Olivia Guest · Ολίβια Γκεστ
LightNews LightNews
banner
olivia.science
Olivia Guest · Ολίβια Γκεστ
@olivia.science
17K followers 2.1K following 6.6K posts
https://olivia.science

assistant professor of computational cognitive science · she/they · cypriot/kıbrıslı/κυπραία · σὺν Ἀθηνᾷ καὶ χεῖρα κίνει
Posts Replies Media Videos
Pinned
olivia.science
Olivia Guest · Ολίβια Γκεστ @olivia.science · Oct 23
CAIL = prerequisite knowledge for a critical perspective, such as to tell apart nonsense hype from true theoretical computer scientific claims. For example, the idea that human-like systems are a sensible or possible goal is the result of circular reasoning and anthropomorphism. olivia.science/ai

🧵
olivia.science
rock paper scissors
danmcquillan.bsky.social Dan McQuillan @danmcquillan.bsky.social · 4h
'AI, Ethics and Society' sounds like a dataset for contrastive learning 🤣
February 10, 2026 at 3:51 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
aleclongstreth.bsky.social
Regulate A.I.!

It took 38 hours to animate this loop (over the last three months) and I'm releasing it with a creative commons license so that others can use it!

© 2025 by Alec Longstreth is licensed under CC BY-NC-ND 4.0. To view a copy of this license, visit creativecommons.org/licenses/by-...
June 30, 2025 at 1:55 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
emp1.bsky.social
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
PALMS: Pavlovian Associative Learning Models Simulator
Simulations are an indispensable step in the cycle of theory development and refinement, helping researchers formulate precise definitions, generate models, and make accurate predictions. This paper i...
arxiv.org
February 10, 2026 at 2:49 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
fintanmallory.com
Only British Vogue has the courage to publish true science as it was practiced in the 18th century Transactions of the Royal Society
Headline from British Vogue I Ate Nothing But Sardines For 3 Days - Here's How My Skin Changed with a photo of a sardine
February 10, 2026 at 10:34 AM
Reposted by Olivia Guest · Ολίβια Γκεστ
anthonymoser.com
This is because a statistical model of word frequency is not a useful tool for modeling the complex interactions of the human body
rocher.lc Luc Rocher @rocher.lc · 1d
⚠️ Despite all the hype, chatbots still make terrible doctors. Out today is the largest user study of language models for medical self-diagnosis. We found that chatbots provide inaccurate and inconsistent answers, and that people are better off using online searches or their own judgment.
Banner image with screenshot of scientific article from nature Medicine, as well as two panels from the study method and results
February 10, 2026 at 2:27 PM
olivia.science
without context this looks severely impish lmao bsky.app/profile/oliv...
olivia.science Olivia Guest · Ολίβια Γκεστ @olivia.science · 28d
Buy a physical book and remove the cover. Vandalism like mess is acceptable when it's for good 🥰
Kitchen table with a nest of books. Another science is possible, cybernetic revolutionaries, the unity of perception, deflating mental representation. 🫡
February 10, 2026 at 2:28 PM
olivia.science
yes, I have known this for a long time also the book by Medina is amazing; she's so great, buy it if you can!

mitpress.mit.edu/978026252596...
Cybernetic Revolutionaries
In Cybernetic Revolutionaries, Eden Medina tells the history of two intersecting utopian visions, one political and one technological. The first was Chile's ...
mitpress.mit.edu
February 10, 2026 at 2:26 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
joshuawcotter.bsky.social
Make Art (2/4)
A dog sits on the ground sculpting a self-portrait, a cat holds a box of art supplies. The cat says “THAT IS A PURPOSE IN DIRECT DIAMETRIC OPPOSITION TO THAT OF CORPORATOCRATIC SYSTEMS:”, the dog adds “OBJECTIFICATION OF THE INDIVIDUAL FOR PROFIT AND CONTROL.”
February 10, 2026 at 12:50 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
joshuawcotter.bsky.social
Make Art (4/4)
Handprints of the 5 animals included in the previous 3 panels, along with text that reads “RESIST COMMODIFICATION. DEFY CORPORATOCRACY. DEFY OLIGARCHY. DEFY THEIR ANTI-HUMAN, ANTI-LIFE AUTHORITARIAN MOVEMENT. MAKE ART.”
February 10, 2026 at 12:54 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
joshuawcotter.bsky.social
Make Art (3/4)
A raccoon draws with its paws on a piece of paper. It says “GENERATIVE A.I. IS PART OF AN AUTHORITARIAN CORPORATOCRATIC EFFORT TO CO-OPT AND COMMODIFY THE CREATIVE PROCESS TO RID THE END RESULT - ART- OF THAT WHICH THREATENS ITS EXISTENCE - HUMANITY -
February 10, 2026 at 12:52 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
joshuawcotter.bsky.social
Make Art (1/4)
A rabbit in overalls and a goat in pants are sitting on the ground drawing pictures. The rabbit is saying “ART POSES A THREAT TO CORPORATOLRATIC SYSTEMS BECAUSE A PURPOSE OF ART IS TO REMIND THE INDIVIDUAL OF THEIR INHERENT AUTONOMY.”, the goat adds “THEIR HUMANITY.”
February 10, 2026 at 12:48 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
karenguzzo.bsky.social
UNC is close to finalizing its policy which outlines permissible circumstances to *SECRETLY RECORD FACULTY IN THE CLASSROOM.*
UNC-Chapel Hill Preps Policy on Recording Professors
The school aims to clarify guidelines after a secret recording controversy and growing fear of surveillance.
www.theassemblync.com
February 10, 2026 at 12:58 PM
olivia.science
have you also seen? bsky.app/profile/clou...
cloudquistador.bsky.social Dwayne Monroe @cloudquistador.bsky.social · 23h
In this video, inspired by @sonjadrimmer.bsky.social I read Evgeny Morozov's essay, 'Socialism after AI' or, as much as I could endure. Morozov writes about 'AI' using glittering generalities, lacking understanding or acknowledgment of the actually existing industry: youtu.be/NUby9eTbsuM
Morozov on AI: A Trip Down Academia Lane
YouTube video by Microsoft - A Materialist Approach
youtu.be
February 10, 2026 at 2:14 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
sexabolition.blog
bsky.app/profile/oliv...

This by Olivia Guest also has a bunch of excellent frameworks for thinking through the labour displacement processes happening.
olivia.science Olivia Guest · Ολίβια Γκεστ @olivia.science · 5h
excellent points, these might be a useful way to dissect it too

1/2

bsky.app/profile/oliv...
olivia.science Olivia Guest · Ολίβια Γκεστ @olivia.science · Jul 29
I split AI into 3 non-mutually exclusive types (see Table 1 above): displacement (harmful), enhancement (beneficial), and/or replacement (neutral) of human cognitive labour. More later possibly, but see Tables 2 to 4 (attached or here: arxiv.org/pdf/2507.19960) for the worked through examples. 2/n
February 10, 2026 at 1:47 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
sexabolition.blog
Just wanted to say thanks to the many actual academics working on AI's deleterious effects on social economies and culture replying with links to their work. I'll just quote skeet them here

(I'm not an academic, I'm a hacker and trans rights researcher with some skin in tech and humanities)
bhaggart.bsky.social Blayne Haggart @bhaggart.bsky.social · 3h
Insightful thread.
I think this argument rhymes with work @ntusikov.bsky.social and I have done using Susan Strange’s knowledge structure concept. AI’s effect on society is primarily in its reconception of knowledge away from “scientific” expertise and toward correlation-driven oracular knowledge.
sexabolition.blog Mallory Moore @sexabolition.blog · 12h
There's a kind of intellectual malaise among a certain kind of socialist that just doesn't understand automation and I think probably comes from not having worked much on or around it.

The prospect of AI as a fantasy capacity to automate labour and free us from drudgery I think corresponds to...
February 10, 2026 at 1:44 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
joannechocolat.bsky.social
Susan Jocelyn Bell Burnell (1943-) is a Northern Irish physicist who, while conducting research for her doctorate, discovered the first radio pulsars in 1967. This discovery earned the Nobel Prize in Physics in 1974, but not to her - the prize was awarded to her (male) supervisor. #CelebratingWomen
February 10, 2026 at 12:44 PM
olivia.science
❤️‍🔥 likewise!
February 10, 2026 at 1:03 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
mongabay.com
[FOUNDER'S BRIEF - @rhettayersbutler.bsky.social]

AI-generated wildlife images are distorting how people understand animals — exaggerating danger, inventing attacks, or turning wild species into “pets.” This erodes trust, fuels fear or sentimentality, and ultimately makes conservation harder.
AI-generated wildlife photos make conservation more difficult
Anyone who looks at a social media feed with any regularity is likely familiar with the deluge of fabricated images and videos now circulating online. Some are harmless curiosities (other than the…
news.mongabay.com
February 10, 2026 at 12:17 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
heldutson.bsky.social
Something compelled me to make this
A screenshot of a mocked-up Guardian headline, which reads: "Ignore the smears: I was never the one who ate a full tub of Lurpack off the kitchen counter." The byline is: "My dog." And there's a picture of my dog, posing with a serious expression, like a journalist. She is a black labrador type dog with brown eyes, grey muzzle and shiny, silky fur on her face, wearing a pink collar. Underneath the headline continues: "Listen, I'm a good girl. I don't know how butter stealers thrive in our political culture, but it's clear there was a failure of moral seriousness." There's then a link to another article, headlined: "Footage of Lola eating the butter, which cost her owners £300, was captured on a dog cam." It's a parody of this article from Wes Streeting in the Guardian: https://www.theguardian.com/commentisfree/2026/feb/09/peter-mandelson-wes-streeting-vetting-moral-seriousness
February 10, 2026 at 10:57 AM
olivia.science
🙃
February 10, 2026 at 12:50 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
cristinarmadeira.bsky.social
If you are silent, you are complicit with nazis.
Choose your side now.
It is impossible to stay impartial.
The image is a screenshot of a quote from goodreads.com attributed to A.R. Moxon. This is the quote: “Historians have a word for Germans who joined the Nazi party, not because they hated Jews, but out of a hope for restored patriotism, or a sense of economic anxiety, or a hope to preserve their religious values, or dislike of their opponents, or raw political opportunism, or convenience, or ignorance, or greed. That word is "Nazi." Nobody cares about their motives anymore. They joined what they joined. They lent their support and their moral approval. And, in so doing, they bound themselves to everything that came after. Who cares any more what particular knot they used in the binding?”
January 29, 2025 at 2:10 PM
olivia.science
thank you!
February 10, 2026 at 12:49 PM
olivia.science
haha I love the 2 minute city with my spoiled ass living next to a supermarket haha
February 10, 2026 at 12:49 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
goldstein.bsky.social
Excellent paper that I missed when it came out.
olivia.science Olivia Guest · Ολίβια Γκεστ @olivia.science · Jul 29
Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX...

Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n
title and abstract from https://arxiv.org/pdf/2507.19960 table 1 from https://arxiv.org/pdf/2507.19960
February 10, 2026 at 12:26 PM
Reposted by Olivia Guest · Ολίβια Γκεστ
altibel.bsky.social
Academics that claim these things can be used for analysing qualitative texts: HOW? Why would it work for your corpus if it doesn't work here?
The disconnect is real.
clapifyoulikeme.favrd.social Mrs. Detective Pikajew, Esq. @clapifyoulikeme.favrd.social · 4d
I’m getting nauseous just reading this
1. Affable's Brief in Support of Its Motion to Dismiss Taking each of Mr. Feldman's three problematic submissions in turn, the Court began with questions about Affable's brief in support of its motion to dismiss. Mr. Feldman offered essentially two explanations for how the brief's citation errors came to be. First, they were the product of AI "hallucinations," i.e., erroneous or non-existent legal authorities. (Tr. 51-52). The Court buys this first explanation; indeed, it is the only explanation the Court credits. 3 Mr. Feldman's second explanation was that the citation errors resulted from his idiosyncratic research process, which involved inserting into his brief unreported cases located via Google Scholar, Google search, or the Al-assisted platform vLex, and then cite-checking those cases by running the brief through a different Al program — either vLex or Paxton Al. (Id. at 37, 47-53). This automated cite check, Mr. Feldman said, introduced citation errors that he mistook as corrections and thus neither reviewed nor fixed. (Id. at 52-53). The Court does not find the second explanation credible, largely because it is not coherent. To that end, the Court will now attempt to flesh out Mr. Feldman's It was not, however, an explanation that Mr. Feldman was eager to concede. (See Tr. 47 The Court: "'You were saying to me, or you were saying to Mr. MacMull, that some o hese cases are the product of hallucinations. Did you not say that? Mr. Feldman: 16 Case 1:24-CV-09429-KPF Document 227 Filed 02/05/26 Page 17 of 33 description of his research process, but doing so is a difficult task because he offered multiple, often inconsistent explanations.
February 10, 2026 at 11:14 AM