Sign in

Guillaume Pasquier

@gap-id.bsky.social
444 followers 565 following 643 posts

Digital librarian at @GVAGradLibrary.bsky.social - MA LISM @SheffieldUni.bsky.social - Open science, RDM, AI luddism, librarianship ethics & dis/misinformation - Posts in FRE/ENG, blogs in French: biblog.ch - Mastodon: glammr.us/@gap

PostsRepliesMedia
Reposted by Guillaume Pasquier
gamera obscura @ohrobin.bsky.social · 28/09/2026
Google earth has updated its satellite images of Gaza. The destruction is almost incomprehensible.
4521331249
Reposted by Guillaume Pasquier
Tom Gauld @tomgauld.bsky.social · 20/09/2026
My latest books cartoon for @theguardian.com
Title: Gentlemen! Are you in want of a wife? Allow our algorithm to assist!

Image: a series of yes/no questions for that lead to different outcomes. 

Questions include “are you rich?”, “are you dashing?”, “are you a rake” and “are you a vicar?”

Outcomes include “You win the most beautiful daughter of the richest man in the county!”, “You remain a bachelor”, “You marry a dowdy cousin” and “You rescue a 23-year-old from Spinsterhood”.
9903268
Reposted by Guillaume Pasquier
Xero Reynolds 41/100 on Patreon @artificer-ind.bsky.social · 09/09/2026
Still patiently waiting to be "left behind" by AI rather than it chasing me across every damn space like a slasher in a low budget flick.
2110743158
Reposted by Guillaume Pasquier
Jules @juleswp.eurosky.social · 18/09/2026
Comment détecter un article généré par IA sur Wikipédia ? C'est désormais notre lot quotidien. 🧵 Un petit fil, à partir d'un exemple concret.
Aperçu d'un article sur Blaise Agresti.

« Blaise Agresti, né le 20 janvier 1971, est un officier supérieur de la Gendarmerie nationale (colonel)[1], guide de haute montagne, auteur et dirigeant d'entreprise français. Spécialiste du secours en montagne et du commandement en situation d'urgence[2], il a dirigé le Peloton de gendarmerie de haute montagne (PGHM) de Chamonix-Mont-Blanc ainsi que le Centre national d'instruction de ski et d'alpinisme de la gendarmerie (CNISAG)[3]. »
21498341
Reposted by Guillaume Pasquier
Dave Willner @dwillner.bsky.social · 17/09/2026
Certain kinds of smart people tend to have a "sexy apocalypse" problem when it comes to thinking about risk. E.g. skynet turning everyone in paperclips, while bad, is also *kind of cool and fun* at the same time. Where the reality is that most actual disasters are both stupid and disappointing.
222572322
Reposted by Guillaume Pasquier
Joseph Cox @josephcox.bsky.social · 18/09/2026
OpenAI and Microsoft admitted LLMs are predatory tech built on “an astonishing theft of unprecedented proportions,” and the “largest theft of labor in human history.” An internal Microsoft document said AI has created a “doom loop” that is killing “the entire web.” www.404media.co/doom-loop-op...
404media.co
‘Doom Loop’: OpenAI and Microsoft Admits LLMs Are Destroying the Web and Built on Theft
"Millions of people around the world will soon consider large models ‘hoovering up’ all their work to be an astonishing theft of unprecedented proportions."
401522708
Reposted by Guillaume Pasquier
Eric Ravenscraft @lordravenscraft.bsky.social · 17/09/2026
holy shit, this is the whole ballgame in a single paragraph gAI systems require mass theft of copyrighted works, for the explicit purpose of replacing the market for the people they stole from, and the people in charge knew what they were doing ahead of time, and did it anyway
1623671114
Reposted by Guillaume Pasquier
MadScientist @mads100tist.bsky.social · 17/09/2026
LLMs are not search engines. They are mathematical hallucinators by default. Any 'search' you do, you'll have to double check anyway, which defeats the purpose of using them in the first place. Search engines can match faster and better because that's what they are designed to do
4378
Reposted by Guillaume Pasquier
Jed Brown @jedbrown.org · 16/09/2026
Oh. That's a heck of a subsidy.
At MIT, Harvard, or any other school, supervising an undergraduate researcher is a lot of work, costs a lot of money, involves a lot of meetings, and the results are not reliably excellent. Asking Claude is easier, faster, and may produce better results. It’s also cheaper, which is no accident: Harvard allows every Harvard student and faculty member to request a Claude token budget of $2,500 per month, even as research funding is continuously threatened and the University lays off staff.
35698186
Reposted by Guillaume Pasquier
MIT Technology Review @technologyreview.com · 17/09/2026
If AI systems are viewed as too advanced to control, the companies that build them can’t held liable for the harms they cause.
trib.al
Debates over AI consciousness are a trap
If AI systems are viewed as too advanced to control, the companies that build them can’t held liable for the harms they cause.
3125
Reposted by Guillaume Pasquier
matilda.science @matilda-science.bsky.social · 14/09/2026
Matilda has reached 170M deduplicated publications, among which 38.5M are full-text searchable. Everything is free, sharable and reusable and will remain as such under our #POSI commitments and following our mission matilda.science/mission?l=en #openscience #opendata
Series of open paper books which represent the open knowledge available through Matilda
02213
Reposted by Guillaume Pasquier
vortex_egg @vortexegg.com · 13/09/2026
Fwiw it's straightforward right now why this is happening; and it's the same reason why the global far right supremacist eliminationist movement is happening, the christian extremist apocalyptic immenantization is happening, etc. People are bug fuck out of their minds in denial of the climate crisis
437269
Reposted by Guillaume Pasquier
sarah zedig 🐐 @hmsnofun.bsky.social · 12/09/2026
hey why is no one talking about the licensing situation with this A24/SCP thing? are they releasing it under the same cc-by-sa license every SCP project is required to have? are they betting creative commons won't hold up in court? seems like the kind of question a journalist should look into idk
deadline.com
SCP Foundation Movie Coming To Theaters In 2027 With A24 Distributing
An SCP Foundation movie is coming to theaters in 2027 via A24. Spooky Pictures and Image Nation Studios announced the project earlier this year.
312550936
Reposted by Guillaume Pasquier
Gramsci at D4RKw0KE_$4$UkE69 @tontonpogo.bsky.social · 11/09/2026
Bah oui les LLM vont tous nous buter Par réchauffement climatique vous étiez où quand on discutait du coût énergétique et environnemental de ces gadgets pourris
3368
Reposted by Guillaume Pasquier
Standplaats Kraków could really use some cash! @standplaatskrk.pl · 09/09/2026
[Left side, an illustration of the "Virgin" character (from the Virgin vs. Chad meme format) modified into a robot. His head is a retro computer monitor with a glowing red dot on the screen, and his body has thin metallic limbs. A computer mouse hangs from his back by a wire, and loose paperclips float around his legs.]

"the virgin "AGI existential risk"

- programmer LARP fantasy that makes coding the single most important skill for saving humanity
- fake and made up
- entirely based on thought experiments
- assumes that there's this vaguely defined concept known as "intelligence" that can be linearly increased as if it were a mathematical quantity
- projects an anthropomorphic drive toward control and power onto algorithms that just do statistics to spit out words; colonialist self-report
- Zero evidence
- sponsored by billionaires who want to make their technology seem more important than it is
- blackmails you from the future using faulty logic; doesn't actually make any sense
- "oh no, i guess we have to build the Doom Machine now. i guess the course of technological progress is inevitable, oh well""

[Right side, an illustration of the "Chad" character, replaced by a globe of the Earth. A large fire burns at the top of the globe, and three black oil derricks are on the yellow landmass, which is shaped like a facial profile. He has muscular blue arms and legs and wears yellow shoes.]

"the Chad Global Warming

- incomprehensibly complex and interconnected polycrisis
- reveals deep-rooted flaws with the dominant politico-economic paradigm; demands a fundamental shift of human value systems by its sheer physical existence
- verified by the entire scientific community
- billionaires don't want you to think about it because it is too scary
- resists ideological simplification
- will actually kill you
- divine retribution for fossil capital's greed and wanton exploitation of the natural world
- doesn't even need intelligence to threaten humanity's survival"
1262123
Guillaume Pasquier @gap-id.bsky.social · 09/09/2026
AI is doing so much harm to the open web and open science... I now have researchers specifically asking for things to be closed off because they don't want their work to be harvested into Frankenstein's POS. F this S
0124
Guillaume Pasquier @gap-id.bsky.social · 09/09/2026
Le pouvoir de nuisance n'est pas une indication de compétence. On peut à la fois penser que des modèles IA peuvent faire des dégâts massifs à nos infrastructures et notre société et être conscient qu'il y a une chance nulle de créer une superintelligence avec des LLM.
000
Reposted by Guillaume Pasquier
Annie, MLIS, Schrödinger's librarian 🍄📚🥌 @dharmascholar.bsky.social · 09/09/2026
#DailySkeletor #affirmation
Skeletor sits in his throne of bones. He leans forward in thought, forearms on his thighs, his left hand cupping his chin. Caption reads: Despite how I feel, I understand this feeling and situation are temporary.
087
Reposted by Guillaume Pasquier
Phillip Maciak @pjmaciak.bsky.social · 04/09/2026
No matter how you refine your questions, the chatbot will generate responses reflecting what the greatest number of users are likely to want to hear, which tends to be close to what they think they already know." A great, painstaking piece. www.hnn.us/article/ai-i...
hnn.us
AI Is an Anti-Primary Source Machine
Real voices from the past are threatened by chatbots designed to amplify — and ultimately become — the loudest voices in the room.
37632
Reposted by Guillaume Pasquier
Nicholas Grossman @nicholasgrossman.bsky.social · 02/09/2026
Republicans who see Canada as an enemy has leapt up this year, similar to how Republicans’ approval of Putin shot up in 2016. Virtually no anti-Canada US constituency before this year, and virtually no pro-Putin constituency before 2016. Out of nowhere. Now millions say that’s their position.
1591555453
Reposted by Guillaume Pasquier
Samuel Moore @samuelmoore.org · 01/09/2026
This is why we had a whole open access movement. You can't leave this stuff up to the good will of organisations who may take access away at a moment's notice.
14520
Reposted by Guillaume Pasquier
Julien Sicot @jsicot.bsky.social · 28/08/2026
Here is how we built a custom middleware github.com/epfllibrary/... to support our mission of showcasing EPFL’s scientific output in our institutional repository, #Infoscience infoscience.epfl.ch. A short thread on our workflow
github.com
GitHub - epfllibrary/infoscience-imports at dev
Infoscience Imports uses modular Python scripts for distinct tasks: harvesting publications and files from different sources : OpenAlex, unpaywall, WoS and Scopus, deduplicating and enriching metad...
143
Reposted by Guillaume Pasquier
Adolfo Neto @adolfoneto.elixiremfoco.com · 28/08/2026
Lots of chips and memories being used and getting more expensive. Lots of AI Data Centers being built or planned. Nothing particularly significant has been added for users, in my opinion. That's what I mean by inefficient And I am not even talking about data workers.
1326
Reposted by Guillaume Pasquier
Richard Sever @richardsever.bsky.social · 27/08/2026
$3.7 billion spent on APCs in 2025. That's ~1000X more than it costs to run bioRxiv/medRxiv. Worth thinking about as we consider how to fund infrastructure... arxiv.org/pdf/2608.16322
arxiv.org
813378
Reposted by Guillaume Pasquier
Annie, MLIS, Schrödinger's librarian 🍄📚🥌 @dharmascholar.bsky.social · 27/08/2026
#DailySkeletor #affirmation
Skeletor stands in a green field. Mountains and a pink sky in the distance. He faces the right side of the image and his head turns, looking over his right shoulder behind him. Caption reads: I will move forward instead of holding on to past mistakes that rob me of my peace.
021
Reposted by Guillaume Pasquier
Paris Marx @parismarx.com · 26/08/2026
The United States is taking aim at Canada once again, with a crushing new set of tariffs. Ultimately, government will need to solve this. But in the same way Canadians are avoiding US alcohol, produce, and other goods, they should get off US tech products too. Luckily, I have a guide for that.
disconnect.blog
Getting off US tech: a guide
I’m in the process of dropping US tech services. Here’s how I did it, and options you should consider.
201576752
Guillaume Pasquier @gap-id.bsky.social · 26/08/2026
New Paradox gateway drug. Your kids will soon all be playing Victoria and Europa Universalis. youtu.be/8i_AqWQ9RwU?...
youtu.be
LEGO Skylines Announcement Trailer
YouTube video by Paradox Interactive
000
Reposted by Guillaume Pasquier
Gramsci at D4RKw0KE_$4$UkE69 @tontonpogo.bsky.social · 25/08/2026
la presse présente la culture war woke comme un truc où les deux camps ont des munitions mais en vrai d'un coté on corrige son savoir précédent à l'aune de nouvelles informations solides, et de l'autre on fait un caprice parce que ce qu'on a retenu par coeur à l'école primaire c'était pour rien
24210
Reposted by Guillaume Pasquier
Wolf @ametria.org · 23/08/2026
More importantly, nobody owns you a deep scientific effort for theories that fail by premise or have been discussed and debunked as nonsense three generations ago.
121114
Reposted by Guillaume Pasquier
BeijingPalmer @beijingpalmer.bsky.social · 23/08/2026
If I turn up at the physics department ranting about the ether nobody is obliged to pretend that I must be given a job.
462217233
Reposted by Guillaume Pasquier
Dr Vanessa @hpsvanessa.bsky.social · 23/08/2026
The loss of search structure is devastating for knowledge retrieval. I teach basic lit searching and it's always felt like a really core transferable skill for students to know boolean operators and "exact phrase searching" and now google, microsoft etc have just killed it to force us to use slopAI.
6643200
Reposted by Guillaume Pasquier
Chris Person @papapishu.bsky.social · 22/08/2026
Sorry chatbot lovers but the waning of Boolean operators in most search functions on websites has made my life actively worse. You can't find anything, you can only have things suggested to you.
10464201089
Reposted by Guillaume Pasquier
Gramsci at D4RKw0KE_$4$UkE69 @tontonpogo.bsky.social · 24/08/2026
Macron brigue un troisième mandat
0214
Reposted by Guillaume Pasquier
Illia Ponomarenko @ioponomarenko.bsky.social · 23/08/2026
HAPPY UKRAINE'S INDEPENDENCE DAY!!!
14869108
Reposted by Guillaume Pasquier
Matt Seybold @mattseybold.bsky.social · 22/08/2026
Dependency is the goal.
152852934
Reposted by Guillaume Pasquier
Big Frickin' Swede @bigfrickinswede.bsky.social · 20/08/2026
Really identifies what is wonderful about writing. Writing is thinking, revealing things you didn't know you knew, or allowing you to clarify and actually say what you always wanted to say. I look at old things I wrote and am often surprised by how smart, knowledgeable, and/or wonderful I was.
112
Reposted by Guillaume Pasquier
Dr. Damien P. Williams, dread portent down from a mountain cave @wolven.blacksky.app · 21/08/2026
The most crucial part in each "AI Winter," & the thing that I genuinely think cannot come soon enough, is when we as a culture remember that we're able to disambiguate the term "AI" & we go back to actually descriptively naming the process(es) we mean. Really can't wait for that to come around again
1230373
Reposted by Guillaume Pasquier
Emily M. Bender @emilymbender.bsky.social · 20/08/2026
A point I frequently make when talking about chatbots is that their output only seems to make sense because we are making sense of it. And worse, we can only do that sense-making by projecting a mind behind the text, something we do reflexively. >>
633468
Reposted by Guillaume Pasquier
John Pfaff @johnpfaff.bsky.social · 19/08/2026
It’s never too late to quit Twitter, even if the effects may take some time to dissipate. There is a loss from it—I felt it for a while. But there are real genuine intellectual risks to being there. And here’s a source for intelligence and vulnerability to propaganda:
theguardian.com
Why smart people are more likely to believe fake news
Research shows that smart people are more susceptible to fake news and conspiracy theories – but why?
316046
Reposted by Guillaume Pasquier
John Pfaff @johnpfaff.bsky.social · 19/08/2026
Evidence is already accruing that spending time on Twitter pulls ppl to the right, bc the algorithm feeds you horrible stuff. And again: academics are likely more vulnerable, bc they convince themselves they are less.
theconversation.com
A few weeks of X’s algorithm can make you more right-wing – and it doesn’t wear off quickly
Elon Musk’s social media platform is boosting conservative content – and it’s having long-lasting effects on how users see the world.
12685327
Reposted by Guillaume Pasquier
Annie, MLIS, Schrödinger's librarian 🍄📚🥌 @dharmascholar.bsky.social · 18/08/2026
#DailySkeletor #affirmation
Skeletor stands, arms raised, elbows bent. His right hand raised, index finger pointing at the sky. Jagged rocks and a purple sky are behind him. Caption reads: I choose to feel amazing today.
0106
Reposted by Guillaume Pasquier
Samuel Moore @samuelmoore.org · 17/08/2026
Helpful reminder from Elsevier that they have the "legal power" to do what they want with your journals. You have the collective power to take your editorial work elsewhere.
16843
Reposted by Guillaume Pasquier
Darius Baktash @sol-cantus.myatproto.social · 14/08/2026
A reminder to people that AI Doomerism exists because "Effective Altruists" (aka rich assholes who don't want to give up their privilege) are unable to comprehend that they're arguing both sides of the same coin in order to maintain the illusion of control they have over the universe.
1125
Reposted by Guillaume Pasquier
Brian C. Keegan, Ph.D. @brianckeegan.com · 12/08/2026
I'd go back to the HOA scolds on Mastodon in a heartbeat over any of the Nazi bar alternatives.
141
Guillaume Pasquier @gap-id.bsky.social · 12/08/2026
(crazy person voice) I would like to state firmly that I am not, in fact, a crazy person. #eclipse
Windows barred with cardboard with holes punched through Punched cardboard close-up Light dots on a white wall
110
Reposted by Guillaume Pasquier
MIT Technology Review @technologyreview.com · 12/08/2026
Marketing AI agents as digital employees may make human workers worse at spotting errors and more likely to offload accountability.
trib.al
AI agents are not your “coworkers”
Marketing AI agents as digital employees may make human workers worse at spotting errors and more likely to offload accountability.
1185
Reposted by Guillaume Pasquier
dell cameron @dell.bsky.social · 11/08/2026
the fact that a billionaire can randomly throw $100,000,000 into a campaign against taxing the rich seems like the best argument yet they can afford it
2848131287
Guillaume Pasquier @gap-id.bsky.social · 11/08/2026
I remember when a journalist I respect(ed) told me about Axios as his inspiration for creating a new outlet in Switzerland. I found it enjoyable at first, then realised maybe not everything could be accurately described through bullet points. Now we see the endgame of that approach.
000
Reposted by Guillaume Pasquier
Brandon Friedman @brandonfriedman.bsky.social · 11/08/2026
Axios writer: I melted my brain with AI. Not only can I no longer read, but I don't even understand the purpose of reading anymore. Here's how you can also do this. www.axios.com/2026/08/11/r...
Axios: AI has become my reading buddy as I've plowed through my summer book list, helping me keep track of plots and characters, unpack themes, and make sense of wordy, old prose.

Why it matters: The way many of us actually read books — a few pages before bed, a chapter at the beach, another on a plane — makes it easy to lose the thread. AI can help you pick it back up without starting over.
📖 I really got into using AI while reading James Fenimore Cooper's "The Last of the Mohicans." The text was far too dense for casual vacation reading. My first prompt asked:

"I'd like a guide as I read The Last of the Mohicans who can help me understand what I've read and where the story is, without giving spoilers or getting ahead of my progress. Can you do this, chapter by chapter?"
The robot agreed: "Cooper's sentences are long and his diction is archaic."
49372171667