Jérôme Denis
@jrmdns.bsky.social
electrical native - prof #sociology & #sts - director of the center for the sociology of innovation (mines paris) New books in 2025: The Care of things (Polity) // Fragilities (MIT Press) Interested in #maintenance #architecture #data #datalabor #city
De l'usage utilitariste des maths comme "benchmark" dans les boîtes d'IA. Pendant ce temps, les effets sur la recherche ne sont, selon les chercheurs, pas inexistants, mais conduisent aussi à des définancement et à des erreurs. next.ink/244924/lia-g...
L’IA générative bouscule la recherche en mathématiques
En mai dernier, OpenAI a annoncé qu’un de ses modèles internes a résolu un des problèmes d’Erdős. De l’extérieur, ça semble impressionnant. L’est-ce…
next.ink
Incredible moment where Fox News puts up a graphic with all the reasons it opposes the DSA and why people should be afraid of them
Found out yesterday that my job is at risk. Feels terrible: multiple video-only "consultations" without even a q and a... Humanities and social sciences overwhelmingly targeted... Whole thing feels deeply ideological and also like payback aimed at the departments with high union density...
University of Exeter in talks to cut about 150 members of staff
The university says it is consulting with colleagues over "limited and specific potential changes".
share.google
McSweeney’s on the economics of the AI bubble. This description of AI driven layoffs is excellent satire.
The purpose of this technology is not to accurately identify people but rather to instill in everyone the fear that they are constantly being watched.
Facial recognition AI misidentifies Utah man in felony vandalism case
A Utah man says he endured months of fear and thousands of dollars in legal expenses after he says facial recognition technology incorrectly identified him as a
kutv.com
This coming Monday we kick off our second instalment of the ST&HV STS School, this time in Seoul. 4 days of seminars, workshops, plenaries and field trips, hosted by the Department of Science Studies at Seoul National University and KAIST Graduate School of Science and Technology Policy. So excited!
This is an important story. And not the first I hear about scholars (all men) who get sucked into a sort of AI-hole, addicted to Claude or whatever, and feeling useless and irrelevant by the very use of it. How sad.
Crystal Fleming クリスタル@alwaystheself.bsky.social · 2mo agoI'm attending a digital humanities event in Montreal. There was a keynote on AI and something about the talk made me wonder if it had been written by Claude. I said as much in the Q&A. As I posed the question, the speaker shifted, looking slightly uncomfortable. What he said next shocked the room +
A Long History: From Universal Language to Artificial Intelligence — Mostafa H. Abdou Technology and Culture, 77(2) muse.jhu.edu/article/988849
If you want to avoid AI Search and metadata shortcomings, you can simply full-text search on 36M works (and counting) with Matilda.
The Scholarly Kitchen@scholarlykitchen.bsky.social · 2mo agoGuest Post — Scholarly AI Search Shortcomings and the Need for Better Metadata scholarlykitchen.sspnet.org/2026/05/29/g...
Hyperscalers as logistical regimes: software containers and the circulation of cloud capital — Nathan Kim (Information, Communication & Society) www.tandfonline.com/doi/full/10....
OpenAI Announces Construction Of New Data Center On Top Of Sick Child
Journal editors and reviewers are "struggling to sort through a flood of material that all seems good enough at first glance to warrant a close read. There is an enormous and growing asymmetry between the time it takes to produce new work and the time it takes a subject-matter expert to vet it."
The Verge@theverge.com · 3mo agoAI research papers are getting better, and it’s a big problem for scientists
Meet the sideshow FREAKS who REFUSE to use the work-stealing, planet-burning, minority-choking, teen-killing, water-guzzling, brain melter that gets most things wrong and makes the user decide whether they give a shit enough to do their job correctly or if they've given up completely on themselves.
Iris van Rooij 💭@irisvanrooij.bsky.social · 3mo agoMeet the academics refusing to use generative AI Researchers say they have their reasons for avoiding AI tools — and they’re sick of arguing about it. www.nature.com/articles/d41... h/t @maartenp.bsky.social
Quel Philippe Aghion doit-on croire ? Le jeune étudiant communiste qui vendait l'Humanité Dimanche ? Celui qui appelait à voter Macron en 2017 ? Ou qui le critiquait un an plus tard dans nos colonnes ? Hélas, quand on défend les mêmes convictions depuis 46 ans, on a parfois du mal à suivre.
🔔 new article! 🔔 I'm quite moved to see this paper published in Computer Supported Cooperative Work (CSCW), a journal I admire since… the late 1990s. open acess >> doi.org/10.1007/s106... #datawork #Covid19 #care #infrastructure #publichealth #sts
Data Trials: Work, Worries and Care Behind the Scenes of a Data-Driven Pandemic - Computer Supported Cooperative Work (CSCW)
In many countries, the data logistics dedicated to generating a national overview of public health were profoundly destabilised in the early months of the COVID-19 pandemic. Drawing on the case of France, this article shows that the (re)centralisation of public health data that took place during these times was far from a top-down initiative. It unfolded in a great variety of sites in the health administration where many people developed an affective relationship with the data they manipulated through daily, and sometimes very mundane, interventions. To highlight frictions, worries and care practices that multiplied in this process, the authors identify what they propose to call three “data trials”: infrastructural, representational and enunciative. These trials shed new light on what has often been described as a data-driven pandemic. Far from technically and epistemically stabilised entities, data were subjected to numerous costly and uncertain operations, carried out by a large number of workers who acknowledged and took care of their fragilities.
doi.org
Une thèse en STS au Centre de sociologie de l'innovation (Mines Paris PSL), ça vous tente ? Nous démarrons une nouvelle campagne de recrutement pour la rentrée 2026-2027. Date limite pour le dépôt des candidatures : vendredi 29 mai. Tout est expliqué ici : www.csi.minesparis.psl.eu/articles-ved...
Appel à candidatures 2026 : recherche doctorale, contrat doctoral
csi.minesparis.psl.eu
New on H-Sci-Med-Tech: Check out Jérôme Baudry (@lhst-epfl.bsky.social)'s rvw of @fdrubio1977.bsky.social, @jrmdns.bsky.social, & David Pontille’s ed vol _Fragilities: Essays on the Politics, Ethics, & Aesthetics of Maintenance & Repair_ 2025 @mitpress.bsky.social Review @hnetreviews.bsky.social
h-net.org
H-Net Reviews@hnetreviews.bsky.social · 3mo agoNew book review: Baudry on Domínguez Rubio, Fernando; Denis, Jérôme; Pontille, David, eds.: _Fragilities: Essays on the Politics, Ethics, and Aesthetics of Maintenance and Repair_. MIT Press, 2025. Published by H-Sci-Med-Tech. Read here: networks.h-net.org/node/20149545
SBES – social, behavioral, and economic sciences – goes to ZERO in 2027. That’s STS, sociology, anthropology, as well as economics. NSF led by a business guy affiliated with Peter Thiel. just a swathe of fucking napalm across the board. No words. We had better win this damn election.
Cole Donovan@colesci.bsky.social · 4mo agoGiving folks a sense of what the cuts to NSF funding in the 2027 President's Budget Request actually looks like. Note the significant cuts across domains--even to research security! STEM and NCSES aren't captured for prior years, so they appear to be cuts, even though they are not.
"Je ne contracte pas." Dans "Paranoid Finance", le sociologue Fabian Muniesa plonge dans l'imaginaire conspirationniste pour interroger les catégories financières sur lesquelles repose notre monde : www.nouvelobs.com/idees/202603...
« Je ne contracte pas » : comment les complotistes nous aident, malgré eux, à comprendre notre monde
Dans « Paranoid Finance », le sociologue Fabian Muniesa plonge dans l’imaginaire financier porté par les mouvements conspirationnistes, comme QAnon ou les « êtres souverains ». Il en tire une philosop...
nouvelobs.com
"Someone decided that deliberation was latency. Someone decided to build a system that produces a thousand targeting decisions per hour and call them high-quality. Someone decided to start this war. Several hundred people are sitting on Capitol Hill, refusing to stop it."
et cette extrême droite proche du peuple, elle est avec nous dans cette pièce ?
Joyeux 18 mars! Il y a 155 ans, Parisiens et Parisiennes se soulevaient contre l'assemblée monarchiste planquée à Versailles et déclaraient LA COMMUNE!!!!!!
🧵 Trump administration AI policy is widely described as deregulatory. This description is misleading. What's happening is not the absence of governance but its rearrangement--intensive state intervention operating through mechanisms we don't typically call regulation. www.science.org/doi/10.1126/...
The mirage of AI deregulation
One of the most interventionist approaches to technology governance in the United States in a generation has cloaked itself in the language of deregulation. In early December 2025, President Donald Tr...
science.org
