Want to wade into the snowy surf of the abyss? Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid: Welcome to the Stubsack, your first port of call for learning fresh Awful youāll near-instantly regret.
Any awful.systems sub may be subsneered in this subthread, techtakes or no.
If your sneer seems higher quality than you thought, feel free to cutānāpaste it into its own post ā thereās no quota for posting and the bar really isnāt that high.
The post Xitter web has spawned soo many āesotericā right wing freaks, but thereās no appropriate sneer-space for them. Iām talking redscare-ish, reality challenged āculture criticsā who write about everything but understand nothing. Iām talking about reply-guys who make the same 6 tweets about the same 3 subjects. Theyāre inescapable at this point, yet I donāt see them mocked (as much as they should be)
Like, there was one dude a while back who insisted that women couldnāt be surgeons because they didnāt believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I canāt escape them, I would love to sneer at them.
(Credit and/or blame to David Gerard for starting this. This was a bit late - I was too busy goofing around on Discord)
https://www.theinformation.com/articles/can-ucla-replace-teaching-assistants-ai
Millerās team also recently used software from startup StackAI to develop an AI-powered app that writes letters of recommendation, saving faculty members time. Faculty type basic details about a student who has requested a letter, such as their grades and accomplishments, and the app writes a draft of the full letter.
AI is āone of those things that you might worry could dehumanize the process of writing recommendation letters, but faculty also say that process [of manually writing the letters] is very labor intensive,ā Miller said. āSo far theyāve gotten a lot out ofā the new app.
Anyone using this thing should be required to serve on the admissions committee. LoRs arenāt for generic B+ students that you donāt even remember, just say no.
I googled stackai, saw their screenshots and had ptsd flashbacks of mid 2000s alteryx. why do we keep reinventing no-code drag-and-drop box-and-arrow crap.
A rare W for the Tennessee General Assembly:
https://www.wkrn.com/news/tennessee-politics/bill-targets-ai-hurt-manipulate/
Orange site mods retitled a post about a16z funding AI slop farms to remove the a16z part.
The mod tried to pretend the reason was that the title was just too damn long and clickbaity. His new title was 1 character shorter than the original.
I donāt love the title but itās the best I could come up with to fit within the 80 character limit.
A half dozen people might still be reading hackernews on punchcards so they ha-
ve no choice but to argue about how to shorten ālongā titles every day.
I went down a punch-card history rabbit hole today on the empirical software engineering discord. TIL:
- IBM leased machines, didnāt sell them
- Lease required using IBM branded punchcard (Razor and Blade Model)
- Until āThe Consent Decree of 1956ā forced them to unbundle https://www.nytimes.com/1994/06/09/business/ibm-and-the-limits-of-a-consent-decree.html
- The competing punch card standard of the time had 90 columns
We have been living in a world 10 columns too short. Think of all the HN headlines we could have had insteadā¦
I imagine it was like VHS vs Beta only with pocket protectors.
- IBM leased machines, didnāt sell them
Good to know that Orange Website is being considerate of us VT220 users. I knew there was a reason why mine has the amber phosphorus.
Lol talk about mixed messages.
[Firefox] will evolve into a modern AI browser
Firefoxās social media account today:
Firefox is not becoming an AI browser.
What the fuck would an āAI browserā even be, let alone a modern one. I know what a web browser is, basically a combined HTTP client and HTML renderer. An AI browser is not something that has a commonly understood meaning, so to claim Firefox or anything else will be one without elaboration is just wankery.
I canāt help but do their dirty work for them and try to imagine what the hell an AI browser would be. Maybe you develop a standard protocol for prompting chatbots and a markup format for displaying responses and an AI browser is a client for that? Or maybe you just put an LLM in the search bar so Mozillaās bullshit machine can give you wrong answers before pressing the return key and having Googleās bullshit machine give you wrong answers. Maybe thereās an about:chatbot page. I think all of these are bad bullshit ideas, but at least theyāre ideas and not just āwhat if we added <latest fad> into <product>ā.
AI Browsers. Metaverse fast food. Blockchain sneakers. Gigwork apartments. Cloud toilets. Big Data headphones. AR chairs. Military grade pianos. 3D books. App drugs. Dotcom condoms. Cyberspace bicycles. Wireless jump ropes. Video silverware. WYSIWYG carpets. Transistor fanny packs. Electromechanical ladders. Atomic flooring. Radio saunas. Horseless glue. Steam pens. Water powered masturbation.
I assume some mesolithic asshole said shit like āwe are transforming our hunter-gatherer settlement to a ācave painting firstā societyā and neighboring community leaders gave that guy like a hundred animal skins each for his insight.
Ryanair now makes you install their app instead of allowing you to just print and scan your ticket at the airport, claiming itās ābetter for our environment (gets rid of 300 tonnes of paper annually).ā Then you log in into the app and you see thereās an update about your flight, but you donāt see what itās about. You need to open an update video, which, of course, is a generated video of an avatar reading it out for you. I bet thatās better for the environment than using some of these weird symbols that I was putting into a box and that have now magically appeared on your screen and are making you feel annoyed (in the future for me, but present for you).
chat is this kafkaesque
New conspiracy theory: Posadist aliens have developed a virus that targets CEOs and makes them hate money.
ā¦.this made me twitch
The ACM has fallen
https://types.pl/@wilbowma/115733550130706711
https://types.pl/@arjen@idf.social/115736432049018026
I became a member this year! and then immediately got to notice how itās oozing out every pore
whatās kinda wild for me is that thereās also an ethics pledge involved, and I do not understand how they square that with the mass theft all LLM services and progress are/is based on
them automatically fucking with authorsā papersā¦.ew
ACM is now showing an AI āsummaryā of a recent paper of mine on the DL instead of the abstract. As an author, I have not granted ACM the right to process my papers in this way, and will not. They should either roll back this (mis)feature or remove my papers from the DL.
Relatedly:
As is typical for educators these days, Heiss was following up on citations in papers to make sure that they led to real sources ā and werenāt fake references supplied by an AI chatbot. Naturally, he caught some of his pupils using generative artificial intelligence to cheat: not only can the bots help write the text, they can supply alleged supporting evidence if asked to back up claims, attributing findings to previously published articles. [ā¦] That in itself wasnāt unusual, however. What Heiss came to realize in the course of vetting these papers was that AI-generated citations have now infested the world of professional scholarship, too. Each time he attempted to track down a bogus source in Google Scholar, he saw that dozens of other published articles had relied on findings from slight variations of the same made-up studies and journals. [ā¦] Thatās because articles which include references to nonexistent research material ā the papers that donāt get flagged and retracted for this use of AI, that is ā are themselves being cited in other papers, which effectively launders their erroneous citations.
a16z funds 1000+ strong phone farm and uses it for mass manufacturing tiktok ai influencers, security turns out to be not good enough https://www.404media.co/hack-reveals-the-a16z-backed-phone-farm-flooding-tiktok-with-ai-influencers/
the usecase is spam:
The hacker also shared a list with me of more than 400 TikTok accounts Doublespeed operates. Around 200 of those were actively promoting products on TikTok, mostly without disclosing the posts were ads, according to 404 Mediaās review of them. Itās not clear if the other 200 accounts ever promoted products or were being āwarmed up,ā as Doublespeed describes the process of making the accounts appear authentic before it starts promoting in order to avoid a ban.
Iāve seen TikTok accounts operated by Doublespeed promote language learning apps, dating apps, a Bible app, supplements, and a massager.
article in large part about our friends
https://bayareacurrent.com/meet-the-new-right-wing-tech-intelligentsia/
some of the people involved with kernel are pretty unhappy about this and claim the piece is in bad faith/factually wrong (see the replies to https://bsky.app/profile/kellypendergrast.bsky.social/post/3ma55xfq7d22y )
Reminder Tivy was the guy behind Phalanx (back in his polyamory microblogging days)~
Shit like Palladium is going to be absolutely hilarious to dig up in the back of a used bookstore 20 years from now
That link canāt be viewed without a bluesky account, btw.
Thanks.
I love the fact that this ādecentralized billionaire-proof open networkā needs a nitter clone.
iāll cut the coiners some slack on this one because requiring a login to view is an account level privacy option. i donāt know what the option is supposed to accomplish. but thatās what it is
you do not, under any circumstances, āgotta hand it to themā
if bsky is supposed to be federated, then it does nothing, but as it is today with 99%+ of users on main instance, it only works as a recruitment tool for bsky
deleted by creator
Larian Studios founder/CEO Swen Vincke is posting through it on Twitter, after the studioās use of plagiarism machines caused significant backlash (to the shock of everyone except Swen).
This is probably Pivot to AI material.
Ah yes, I love the smell of burning bridges in the evening. Fuck. And I was getting excited about Divinity! Well, guess that means more money to spend on other things.
Rewatched Dr. Geoff Lindseyās video about deaccenting in English language and how āAIā speech synthesizers and youtubers tend to get it wrong. In the case of latter, itās usually due to reading from a script or being an L2 English speaker whose native language doesnāt use destressing.
It reminded me of a particular line in Portal
spoilers for Portal (2007 puzzle game)
GLaDOS: (with a deeper, more seductive, slightly less monotone voice than unti now) āGood news: I figured out what that thing you just incinerated did. It was a morality core they installed after I flooded the Enrichment Center with a deadly neurotoxin to make me stop flooding the Enrichment Center with a deadly neurotoxin.ā
The words āthe Enrichment Center with a deadly neurotoxinā are spoken with the exact same intonation both times, which helps maintain the robotic affect in GLaDOSās voice even after it shifts to be slightly more expressive.
Now Iām wondering if people whose native language lacks deaccenting even find the line funny. To me itās hilarious to repeat a part of a sentence without changing its stress because in English and Finnish itās unusual to repeat a part of a sentence without changing its stress.
It is not lost on me that the fictional evil AI was written with a quirk in its speech to make it sound more alien and unsettling, and real life computer speech has the same quirk, which makes it sound more alien and unsettling.
To me itās hilarious to repeat a part of a sentence without changing its stress because in English and Finnish itās unusual to repeat a part of a sentence without changing its stress.
Not a native speaker of either language but I read this in my mind without changing its stress in the part where it repeated āwithout changing its stressā.
Thatās interesting. If I werenāt going for a comical effect Iād try and rephrase the sentence, probably with a relative pronoun or something similar, but if unable to do so* Iād probably deemphasize the whole phrase the second time I say it. Though in terms of multi-word phrased, I think intonation would be the more accurate word to use than stress per se.
*āTo do soā would be another way to avoid repetition
A bunch of big hollywood people have started ācreatorās coalition on AIā. A snarky summary of their mission: āletās figure out how we can work with the fascists!ā
Introducing the Palantir shit sandwich combo: Get a cover up for the CEO tweaking out and start laying the groundwork for the AGI godās priest class absolutely free!
https://mashable.com/article/palantir-ceo-neurodivergent
TL;DR- Palantir CEO tweaks out during an interview. Definitely not any drugs guys, heās just neurodivergent! But the good, corporate approved kind. The kind that has extra special powers that make them good at AI. Theyāre so good at AI, and AI is the future, so Palantir is starting a group of neurodivergents hand picked by the CEO (to lead humanity under their totally imminent new AI god). He totally wasnāt tweaking out. Heās never even heard of cocaine! Or billionaire designer drugs! Never ever!
Edit: To be clear, no hate against neurodivergence, or skepticism about it in general. Iām neurodivergent. And yeah, some types of neurodivergence tend to result in people predisposed to working in tech.
But if youāre the fucking CEO of Palantir, surely youāve been through training for public appearances. Itās funnier that it didnāt take, but this is clearly just an excuse.
I strongly feel that itās an attempt to start normalizing the elevation of certain people into positions of power based off vague characteristics they were born with.
Lemmy post that pointed me to this: https://sh.itjust.works/post/51704917
I feel bad for the gullible ND people who spend time applying to this thinking they might have a chance and it isnāt a high level coverup attempt.
Otoh, somebody should take some fun drugs and tape their interviews, see how it works out. Are there any Hunter S Tech journalists around?
Jesus. This being 2025 of course he had to clarify that itās definitely not DEI. Also it really grinds me gears to see hyperfocus listed as one of the ābeneficialā aspects because thereās no way itās not exploitative. Hey, so you know how sometimes you get so caught up in a project you forget to eat? Just so you know, you could starve on the clock. For me.
Merriam-Websterās human editors have chosen slop as the 2025 Word of the Year. We define slop as ādigital content of low quality that is produced usually in quantity by means of artificial intelligence.ā All that stuff dumped on our screens, captured in just four letters: the English language came through again.
https://www.merriam-webster.com/wordplay/word-of-the-year?2025-share
Can I just take a moment to appreciate Merriam-Webster for coming in clutch with the confirmation that weāre not misunderstanding the ā6-7ā meme that the kids have been throwing around?
did they use ai to make the image? āwood of of yearā
Itās most obvious on the cat which is all around nightmare material.
The image also comes with alt text:
a bizarre collection of ai-generated illustrations including a sign that reads wood of of year and a chyron that reads breaking news
An academic sneer delivered through the arXiv-o-tube:
Large Language Models are useless for linguistics, as they are probabilistic models that require a vast amount of data to analyse externalized strings of words. In contrast, human language is underpinned by a mind-internal computational system that recursively generates hierarchical thought structures. The language system grows with minimal external input and can readily distinguish between real language and impossible languages.
Sadly, itās a Chomskian paper, and those are just too weak for today. Also, I think itās sloppy and too Eurocentric. Here are some of the biggest gaffes or stretches I found by skimming Moroās $30 book, which I obtained by asking a shadow library for āimpossible languagesā (ISBN doesnāt work for some reason):
book review of Impossible Languages (Moro, 2016)
- Moro claims that itās impossible for a natlang to have free word order. Thereās many counterexamples which could be argued, like Arabic or Mandarin, but I think that the best counterexample is Latin, which has Latinate (free) word order. On one hand, of course word order matters for parsers, but on the other hand the Transformers architecture attends without ordering, so this isnāt really an issue for machines. Ironically, on p73-74, Moro rearranges the word order of a Latin phrase while translating it, suggesting either a use of machine translation or an implicit acceptance of Latin (lack of) word order. I could be harsher here; it seems like Moro draws mostly from modern Romance and Germanic languages to make their points about word order, and the sensitivity of English and Italian to word order doesnāt imply a universality.
- Speaking of universality, both the generative-grammar and universal-grammar hypotheses are assumed. By āimpossibleā Moro means a non-recursive language with a non-context-free grammar, or perhaps a language failing to satisfy some nebulous geometric requirements.
- Moro claims that sentences without truth values are lacking semantics. Gƶdel and Tarski are completely unmentioned; Moro ignores any sort of computability of truth values.
- Russellās paradox is indirectly mentioned and incorrectly analyzed; Moro claims that Russell fixed Fregeās system by redefining the copula, but Russell and others actually refined the notion of building sets.
- It is claimed that Brocaās area uniquely lights up for recursive patterns but not patterns which depend on linear word order (e.g. a rule that a sentence is negated iff the fourth word is ānoā), so that Brocaās area canāt do context-sensitive processing. But humans clearly do XOR when counting nested negations in many languages and can internalize that XOR so that they can handle utterances consisting of many repetitions of e.g. ānot notā.
- Moro mentions Esperanto and VolapĆ¼k as auxlangs in their chapter on conlangs. They completely fail to recognize the past century of applied research: Interlingue and Interlingua, Loglan and Lojban, LĆ”adan, etc.
- Sanskrit is Indo-European. Also, thatās not how junk DNA works; it genuinely isnāt coding or active. Also also, thatās not how Turing patterns work; they are genuine cellular automata and itās not merely an analogy.
I think that Moroās strongest point, on which they spend an entire chapter reviewing fairly solid neuroscience, is that natural language is spoken and heard, such that a proper language model must be simultaneously acoustic and textual. But because they donāt address computability theory at all, they completely fail to address the modern critique that machines can learn any learnable system, including grammars; they worst that they can say is that itās literally not a human.
Plus, natural languages are not necessarily spoken nor heard; sign language is gestured (signed) and seen and many, mutually-incompatible sign languages have arisen over just the last few hundred years. Is this just me being pedantic or does Moro not address them at all in their book?
