal sense — appear to be saving the US economy right now."

“AI safety requires more than just preventing obviously harmful outputs. Systems must be capable of understanding context, maintaining consistency across languages… and navigating the subtle boundary between helpful and harmful assistance,” write Jen Weedon, Theodora Skeadas, and Sarah Amos.

No, you do not need to add these technologies to your repertoire - especially to write IEPs - and rather than invest in and solidify Big Tech's capture of education, may I suggest using that money to support relationships between teachers and students unmediated by OpenAI, Microsoft, and Anthropic?

Is this a record.

It’s official - nothing is sacred. Herzog should sue

Tesla influencers testing full self-driving made it 60 miles on a cross USA trip before the Model Y hit an object that broke the sway bar bracket and damaged the suspension.

In 3.7: "LLMs do not improve one's writing ability much like taking a taxi does not improve one's driving ability." Love this! I tell my physics students, watching a football game doesn't make them better at playing, and watching me do examples isn't enough to build their problem solving abilities.

sorry pedo guy, you really did ask for it

Cool new way for cars to kill people

A parent with a child at the hacked nursery chain "viewed the cyber-attack as an inherent risk of using any app, and considered the opportunity to gain real-time information on his child, such as what they had eaten, worth it" 🤔

groimes, oim een theh eepsteen foiles

This has been pushed to the background, but pay attention: Right now, a stack of bills aimed at reining in AI companies and big tech sits on Gavin Newsom's desk in California. These are the bills Silicon Valley and the GOP tried to ban outright. Now they await Newsom's signature—or veto.

3/ Public policy requires different standards than consumer technology. Pedro Tavares explores how to distinguish between AI that serves the public interest and AI that silently undermines it.

Oh boy! AI can hallucinate bullshit at me when I wake up based on what I was talking about yesterday! And all I have to do is let it listen in on all my conversations! It's really too bad a) I never asked for that feature, and b) I'd never, ever, let OpenAI anywhere near anything private

i mean, AI is arguably racking up more student debt than anyone ever thought possible, and it's very disappointing

The company acknowledges that Pulse "won't always get things right" and may suggest tips for completed projects or miss relevant topics entirely. Translation: OpenAI expands its range of spyware

Saw someone despairing over the state of gaming due to the behavior of the "average gamer" as indicated by grotesque posts on twitter, and it's like, you know the site is designed to boost people and bots like that in order produce exactly this reaction in people like you, right?

“I want AI to do my laundry and dishes so that I can do art and writing,” as the from SF author Joanna Maciejewska memorably put it, “not for AI to do my art and writing so that I can do my laundry and dishes.” These stories show what happens when it’s the other way around."

It's highly effective.

How many times would you like to be punched in the head? v How many times would you like to be punched in the head?

“Hundreds of billions are being invested in a technology with an uncertain use case. Enormous data centres are sucking up power and potentially derailing progress in emissions reductions. Plus concerns the entire US economy is propped up by the AI investment bubble.” www.abc.net.au/news/science...

🤷🏿‍♂️

Got a new piece in @goodtape.com on some of the best journos & podcasters covering AI right now: like @bcmerchant.bsky.social @parismarx.com @404media.co @edzitron.com & more. Check it ouuuut....

lol someone's set www.howtoprogramwithai.com to 302 redirect to pivot-to-ai.com

there is going to be an unstoppable, ultra-massive tsunami of AI-for-good type stuff, and particularly 'AI for climate' at new york climate week this one has will i am and google aiforgood.itu.int/event/ai-for...

Important point in Section 3.4 on anthropomorphizing terms used as metaphor or jargon. Phrases "like train, learn, hallucinate, reason [...] result in distorting how we perceive these machines: humanizing them while dehumanizing us" There's a lack of scientific rigor when we use words carelessly.

See this quoted thread for highlights from the lead author of this important paper! My own key thoughts and connections will be in a thread right here, too. 🧵

I have *never* used ChatGPT. Not once. I’m not saying I’m better than you… oh fuck it, I’m better than you.

AI companies will make, in the most optimistic estimate, $60 billion in revenue in total, and that includes Microsoft, OpenAI, CoreWeave, Anthropic, Amazon, Meta and Google. I don’t know what universe they’re gonna make $1.2 trillion in revenue in

Paulina sends word: She is very touched by the burst of financial support this thread is generating. "Just wow," she says. "Much gratitude." We're going to work to raise funds via a nonprofit to issue "Cyberselfish" as a proper e-book. It's not very expensive. Details to come.

"Consider the implications if ChatGPT started saying “I don’t know” to even 30% of queries ... Users accustomed to receiving confident answers to virtually any question would likely abandon such systems rapidly."

A month after my last skeet, the subreddit "My Boyfriend is AI" now has 88,000 members and is the subject of an MIT study that found that "AI companionship emerges unintentionally through functional use rather than deliberate seeking." arxiv.org/html/2509.11...

2/ In 2001, Borsook said tech "libertarianism" reflected an adolescent mindset, with a craving for unchecked independence & resistance to constraint. She warned that tech libertarians wanted an anti-human world that worked more like a computer. From "Cyberselfish," a book based on her 90s writing:

Half of Australians have no idea that Google and other search engines will be forced to check the age of their users & change their service based on that info. This polling comes as an inquiry into Australia's online safety codes & the teen social media ban kicks off today www.crikey.com.au/20...

Another case of “techno-optimist offsets”: i.e. it’s fine for AI to drive up emissions today cause it’ll fix everything in the future.

"The FT sought to categorise the expected positive benefits of the technology. Most of the anticipated benefits, such as increased productivity, were vaguely stated and harder to categorise than the risks" www.ft.com/content/e93e... (note: Entergy is the fossil fuel company supplying Meta)

Large language models have a pretty fundamental flaw: they don't distinguish between data and code. On its own, that might be embarrassing (think helpbots writing dirty limericks) but get the wrong set of capabilities in the mix and a fancy new AI tool can become a dangerous security vulnerability.

The bots are getting worse.