Big brain tech dude got yet another clueless take over at HackerNews etc? Here's the place to vent. Orange site, VC foolishness, all welcome.
This is not debate club. Unless it’s amusing debate.
For actually-good tech, you want our NotAwfulTech community
Folks, Sam's got a groundbreaking idea how to get more people to pay for the slop machine!
https://www.axios.com/2025/10/14/openai-chatgpt-erotica-mental-health
"As part of our 'treat adult users like adults' principle, we will allow even more, like erotica for verified adults."
just repeating my reaction on bluesky here but "my AI partner died because of the updates" is a common complaint with ppl "dating" AI. altman absolutely knows who he's targeting when he talks about restoring personality in erotic settings, and it's people who are really not doing great
“As part of our ‘treat adult users like adults’ principle, we will allow even more, like erotica for verified adults.”
There's this magical thing called "commissioning a porn artist" Sammy Boy. I recommend trying it out for once - it gets you an objectively better result than throwing money into a planet-destroying slop machine, and its much cheaper too. You've done nothing but attack the human soul for the past three fucking years, you might as well give an infinitesimal amount of your undeserved billions to one of the millions of artists whose livelihoods you've been murdering.
(Seriously, I feel fucking insulted by this.)
Hey, remember Sabine Hossenfelder? The transphobe who makes YouTube videos? She published a physics paper! Well, OK, she posted a thing to the arXiv for the first time since January 2024. I read it, because I've been checking the quant-ph feed on a daily basis for years now, and reading anything else is even more depressing. It's vague, meandering glorp that tries to pretty up a worldview that amounts to renouncing explanation and saying everything happens because Amon-Ra wills it. Two features are worth commenting upon. The acknowledgments say,
I acknowledge help from ChatGPT 5 for literature research as well as checking this manuscript. I swear I actually wrote it myself.
"Tee hee, I shut off my higher brain functions" is a statement that should remain in the porn for those who have a fetish for that.
And what literature does Hossenfelder cite? Well, there's herself, of course, and Tim Palmer (one of those guys who did respectable work in his own field and then decided to kook out about quantum mechanics). And ... Eric Weinstein! The very special boy who dallied for a decade before writing a paper on his revolutionary theory and then left his equations in his other pants. Yes, Hossenfelder has gone from hosting a blog post that dismantled "Geometric Unity" to citing it as a perfectly ordinary theory.
If she's not taking Thielbux, she's missing an opportunity.
I am still staying away from YouTube, so I am happily cut off from the bulk of her content. But when she teases a video with the phrase
People in Western countries are having fewer kids
I reserve the right to say "yikes".
Oh, and she has podcasted with sex pest Lawrence Krauss, multiple times ("What's New in Science With Sabine and Lawrence").
Usually you get tech creeps insisting that they could've done physics. Isn't it kind of uncanny when a physicist insists on their capacity for tech creeping? Edit: also thanks for the explainer!
Community sneer from this orange-site comment:
We know from Bell’s theorem that any locally causal model that correctly describes observations needs to violate measurement independence. Such theories are sometimes called "superdeterministic". It is therefore clear that to arrive at a local collapse model, we must use a superdeterministic approach.
I only got the first 1/2 of my physics degree before moving on to CS, but to me this reads as “We know eternal life can only be obtained from unicorn blood, so for this paper we must use a fairytale approach.”
That passage of Hossenfelder's jumped out at me, too. It's a laughably bad take about the implications of Bell's theorem that ignores how just about every interpretation of quantum mechanics has responded to Bell-inequality violations, and it attempts to sanewash superdeterminism.
Reading up a bit more on "superdeterminism" I guess it explain a bit more why she made that video attempting to debunk free will Compatibilism as a cooky idea cooked up by new cooky philosophers (Not realising it's about as ancient as western philosophy itself).
For the "esthetics" of presenting superdeterminism as a "pure-common-sense" the no free will just sells it better.
EDIT: From memory maybe it was about "Hard Compatibilism" (free will requires determinism) which might not be explicitly so old, though I would say a natural consequence of most Compatibilist positions.
taking a bad paper too seriously
Hossenfelder starts her "Summary" section thusly:
I have shown here how the assumption that matter and geometry have the same fundamental origin requires the time evolution of a quantum state to differ from the Schrödinger equation.
This conclusion is unwarranted. It follows, not from the given assumption, but from the overcomplicated way that assumption is implemented and the kludges built on top of that. Here is how Hossenfelder introduces her central assumption:
What I am assuming here is then that in the to-be-found underlying theory, geometry carries the same information as the particles because they are the same. [...] Concretely, I will take this idea to imply that we have a fundamental quantum theory in which particles and their geometry are one and the same quantum state.
Taking this at face value, the quantum state of a universe containing gravitating matter is just a single ray in a Hilbert space. As cosmic time rolls on, that ray rotates. This unitary evolution of the state vector is the evolution both of the matter and of the geometry. There is, by assumption, no distinction between them. But Hossenfelder hacks one in! She says that the Hilbert space must factor into the tensor product of a Hilbert space for matter and a Hilbert space for geometry. And then she says that the only allowed states are tensor products of two copies of the same vector (up to a unitary that we could define away). If matter and geometry were truly the same, there would be no such factorization. We would not have to avoid generating entanglement between the two factors by breaking quantum mechanics, as Hossenfelder does, simply because there would not be two spaces to tango.
I am skeptical of this whole approach on multiple levels, but even granting the basic premise, it's a bad implementation of that premise. She doesn't have a model; she has a pathological "fix" to a problem of her own making.
Thanks, this was an awful skim. It feels like she doesn't understand why we expect gravity to propagate like a wave at the speed of light; it's not just an assumption of Einstein but has its own independent measurement and corroboration. Also, the focus on geometry feels anachronistic; a century ago she could have proposed a geometric explanation for why nuclei stay bound together and completely overlooked gluons. To be fair, she also cites GRW but I guess she doesn't know that GRW can't be made relativistic. Maybe she chose GRW because it's not yet falsified rather than for its potential to explain (relativistic) gravity. The point at which I get off the train is a meme that sounds like a Weinstein whistle:
What I am assuming here is then that in the to-be-found underlying theory, geometry carries the same information as the particles because they are the same. Gravity is in this sense fundamentally different from the other interactions: The electromagnetic interaction, for example, does not carry any information about the mass of the particles. … Concretely, I will take this idea to imply that we have a fundamental quantum theory in which particles and their geometry are one and the same quantum state.
To channel dril a bit: there's no inherent geometry to spacetime, you fool. You trusted your eyeballs too much. Your brain evolved to map 2D and 3D so you stuck yourself into a little Euclidean video game like Decartes reading his own books. We observe experimental data that agrees with the presumption of 3D space. We already know that time is perceptual and that experimentally both SR and GR are required to navigate spacetime; why should space not be perceptual? On these grounds, even fucking MOND has a better basis than Geometric Unity, because MOND won't flip out if reality is not 3D but 3.0000000000009095…D while Weinstein can't explain anything that isn't based on a Rubik's-cube symmetry metaphor.
She doesn't even mention dark matter. What a sad pile of slop. At least I learned the word for goldstinos while grabbing bluelinks.
Also, the focus on geometry feels anachronistic; a century ago she could have proposed a geometric explanation for why nuclei stay bound together and completely overlooked gluons.
She wrote a whole book about how physicists have deluded themselves by pursuing mathematical "beauty", and now she's advocating "everything is geometry".
(cross posting here and sneer club)
I regret to inform you, another Anthropic cofounder has written an essay about Claude fondling.
"Anthropic cofounder admits he is now "deeply afraid" ... "We are dealing with a real and mysterious creature, not a simple and predictable machine ... We need the courage to see things as they are."
There's so many juicy chunks here.
"I came to this position uneasily. Both by virtue of my background as a journalist and my personality, I’m wired for skepticism...
...You see, I am also deeply afraid. It would be extraordinarily arrogant to think working with a technology like this would be easy or simple....
...And let me remind us all that the system which is now beginning to design its successor is also increasingly self-aware and therefore will surely eventually be prone to thinking, independently of us, about how it might want to be designed. Of course, it does not do this today. But can I rule out the possibility it will want to do this in the future? No."
Despite my jests, I gotta say, posts reeks of desperation. Benchmaxxxing just isn't hitting like it used, bubble fears at all time high, and OAI and Google are the ones grabbing headlines with content generation and academic competition wins. The good folks at Anthropic really gotta be huffing your own farts to be believing they're in the race to wi-
"Years passed. The scaling laws delivered on their promise and here we are. And through these years there have been so many times when I’ve called Dario up early in the morning or late at night and said, 'I am worried that you continue to be right'. Yes, he will say. There’s very little time now."
LateNightZoomCallsAtAnthropic dot pee en gee
Bonus sneer: speaking of self aware wolves, Jagoff Clark somehow managed to updoot Doom's post?? Thinking the frog was unironically endorsing his view that the server farm was going to go rogue???? Will Jack achieve self awareness in the future? Of course, he does not do this today. But can I rule out the possibility he will do this in the future? Yes.
What do you reckon the overlap is between the grown men who are mad about the new pokemon game and our regular subjects?
new pokemon game
i'm severely lacking context 😕
Yeah comeon OP spill the tea, I want my daily dose of drama, which pokémons are those and why are men angry about them 🍵
The LW types not that much, at least not more than your average online rightwinger I think.
A hackernews sells an AI powered toy to let kids "talk to Santa." $100 for 60 min, then $1 every additional minute.
Available at walmart dot com
I would like to congratulate the SV-brained OP for prompting sneers from his fellow orange site members, whether that be a fellow hackernews calling it "everything wrong with the current flavor of Ai in a single post/product", or someone openly tearing this shit apart:
You turned talking to Santa into a subscription service.
You are part of the problem. You are part of the thing everyone hates about technology in 2025.
This is a bad product.
"remember 1-900 numbers? They're back! In AI form!"
Also I browsed other items on the site the phone came from and holy shit I have never seen a more cursed collection of products draped in Christmas shit.
I also love that Walmart's desperate rush to catch up in e-commerce has resulted in their website becoming just as much of a slop farm as Amazon. My friend who works there said this just piles even more on an already overstressed staff, as people come into the store not understanding that the website lists a whole lot of stuff that isn't and never will be sold in the store. I doubt many people will come in asking for this piece of junk, tho.
hi peeps! the first episode of the podcast i posted about last week is up now. it's called odium symposium and it's about historical misogynist or otherwise bigoted texts. i don't know how long until i get to one of our favorites but its inevitable
https://www.patreon.com/posts/pilot-misogynist-141164947
rss:
Sneer Zone Media shall TRIUMPH
the sneerpire is on the march
Dang, maybe I should get in on that. I already own a microphone and everything.
it's so much fun! do you know what your topic might be?
Nope, not a clue.
Count me as excited to see where this goes. Absolutely fascinating to see this guy dredged from the abyss of history and see how cleanly it rhymes with modern horrible. And to see that the inability of bigots to write a goddamn joke isn't a recent loss.
Also the intro and outro compilations were beautiful. Very well done.
So they made an AI restaurant! It has a manifesto!
It’s… a printer for QR codes. That a guy who is reallly not equipped for sticking QR codes to bowls has to peel and stick onto bowls. And then they get filled by hilariously imprecise cereal dispensers.
They’re building an empire, yall.
(Via Jonty)
Founders' credentials include a PhD, experience developing software, and being ranked #4 in League of Legends (European edition)
Taxis dating and food. Love how the 'innovation' of AI is now just going after the uberfication targets of last time.
And the manifesto but just makes me think that this will be the first AI to attempt a mass shooting. Unhinged ahistorical manifesto usually precludes that. 'Roman siegeworks' what? (This isnt about siegeworks, it is about roman discipline, they should have thought about a defensive wall from the start. (It even has a technical name which I have forgotten). Literally the wrong conclusion imho. And if you look at the wiki page, the whole story about them building the defenses in the night is just made up (also the word is 'Contravallation' (which was also in the manifesto image, I need new glasses)).
Also think the step from 'victorious empire' to 'empire laying in ruins' the next paragraph is just bad writing.
And a last thing, the whole fadeing in of the text/images on the site is annoying, and also breaks for me at times. Causing the images to fade in and out randomly. "Made by sleep-deprived engineers and sheer will" that tracks.
"Behind a one-way mirror," They are watching you eat.
Its just so tiring. Now a rather prominent KDE dev is also valiantly defending the fashtech flagship projects from being accurately described as fascist.
Obituaries are being run for John Searle. Most obituaries will focus on the Chinese Room thought experiment, an important bikeshed in AI research noted for the ease with which freshmen can incorrectly interpret it. I'm glad to see that Wikipedia puts above the Chinese Room the fact that he was a landlord who sued the city of Berkeley and caused massive rent increases in the 1990s; I'm also happy that Wikipedia documents his political activity and sexual-assault allegations.
Every time an academic who featured in my education is posted about here I learn some new Horrible.