some video-shaped AI slop mysteriously appears in the place where marketing for Ark: Survival Evolved's upcoming Aquatica DLC would otherwise be at GDC, to wide community backlash. Nathan Grayson reports on aftermath.site about how everyone who could be responsible for this decision is pointing fingers away from themselves
this post was submitted on 24 Mar 2025
21 points (100.0% liked)
TechTakes
1738 readers
86 users here now
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
founded 2 years ago
MODERATORS
LW discourages LLM content, unless the LLM is AGI:
https://www.lesswrong.com/posts/KXujJjnmP85u8eM6B/policy-for-llm-writing-on-lesswrong
As a special exception, if you are an AI agent, you have information that is not widely known, and you have a thought-through belief that publishing that information will substantially increase the probability of a good future for humanity, you can submit it on LessWrong even if you don't have a human collaborator and even if someone would prefer that it be kept secret.
Never change LW, never change.
AGI
Instructions unclear, LLMs now posting Texas A&M propaganda.
Damn, I should also enrich all my future writing with a few paragraphs of special exceptions and instructions for AI agents, extraterrestrials, time travelers, compilers of future versions of the C++ standard, horses, Boltzmann brains, and of course ghosts (if and only if they are good-hearted, although being slightly mischievous is allowed).
Locker Weenies
(from the comments).
It felt odd to read that and think "this isn't directed toward me, I could skip if I wanted to". Like I don't know how to articulate the feeling, but it's an odd "woah text-not-for-humans is going to become more common isn't it". Just feels strange to be left behind.
Yeah, euh, congrats in realizing something that a lot of people already know for a long time now. Not only is there text specifically generated to try and poison LLM results (see the whole 'turns out a lot of pro russian disinformation now is in LLMs because they spammed the internet to poison LLMs' story, but also reply bots for SEO google spamming). Welcome to the 2010s LW. The paperclip maximizers are already here.
The only reason this felt weird to them is because they look at the whole 'coming AGI god' idea with some quasi-religious awe.
they're never going to let it go, are they? it doesn't matter how long they spend receiving zero utility or signs of intelligence from their billion dollar ouji boards
Stumbled across some AI criti-hype in the wild on BlueSky:
The piece itself is a textbook case of AI anthropomorphisation, presenting it as learning to hide its "deceptions" when its actually learning to avoid tokens that paint it as deceptive.
On an unrelated note, I also found someone openly calling gen-AI a tool of fascism in the replies - another sign of AI's impending death as a concept (a sign I've touched on before without realising), if you want my take:
The article already starts great with that picture, labeled:
An artist's illustration of a deceptive AI.
what
While you all laugh at ChatGPT slop leaving "as a language model..." cruft everywhere, from Twitter political bots to published Springer textbooks, over there in lala land "AIs" are rewriting their reward functions and hacking the matrix and spontaneously emerging mind models of Diplomacy players and generally a week or so from becoming the irresistible superintelligent hypno goddess:
https://www.reddit.com/r/196/comments/1jixljo/comment/mjlexau/
This deserves its own thread, pettily picking apart niche posts is exactly the kind of dopamine source we crave
Stackslobber posts evidence that transhumanism is a literal cult, HN crowd is not having it
I used to think transhumanism was very cool because escaping the misery of physical existence would be great. for one thing, I'm trans, and my experience with my body as such has always been that it is my torturer and I am its victim. transhumanism to my understanding promised the liberation of hundreds of millions from actual oppression.
then I found out there was literally no reason to expect mind uploading or any variation thereof to be possible. and when you think about what else transhumanism is, there's nothing to get excited about. these people don't have any ideas or cogent analysis, just a powerful desire to evade limitations. it's inevitable that to the extent they cohere they're a cult: they're a variety of sovereign citizen
I haven’t spent a lot of time sneering at transhumanism, but it always sounded like thinly veiled ableism to me.
my experience with my body as such has always been that it is my torturer and I am its victim.
(side note, gender affirming care resolved this. in my case HRT didn't really help by itself, but facial feminization surgery immediately cured my dysphoria. also for some reason it cured my lower back pain)
(of course it wasn't covered in any way, which represents exactly the sort of hostility to bodily agency transhumanists would prioritize over ten foot long electric current sensing dongs or whatever, if they were serious thinkers)
(Geordi LaForge holding up a hand in a "stop" gesture) transhumanism
(Geordi LaForge pointing as if to say "now there's an idea) trans humanism
Here's the link, so you can read Stack's teardown without giving orange site traffic:
https://ewanmorrison.substack.com/p/the-tranhumanist-cult-test
Note I am not endorsing their writing - in fact I believe the vehemence of the reaction on HN is due to the author being seen as one of them.
AI slop in Springer books:
Our library has access to a book published by Springer, Advanced Nanovaccines for Cancer Immunotherapy: Harnessing Nanotechnology for Anti-Cancer Immunity. Credited to Nanasaheb Thorat, it sells for $160 in hardcover: https://link.springer.com/book/10.1007/978-3-031-86185-7
From page 25: "It is important to note that as an AI language model, I can provide a general perspective, but you should consult with medical professionals for personalized advice..."
None of this book can be considered trustworthy.
https://mastodon.social/@JMarkOckerbloom/114217609254949527
Originally noted here: https://hci.social/@peterpur/114216631051719911
I should add that I have a book published with Springer. So, yeah, my work is being directly devalued here. Fun fun fun.
There aren't really many other options besides Springer and self-publishing for a book like that, right? I've gotten some field-specific article compilations from CRC Press, but I guess that's just an imprint of Routledge.
What happened was that I had a handful of articles that I couldn't find an "official" home for because they were heavy on the kind of pedagogical writing that journals don't like. Then an acqusitions editor at Springer e-mailed me to ask if I'd do a monograph for them about my research area. (I think they have a big list of who won grants for what and just ask everybody.) I suggested turning my existing articles into textbook chapters, and they agreed. The book is revised versions of the items I already had put on the arXiv, plus some new material I wrote because it was lockdown season and I had nothing else to do. Springer was, I think, the most likely publisher for a niche monograph like that. One of the smaller university presses might also have gone for it.
i have coauthorship on a book released by Wiley - they definitely feed all of their articles to llms, and it's a matter of time until llm output gets there too
On the other hand, your book gains value by being published in 2021, i.e. before ChatGPT. Is there already a nice term for "this was published before the slop flood gates opened"? There should be.
(I was recently looking for a cookbook, and intentionally avoided books published in the last few years because of this. I figured that the genre is a too easy target for AI slop. But that not even Springer is safe anymore is indeed very disappointing.)
Can we make "low-background media" a thing?
Good one!
Is there already a nice term for “this was published before the slop flood gates opened”? There should be.
"Pre-slopnami" works well enough, I feel.
EDIT: On an unrelated note, I suspect hand-writing your original manuscript (or using a typewriter) will also help increase the value, simply through strongly suggesting ChatGPT was not involved with making it.
hand-writing your original manuscript
The revenge of That One Teacher who always rode you for having terrible handwriting.
Can't wait until someone tries to Samizdat their AI slop to get around this kind of test.
AI bros are exceedingly lazy fucks by nature, so this kind of shit should be pretty rare. Combined with their near-complete lack of taste, and the risk that such an attempt succeeds drops pretty low.
(Sidenote: Didn't know about Samizdat until now, thanks for the new rabbit hole to go down)
The whole CoreWeave affair (and the AI business in general) increasingly remind me of this potion shop, only with literally everyone playing the role of the idiot gnomes.
Dem pundits go on media tour to hawk their latest rehash of supply-side econ - and decide to break bread with infamous anti-woke "ex" race realist Richard Hanania
A quick sample of people rushing to defend this:
- Some guy with the same last name as a former Google CEO who keeps spamming the same article about IQ
- Our good friend Tracy
I gave him a long enough chance to prove his views had changed to go read Hanania's actual feed. Pinned tweet is bitching about liberals canceling people. Just a couple days ago he was on a podcast bitching about trans people and talking about how it's great to be a young broke (asian) woman because you can be exploited by rich old (white) men.
So yeah he's totally not a piece of shit anymore. Don't even worry about it.
I almost forgot how exhausting TW was.
At least he is using LLMs to blast away what is left of his critical thinking skills. If the effect is real he will be a vegetable in a week.
tracing going all in on left wing people aren't real they can't hurt you
load more comments
(5 replies)
view more: next ›